diff options
383 files changed, 14813 insertions, 9830 deletions
diff --git a/Documentation/RCU/NMI-RCU.txt b/Documentation/RCU/NMI-RCU.txt index a6d32e65d222..a8536cb88091 100644 --- a/Documentation/RCU/NMI-RCU.txt +++ b/Documentation/RCU/NMI-RCU.txt @@ -34,7 +34,7 @@ NMI handler. cpu = smp_processor_id(); ++nmi_count(cpu); - if (!rcu_dereference(nmi_callback)(regs, cpu)) + if (!rcu_dereference_sched(nmi_callback)(regs, cpu)) default_do_nmi(regs); nmi_exit(); @@ -47,12 +47,13 @@ function pointer. If this handler returns zero, do_nmi() invokes the default_do_nmi() function to handle a machine-specific NMI. Finally, preemption is restored. -Strictly speaking, rcu_dereference() is not needed, since this code runs -only on i386, which does not need rcu_dereference() anyway. However, -it is a good documentation aid, particularly for anyone attempting to -do something similar on Alpha. +In theory, rcu_dereference_sched() is not needed, since this code runs +only on i386, which in theory does not need rcu_dereference_sched() +anyway. However, in practice it is a good documentation aid, particularly +for anyone attempting to do something similar on Alpha or on systems +with aggressive optimizing compilers. -Quick Quiz: Why might the rcu_dereference() be necessary on Alpha, +Quick Quiz: Why might the rcu_dereference_sched() be necessary on Alpha, given that the code referenced by the pointer is read-only? @@ -99,17 +100,21 @@ invoke irq_enter() and irq_exit() on NMI entry and exit, respectively. Answer to Quick Quiz - Why might the rcu_dereference() be necessary on Alpha, given + Why might the rcu_dereference_sched() be necessary on Alpha, given that the code referenced by the pointer is read-only? Answer: The caller to set_nmi_callback() might well have - initialized some data that is to be used by the - new NMI handler. In this case, the rcu_dereference() - would be needed, because otherwise a CPU that received - an NMI just after the new handler was set might see - the pointer to the new NMI handler, but the old - pre-initialized version of the handler's data. - - More important, the rcu_dereference() makes it clear - to someone reading the code that the pointer is being - protected by RCU. + initialized some data that is to be used by the new NMI + handler. In this case, the rcu_dereference_sched() would + be needed, because otherwise a CPU that received an NMI + just after the new handler was set might see the pointer + to the new NMI handler, but the old pre-initialized + version of the handler's data. + + This same sad story can happen on other CPUs when using + a compiler with aggressive pointer-value speculation + optimizations. + + More important, the rcu_dereference_sched() makes it + clear to someone reading the code that the pointer is + being protected by RCU-sched. diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt index cbc180f90194..790d1a812376 100644 --- a/Documentation/RCU/checklist.txt +++ b/Documentation/RCU/checklist.txt @@ -260,7 +260,8 @@ over a rather long period of time, but improvements are always welcome! The reason that it is permissible to use RCU list-traversal primitives when the update-side lock is held is that doing so can be quite helpful in reducing code bloat when common code is - shared between readers and updaters. + shared between readers and updaters. Additional primitives + are provided for this case, as discussed in lockdep.txt. 10. Conversely, if you are in an RCU read-side critical section, and you don't hold the appropriate update-side lock, you -must- @@ -344,8 +345,8 @@ over a rather long period of time, but improvements are always welcome! requiring SRCU's read-side deadlock immunity or low read-side realtime latency. - Note that, rcu_assign_pointer() and rcu_dereference() relate to - SRCU just as they do to other forms of RCU. + Note that, rcu_assign_pointer() relates to SRCU just as they do + to other forms of RCU. 15. The whole point of call_rcu(), synchronize_rcu(), and friends is to wait until all pre-existing readers have finished before diff --git a/Documentation/RCU/lockdep.txt b/Documentation/RCU/lockdep.txt index fe24b58627bd..d7a49b2f6994 100644 --- a/Documentation/RCU/lockdep.txt +++ b/Documentation/RCU/lockdep.txt @@ -32,9 +32,20 @@ checking of rcu_dereference() primitives: srcu_dereference(p, sp): Check for SRCU read-side critical section. rcu_dereference_check(p, c): - Use explicit check expression "c". + Use explicit check expression "c". This is useful in + code that is invoked by both readers and updaters. rcu_dereference_raw(p) Don't check. (Use sparingly, if at all.) + rcu_dereference_protected(p, c): + Use explicit check expression "c", and omit all barriers + and compiler constraints. This is useful when the data + structure cannot change, for example, in code that is + invoked only by updaters. + rcu_access_pointer(p): + Return the value of the pointer and omit all barriers, + but retain the compiler constraints that prevent duplicating + or coalescsing. This is useful when when testing the + value of the pointer itself, for example, against NULL. The rcu_dereference_check() check expression can be any boolean expression, but would normally include one of the rcu_read_lock_held() @@ -59,7 +70,20 @@ In case (1), the pointer is picked up in an RCU-safe manner for vanilla RCU read-side critical sections, in case (2) the ->file_lock prevents any change from taking place, and finally, in case (3) the current task is the only task accessing the file_struct, again preventing any change -from taking place. +from taking place. If the above statement was invoked only from updater +code, it could instead be written as follows: + + file = rcu_dereference_protected(fdt->fd[fd], + lockdep_is_held(&files->file_lock) || + atomic_read(&files->count) == 1); + +This would verify cases #2 and #3 above, and furthermore lockdep would +complain if this was used in an RCU read-side critical section unless one +of these two cases held. Because rcu_dereference_protected() omits all +barriers and compiler constraints, it generates better code than do the +other flavors of rcu_dereference(). On the other hand, it is illegal +to use rcu_dereference_protected() if either the RCU-protected pointer +or the RCU-protected data that it points to can change concurrently. There are currently only "universal" versions of the rcu_assign_pointer() and RCU list-/tree-traversal primitives, which do not (yet) check for diff --git a/Documentation/RCU/whatisRCU.txt b/Documentation/RCU/whatisRCU.txt index 1dc00ee97163..cfaac34c4557 100644 --- a/Documentation/RCU/whatisRCU.txt +++ b/Documentation/RCU/whatisRCU.txt @@ -840,6 +840,12 @@ SRCU: Initialization/cleanup init_srcu_struct cleanup_srcu_struct +All: lockdep-checked RCU-protected pointer access + + rcu_dereference_check + rcu_dereference_protected + rcu_access_pointer + See the comment headers in the source code (or the docbook generated from them) for more information. diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 9c31c2e63684..d075c31c1386 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -564,6 +564,16 @@ Who: Avi Kivity <avi@redhat.com> ---------------------------- +What: xtime, wall_to_monotonic +When: 2.6.36+ +Files: kernel/time/timekeeping.c include/linux/time.h +Why: Cleaning up timekeeping internal values. Please use + existing timekeeping accessor functions to access + the equivalent functionality. +Who: John Stultz <johnstul@us.ibm.com> + +---------------------------- + What: KVM kernel-allocated memory slots When: July 2010 Why: Since 2.6.25, kvm supports user-allocated memory slots, which are diff --git a/Documentation/input/multi-touch-protocol.txt b/Documentation/input/multi-touch-protocol.txt index 8490480ce432..c0fc1c75fd88 100644 --- a/Documentation/input/multi-touch-protocol.txt +++ b/Documentation/input/multi-touch-protocol.txt @@ -68,6 +68,22 @@ like: SYN_MT_REPORT SYN_REPORT +Here is the sequence after lifting one of the fingers: + + ABS_MT_POSITION_X + ABS_MT_POSITION_Y + SYN_MT_REPORT + SYN_REPORT + +And here is the sequence after lifting the remaining finger: + + SYN_MT_REPORT + SYN_REPORT + +If the driver reports one of BTN_TOUCH or ABS_PRESSURE in addition to the +ABS_MT events, the last SYN_MT_REPORT event may be omitted. Otherwise, the +last SYN_REPORT will be dropped by the input core, resulting in no +zero-finger event reaching userland. Event Semantics --------------- @@ -217,11 +233,6 @@ where examples can be found. difference between the contact position and the approaching tool position could be used to derive tilt. [2] The list can of course be extended. -[3] The multi-touch X driver is currently in the prototyping stage. At the -time of writing (April 2009), the MT protocol is not yet merged, and the -prototype implements finger matching, basic mouse support and two-finger -scrolling. The project aims at improving the quality of current multi-touch -functionality available in the Synaptics X driver, and in addition -implement more advanced gestures. +[3] Multitouch X driver project: http://bitmath.org/code/multitouch/. [4] See the section on event computation. [5] See the section on finger tracking. diff --git a/Documentation/intel_txt.txt b/Documentation/intel_txt.txt index f40a1f030019..87c8990dbbd9 100644 --- a/Documentation/intel_txt.txt +++ b/Documentation/intel_txt.txt @@ -161,13 +161,15 @@ o In order to put a system into any of the sleep states after a TXT has been restored, it will restore the TPM PCRs and then transfer control back to the kernel's S3 resume vector. In order to preserve system integrity across S3, the kernel - provides tboot with a set of memory ranges (kernel - code/data/bss, S3 resume code, and AP trampoline) that tboot - will calculate a MAC (message authentication code) over and then - seal with the TPM. On resume and once the measured environment - has been re-established, tboot will re-calculate the MAC and - verify it against the sealed value. Tboot's policy determines - what happens if the verification fails. + provides tboot with a set of memory ranges (RAM and RESERVED_KERN + in the e820 table, but not any memory that BIOS might alter over + the S3 transition) that tboot will calculate a MAC (message + authentication code) over and then seal with the TPM. On resume + and once the measured environment has been re-established, tboot + will re-calculate the MAC and verify it against the sealed value. + Tboot's policy determines what happens if the verification fails. + Note that the c/s 194 of tboot which has the new MAC code supports + this. That's pretty much it for TXT support. diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e4cbca58536c..e2202e93b148 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -320,11 +320,6 @@ and is between 256 and 4096 characters. It is defined in the file amd_iommu= [HW,X86-84] Pass parameters to the AMD IOMMU driver in the system. Possible values are: - isolate - enable device isolation (each device, as far - as possible, will get its own protection - domain) [default] - share - put every device behind one IOMMU into the - same protection domain fullflush - enable flushing of IO/TLB entries when they are unmapped. Otherwise they are flushed before they will be reused, which diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index 2f9115c0ae62..61c291cddf18 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -165,8 +165,8 @@ the user entry_handler invocation is also skipped. 1.4 How Does Jump Optimization Work? -If you configured your kernel with CONFIG_OPTPROBES=y (currently -this option is supported on x86/x86-64, non-preemptive kernel) and +If your kernel is built with CONFIG_OPTPROBES=y (currently this flag +is automatically set 'y' on x86/x86-64, non-preemptive kernel) and the "debug.kprobes_optimization" kernel parameter is set to 1 (see sysctl(8)), Kprobes tries to reduce probe-hit overhead by using a jump instruction instead of a breakpoint instruction at each probepoint. @@ -271,8 +271,6 @@ tweak the kernel's execution path, you need to suppress optimization, using one of the following techniques: - Specify an empty function for the kprobe's post_handler or break_handler. or -- Config CONFIG_OPTPROBES=n. - or - Execute 'sysctl -w debug.kprobes_optimization=n' 2. Architectures Supported @@ -307,10 +305,6 @@ it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO), so you can use "objdump -d -l vmlinux" to see the source-to-object code mapping. -If you want to reduce probing overhead, set "Kprobes jump optimization -support" (CONFIG_OPTPROBES) to "y". You can find this option under the -"Kprobes" line. - 4. API Reference The Kprobes API includes a "register" function and an "unregister" diff --git a/Documentation/networking/timestamping.txt b/Documentation/networking/timestamping.txt index 0e58b4539176..e8c8f4f06c67 100644 --- a/Documentation/networking/timestamping.txt +++ b/Documentation/networking/timestamping.txt @@ -41,11 +41,12 @@ SOF_TIMESTAMPING_SOFTWARE: return system time stamp generated in SOF_TIMESTAMPING_TX/RX determine how time stamps are generated. SOF_TIMESTAMPING_RAW/SYS determine how they are reported in the following control message: - struct scm_timestamping { - struct timespec systime; - struct timespec hwtimetrans; - struct timespec hwtimeraw; - }; + +struct scm_timestamping { + struct timespec systime; + struct timespec hwtimetrans; + struct timespec hwtimeraw; +}; recvmsg() can be used to get this control message for regular incoming packets. For send time stamps the outgoing packet is looped back to @@ -87,12 +88,13 @@ by the network device and will be empty without that support. SIOCSHWTSTAMP: Hardware time stamping must also be initialized for each device driver -that is expected to do hardware time stamping. The parameter is: +that is expected to do hardware time stamping. The parameter is defined in +/include/linux/net_tstamp.h as: struct hwtstamp_config { - int flags; /* no flags defined right now, must be zero */ - int tx_type; /* HWTSTAMP_TX_* */ - int rx_filter; /* HWTSTAMP_FILTER_* */ + int flags; /* no flags defined right now, must be zero */ + int tx_type; /* HWTSTAMP_TX_* */ + int rx_filter; /* HWTSTAMP_FILTER_* */ }; Desired behavior is passed into the kernel and to a specific device by @@ -139,42 +141,56 @@ enum { /* time stamp any incoming packet */ HWTSTAMP_FILTER_ALL, - /* return value: time stamp all packets requested plus some others */ - HWTSTAMP_FILTER_SOME, + /* return value: time stamp all packets requested plus some others */ + HWTSTAMP_FILTER_SOME, /* PTP v1, UDP, any kind of event packet */ HWTSTAMP_FILTER_PTP_V1_L4_EVENT, - ... + /* for the complete list of values, please check + * the include file /include/linux/net_tstamp.h + */ }; DEVICE IMPLEMENTATION A driver which supports hardware time stamping must support the -SIOCSHWTSTAMP ioctl. Time stamps for received packets must be stored -in the skb with skb_hwtstamp_set(). +SIOCSHWTSTAMP ioctl and update the supplied struct hwtstamp_config with +the actual values as described in the section on SIOCSHWTSTAMP. + +Time stamps for received packets must be stored in the skb. To get a pointer +to the shared time stamp structure of the skb call skb_hwtstamps(). Then +set the time stamps in the structure: + +struct skb_shared_hwtstamps { + /* hardware time stamp transformed into duration + * since arbitrary point in time + */ + ktime_t hwtstamp; + ktime_t syststamp; /* hwtstamp transformed to system time base */ +}; Time stamps for outgoing packets are to be generated as follows: -- In hard_start_xmit(), check if skb_hwtstamp_check_tx_hardware() - returns non-zero. If yes, then the driver is expected - to do hardware time stamping. +- In hard_start_xmit(), check if skb_tx(skb)->hardware is set no-zero. + If yes, then the driver is expected to do hardware time stamping. - If this is possible for the skb and requested, then declare - that the driver is doing the time stamping by calling - skb_hwtstamp_tx_in_progress(). A driver not supporting - hardware time stamping doesn't do that. A driver must never - touch sk_buff::tstamp! It is used to store how time stamping - for an outgoing packets is to be done. + that the driver is doing the time stamping by setting the field + skb_tx(skb)->in_progress non-zero. You might want to keep a pointer + to the associated skb for the next step and not free the skb. A driver + not supporting hardware time stamping doesn't do that. A driver must + never touch sk_buff::tstamp! It is used to store software generated + time stamps by the network subsystem. - As soon as the driver has sent the packet and/or obtained a hardware time stamp for it, it passes the time stamp back by calling skb_hwtstamp_tx() with the original skb, the raw - hardware time stamp and a handle to the device (necessary - to convert the hardware time stamp to system time). If obtaining - the hardware time stamp somehow fails, then the driver should - not fall back to software time stamping. The rationale is that - this would occur at a later time in the processing pipeline - than other software time stamping and therefore could lead - to unexpected deltas between time stamps. -- If the driver did not call skb_hwtstamp_tx_in_progress(), then + hardware time stamp. skb_hwtstamp_tx() clones the original skb and + adds the timestamps, therefore the original skb has to be freed now. + If obtaining the hardware time stamp somehow fails, then the driver + should not fall back to software time stamping. The rationale is that + this would occur at a later time in the processing pipeline than other + software time stamping and therefore could lead to unexpected deltas + between time stamps. +- If the driver did not call set skb_tx(skb)->in_progress, then dev_hard_start_xmit() checks whether software time stamping is wanted as fallback and potentially generates the time stamp. diff --git a/Documentation/rbtree.txt b/Documentation/rbtree.txt index aae8355d3166..221f38be98f4 100644 --- a/Documentation/rbtree.txt +++ b/Documentation/rbtree.txt @@ -190,3 +190,61 @@ Example: for (node = rb_first(&mytree); node; node = rb_next(node)) printk("key=%s\n", rb_entry(node, struct mytype, node)->keystring); +Support for Augmented rbtrees +----------------------------- + +Augmented rbtree is an rbtree with "some" additional data stored in each node. +This data can be used to augment some new functionality to rbtree. +Augmented rbtree is an optional feature built on top of basic rbtree +infrastructure. rbtree user who wants this feature will have an augment +callback function in rb_root initialized. + +This callback function will be called from rbtree core routines whenever +a node has a change in one or both of its children. It is the responsibility +of the callback function to recalculate the additional data that is in the +rb node using new children information. Note that if this new additional +data affects the parent node's additional data, then callback function has +to handle it and do the recursive updates. + + +Interval tree is an example of augmented rb tree. Reference - +"Introduction to Algorithms" by Cormen, Leiserson, Rivest and Stein. +More details about interval trees: + +Classical rbtree has a single key and it cannot be directly used to store +interval ranges like [lo:hi] and do a quick lookup for any overlap with a new +lo:hi or to find whether there is an exact match for a new lo:hi. + +However, rbtree can be augmented to store such interval ranges in a structured +way making it possible to do efficient lookup and exact match. + +This "extra information" stored in each node is the maximum hi +(max_hi) value among all the nodes that are its descendents. This +information can be maintained at each node just be looking at the node +and its immediate children. And this will be used in O(log n) lookup +for lowest match (lowest start address among all possible matches) +with something like: + +find_lowest_match(lo, hi, node) +{ + lowest_match = NULL; + while (node) { + if (max_hi(node->left) > lo) { + // Lowest overlap if any must be on left side + node = node->left; + } else if (overlap(lo, hi, node)) { + lowest_match = node; + break; + } else if (lo > node->lo) { + // Lowest overlap if any must be on right side + node = node->right; + } else { + break; + } + } + return lowest_match; +} + +Finding exact match will be to first find lowest match and then to follow +successor nodes looking for exact match, until the start of a node is beyond +the hi value we are looking for. diff --git a/Documentation/scheduler/sched-design-CFS.txt b/Documentation/scheduler/sched-design-CFS.txt index 6f33593e59e2..8239ebbcddce 100644 --- a/Documentation/scheduler/sched-design-CFS.txt +++ b/Documentation/scheduler/sched-design-CFS.txt @@ -211,7 +211,7 @@ provide fair CPU time to each such task group. For example, it may be desirable to first provide fair CPU time to each user on the system and then to each task belonging to a user. -CONFIG_GROUP_SCHED strives to achieve exactly that. It lets tasks to be +CONFIG_CGROUP_SCHED strives to achieve exactly that. It lets tasks to be grouped and divides CPU time fairly among such groups. CONFIG_RT_GROUP_SCHED permits to group real-time (i.e., SCHED_FIFO and @@ -220,38 +220,11 @@ SCHED_RR) tasks. CONFIG_FAIR_GROUP_SCHED permits to group CFS (i.e., SCHED_NORMAL and SCHED_BATCH) tasks. -At present, there are two (mutually exclusive) mechanisms to group tasks for -CPU bandwidth control purposes: - - - Based on user id (CONFIG_USER_SCHED) - - With this option, tasks are grouped according to their user id. - - - Based on "cgroup" pseudo filesystem (CONFIG_CGROUP_SCHED) - - This options needs CONFIG_CGROUPS to be defined, and lets the administrator + These options need CONFIG_CGROUPS to be defined, and let the administrator create arbitrary groups of tasks, using the "cgroup" pseudo filesystem. See Documentation/cgroups/cgroups.txt for more information about this filesystem. -Only one of these options to group tasks can be chosen and not both. - -When CONFIG_USER_SCHED is defined, a directory is created in sysfs for each new -user and a "cpu_share" file is added in that directory. - - # cd /sys/kernel/uids - # cat 512/cpu_share # Display user 512's CPU share - 1024 - # echo 2048 > 512/cpu_share # Modify user 512's CPU share - # cat 512/cpu_share # Display user 512's CPU share - 2048 - # - -CPU bandwidth between two users is divided in the ratio of their CPU shares. -For example: if you would like user "root" to get twice the bandwidth of user -"guest," then set the cpu_share for both the users such that "root"'s cpu_share -is twice "guest"'s cpu_share. - -When CONFIG_CGROUP_SCHED is defined, a "cpu.shares" file is created for each +When CONFIG_FAIR_GROUP_SCHED is defined, a "cpu.shares" file is created for each group created using the pseudo filesystem. See example steps below to create task groups and modify their CPU share using the "cgroups" pseudo filesystem. @@ -273,24 +246,3 @@ task groups and modify their CPU share using the "cgroups" pseudo filesystem. # #Launch gmplayer (or your favourite movie player) # echo <movie_player_pid> > multimedia/tasks - -8. Implementation note: user namespaces - -User namespaces are intended to be hierarchical. But they are currently -only partially implemented. Each of those has ramifications for CFS. - -First, since user namespaces are hierarchical, the /sys/kernel/uids -presentation is inadequate. Eventually we will likely want to use sysfs -tagging to provide private views of /sys/kernel/uids within each user -namespace. - -Second, the hierarchical nature is intended to support completely -unprivileged use of user namespaces. So if using user groups, then -we want the users in a user namespace to be children of the user -who created it. - -That is currently unimplemented. So instead, every user in a new -user namespace will receive 1024 shares just like any user in the -initial user namespace. Note that at the moment creation of a new -user namespace requires each of CAP_SYS_ADMIN, CAP_SETUID, and -CAP_SETGID. diff --git a/Documentation/scheduler/sched-rt-group.txt b/Documentation/scheduler/sched-rt-group.txt index 86eabe6c3419..605b0d40329d 100644 --- a/Documentation/scheduler/sched-rt-group.txt +++ b/Documentation/scheduler/sched-rt-group.txt @@ -126,23 +126,12 @@ priority! 2.3 Basis for grouping tasks ---------------------------- -There are two compile-time settings for allocating CPU bandwidth. These are -configured using the "Basis for grouping tasks" multiple choice menu under -General setup > Group CPU Scheduler: - -a. CONFIG_USER_SCHED (aka "Basis for grouping tasks" = "user id") - -This lets you use the virtual files under -"/sys/kernel/uids/<uid>/cpu_rt_runtime_us" to control he CPU time reserved for -each user . - -The other option is: - -.o CONFIG_CGROUP_SCHED (aka "Basis for grouping tasks" = "Control groups") +Enabling CONFIG_RT_GROUP_SCHED lets you explicitly allocate real +CPU bandwidth to task groups. This uses the /cgroup virtual file system and "/cgroup/<cgroup>/cpu.rt_runtime_us" to control the CPU time reserved for each -control group instead. +control group. For more information on working with control groups, you should read Documentation/cgroups/cgroups.txt as well. @@ -161,8 +150,7 @@ For now, this can be simplified to just the following (but see Future plans): =============== There is work in progress to make the scheduling period for each group -("/sys/kernel/uids/<uid>/cpu_rt_period_us" or -"/cgroup/<cgroup>/cpu.rt_period_us" respectively) configurable as well. +("/cgroup/<cgroup>/cpu.rt_period_us") configurable as well. The constraint on the period is that a subgroup must have a smaller or equal period to its parent. But realistically its not very useful _yet_ diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt index a9100b28eb84..ec94748ae65b 100644 --- a/Documentation/trace/kprobetrace.txt +++ b/Documentation/trace/kprobetrace.txt @@ -40,7 +40,9 @@ Synopsis of kprobe_events $stack : Fetch stack address. $retval : Fetch return value.(*) +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**) - NAME=FETCHARG: Set NAME as the argument name of FETCHARG. + NAME=FETCHARG : Set NAME as the argument name of FETCHARG. + FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types + (u8/u16/u32/u64/s8/s16/s32/s64) are supported. (*) only for return probe. (**) this is useful for fetching a field of data structures. diff --git a/MAINTAINERS b/MAINTAINERS index 5b422908d0f3..70eef8a9ae17 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -485,8 +485,8 @@ S: Maintained F: drivers/input/mouse/bcm5974.c APPLE SMC DRIVER -M: Nicolas Boichat <nicolas@boichat.ch> -L: mactel-linux-devel@lists.sourceforge.net +M: Henrik Rydberg <rydberg@euromail.se> +L: lm-sensors@lm-sensors.org S: Maintained F: drivers/hwmon/applesmc.c @@ -2953,6 +2953,17 @@ S: Odd Fixes F: Documentation/networking/README.ipw2200 F: drivers/net/wireless/ipw2x00/ipw2200.* +INTEL(R) TRUSTED EXECUTION TECHNOLOGY (TXT) +M: Joseph Cihula <joseph.cihula@intel.com> +M: Shane Wang <shane.wang@intel.com> +L: tboot-devel@lists.sourceforge.net +W: http://tboot.sourceforge.net +T: Mercurial http://www.bughost.org/repos.hg/tboot.hg +S: Supported +F: Documentation/intel_txt.txt +F: include/linux/tboot.h +F: arch/x86/kernel/tboot.c + INTEL WIRELESS WIMAX CONNECTION 2400 M: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com> M: linux-wimax@intel.com @@ -4353,13 +4364,13 @@ M: Paul Mackerras <paulus@samba.org> M: Ingo Molnar <mingo@elte.hu> M: Arnaldo Carvalho de Melo <acme@redhat.com> S: Supported -F: kernel/perf_event.c +F: kernel/perf_event*.c F: include/linux/perf_event.h -F: arch/*/kernel/perf_event.c -F: arch/*/kernel/*/perf_event.c -F: arch/*/kernel/*/*/perf_event.c +F: arch/*/kernel/perf_event*.c +F: arch/*/kernel/*/perf_event*.c +F: arch/*/kernel/*/*/perf_event*.c F: arch/*/include/asm/perf_event.h -F: arch/*/lib/perf_event.c +F: arch/*/lib/perf_event*.c F: arch/*/kernel/perf_callchain.c F: tools/perf/ diff --git a/arch/Kconfig b/arch/Kconfig index e5eb1337a537..f06010fb4838 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -42,15 +42,10 @@ config KPROBES If in doubt, say "N". config OPTPROBES - bool "Kprobes jump optimization support (EXPERIMENTAL)" - default y - depends on KPROBES + def_bool y + depends on KPROBES && HAVE_OPTPROBES depends on !PREEMPT - depends on HAVE_OPTPROBES select KALLSYMS_ALL - help - This option will allow kprobes to optimize breakpoint to - a jump for reducing its overhead. config HAVE_EFFICIENT_UNALIGNED_ACCESS bool diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 75291fdd379f..b7193986cbf9 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -55,6 +55,9 @@ config ARCH_USES_GETTIMEOFFSET bool default y +config GENERIC_CMOS_UPDATE + def_bool y + config ZONE_DMA bool default y diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c index 5d0826654c61..5465e932e568 100644 --- a/arch/alpha/kernel/time.c +++ b/arch/alpha/kernel/time.c @@ -75,8 +75,6 @@ static struct { __u32 last_time; /* ticks/cycle * 2^48 */ unsigned long scaled_ticks_per_cycle; - /* last time the CMOS clock got updated */ - time_t last_rtc_update; /* partial unused tick */ unsigned long partial_tick; } state; @@ -91,6 +89,52 @@ static inline __u32 rpcc(void) return result; } +int update_persistent_clock(struct timespec now) +{ + return set_rtc_mmss(now.tv_sec); +} + +void read_persistent_clock(struct timespec *ts) +{ + unsigned int year, mon, day, hour, min, sec, epoch; + + sec = CMOS_READ(RTC_SECONDS); + min = CMOS_READ(RTC_MINUTES); + hour = CMOS_READ(RTC_HOURS); + day = CMOS_READ(RTC_DAY_OF_MONTH); + mon = CMOS_READ(RTC_MONTH); + year = CMOS_READ(RTC_YEAR); + + if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) { + sec = bcd2bin(sec); + min = bcd2bin(min); + hour = bcd2bin(hour); + day = bcd2bin(day); + mon = bcd2bin(mon); + year = bcd2bin(year); + } + + /* PC-like is standard; used for year >= 70 */ + epoch = 1900; + if (year < 20) + epoch = 2000; + else if (year >= 20 && year < 48) + /* NT epoch */ + epoch = 1980; + else if (year >= 48 && year < 70) + /* Digital UNIX epoch */ + epoch = 1952; + + printk(KERN_INFO "Using epoch = %d\n", epoch); + + if ((year += epoch) < 1970) + year += 100; + + ts->tv_sec = mktime(year, mon, day, hour, min, sec); +} + + + /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -123,19 +167,6 @@ irqreturn_t timer_interrupt(int irq, void *dev) if (nticks) do_timer(nticks); - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if (ntp_synced() - && xtime.tv_sec > state.last_rtc_update + 660 - && xtime.tv_nsec >= 500000 - ((unsigned) TICK_SIZE) / 2 - && xtime.tv_nsec <= 500000 + ((unsigned) TICK_SIZE) / 2) { - int tmp = set_rtc_mmss(xtime.tv_sec); - state.last_rtc_update = xtime.tv_sec - (tmp ? 600 : 0); - } - write_sequnlock(&xtime_lock); #ifndef CONFIG_SMP @@ -304,7 +335,7 @@ rpcc_after_update_in_progress(void) void __init time_init(void) { - unsigned int year, mon, day, hour, min, sec, cc1, cc2, epoch; + unsigned int cc1, cc2; unsigned long cycle_freq, tolerance; long diff; @@ -348,43 +379,6 @@ time_init(void) bogomips yet, but this is close on a 500Mhz box. */ __delay(1000000); - sec = CMOS_READ(RTC_SECONDS); - min = CMOS_READ(RTC_MINUTES); - hour = CMOS_READ(RTC_HOURS); - day = CMOS_READ(RTC_DAY_OF_MONTH); - mon = CMOS_READ(RTC_MONTH); - year = CMOS_READ(RTC_YEAR); - - if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) { - sec = bcd2bin(sec); - min = bcd2bin(min); - hour = bcd2bin(hour); - day = bcd2bin(day); - mon = bcd2bin(mon); - year = bcd2bin(year); - } - - /* PC-like is standard; used for year >= 70 */ - epoch = 1900; - if (year < 20) - epoch = 2000; - else if (year >= 20 && year < 48) - /* NT epoch */ - epoch = 1980; - else if (year >= 48 && year < 70) - /* Digital UNIX epoch */ - epoch = 1952; - - printk(KERN_INFO "Using epoch = %d\n", epoch); - - if ((year += epoch) < 1970) - year += 100; - - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); - xtime.tv_nsec = 0; - - wall_to_monotonic.tv_sec -= xtime.tv_sec; - wall_to_monotonic.tv_nsec = 0; if (HZ > (1<<16)) { extern void __you_loose (void); @@ -394,7 +388,6 @@ time_init(void) state.last_time = cc1; state.scaled_ticks_per_cycle = ((unsigned long) HZ << FIX_SHIFT) / cycle_freq; - state.last_rtc_update = 0; state.partial_tick = 0L; /* Startup the timer source. */ diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c index f27aa3b259fa..a047d69d17e2 100644 --- a/arch/avr32/kernel/time.c +++ b/arch/avr32/kernel/time.c @@ -110,17 +110,17 @@ static struct clock_event_device comparator = { .set_mode = comparator_mode, }; +void read_persistent_clock(struct timespec *ts) +{ + ts->tv.sec = mktime(2007, 1, 1, 0, 0, 0); + ts->tv_nsec = 0; +} + void __init time_init(void) { unsigned long counter_hz; int ret; - xtime.tv_sec = mktime(2007, 1, 1, 0, 0, 0); - xtime.tv_nsec = 0; - - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - /* figure rate for counter */ counter_hz = clk_get_rate(boot_cpu_data.clk); counter.mult = clocksource_hz2mult(counter_hz, counter.shift); diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c index cb7a01d4f009..8c9a43daf80f 100644 --- a/arch/blackfin/kernel/time-ts.c +++ b/arch/blackfin/kernel/time-ts.c @@ -353,9 +353,15 @@ void bfin_coretmr_clockevent_init(void) #endif /* CONFIG_TICKSOURCE_CORETMR */ -void __init time_init(void) +void read_persistent_clock(struct timespec *ts) { time_t secs_since_1970 = (365 * 37 + 9) * 24 * 60 * 60; /* 1 Jan 2007 */ + ts->tv_sec = secs_since_1970; + ts->tv_nsec = 0; +} + +void __init time_init(void) +{ #ifdef CONFIG_RTC_DRV_BFIN /* [#2663] hack to filter junk RTC values that would cause @@ -368,11 +374,6 @@ void __init time_init(void) } #endif - /* Initialize xtime. From now on, xtime is updated with timer interrupts */ - xtime.tv_sec = secs_since_1970; - xtime.tv_nsec = 0; - set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); - bfin_cs_cycles_init(); bfin_cs_gptimer0_init(); diff --git a/arch/blackfin/kernel/time.c b/arch/blackfin/kernel/time.c index 13c1ee3e6408..c9113619029f 100644 --- a/arch/blackfin/kernel/time.c +++ b/arch/blackfin/kernel/time.c @@ -112,11 +112,6 @@ u32 arch_gettimeoffset(void) } #endif -static inline int set_rtc_mmss(unsigned long nowtime) -{ - return 0; -} - /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -126,29 +121,8 @@ __attribute__((l1_text)) #endif irqreturn_t timer_interrupt(int irq, void *dummy) { - /* last time the cmos clock got updated */ - static long last_rtc_update; - write_seqlock(&xtime_lock); do_timer(1); - - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / NSEC_PER_USEC) >= - 500000 - ((unsigned)TICK_SIZE) / 2 - && (xtime.tv_nsec / NSEC_PER_USEC) <= - 500000 + ((unsigned)TICK_SIZE) / 2) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - /* Do it again in 60s. */ - last_rtc_update = xtime.tv_sec - 600; - } write_sequnlock(&xtime_lock); #ifdef CONFIG_IPIPE @@ -161,10 +135,15 @@ irqreturn_t timer_interrupt(int irq, void *dummy) return IRQ_HANDLED; } -void __init time_init(void) +void read_persistent_clock(struct timespec *ts) { time_t secs_since_1970 = (365 * 37 + 9) * 24 * 60 * 60; /* 1 Jan 2007 */ + ts->tv_sec = secs_since_1970; + ts->tv_nsec = 0; +} +void __init time_init(void) +{ #ifdef CONFIG_RTC_DRV_BFIN /* [#2663] hack to filter junk RTC values that would cause * userspace to have to deal with time values greater than @@ -176,11 +155,5 @@ void __init time_init(void) } #endif - /* Initialize xtime. From now on, xtime is updated with timer interrupts */ - xtime.tv_sec = secs_since_1970; - xtime.tv_nsec = 0; - - wall_to_monotonic.tv_sec = -xtime.tv_sec; - time_sched_init(timer_interrupt); } diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index 059eac6abda1..e25bf4440b51 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -23,6 +23,9 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_TIME def_bool y +config GENERIC_CMOS_UPDATE + def_bool y + config ARCH_USES_GETTIMEOFFSET def_bool y diff --git a/arch/cris/arch-v10/kernel/time.c b/arch/cris/arch-v10/kernel/time.c index 31ca1418d5a7..30adae594aef 100644 --- a/arch/cris/arch-v10/kernel/time.c +++ b/arch/cris/arch-v10/kernel/time.c @@ -26,7 +26,6 @@ /* it will make jiffies at 96 hz instead of 100 hz though */ #undef USE_CASCADE_TIMERS -extern void update_xtime_from_cmos(void); extern int set_rtc_mmss(unsigned long nowtime); extern int have_rtc; @@ -188,8 +187,6 @@ stop_watchdog(void) #endif } -/* last time the cmos clock got updated */ -static long last_rtc_update = 0; /* * timer_interrupt() needs to keep up the real-time clock, @@ -232,24 +229,6 @@ timer_interrupt(int irq, void *dev_id) do_timer(1); cris_do_profile(regs); /* Save profiling information */ - - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - * - * The division here is not time critical since it will run once in - * 11 minutes - */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - (tick_nsec / 1000) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + (tick_nsec / 1000) / 2) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ - } return IRQ_HANDLED; } @@ -274,22 +253,10 @@ time_init(void) */ loops_per_usec = 50; - if(RTC_INIT() < 0) { - /* no RTC, start at 1980 */ - xtime.tv_sec = 0; - xtime.tv_nsec = 0; + if(RTC_INIT() < 0) have_rtc = 0; - } else { - /* get the current time */ + else have_rtc = 1; - update_xtime_from_cmos(); - } - - /* - * Initialize wall_to_monotonic such that adding it to xtime will yield zero, the - * tv_nsec field must be normalized (i.e., 0 <= nsec < NSEC_PER_SEC). - */ - set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); /* Setup the etrax timers * Base frequency is 25000 hz, divider 250 -> 100 HZ diff --git a/arch/cris/arch-v32/kernel/time.c b/arch/cris/arch-v32/kernel/time.c index b1920d8de403..1ee0e1010228 100644 --- a/arch/cris/arch-v32/kernel/time.c +++ b/arch/cris/arch-v32/kernel/time.c @@ -44,7 +44,6 @@ unsigned long timer_regs[NR_CPUS] = #endif }; -extern void update_xtime_from_cmos(void); extern int set_rtc_mmss(unsigned long nowtime); extern int have_rtc; @@ -198,9 +197,6 @@ handle_watchdog_bite(struct pt_regs* regs) #endif } -/* Last time the cmos clock got updated. */ -static long last_rtc_update = 0; - /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick. @@ -238,25 +234,6 @@ timer_interrupt(int irq, void *dev_id) /* Call the real timer interrupt handler */ do_timer(1); - - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - * - * The division here is not time critical since it will run once in - * 11 minutes - */ - if ((time_status & STA_UNSYNC) == 0 && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - (tick_nsec / 1000) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + (tick_nsec / 1000) / 2) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - /* Do it again in 60 s */ - last_rtc_update = xtime.tv_sec - 600; - } return IRQ_HANDLED; } @@ -309,23 +286,10 @@ time_init(void) */ loops_per_usec = 50; - if(RTC_INIT() < 0) { - /* No RTC, start at 1980 */ - xtime.tv_sec = 0; - xtime.tv_nsec = 0; + if(RTC_INIT() < 0) have_rtc = 0; - } else { - /* Get the current time */ + else have_rtc = 1; - update_xtime_from_cmos(); - } - - /* - * Initialize wall_to_monotonic such that adding it to - * xtime will yield zero, the tv_nsec field must be normalized - * (i.e., 0 <= nsec < NSEC_PER_SEC). - */ - set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); /* Start CPU local timer. */ cris_timer_init(); diff --git a/arch/cris/kernel/time.c b/arch/cris/kernel/time.c index a05dd31f3efb..c72730d20ef6 100644 --- a/arch/cris/kernel/time.c +++ b/arch/cris/kernel/time.c @@ -98,6 +98,8 @@ unsigned long get_cmos_time(void) { unsigned int year, mon, day, hour, min, sec; + if(!have_rtc) + return 0; sec = CMOS_READ(RTC_SECONDS); min = CMOS_READ(RTC_MINUTES); @@ -119,19 +121,19 @@ get_cmos_time(void) return mktime(year, mon, day, hour, min, sec); } -/* update xtime from the CMOS settings. used when /dev/rtc gets a SET_TIME. - * TODO: this doesn't reset the fancy NTP phase stuff as do_settimeofday does. - */ -void -update_xtime_from_cmos(void) +int update_persistent_clock(struct timespec now) { - if(have_rtc) { - xtime.tv_sec = get_cmos_time(); - xtime.tv_nsec = 0; - } + return set_rtc_mmss(now.tv_sec); } +void read_persistent_clock(struct timespec *ts) +{ + ts->tv_sec = get_cmos_time(); + ts->tv_nsec = 0; +} + + extern void cris_profile_sample(struct pt_regs* regs); void diff --git a/arch/frv/kernel/time.c b/arch/frv/kernel/time.c index fb0ce7577225..0ddbbae83cb2 100644 --- a/arch/frv/kernel/time.c +++ b/arch/frv/kernel/time.c @@ -48,20 +48,12 @@ static struct irqaction timer_irq = { .name = "timer", }; -static inline int set_rtc_mmss(unsigned long nowtime) -{ - return -1; -} - /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick */ static irqreturn_t timer_interrupt(int irq, void *dummy) { - /* last time the cmos clock got updated */ - static long last_rtc_update = 0; - profile_tick(CPU_PROFILING); /* * Here we are in the timer irq handler. We just have irqs locally @@ -74,22 +66,6 @@ static irqreturn_t timer_interrupt(int irq, void *dummy) do_timer(1); - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2 - ) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ - } - #ifdef CONFIG_HEARTBEAT static unsigned short n; n++; @@ -119,7 +95,8 @@ void time_divisor_init(void) __set_TCSR_DATA(0, base >> 8); } -void time_init(void) + +void read_persistent_clock(struct timespec *ts) { unsigned int year, mon, day, hour, min, sec; @@ -135,9 +112,12 @@ void time_init(void) if ((year += 1900) < 1970) year += 100; - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); - xtime.tv_nsec = 0; + ts->tv_sec = mktime(year, mon, day, hour, min, sec); + ts->tv_nsec = 0; +} +void time_init(void) +{ /* install scheduling interrupt handler */ setup_irq(IRQ_CPU_TIMER0, &timer_irq); diff --git a/arch/h8300/kernel/time.c b/arch/h8300/kernel/time.c index 7f2d6cfbb4b6..165005aff9df 100644 --- a/arch/h8300/kernel/time.c +++ b/arch/h8300/kernel/time.c @@ -41,7 +41,7 @@ void h8300_timer_tick(void) update_process_times(user_mode(get_irq_regs())); } -void __init time_init(void) +void read_persistent_clock(struct timespec *ts) { unsigned int year, mon, day, hour, min, sec; @@ -56,8 +56,12 @@ void __init time_init(void) #endif if ((year += 1900) < 1970) year += 100; - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); - xtime.tv_nsec = 0; + ts->tv_sec = mktime(year, mon, day, hour, min, sec); + ts->tv_nsec = 0; +} + +void __init time_init(void) +{ h8300_timer_setup(); } diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 47a192781b0a..653b3c46ea82 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -430,18 +430,16 @@ static int __init rtc_init(void) } module_init(rtc_init); +void read_persistent_clock(struct timespec *ts) +{ + efi_gettimeofday(ts); +} + void __init time_init (void) { register_percpu_irq(IA64_TIMER_VECTOR, &timer_irqaction); - efi_gettimeofday(&xtime); ia64_init_itm(); - - /* - * Initialize wall_to_monotonic such that adding it to xtime will yield zero, the - * tv_nsec field must be normalized (i.e., 0 <= nsec < NSEC_PER_SEC). - */ - set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); } /* diff --git a/arch/m32r/kernel/time.c b/arch/m32r/kernel/time.c index 9cedcef11575..bda86820bffd 100644 --- a/arch/m32r/kernel/time.c +++ b/arch/m32r/kernel/time.c @@ -106,24 +106,6 @@ u32 arch_gettimeoffset(void) } /* - * In order to set the CMOS clock precisely, set_rtc_mmss has to be - * called 500 ms after the second nowtime has started, because when - * nowtime is written into the registers of the CMOS clock, it will - * jump to the next second precisely 500 ms later. Check the Motorola - * MC146818A or Dallas DS12887 data sheet for details. - * - * BUG: This routine does not handle hour overflow properly; it just - * sets the minutes. Usually you won't notice until after reboot! - */ -static inline int set_rtc_mmss(unsigned long nowtime) -{ - return 0; -} - -/* last time the cmos clock got updated */ -static long last_rtc_update = 0; - -/* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick */ @@ -138,23 +120,6 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id) #ifndef CONFIG_SMP update_process_times(user_mode(get_irq_regs())); #endif - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - write_seqlock(&xtime_lock); - if (ntp_synced() - && xtime.tv_sec > last_rtc_update + 660 - && (xtime.tv_nsec / 1000) >= 500000 - ((unsigned)TICK_SIZE) / 2 - && (xtime.tv_nsec / 1000) <= 500000 + ((unsigned)TICK_SIZE) / 2) - { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else /* do it again in 60 s */ - last_rtc_update = xtime.tv_sec - 600; - } - write_sequnlock(&xtime_lock); /* As we return to user mode fire off the other CPU schedulers.. this is basically because we don't yet share IRQ's around. This message is rigged to be safe on the 386 - basically it's @@ -174,7 +139,7 @@ static struct irqaction irq0 = { .name = "MFT2", }; -void __init time_init(void) +void read_persistent_clock(struct timespec *ts) { unsigned int epoch, year, mon, day, hour, min, sec; @@ -194,11 +159,13 @@ void __init time_init(void) epoch = 1952; year += epoch; - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); - xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ); - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); + ts->tv_sec = mktime(year, mon, day, hour, min, sec); + ts->tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ); +} + +void __init time_init(void) +{ #if defined(CONFIG_CHIP_M32102) || defined(CONFIG_CHIP_XNUX2) \ || defined(CONFIG_CHIP_VDEC2) || defined(CONFIG_CHIP_M32700) \ || defined(CONFIG_CHIP_OPSP) || defined(CONFIG_CHIP_M32104) diff --git a/arch/m68k/include/asm/atomic_mm.h b/arch/m68k/include/asm/atomic_mm.h index 88b7af20a996..d9d2ed647435 100644 --- a/arch/m68k/include/asm/atomic_mm.h +++ b/arch/m68k/include/asm/atomic_mm.h @@ -148,14 +148,18 @@ static inline int atomic_xchg(atomic_t *v, int new) static inline int atomic_sub_and_test(int i, atomic_t *v) { char c; - __asm__ __volatile__("subl %2,%1; seq %0" : "=d" (c), "+m" (*v): "g" (i)); + __asm__ __volatile__("subl %2,%1; seq %0" + : "=d" (c), "+m" (*v) + : "id" (i)); return c != 0; } static inline int atomic_add_negative(int i, atomic_t *v) { char c; - __asm__ __volatile__("addl %2,%1; smi %0" : "=d" (c), "+m" (*v): "g" (i)); + __asm__ __volatile__("addl %2,%1; smi %0" + : "=d" (c), "+m" (*v) + : "id" (i)); return c != 0; } diff --git a/arch/m68k/include/asm/sigcontext.h b/arch/m68k/include/asm/sigcontext.h index 1320eaa4cc2a..a29dd74a17cb 100644 --- a/arch/m68k/include/asm/sigcontext.h +++ b/arch/m68k/include/asm/sigcontext.h @@ -17,13 +17,11 @@ struct sigcontext { #ifndef __uClinux__ # ifdef __mcoldfire__ unsigned long sc_fpregs[2][2]; /* room for two fp registers */ - unsigned long sc_fpcntl[3]; - unsigned char sc_fpstate[16+6*8]; # else unsigned long sc_fpregs[2*3]; /* room for two fp registers */ +# endif unsigned long sc_fpcntl[3]; unsigned char sc_fpstate[216]; -# endif #endif }; diff --git a/arch/m68k/kernel/time.c b/arch/m68k/kernel/time.c index 17dc2a31a7ca..4926b3856c15 100644 --- a/arch/m68k/kernel/time.c +++ b/arch/m68k/kernel/time.c @@ -73,21 +73,24 @@ static irqreturn_t timer_interrupt(int irq, void *dummy) return IRQ_HANDLED; } -void __init time_init(void) +void read_persistent_clock(struct timespec *ts) { struct rtc_time time; + ts->tv_sec = 0; + ts->tv_nsec = 0; if (mach_hwclk) { mach_hwclk(0, &time); if ((time.tm_year += 1900) < 1970) time.tm_year += 100; - xtime.tv_sec = mktime(time.tm_year, time.tm_mon, time.tm_mday, + ts->tv_sec = mktime(time.tm_year, time.tm_mon, time.tm_mday, time.tm_hour, time.tm_min, time.tm_sec); - xtime.tv_nsec = 0; } - wall_to_monotonic.tv_sec = -xtime.tv_sec; +} +void __init time_init(void) +{ mach_sched_init(timer_interrupt); } diff --git a/arch/mips/include/asm/i8253.h b/arch/mips/include/asm/i8253.h index 032ca73f181b..48bb82372994 100644 --- a/arch/mips/include/asm/i8253.h +++ b/arch/mips/include/asm/i8253.h @@ -12,7 +12,7 @@ #define PIT_CH0 0x40 #define PIT_CH2 0x42 -extern spinlock_t i8253_lock; +extern raw_spinlock_t i8253_lock; extern void setup_pit_timer(void); diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c index ed5c441615e4..94794062a177 100644 --- a/arch/mips/kernel/i8253.c +++ b/arch/mips/kernel/i8253.c @@ -15,7 +15,7 @@ #include <asm/io.h> #include <asm/time.h> -DEFINE_SPINLOCK(i8253_lock); +DEFINE_RAW_SPINLOCK(i8253_lock); EXPORT_SYMBOL(i8253_lock); /* @@ -26,7 +26,7 @@ EXPORT_SYMBOL(i8253_lock); static void init_pit_timer(enum clock_event_mode mode, struct clock_event_device *evt) { - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); switch(mode) { case CLOCK_EVT_MODE_PERIODIC: @@ -55,7 +55,7 @@ static void init_pit_timer(enum clock_event_mode mode, /* Nothing to do here */ break; } - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); } /* @@ -65,10 +65,10 @@ static void init_pit_timer(enum clock_event_mode mode, */ static int pit_next_event(unsigned long delta, struct clock_event_device *evt) { - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); outb_p(delta & 0xff , PIT_CH0); /* LSB */ outb(delta >> 8 , PIT_CH0); /* MSB */ - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); return 0; } @@ -137,7 +137,7 @@ static cycle_t pit_read(struct clocksource *cs) static int old_count; static u32 old_jifs; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); /* * Although our caller may have the read side of xtime_lock, * this is now a seqlock, and we are cheating in this routine @@ -183,7 +183,7 @@ static cycle_t pit_read(struct clocksource *cs) old_count = count; old_jifs = jifs; - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); count = (LATCH - 1) - count; diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig index 89faacad5d17..1c4565a9102b 100644 --- a/arch/mn10300/Kconfig +++ b/arch/mn10300/Kconfig @@ -37,6 +37,9 @@ config GENERIC_HARDIRQS_NO__DO_IRQ config GENERIC_CALIBRATE_DELAY def_bool y +config GENERIC_CMOS_UPDATE + def_bool y + config GENERIC_FIND_NEXT_BIT def_bool y diff --git a/arch/mn10300/kernel/rtc.c b/arch/mn10300/kernel/rtc.c index 7978470b5749..815a933aafa8 100644 --- a/arch/mn10300/kernel/rtc.c +++ b/arch/mn10300/kernel/rtc.c @@ -26,17 +26,15 @@ static long last_rtc_update; /* time for RTC to update itself in ioclks */ static unsigned long mn10300_rtc_update_period; -/* - * read the current RTC time - */ -unsigned long __init get_initial_rtc_time(void) +void read_persistent_clock(struct timespec *ts) { struct rtc_time tm; get_rtc_time(&tm); - return mktime(tm.tm_year, tm.tm_mon, tm.tm_mday, + ts->tv_sec = mktime(tm.tm_year, tm.tm_mon, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec); + ts->tv_nsec = 0; } /* @@ -110,24 +108,9 @@ static int set_rtc_mmss(unsigned long nowtime) return retval; } -void check_rtc_time(void) +int update_persistent_clock(struct timespec now) { - /* the RTC clock just finished ticking over again this second - * - if we have an externally synchronized Linux clock, then update - * RTC clock accordingly every ~11 minutes. set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if ((time_status & STA_UNSYNC) == 0 && - xtime.tv_sec > last_rtc_update + 660 && - xtime.tv_nsec / 1000 >= 500000 - ((unsigned) TICK_SIZE) / 2 && - xtime.tv_nsec / 1000 <= 500000 + ((unsigned) TICK_SIZE) / 2 - ) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - /* do it again in 60s */ - last_rtc_update = xtime.tv_sec - 600; - } + return set_rtc_mms(now.tv_sec); } /* diff --git a/arch/mn10300/kernel/time.c b/arch/mn10300/kernel/time.c index 395caf01b909..8f7f6d22783d 100644 --- a/arch/mn10300/kernel/time.c +++ b/arch/mn10300/kernel/time.c @@ -111,7 +111,6 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id) /* advance the kernel's time tracking system */ profile_tick(CPU_PROFILING); do_timer(1); - check_rtc_time(); } write_sequnlock(&xtime_lock); @@ -139,9 +138,6 @@ void __init time_init(void) " (calibrated against RTC)\n", MN10300_TSCCLK / 1000000, (MN10300_TSCCLK / 10000) % 100); - xtime.tv_sec = get_initial_rtc_time(); - xtime.tv_nsec = 0; - mn10300_last_tsc = TMTSCBC; /* use timer 0 & 1 cascaded to tick at as close to HZ as possible */ diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 6db513674050..d6781ce687e2 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -37,6 +37,9 @@ config SPARC64 def_bool 64BIT select ARCH_SUPPORTS_MSI select HAVE_FUNCTION_TRACER + select HAVE_FUNCTION_GRAPH_TRACER + select HAVE_FUNCTION_GRAPH_FP_TEST + select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_KRETPROBES select HAVE_KPROBES select HAVE_LMB @@ -72,7 +75,7 @@ config ARCH_USES_GETTIMEOFFSET config GENERIC_CMOS_UPDATE bool - default y if SPARC64 + default y config GENERIC_CLOCKEVENTS bool diff --git a/arch/sparc/Kconfig.debug b/arch/sparc/Kconfig.debug index 9d3c889718ac..1b4a831565f9 100644 --- a/arch/sparc/Kconfig.debug +++ b/arch/sparc/Kconfig.debug @@ -19,13 +19,10 @@ config DEBUG_DCFLUSH bool "D-cache flush debugging" depends on SPARC64 && DEBUG_KERNEL -config STACK_DEBUG - bool "Stack Overflow Detection Support" - config MCOUNT bool depends on SPARC64 - depends on STACK_DEBUG || FUNCTION_TRACER + depends on FUNCTION_TRACER default y config FRAME_POINTER diff --git a/arch/sparc/include/asm/cpudata_64.h b/arch/sparc/include/asm/cpudata_64.h index 926397d345ff..050ef35b9dcf 100644 --- a/arch/sparc/include/asm/cpudata_64.h +++ b/arch/sparc/include/asm/cpudata_64.h @@ -17,7 +17,7 @@ typedef struct { unsigned int __nmi_count; unsigned long clock_tick; /* %tick's per second */ unsigned long __pad; - unsigned int __pad1; + unsigned int irq0_irqs; unsigned int __pad2; /* Dcache line 2, rarely used */ diff --git a/arch/sparc/include/asm/irqflags_64.h b/arch/sparc/include/asm/irqflags_64.h index 8b49bf920df3..bfa1ea45b4cd 100644 --- a/arch/sparc/include/asm/irqflags_64.h +++ b/arch/sparc/include/asm/irqflags_64.h @@ -76,9 +76,26 @@ static inline int raw_irqs_disabled(void) */ static inline unsigned long __raw_local_irq_save(void) { - unsigned long flags = __raw_local_save_flags(); - - raw_local_irq_disable(); + unsigned long flags, tmp; + + /* Disable interrupts to PIL_NORMAL_MAX unless we already + * are using PIL_NMI, in which case PIL_NMI is retained. + * + * The only values we ever program into the %pil are 0, + * PIL_NORMAL_MAX and PIL_NMI. + * + * Since PIL_NMI is the largest %pil value and all bits are + * set in it (0xf), it doesn't matter what PIL_NORMAL_MAX + * actually is. + */ + __asm__ __volatile__( + "rdpr %%pil, %0\n\t" + "or %0, %2, %1\n\t" + "wrpr %1, 0x0, %%pil" + : "=r" (flags), "=r" (tmp) + : "i" (PIL_NORMAL_MAX) + : "memory" + ); return flags; } diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile index c6316142db4e..0c2dc1f24a9a 100644 --- a/arch/sparc/kernel/Makefile +++ b/arch/sparc/kernel/Makefile @@ -13,6 +13,14 @@ extra-y += init_task.o CPPFLAGS_vmlinux.lds := -Usparc -m$(BITS) extra-y += vmlinux.lds +ifdef CONFIG_FUNCTION_TRACER +# Do not profile debug and lowlevel utilities +CFLAGS_REMOVE_ftrace.o := -pg +CFLAGS_REMOVE_time_$(BITS).o := -pg +CFLAGS_REMOVE_perf_event.o := -pg +CFLAGS_REMOVE_pcr.o := -pg +endif + obj-$(CONFIG_SPARC32) += entry.o wof.o wuf.o obj-$(CONFIG_SPARC32) += etrap_32.o obj-$(CONFIG_SPARC32) += rtrap_32.o @@ -85,7 +93,7 @@ obj-$(CONFIG_KGDB) += kgdb_$(BITS).o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o -CFLAGS_REMOVE_ftrace.o := -pg +obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_EARLYFB) += btext.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/arch/sparc/kernel/ftrace.c b/arch/sparc/kernel/ftrace.c index 9103a56b39e8..03ab022e51c5 100644 --- a/arch/sparc/kernel/ftrace.c +++ b/arch/sparc/kernel/ftrace.c @@ -13,7 +13,7 @@ static const u32 ftrace_nop = 0x01000000; static u32 ftrace_call_replace(unsigned long ip, unsigned long addr) { - static u32 call; + u32 call; s32 off; off = ((s32)addr - (s32)ip); @@ -91,3 +91,61 @@ int __init ftrace_dyn_arch_init(void *data) return 0; } #endif + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + +#ifdef CONFIG_DYNAMIC_FTRACE +extern void ftrace_graph_call(void); + +int ftrace_enable_ftrace_graph_caller(void) +{ + unsigned long ip = (unsigned long)(&ftrace_graph_call); + u32 old, new; + + old = *(u32 *) &ftrace_graph_call; + new = ftrace_call_replace(ip, (unsigned long) &ftrace_graph_caller); + return ftrace_modify_code(ip, old, new); +} + +int ftrace_disable_ftrace_graph_caller(void) +{ + unsigned long ip = (unsigned long)(&ftrace_graph_call); + u32 old, new; + + old = *(u32 *) &ftrace_graph_call; + new = ftrace_call_replace(ip, (unsigned long) &ftrace_stub); + + return ftrace_modify_code(ip, old, new); +} + +#endif /* !CONFIG_DYNAMIC_FTRACE */ + +/* + * Hook the return address and push it in the stack of return addrs + * in current thread info. + */ +unsigned long prepare_ftrace_return(unsigned long parent, + unsigned long self_addr, + unsigned long frame_pointer) +{ + unsigned long return_hooker = (unsigned long) &return_to_handler; + struct ftrace_graph_ent trace; + + if (unlikely(atomic_read(¤t->tracing_graph_pause))) + return parent + 8UL; + + if (ftrace_push_return_trace(parent, self_addr, &trace.depth, + frame_pointer) == -EBUSY) + return parent + 8UL; + + trace.func = self_addr; + + /* Only trace if the calling function expects to */ + if (!ftrace_graph_entry(&trace)) { + current->curr_ret_stack--; + return parent + 8UL; + } + + return return_hooker; +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index e1cbdb94d97b..454ce3a25273 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -20,6 +20,7 @@ #include <linux/delay.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/ftrace.h> #include <linux/irq.h> #include <asm/ptrace.h> @@ -647,6 +648,14 @@ unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino) bucket = kzalloc(sizeof(struct ino_bucket), GFP_ATOMIC); if (unlikely(!bucket)) return 0; + + /* The only reference we store to the IRQ bucket is + * by physical address which kmemleak can't see, tell + * it that this object explicitly is not a leak and + * should be scanned. + */ + kmemleak_not_leak(bucket); + __flush_dcache_range((unsigned long) bucket, ((unsigned long) bucket + sizeof(struct ino_bucket))); @@ -721,7 +730,7 @@ static __attribute__((always_inline)) void restore_hardirq_stack(void *orig_sp) __asm__ __volatile__("mov %0, %%sp" : : "r" (orig_sp)); } -void handler_irq(int irq, struct pt_regs *regs) +void __irq_entry handler_irq(int irq, struct pt_regs *regs) { unsigned long pstate, bucket_pa; struct pt_regs *old_regs; diff --git a/arch/sparc/kernel/kgdb_64.c b/arch/sparc/kernel/kgdb_64.c index f5a0fd490b59..0a2bd0f99fc1 100644 --- a/arch/sparc/kernel/kgdb_64.c +++ b/arch/sparc/kernel/kgdb_64.c @@ -5,6 +5,7 @@ #include <linux/kgdb.h> #include <linux/kdebug.h> +#include <linux/ftrace.h> #include <asm/kdebug.h> #include <asm/ptrace.h> @@ -108,7 +109,7 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) } #ifdef CONFIG_SMP -void smp_kgdb_capture_client(int irq, struct pt_regs *regs) +void __irq_entry smp_kgdb_capture_client(int irq, struct pt_regs *regs) { unsigned long flags; diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index b287b62c7ea3..75a3d1a25356 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -92,7 +92,6 @@ static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) { unsigned int sum, touched = 0; - int cpu = smp_processor_id(); clear_softint(1 << irq); @@ -106,7 +105,7 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) else pcr_ops->write(PCR_PIC_PRIV); - sum = kstat_irqs_cpu(0, cpu); + sum = local_cpu_data().irq0_irqs; if (__get_cpu_var(nmi_touch)) { __get_cpu_var(nmi_touch) = 0; touched = 1; diff --git a/arch/sparc/kernel/pci_common.c b/arch/sparc/kernel/pci_common.c index b775658a927d..8a000583b5cf 100644 --- a/arch/sparc/kernel/pci_common.c +++ b/arch/sparc/kernel/pci_common.c @@ -371,14 +371,19 @@ static void pci_register_iommu_region(struct pci_pbm_info *pbm) struct resource *rp = kzalloc(sizeof(*rp), GFP_KERNEL); if (!rp) { - prom_printf("Cannot allocate IOMMU resource.\n"); - prom_halt(); + pr_info("%s: Cannot allocate IOMMU resource.\n", + pbm->name); + return; } rp->name = "IOMMU"; rp->start = pbm->mem_space.start + (unsigned long) vdma[0]; rp->end = rp->start + (unsigned long) vdma[1] - 1UL; rp->flags = IORESOURCE_BUSY; - request_resource(&pbm->mem_space, rp); + if (request_resource(&pbm->mem_space, rp)) { + pr_info("%s: Unable to request IOMMU resource.\n", + pbm->name); + kfree(rp); + } } } diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index 2d94e7a03af5..c4a6a50b4849 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -8,6 +8,7 @@ #include <linux/irq.h> #include <linux/perf_event.h> +#include <linux/ftrace.h> #include <asm/pil.h> #include <asm/pcr.h> @@ -34,7 +35,7 @@ unsigned int picl_shift; * Therefore in such situations we defer the work by signalling * a lower level cpu IRQ. */ -void deferred_pcr_work_irq(int irq, struct pt_regs *regs) +void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs) { struct pt_regs *old_regs; diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index 4c5334528109..b6a2b8f47040 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -22,6 +22,7 @@ #include <linux/profile.h> #include <linux/bootmem.h> #include <linux/vmalloc.h> +#include <linux/ftrace.h> #include <linux/cpu.h> #include <linux/slab.h> @@ -823,13 +824,13 @@ void arch_send_call_function_single_ipi(int cpu) &cpumask_of_cpu(cpu)); } -void smp_call_function_client(int irq, struct pt_regs *regs) +void __irq_entry smp_call_function_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); generic_smp_call_function_interrupt(); } -void smp_call_function_single_client(int irq, struct pt_regs *regs) +void __irq_entry smp_call_function_single_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); generic_smp_call_function_single_interrupt(); @@ -965,7 +966,7 @@ void flush_dcache_page_all(struct mm_struct *mm, struct page *page) put_cpu(); } -void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) +void __irq_entry smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) { struct mm_struct *mm; unsigned long flags; @@ -1149,7 +1150,7 @@ void smp_release(void) */ extern void prom_world(int); -void smp_penguin_jailcell(int irq, struct pt_regs *regs) +void __irq_entry smp_penguin_jailcell(int irq, struct pt_regs *regs) { clear_softint(1 << irq); @@ -1365,7 +1366,7 @@ void smp_send_reschedule(int cpu) &cpumask_of_cpu(cpu)); } -void smp_receive_signal_client(int irq, struct pt_regs *regs) +void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs) { clear_softint(1 << irq); } diff --git a/arch/sparc/kernel/time_32.c b/arch/sparc/kernel/time_32.c index 0d4c09b15efc..4453003032b5 100644 --- a/arch/sparc/kernel/time_32.c +++ b/arch/sparc/kernel/time_32.c @@ -78,6 +78,11 @@ __volatile__ unsigned int *master_l10_counter; u32 (*do_arch_gettimeoffset)(void); +int update_persistent_clock(struct timespec now) +{ + return set_rtc_mmss(now.tv_sec); +} + /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -87,9 +92,6 @@ u32 (*do_arch_gettimeoffset)(void); static irqreturn_t timer_interrupt(int dummy, void *dev_id) { - /* last time the cmos clock got updated */ - static long last_rtc_update; - #ifndef CONFIG_SMP profile_tick(CPU_PROFILING); #endif @@ -101,16 +103,6 @@ static irqreturn_t timer_interrupt(int dummy, void *dev_id) do_timer(1); - /* Determine when to update the Mostek clock. */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ - } write_sequnlock(&xtime_lock); #ifndef CONFIG_SMP diff --git a/arch/sparc/kernel/time_64.c b/arch/sparc/kernel/time_64.c index 67e165102885..c7bbe6cf7b85 100644 --- a/arch/sparc/kernel/time_64.c +++ b/arch/sparc/kernel/time_64.c @@ -35,6 +35,7 @@ #include <linux/clocksource.h> #include <linux/of_device.h> #include <linux/platform_device.h> +#include <linux/ftrace.h> #include <asm/oplib.h> #include <asm/timer.h> @@ -717,7 +718,7 @@ static struct clock_event_device sparc64_clockevent = { }; static DEFINE_PER_CPU(struct clock_event_device, sparc64_events); -void timer_interrupt(int irq, struct pt_regs *regs) +void __irq_entry timer_interrupt(int irq, struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); unsigned long tick_mask = tick_ops->softint_mask; @@ -728,6 +729,7 @@ void timer_interrupt(int irq, struct pt_regs *regs) irq_enter(); + local_cpu_data().irq0_irqs++; kstat_incr_irqs_this_cpu(0, irq_to_desc(0)); if (unlikely(!evt->event_handler)) { diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 837dfc2390d6..9da57f032983 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -2203,27 +2203,6 @@ void dump_stack(void) EXPORT_SYMBOL(dump_stack); -static inline int is_kernel_stack(struct task_struct *task, - struct reg_window *rw) -{ - unsigned long rw_addr = (unsigned long) rw; - unsigned long thread_base, thread_end; - - if (rw_addr < PAGE_OFFSET) { - if (task != &init_task) - return 0; - } - - thread_base = (unsigned long) task_stack_page(task); - thread_end = thread_base + sizeof(union thread_union); - if (rw_addr >= thread_base && - rw_addr < thread_end && - !(rw_addr & 0x7UL)) - return 1; - - return 0; -} - static inline struct reg_window *kernel_stack_up(struct reg_window *rw) { unsigned long fp = rw->ins[6]; @@ -2252,6 +2231,7 @@ void die_if_kernel(char *str, struct pt_regs *regs) show_regs(regs); add_taint(TAINT_DIE); if (regs->tstate & TSTATE_PRIV) { + struct thread_info *tp = current_thread_info(); struct reg_window *rw = (struct reg_window *) (regs->u_regs[UREG_FP] + STACK_BIAS); @@ -2259,8 +2239,8 @@ void die_if_kernel(char *str, struct pt_regs *regs) * find some badly aligned kernel stack. */ while (rw && - count++ < 30&& - is_kernel_stack(current, rw)) { + count++ < 30 && + kstack_valid(tp, (unsigned long) rw)) { printk("Caller[%016lx]: %pS\n", rw->ins[7], (void *) rw->ins[7]); diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 4e5992593967..0c1e6783657f 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -46,11 +46,16 @@ SECTIONS SCHED_TEXT LOCK_TEXT KPROBES_TEXT + IRQENTRY_TEXT *(.gnu.warning) } = 0 _etext = .; RO_DATA(PAGE_SIZE) + + /* Start of data section */ + _sdata = .; + .data1 : { *(.data1) } diff --git a/arch/sparc/lib/mcount.S b/arch/sparc/lib/mcount.S index 24b8b12deed2..3753e3c6e176 100644 --- a/arch/sparc/lib/mcount.S +++ b/arch/sparc/lib/mcount.S @@ -7,26 +7,11 @@ #include <linux/linkage.h> -#include <asm/ptrace.h> -#include <asm/thread_info.h> - /* * This is the main variant and is called by C code. GCC's -pg option * automatically instruments every C function with a call to this. */ -#ifdef CONFIG_STACK_DEBUG - -#define OVSTACKSIZE 4096 /* lets hope this is enough */ - - .data - .align 8 -panicstring: - .asciz "Stack overflow\n" - .align 8 -ovstack: - .skip OVSTACKSIZE -#endif .text .align 32 .globl _mcount @@ -35,84 +20,48 @@ ovstack: .type mcount,#function _mcount: mcount: -#ifdef CONFIG_STACK_DEBUG - /* - * Check whether %sp is dangerously low. - */ - ldub [%g6 + TI_FPDEPTH], %g1 - srl %g1, 1, %g3 - add %g3, 1, %g3 - sllx %g3, 8, %g3 ! each fpregs frame is 256b - add %g3, 192, %g3 - add %g6, %g3, %g3 ! where does task_struct+frame end? - sub %g3, STACK_BIAS, %g3 - cmp %sp, %g3 - bg,pt %xcc, 1f - nop - lduh [%g6 + TI_CPU], %g1 - sethi %hi(hardirq_stack), %g3 - or %g3, %lo(hardirq_stack), %g3 - sllx %g1, 3, %g1 - ldx [%g3 + %g1], %g7 - sub %g7, STACK_BIAS, %g7 - cmp %sp, %g7 - bleu,pt %xcc, 2f - sethi %hi(THREAD_SIZE), %g3 - add %g7, %g3, %g7 - cmp %sp, %g7 - blu,pn %xcc, 1f -2: sethi %hi(softirq_stack), %g3 - or %g3, %lo(softirq_stack), %g3 - ldx [%g3 + %g1], %g7 - sub %g7, STACK_BIAS, %g7 - cmp %sp, %g7 - bleu,pt %xcc, 3f - sethi %hi(THREAD_SIZE), %g3 - add %g7, %g3, %g7 - cmp %sp, %g7 - blu,pn %xcc, 1f - nop - /* If we are already on ovstack, don't hop onto it - * again, we are already trying to output the stack overflow - * message. - */ -3: sethi %hi(ovstack), %g7 ! cant move to panic stack fast enough - or %g7, %lo(ovstack), %g7 - add %g7, OVSTACKSIZE, %g3 - sub %g3, STACK_BIAS + 192, %g3 - sub %g7, STACK_BIAS, %g7 - cmp %sp, %g7 - blu,pn %xcc, 2f - cmp %sp, %g3 - bleu,pn %xcc, 1f - nop -2: mov %g3, %sp - sethi %hi(panicstring), %g3 - call prom_printf - or %g3, %lo(panicstring), %o0 - call prom_halt - nop -1: -#endif #ifdef CONFIG_FUNCTION_TRACER #ifdef CONFIG_DYNAMIC_FTRACE - mov %o7, %o0 - .globl mcount_call -mcount_call: - call ftrace_stub - mov %o0, %o7 + /* Do nothing, the retl/nop below is all we need. */ #else - sethi %hi(ftrace_trace_function), %g1 + sethi %hi(function_trace_stop), %g1 + lduw [%g1 + %lo(function_trace_stop)], %g2 + brnz,pn %g2, 2f + sethi %hi(ftrace_trace_function), %g1 sethi %hi(ftrace_stub), %g2 ldx [%g1 + %lo(ftrace_trace_function)], %g1 or %g2, %lo(ftrace_stub), %g2 cmp %g1, %g2 be,pn %icc, 1f - mov %i7, %o1 - jmpl %g1, %g0 - mov %o7, %o0 + mov %i7, %g3 + save %sp, -128, %sp + mov %g3, %o1 + jmpl %g1, %o7 + mov %i7, %o0 + ret + restore /* not reached */ 1: +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + sethi %hi(ftrace_graph_return), %g1 + ldx [%g1 + %lo(ftrace_graph_return)], %g3 + cmp %g2, %g3 + bne,pn %xcc, 5f + sethi %hi(ftrace_graph_entry_stub), %g2 + sethi %hi(ftrace_graph_entry), %g1 + or %g2, %lo(ftrace_graph_entry_stub), %g2 + ldx [%g1 + %lo(ftrace_graph_entry)], %g1 + cmp %g1, %g2 + be,pt %xcc, 2f + nop +5: mov %i7, %g2 + mov %fp, %g3 + save %sp, -128, %sp + mov %g2, %l0 + ba,pt %xcc, ftrace_graph_caller + mov %g3, %l1 +#endif +2: #endif #endif retl @@ -131,14 +80,50 @@ ftrace_stub: .globl ftrace_caller .type ftrace_caller,#function ftrace_caller: - mov %i7, %o1 - mov %o7, %o0 + sethi %hi(function_trace_stop), %g1 + mov %i7, %g2 + lduw [%g1 + %lo(function_trace_stop)], %g1 + brnz,pn %g1, ftrace_stub + mov %fp, %g3 + save %sp, -128, %sp + mov %g2, %o1 + mov %g2, %l0 + mov %g3, %l1 .globl ftrace_call ftrace_call: call ftrace_stub - mov %o0, %o7 - retl + mov %i7, %o0 +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + .globl ftrace_graph_call +ftrace_graph_call: + call ftrace_stub nop +#endif + ret + restore +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + .size ftrace_graph_call,.-ftrace_graph_call +#endif + .size ftrace_call,.-ftrace_call .size ftrace_caller,.-ftrace_caller #endif #endif + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +ENTRY(ftrace_graph_caller) + mov %l0, %o0 + mov %i7, %o1 + call prepare_ftrace_return + mov %l1, %o2 + ret + restore %o0, -8, %i7 +END(ftrace_graph_caller) + +ENTRY(return_to_handler) + save %sp, -128, %sp + call ftrace_return_to_handler + mov %fp, %o0 + jmpl %o0 + 8, %g0 + restore +END(return_to_handler) +#endif diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9458685902bd..d3b7bb34ea01 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -54,10 +54,14 @@ config X86 select HAVE_KERNEL_LZO select HAVE_HW_BREAKPOINT select PERF_EVENTS + select PERF_EVENTS_NMI select ANON_INODES select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER +config INSTRUCTION_DECODER + def_bool (KPROBES || PERF_EVENTS) + config OUTPUT_FORMAT string default "elf32-i386" if X86_32 diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index a19829374e6a..918fbb1855cc 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -502,23 +502,3 @@ config CPU_SUP_UMC_32 CPU might render the kernel unbootable. If unsure, say N. - -config X86_DS - def_bool X86_PTRACE_BTS - depends on X86_DEBUGCTLMSR - select HAVE_HW_BRANCH_TRACER - -config X86_PTRACE_BTS - bool "Branch Trace Store" - default y - depends on X86_DEBUGCTLMSR - depends on BROKEN - ---help--- - This adds a ptrace interface to the hardware's branch trace store. - - Debuggers may use it to collect an execution trace of the debugged - application in order to answer the question 'how did I get here?'. - Debuggers may trace user mode as well as kernel mode. - - Say Y unless there is no application development on this machine - and you want to save a small amount of code size. diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index bc01e3ebfeb2..bd58c8abbfbd 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -174,15 +174,6 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. -config X86_DS_SELFTEST - bool "DS selftest" - default y - depends on DEBUG_KERNEL - depends on X86_DS - ---help--- - Perform Debug Store selftests at boot time. - If in doubt, say "N". - config HAVE_MMIOTRACE_SUPPORT def_bool y diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h index ba19ad4c47d0..86a0ff0aeac7 100644 --- a/arch/x86/include/asm/amd_iommu_types.h +++ b/arch/x86/include/asm/amd_iommu_types.h @@ -21,6 +21,7 @@ #define _ASM_X86_AMD_IOMMU_TYPES_H #include <linux/types.h> +#include <linux/mutex.h> #include <linux/list.h> #include <linux/spinlock.h> @@ -140,6 +141,7 @@ /* constants to configure the command buffer */ #define CMD_BUFFER_SIZE 8192 +#define CMD_BUFFER_UNINITIALIZED 1 #define CMD_BUFFER_ENTRIES 512 #define MMIO_CMD_SIZE_SHIFT 56 #define MMIO_CMD_SIZE_512 (0x9ULL << MMIO_CMD_SIZE_SHIFT) @@ -237,6 +239,7 @@ struct protection_domain { struct list_head list; /* for list of all protection domains */ struct list_head dev_list; /* List of all devices in this domain */ spinlock_t lock; /* mostly used to lock the page table*/ + struct mutex api_lock; /* protect page tables in the iommu-api path */ u16 id; /* the domain id written to the device table */ int mode; /* paging mode (0-6 levels) */ u64 *pt_root; /* page table root pointer */ diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index b4ac2cdcb64f..1fa03e04ae44 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -373,6 +373,7 @@ extern atomic_t init_deasserted; extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip); #endif +#ifdef CONFIG_X86_LOCAL_APIC static inline u32 apic_read(u32 reg) { return apic->read(reg); @@ -403,10 +404,19 @@ static inline u32 safe_apic_wait_icr_idle(void) return apic->safe_wait_icr_idle(); } +#else /* CONFIG_X86_LOCAL_APIC */ + +static inline u32 apic_read(u32 reg) { return 0; } +static inline void apic_write(u32 reg, u32 val) { } +static inline u64 apic_icr_read(void) { return 0; } +static inline void apic_icr_write(u32 low, u32 high) { } +static inline void apic_wait_icr_idle(void) { } +static inline u32 safe_apic_wait_icr_idle(void) { return 0; } + +#endif /* CONFIG_X86_LOCAL_APIC */ static inline void ack_APIC_irq(void) { -#ifdef CONFIG_X86_LOCAL_APIC /* * ack_APIC_irq() actually gets compiled as a single instruction * ... yummie. @@ -414,7 +424,6 @@ static inline void ack_APIC_irq(void) /* Docs say use 0 for future compatibility */ apic_write(APIC_EOI, 0); -#endif } static inline unsigned default_get_apic_id(unsigned long x) diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 0cd82d068613..630e623f61e0 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -161,6 +161,7 @@ */ #define X86_FEATURE_IDA (7*32+ 0) /* Intel Dynamic Acceleration */ #define X86_FEATURE_ARAT (7*32+ 1) /* Always Running APIC Timer */ +#define X86_FEATURE_CPB (7*32+ 2) /* AMD Core Performance Boost */ /* Virtualization flags: Linux defined */ #define X86_FEATURE_TPR_SHADOW (8*32+ 0) /* Intel TPR Shadow */ diff --git a/arch/x86/include/asm/ds.h b/arch/x86/include/asm/ds.h deleted file mode 100644 index 70dac199b093..000000000000 --- a/arch/x86/include/asm/ds.h +++ /dev/null @@ -1,302 +0,0 @@ -/* - * Debug Store (DS) support - * - * This provides a low-level interface to the hardware's Debug Store - * feature that is used for branch trace store (BTS) and - * precise-event based sampling (PEBS). - * - * It manages: - * - DS and BTS hardware configuration - * - buffer overflow handling (to be done) - * - buffer access - * - * It does not do: - * - security checking (is the caller allowed to trace the task) - * - buffer allocation (memory accounting) - * - * - * Copyright (C) 2007-2009 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, 2007-2009 - */ - -#ifndef _ASM_X86_DS_H -#define _ASM_X86_DS_H - - -#include <linux/types.h> -#include <linux/init.h> -#include <linux/err.h> - - -#ifdef CONFIG_X86_DS - -struct task_struct; -struct ds_context; -struct ds_tracer; -struct bts_tracer; -struct pebs_tracer; - -typedef void (*bts_ovfl_callback_t)(struct bts_tracer *); -typedef void (*pebs_ovfl_callback_t)(struct pebs_tracer *); - - -/* - * A list of features plus corresponding macros to talk about them in - * the ds_request function's flags parameter. - * - * We use the enum to index an array of corresponding control bits; - * we use the macro to index a flags bit-vector. - */ -enum ds_feature { - dsf_bts = 0, - dsf_bts_kernel, -#define BTS_KERNEL (1 << dsf_bts_kernel) - /* trace kernel-mode branches */ - - dsf_bts_user, -#define BTS_USER (1 << dsf_bts_user) - /* trace user-mode branches */ - - dsf_bts_overflow, - dsf_bts_max, - dsf_pebs = dsf_bts_max, - - dsf_pebs_max, - dsf_ctl_max = dsf_pebs_max, - dsf_bts_timestamps = dsf_ctl_max, -#define BTS_TIMESTAMPS (1 << dsf_bts_timestamps) - /* add timestamps into BTS trace */ - -#define BTS_USER_FLAGS (BTS_KERNEL | BTS_USER | BTS_TIMESTAMPS) -}; - - -/* - * Request BTS or PEBS - * - * Due to alignement constraints, the actual buffer may be slightly - * smaller than the requested or provided buffer. - * - * Returns a pointer to a tracer structure on success, or - * ERR_PTR(errcode) on failure. - * - * The interrupt threshold is independent from the overflow callback - * to allow users to use their own overflow interrupt handling mechanism. - * - * The function might sleep. - * - * task: the task to request recording for - * cpu: the cpu to request recording for - * base: the base pointer for the (non-pageable) buffer; - * size: the size of the provided buffer in bytes - * ovfl: pointer to a function to be called on buffer overflow; - * NULL if cyclic buffer requested - * th: the interrupt threshold in records from the end of the buffer; - * -1 if no interrupt threshold is requested. - * flags: a bit-mask of the above flags - */ -extern struct bts_tracer *ds_request_bts_task(struct task_struct *task, - void *base, size_t size, - bts_ovfl_callback_t ovfl, - size_t th, unsigned int flags); -extern struct bts_tracer *ds_request_bts_cpu(int cpu, void *base, size_t size, - bts_ovfl_callback_t ovfl, - size_t th, unsigned int flags); -extern struct pebs_tracer *ds_request_pebs_task(struct task_struct *task, - void *base, size_t size, - pebs_ovfl_callback_t ovfl, - size_t th, unsigned int flags); -extern struct pebs_tracer *ds_request_pebs_cpu(int cpu, - void *base, size_t size, - pebs_ovfl_callback_t ovfl, - size_t th, unsigned int flags); - -/* - * Release BTS or PEBS resources - * Suspend and resume BTS or PEBS tracing - * - * Must be called with irq's enabled. - * - * tracer: the tracer handle returned from ds_request_~() - */ -extern void ds_release_bts(struct bts_tracer *tracer); -extern void ds_suspend_bts(struct bts_tracer *tracer); -extern void ds_resume_bts(struct bts_tracer *tracer); -extern void ds_release_pebs(struct pebs_tracer *tracer); -extern void ds_suspend_pebs(struct pebs_tracer *tracer); -extern void ds_resume_pebs(struct pebs_tracer *tracer); - -/* - * Release BTS or PEBS resources - * Suspend and resume BTS or PEBS tracing - * - * Cpu tracers must call this on the traced cpu. - * Task tracers must call ds_release_~_noirq() for themselves. - * - * May be called with irq's disabled. - * - * Returns 0 if successful; - * -EPERM if the cpu tracer does not trace the current cpu. - * -EPERM if the task tracer does not trace itself. - * - * tracer: the tracer handle returned from ds_request_~() - */ -extern int ds_release_bts_noirq(struct bts_tracer *tracer); -extern int ds_suspend_bts_noirq(struct bts_tracer *tracer); -extern int ds_resume_bts_noirq(struct bts_tracer *tracer); -extern int ds_release_pebs_noirq(struct pebs_tracer *tracer); -extern int ds_suspend_pebs_noirq(struct pebs_tracer *tracer); -extern int ds_resume_pebs_noirq(struct pebs_tracer *tracer); - - -/* - * The raw DS buffer state as it is used for BTS and PEBS recording. - * - * This is the low-level, arch-dependent interface for working - * directly on the raw trace data. - */ -struct ds_trace { - /* the number of bts/pebs records */ - size_t n; - /* the size of a bts/pebs record in bytes */ - size_t size; - /* pointers into the raw buffer: - - to the first entry */ - void *begin; - /* - one beyond the last entry */ - void *end; - /* - one beyond the newest entry */ - void *top; - /* - the interrupt threshold */ - void *ith; - /* flags given on ds_request() */ - unsigned int flags; -}; - -/* - * An arch-independent view on branch trace data. - */ -enum bts_qualifier { - bts_invalid, -#define BTS_INVALID bts_invalid - - bts_branch, -#define BTS_BRANCH bts_branch - - bts_task_arrives, -#define BTS_TASK_ARRIVES bts_task_arrives - - bts_task_departs, -#define BTS_TASK_DEPARTS bts_task_departs - - bts_qual_bit_size = 4, - bts_qual_max = (1 << bts_qual_bit_size), -}; - -struct bts_struct { - __u64 qualifier; - union { - /* BTS_BRANCH */ - struct { - __u64 from; - __u64 to; - } lbr; - /* BTS_TASK_ARRIVES or BTS_TASK_DEPARTS */ - struct { - __u64 clock; - pid_t pid; - } event; - } variant; -}; - - -/* - * The BTS state. - * - * This gives access to the raw DS state and adds functions to provide - * an arch-independent view of the BTS data. - */ -struct bts_trace { - struct ds_trace ds; - - int (*read)(struct bts_tracer *tracer, const void *at, - struct bts_struct *out); - int (*write)(struct bts_tracer *tracer, const struct bts_struct *in); -}; - - -/* - * The PEBS state. - * - * This gives access to the raw DS state and the PEBS-specific counter - * reset value. - */ -struct pebs_trace { - struct ds_trace ds; - - /* the number of valid counters in the below array */ - unsigned int counters; - -#define MAX_PEBS_COUNTERS 4 - /* the counter reset value */ - unsigned long long counter_reset[MAX_PEBS_COUNTERS]; -}; - - -/* - * Read the BTS or PEBS trace. - * - * Returns a view on the trace collected for the parameter tracer. - * - * The view remains valid as long as the traced task is not running or - * the tracer is suspended. - * Writes into the trace buffer are not reflected. - * - * tracer: the tracer handle returned from ds_request_~() - */ -extern const struct bts_trace *ds_read_bts(struct bts_tracer *tracer); -extern const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer); - - -/* - * Reset the write pointer of the BTS/PEBS buffer. - * - * Returns 0 on success; -Eerrno on error - * - * tracer: the tracer handle returned from ds_request_~() - */ -extern int ds_reset_bts(struct bts_tracer *tracer); -extern int ds_reset_pebs(struct pebs_tracer *tracer); - -/* - * Set the PEBS counter reset value. - * - * Returns 0 on success; -Eerrno on error - * - * tracer: the tracer handle returned from ds_request_pebs() - * counter: the index of the counter - * value: the new counter reset value - */ -extern int ds_set_pebs_reset(struct pebs_tracer *tracer, - unsigned int counter, u64 value); - -/* - * Initialization - */ -struct cpuinfo_x86; -extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *); - -/* - * Context switch work - */ -extern void ds_switch_to(struct task_struct *prev, struct task_struct *next); - -#else /* CONFIG_X86_DS */ - -struct cpuinfo_x86; -static inline void __cpuinit ds_init_intel(struct cpuinfo_x86 *ignored) {} -static inline void ds_switch_to(struct task_struct *prev, - struct task_struct *next) {} - -#endif /* CONFIG_X86_DS */ -#endif /* _ASM_X86_DS_H */ diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 0e22296790d3..ec8a52d14ab1 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -45,7 +45,12 @@ #define E820_NVS 4 #define E820_UNUSABLE 5 -/* reserved RAM used by kernel itself */ +/* + * reserved RAM used by kernel itself + * if CONFIG_INTEL_TXT is enabled, memory of this type will be + * included in the S3 integrity calculation and so should not include + * any memory that BIOS might alter over the S3 transition + */ #define E820_RESERVED_KERN 128 #ifndef __ASSEMBLY__ diff --git a/arch/x86/include/asm/i8253.h b/arch/x86/include/asm/i8253.h index 1edbf89680fd..fc1f579fb965 100644 --- a/arch/x86/include/asm/i8253.h +++ b/arch/x86/include/asm/i8253.h @@ -6,7 +6,7 @@ #define PIT_CH0 0x40 #define PIT_CH2 0x42 -extern spinlock_t i8253_lock; +extern raw_spinlock_t i8253_lock; extern struct clock_event_device *global_clock_event; diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h index 96c2e0ad04ca..88c765e16410 100644 --- a/arch/x86/include/asm/insn.h +++ b/arch/x86/include/asm/insn.h @@ -68,6 +68,8 @@ struct insn { const insn_byte_t *next_byte; }; +#define MAX_INSN_SIZE 16 + #define X86_MODRM_MOD(modrm) (((modrm) & 0xc0) >> 6) #define X86_MODRM_REG(modrm) (((modrm) & 0x38) >> 3) #define X86_MODRM_RM(modrm) ((modrm) & 0x07) diff --git a/arch/x86/include/asm/kprobes.h b/arch/x86/include/asm/kprobes.h index 4ffa345a8ccb..547882539157 100644 --- a/arch/x86/include/asm/kprobes.h +++ b/arch/x86/include/asm/kprobes.h @@ -24,6 +24,7 @@ #include <linux/types.h> #include <linux/ptrace.h> #include <linux/percpu.h> +#include <asm/insn.h> #define __ARCH_WANT_KPROBES_INSN_SLOT @@ -36,7 +37,6 @@ typedef u8 kprobe_opcode_t; #define RELATIVEJUMP_SIZE 5 #define RELATIVECALL_OPCODE 0xe8 #define RELATIVE_ADDR_SIZE 4 -#define MAX_INSN_SIZE 16 #define MAX_STACK_SIZE 64 #define MIN_STACK_SIZE(ADDR) \ (((MAX_STACK_SIZE) < (((unsigned long)current_thread_info()) + \ diff --git a/arch/x86/include/asm/lguest_hcall.h b/arch/x86/include/asm/lguest_hcall.h index ba0eed8aa1a6..b60f2924c413 100644 --- a/arch/x86/include/asm/lguest_hcall.h +++ b/arch/x86/include/asm/lguest_hcall.h @@ -28,22 +28,39 @@ #ifndef __ASSEMBLY__ #include <asm/hw_irq.h> -#include <asm/kvm_para.h> /*G:030 * But first, how does our Guest contact the Host to ask for privileged * operations? There are two ways: the direct way is to make a "hypercall", * to make requests of the Host Itself. * - * We use the KVM hypercall mechanism, though completely different hypercall - * numbers. Seventeen hypercalls are available: the hypercall number is put in - * the %eax register, and the arguments (when required) are placed in %ebx, - * %ecx, %edx and %esi. If a return value makes sense, it's returned in %eax. + * Our hypercall mechanism uses the highest unused trap code (traps 32 and + * above are used by real hardware interrupts). Seventeen hypercalls are + * available: the hypercall number is put in the %eax register, and the + * arguments (when required) are placed in %ebx, %ecx, %edx and %esi. + * If a return value makes sense, it's returned in %eax. * * Grossly invalid calls result in Sudden Death at the hands of the vengeful * Host, rather than returning failure. This reflects Winston Churchill's * definition of a gentleman: "someone who is only rude intentionally". -:*/ + */ +static inline unsigned long +hcall(unsigned long call, + unsigned long arg1, unsigned long arg2, unsigned long arg3, + unsigned long arg4) +{ + /* "int" is the Intel instruction to trigger a trap. */ + asm volatile("int $" __stringify(LGUEST_TRAP_ENTRY) + /* The call in %eax (aka "a") might be overwritten */ + : "=a"(call) + /* The arguments are in %eax, %ebx, %ecx, %edx & %esi */ + : "a"(call), "b"(arg1), "c"(arg2), "d"(arg3), "S"(arg4) + /* "memory" means this might write somewhere in memory. + * This isn't true for all calls, but it's safe to tell + * gcc that it might happen so it doesn't get clever. */ + : "memory"); + return call; +} /* Can't use our min() macro here: needs to be a constant */ #define LGUEST_IRQS (NR_IRQS < 32 ? NR_IRQS: 32) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 4604e6a54d36..bc473acfa7f9 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -71,11 +71,14 @@ #define MSR_IA32_LASTINTTOIP 0x000001de /* DEBUGCTLMSR bits (others vary by model): */ -#define _DEBUGCTLMSR_LBR 0 /* last branch recording */ -#define _DEBUGCTLMSR_BTF 1 /* single-step on branches */ - -#define DEBUGCTLMSR_LBR (1UL << _DEBUGCTLMSR_LBR) -#define DEBUGCTLMSR_BTF (1UL << _DEBUGCTLMSR_BTF) +#define DEBUGCTLMSR_LBR (1UL << 0) /* last branch recording */ +#define DEBUGCTLMSR_BTF (1UL << 1) /* single-step on branches */ +#define DEBUGCTLMSR_TR (1UL << 6) +#define DEBUGCTLMSR_BTS (1UL << 7) +#define DEBUGCTLMSR_BTINT (1UL << 8) +#define DEBUGCTLMSR_BTS_OFF_OS (1UL << 9) +#define DEBUGCTLMSR_BTS_OFF_USR (1UL << 10) +#define DEBUGCTLMSR_FREEZE_LBRS_ON_PMI (1UL << 11) #define MSR_IA32_MC0_CTL 0x00000400 #define MSR_IA32_MC0_STATUS 0x00000401 @@ -359,6 +362,8 @@ #define MSR_P4_U2L_ESCR0 0x000003b0 #define MSR_P4_U2L_ESCR1 0x000003b1 +#define MSR_P4_PEBS_MATRIX_VERT 0x000003f2 + /* Intel Core-based CPU performance counters */ #define MSR_CORE_PERF_FIXED_CTR0 0x00000309 #define MSR_CORE_PERF_FIXED_CTR1 0x0000030a diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index 93da9c3f3341..5b41b0feb6db 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -17,7 +17,9 @@ int do_nmi_callback(struct pt_regs *regs, int cpu); extern void die_nmi(char *str, struct pt_regs *regs, int do_panic); extern int check_nmi_watchdog(void); +#if !defined(CONFIG_NMI_WATCHDOG) extern int nmi_watchdog_enabled; +#endif extern int avail_to_resrv_perfctr_nmi_bit(unsigned int); extern int reserve_perfctr_nmi(unsigned int); extern void release_perfctr_nmi(unsigned int); diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index db6109a885a7..f6d43dbfd8e7 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -5,7 +5,7 @@ * Performance event hw details: */ -#define X86_PMC_MAX_GENERIC 8 +#define X86_PMC_MAX_GENERIC 32 #define X86_PMC_MAX_FIXED 3 #define X86_PMC_IDX_GENERIC 0 @@ -18,39 +18,31 @@ #define MSR_ARCH_PERFMON_EVENTSEL0 0x186 #define MSR_ARCH_PERFMON_EVENTSEL1 0x187 -#define ARCH_PERFMON_EVENTSEL_ENABLE (1 << 22) -#define ARCH_PERFMON_EVENTSEL_ANY (1 << 21) -#define ARCH_PERFMON_EVENTSEL_INT (1 << 20) -#define ARCH_PERFMON_EVENTSEL_OS (1 << 17) -#define ARCH_PERFMON_EVENTSEL_USR (1 << 16) - -/* - * Includes eventsel and unit mask as well: - */ - - -#define INTEL_ARCH_EVTSEL_MASK 0x000000FFULL -#define INTEL_ARCH_UNIT_MASK 0x0000FF00ULL -#define INTEL_ARCH_EDGE_MASK 0x00040000ULL -#define INTEL_ARCH_INV_MASK 0x00800000ULL -#define INTEL_ARCH_CNT_MASK 0xFF000000ULL -#define INTEL_ARCH_EVENT_MASK (INTEL_ARCH_UNIT_MASK|INTEL_ARCH_EVTSEL_MASK) - -/* - * filter mask to validate fixed counter events. - * the following filters disqualify for fixed counters: - * - inv - * - edge - * - cnt-mask - * The other filters are supported by fixed counters. - * The any-thread option is supported starting with v3. - */ -#define INTEL_ARCH_FIXED_MASK \ - (INTEL_ARCH_CNT_MASK| \ - INTEL_ARCH_INV_MASK| \ - INTEL_ARCH_EDGE_MASK|\ - INTEL_ARCH_UNIT_MASK|\ - INTEL_ARCH_EVENT_MASK) +#define ARCH_PERFMON_EVENTSEL_EVENT 0x000000FFULL +#define ARCH_PERFMON_EVENTSEL_UMASK 0x0000FF00ULL +#define ARCH_PERFMON_EVENTSEL_USR (1ULL << 16) +#define ARCH_PERFMON_EVENTSEL_OS (1ULL << 17) +#define ARCH_PERFMON_EVENTSEL_EDGE (1ULL << 18) +#define ARCH_PERFMON_EVENTSEL_INT (1ULL << 20) +#define ARCH_PERFMON_EVENTSEL_ANY (1ULL << 21) +#define ARCH_PERFMON_EVENTSEL_ENABLE (1ULL << 22) +#define ARCH_PERFMON_EVENTSEL_INV (1ULL << 23) +#define ARCH_PERFMON_EVENTSEL_CMASK 0xFF000000ULL + +#define AMD64_EVENTSEL_EVENT \ + (ARCH_PERFMON_EVENTSEL_EVENT | (0x0FULL << 32)) +#define INTEL_ARCH_EVENT_MASK \ + (ARCH_PERFMON_EVENTSEL_UMASK | ARCH_PERFMON_EVENTSEL_EVENT) + +#define X86_RAW_EVENT_MASK \ + (ARCH_PERFMON_EVENTSEL_EVENT | \ + ARCH_PERFMON_EVENTSEL_UMASK | \ + ARCH_PERFMON_EVENTSEL_EDGE | \ + ARCH_PERFMON_EVENTSEL_INV | \ + ARCH_PERFMON_EVENTSEL_CMASK) +#define AMD64_RAW_EVENT_MASK \ + (X86_RAW_EVENT_MASK | \ + AMD64_EVENTSEL_EVENT) #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL 0x3c #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK (0x00 << 8) @@ -67,7 +59,7 @@ union cpuid10_eax { struct { unsigned int version_id:8; - unsigned int num_events:8; + unsigned int num_counters:8; unsigned int bit_width:8; unsigned int mask_length:8; } split; @@ -76,7 +68,7 @@ union cpuid10_eax { union cpuid10_edx { struct { - unsigned int num_events_fixed:4; + unsigned int num_counters_fixed:4; unsigned int reserved:28; } split; unsigned int full; @@ -136,6 +128,25 @@ extern void perf_events_lapic_init(void); #define PERF_EVENT_INDEX_OFFSET 0 +/* + * Abuse bit 3 of the cpu eflags register to indicate proper PEBS IP fixups. + * This flag is otherwise unused and ABI specified to be 0, so nobody should + * care what we do with it. + */ +#define PERF_EFLAGS_EXACT (1UL << 3) + +#define perf_misc_flags(regs) \ +({ int misc = 0; \ + if (user_mode(regs)) \ + misc |= PERF_RECORD_MISC_USER; \ + else \ + misc |= PERF_RECORD_MISC_KERNEL; \ + if (regs->flags & PERF_EFLAGS_EXACT) \ + misc |= PERF_RECORD_MISC_EXACT; \ + misc; }) + +#define perf_instruction_pointer(regs) ((regs)->ip) + #else static inline void init_hw_perf_events(void) { } static inline void perf_events_lapic_init(void) { } diff --git a/arch/x86/include/asm/perf_event_p4.h b/arch/x86/include/asm/perf_event_p4.h new file mode 100644 index 000000000000..b05400a542ff --- /dev/null +++ b/arch/x86/include/asm/perf_event_p4.h @@ -0,0 +1,794 @@ +/* + * Netburst Perfomance Events (P4, old Xeon) + */ + +#ifndef PERF_EVENT_P4_H +#define PERF_EVENT_P4_H + +#include <linux/cpu.h> +#include <linux/bitops.h> + +/* + * NetBurst has perfomance MSRs shared between + * threads if HT is turned on, ie for both logical + * processors (mem: in turn in Atom with HT support + * perf-MSRs are not shared and every thread has its + * own perf-MSRs set) + */ +#define ARCH_P4_TOTAL_ESCR (46) +#define ARCH_P4_RESERVED_ESCR (2) /* IQ_ESCR(0,1) not always present */ +#define ARCH_P4_MAX_ESCR (ARCH_P4_TOTAL_ESCR - ARCH_P4_RESERVED_ESCR) +#define ARCH_P4_MAX_CCCR (18) +#define ARCH_P4_MAX_COUNTER (ARCH_P4_MAX_CCCR / 2) + +#define P4_ESCR_EVENT_MASK 0x7e000000U +#define P4_ESCR_EVENT_SHIFT 25 +#define P4_ESCR_EVENTMASK_MASK 0x01fffe00U +#define P4_ESCR_EVENTMASK_SHIFT 9 +#define P4_ESCR_TAG_MASK 0x000001e0U +#define P4_ESCR_TAG_SHIFT 5 +#define P4_ESCR_TAG_ENABLE 0x00000010U +#define P4_ESCR_T0_OS 0x00000008U +#define P4_ESCR_T0_USR 0x00000004U +#define P4_ESCR_T1_OS 0x00000002U +#define P4_ESCR_T1_USR 0x00000001U + +#define P4_ESCR_EVENT(v) ((v) << P4_ESCR_EVENT_SHIFT) +#define P4_ESCR_EMASK(v) ((v) << P4_ESCR_EVENTMASK_SHIFT) +#define P4_ESCR_TAG(v) ((v) << P4_ESCR_TAG_SHIFT) + +/* Non HT mask */ +#define P4_ESCR_MASK \ + (P4_ESCR_EVENT_MASK | \ + P4_ESCR_EVENTMASK_MASK | \ + P4_ESCR_TAG_MASK | \ + P4_ESCR_TAG_ENABLE | \ + P4_ESCR_T0_OS | \ + P4_ESCR_T0_USR) + +/* HT mask */ +#define P4_ESCR_MASK_HT \ + (P4_ESCR_MASK | P4_ESCR_T1_OS | P4_ESCR_T1_USR) + +#define P4_CCCR_OVF 0x80000000U +#define P4_CCCR_CASCADE 0x40000000U +#define P4_CCCR_OVF_PMI_T0 0x04000000U +#define P4_CCCR_OVF_PMI_T1 0x08000000U +#define P4_CCCR_FORCE_OVF 0x02000000U +#define P4_CCCR_EDGE 0x01000000U +#define P4_CCCR_THRESHOLD_MASK 0x00f00000U +#define P4_CCCR_THRESHOLD_SHIFT 20 +#define P4_CCCR_COMPLEMENT 0x00080000U +#define P4_CCCR_COMPARE 0x00040000U +#define P4_CCCR_ESCR_SELECT_MASK 0x0000e000U +#define P4_CCCR_ESCR_SELECT_SHIFT 13 +#define P4_CCCR_ENABLE 0x00001000U +#define P4_CCCR_THREAD_SINGLE 0x00010000U +#define P4_CCCR_THREAD_BOTH 0x00020000U +#define P4_CCCR_THREAD_ANY 0x00030000U +#define P4_CCCR_RESERVED 0x00000fffU + +#define P4_CCCR_THRESHOLD(v) ((v) << P4_CCCR_THRESHOLD_SHIFT) +#define P4_CCCR_ESEL(v) ((v) << P4_CCCR_ESCR_SELECT_SHIFT) + +/* Custom bits in reerved CCCR area */ +#define P4_CCCR_CACHE_OPS_MASK 0x0000003fU + + +/* Non HT mask */ +#define P4_CCCR_MASK \ + (P4_CCCR_OVF | \ + P4_CCCR_CASCADE | \ + P4_CCCR_OVF_PMI_T0 | \ + P4_CCCR_FORCE_OVF | \ + P4_CCCR_EDGE | \ + P4_CCCR_THRESHOLD_MASK | \ + P4_CCCR_COMPLEMENT | \ + P4_CCCR_COMPARE | \ + P4_CCCR_ESCR_SELECT_MASK | \ + P4_CCCR_ENABLE) + +/* HT mask */ +#define P4_CCCR_MASK_HT (P4_CCCR_MASK | P4_CCCR_THREAD_ANY) + +#define P4_GEN_ESCR_EMASK(class, name, bit) \ + class##__##name = ((1 << bit) << P4_ESCR_EVENTMASK_SHIFT) +#define P4_ESCR_EMASK_BIT(class, name) class##__##name + +/* + * config field is 64bit width and consists of + * HT << 63 | ESCR << 32 | CCCR + * where HT is HyperThreading bit (since ESCR + * has it reserved we may use it for own purpose) + * + * note that this is NOT the addresses of respective + * ESCR and CCCR but rather an only packed value should + * be unpacked and written to a proper addresses + * + * the base idea is to pack as much info as + * possible + */ +#define p4_config_pack_escr(v) (((u64)(v)) << 32) +#define p4_config_pack_cccr(v) (((u64)(v)) & 0xffffffffULL) +#define p4_config_unpack_escr(v) (((u64)(v)) >> 32) +#define p4_config_unpack_cccr(v) (((u64)(v)) & 0xffffffffULL) + +#define p4_config_unpack_emask(v) \ + ({ \ + u32 t = p4_config_unpack_escr((v)); \ + t = t & P4_ESCR_EVENTMASK_MASK; \ + t = t >> P4_ESCR_EVENTMASK_SHIFT; \ + t; \ + }) + +#define p4_config_unpack_event(v) \ + ({ \ + u32 t = p4_config_unpack_escr((v)); \ + t = t & P4_ESCR_EVENT_MASK; \ + t = t >> P4_ESCR_EVENT_SHIFT; \ + t; \ + }) + +#define p4_config_unpack_cache_event(v) (((u64)(v)) & P4_CCCR_CACHE_OPS_MASK) + +#define P4_CONFIG_HT_SHIFT 63 +#define P4_CONFIG_HT (1ULL << P4_CONFIG_HT_SHIFT) + +static inline bool p4_is_event_cascaded(u64 config) +{ + u32 cccr = p4_config_unpack_cccr(config); + return !!(cccr & P4_CCCR_CASCADE); +} + +static inline int p4_ht_config_thread(u64 config) +{ + return !!(config & P4_CONFIG_HT); +} + +static inline u64 p4_set_ht_bit(u64 config) +{ + return config | P4_CONFIG_HT; +} + +static inline u64 p4_clear_ht_bit(u64 config) +{ + return config & ~P4_CONFIG_HT; +} + +static inline int p4_ht_active(void) +{ +#ifdef CONFIG_SMP + return smp_num_siblings > 1; +#endif + return 0; +} + +static inline int p4_ht_thread(int cpu) +{ +#ifdef CONFIG_SMP + if (smp_num_siblings == 2) + return cpu != cpumask_first(__get_cpu_var(cpu_sibling_map)); +#endif + return 0; +} + +static inline int p4_should_swap_ts(u64 config, int cpu) +{ + return p4_ht_config_thread(config) ^ p4_ht_thread(cpu); +} + +static inline u32 p4_default_cccr_conf(int cpu) +{ + /* + * Note that P4_CCCR_THREAD_ANY is "required" on + * non-HT machines (on HT machines we count TS events + * regardless the state of second logical processor + */ + u32 cccr = P4_CCCR_THREAD_ANY; + + if (!p4_ht_thread(cpu)) + cccr |= P4_CCCR_OVF_PMI_T0; + else + cccr |= P4_CCCR_OVF_PMI_T1; + + return cccr; +} + +static inline u32 p4_default_escr_conf(int cpu, int exclude_os, int exclude_usr) +{ + u32 escr = 0; + + if (!p4_ht_thread(cpu)) { + if (!exclude_os) + escr |= P4_ESCR_T0_OS; + if (!exclude_usr) + escr |= P4_ESCR_T0_USR; + } else { + if (!exclude_os) + escr |= P4_ESCR_T1_OS; + if (!exclude_usr) + escr |= P4_ESCR_T1_USR; + } + + return escr; +} + +enum P4_EVENTS { + P4_EVENT_TC_DELIVER_MODE, + P4_EVENT_BPU_FETCH_REQUEST, + P4_EVENT_ITLB_REFERENCE, + P4_EVENT_MEMORY_CANCEL, + P4_EVENT_MEMORY_COMPLETE, + P4_EVENT_LOAD_PORT_REPLAY, + P4_EVENT_STORE_PORT_REPLAY, + P4_EVENT_MOB_LOAD_REPLAY, + P4_EVENT_PAGE_WALK_TYPE, + P4_EVENT_BSQ_CACHE_REFERENCE, + P4_EVENT_IOQ_ALLOCATION, + P4_EVENT_IOQ_ACTIVE_ENTRIES, + P4_EVENT_FSB_DATA_ACTIVITY, + P4_EVENT_BSQ_ALLOCATION, + P4_EVENT_BSQ_ACTIVE_ENTRIES, + P4_EVENT_SSE_INPUT_ASSIST, + P4_EVENT_PACKED_SP_UOP, + P4_EVENT_PACKED_DP_UOP, + P4_EVENT_SCALAR_SP_UOP, + P4_EVENT_SCALAR_DP_UOP, + P4_EVENT_64BIT_MMX_UOP, + P4_EVENT_128BIT_MMX_UOP, + P4_EVENT_X87_FP_UOP, + P4_EVENT_TC_MISC, + P4_EVENT_GLOBAL_POWER_EVENTS, + P4_EVENT_TC_MS_XFER, + P4_EVENT_UOP_QUEUE_WRITES, + P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, + P4_EVENT_RETIRED_BRANCH_TYPE, + P4_EVENT_RESOURCE_STALL, + P4_EVENT_WC_BUFFER, + P4_EVENT_B2B_CYCLES, + P4_EVENT_BNR, + P4_EVENT_SNOOP, + P4_EVENT_RESPONSE, + P4_EVENT_FRONT_END_EVENT, + P4_EVENT_EXECUTION_EVENT, + P4_EVENT_REPLAY_EVENT, + P4_EVENT_INSTR_RETIRED, + P4_EVENT_UOPS_RETIRED, + P4_EVENT_UOP_TYPE, + P4_EVENT_BRANCH_RETIRED, + P4_EVENT_MISPRED_BRANCH_RETIRED, + P4_EVENT_X87_ASSIST, + P4_EVENT_MACHINE_CLEAR, + P4_EVENT_INSTR_COMPLETED, +}; + +#define P4_OPCODE(event) event##_OPCODE +#define P4_OPCODE_ESEL(opcode) ((opcode & 0x00ff) >> 0) +#define P4_OPCODE_EVNT(opcode) ((opcode & 0xff00) >> 8) +#define P4_OPCODE_PACK(event, sel) (((event) << 8) | sel) + +/* + * Comments below the event represent ESCR restriction + * for this event and counter index per ESCR + * + * MSR_P4_IQ_ESCR0 and MSR_P4_IQ_ESCR1 are available only on early + * processor builds (family 0FH, models 01H-02H). These MSRs + * are not available on later versions, so that we don't use + * them completely + * + * Also note that CCCR1 do not have P4_CCCR_ENABLE bit properly + * working so that we should not use this CCCR and respective + * counter as result + */ +enum P4_EVENT_OPCODES { + P4_OPCODE(P4_EVENT_TC_DELIVER_MODE) = P4_OPCODE_PACK(0x01, 0x01), + /* + * MSR_P4_TC_ESCR0: 4, 5 + * MSR_P4_TC_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST) = P4_OPCODE_PACK(0x03, 0x00), + /* + * MSR_P4_BPU_ESCR0: 0, 1 + * MSR_P4_BPU_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_ITLB_REFERENCE) = P4_OPCODE_PACK(0x18, 0x03), + /* + * MSR_P4_ITLB_ESCR0: 0, 1 + * MSR_P4_ITLB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_MEMORY_CANCEL) = P4_OPCODE_PACK(0x02, 0x05), + /* + * MSR_P4_DAC_ESCR0: 8, 9 + * MSR_P4_DAC_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_MEMORY_COMPLETE) = P4_OPCODE_PACK(0x08, 0x02), + /* + * MSR_P4_SAAT_ESCR0: 8, 9 + * MSR_P4_SAAT_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY) = P4_OPCODE_PACK(0x04, 0x02), + /* + * MSR_P4_SAAT_ESCR0: 8, 9 + * MSR_P4_SAAT_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY) = P4_OPCODE_PACK(0x05, 0x02), + /* + * MSR_P4_SAAT_ESCR0: 8, 9 + * MSR_P4_SAAT_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY) = P4_OPCODE_PACK(0x03, 0x02), + /* + * MSR_P4_MOB_ESCR0: 0, 1 + * MSR_P4_MOB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE) = P4_OPCODE_PACK(0x01, 0x04), + /* + * MSR_P4_PMH_ESCR0: 0, 1 + * MSR_P4_PMH_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE) = P4_OPCODE_PACK(0x0c, 0x07), + /* + * MSR_P4_BSU_ESCR0: 0, 1 + * MSR_P4_BSU_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_IOQ_ALLOCATION) = P4_OPCODE_PACK(0x03, 0x06), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES) = P4_OPCODE_PACK(0x1a, 0x06), + /* + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY) = P4_OPCODE_PACK(0x17, 0x06), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_BSQ_ALLOCATION) = P4_OPCODE_PACK(0x05, 0x07), + /* + * MSR_P4_BSU_ESCR0: 0, 1 + */ + + P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES) = P4_OPCODE_PACK(0x06, 0x07), + /* + * NOTE: no ESCR name in docs, it's guessed + * MSR_P4_BSU_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST) = P4_OPCODE_PACK(0x34, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_PACKED_SP_UOP) = P4_OPCODE_PACK(0x08, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_PACKED_DP_UOP) = P4_OPCODE_PACK(0x0c, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_SCALAR_SP_UOP) = P4_OPCODE_PACK(0x0a, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_SCALAR_DP_UOP) = P4_OPCODE_PACK(0x0e, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_64BIT_MMX_UOP) = P4_OPCODE_PACK(0x02, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_128BIT_MMX_UOP) = P4_OPCODE_PACK(0x1a, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_X87_FP_UOP) = P4_OPCODE_PACK(0x04, 0x01), + /* + * MSR_P4_FIRM_ESCR0: 8, 9 + * MSR_P4_FIRM_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_TC_MISC) = P4_OPCODE_PACK(0x06, 0x01), + /* + * MSR_P4_TC_ESCR0: 4, 5 + * MSR_P4_TC_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS) = P4_OPCODE_PACK(0x13, 0x06), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_TC_MS_XFER) = P4_OPCODE_PACK(0x05, 0x00), + /* + * MSR_P4_MS_ESCR0: 4, 5 + * MSR_P4_MS_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES) = P4_OPCODE_PACK(0x09, 0x00), + /* + * MSR_P4_MS_ESCR0: 4, 5 + * MSR_P4_MS_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE) = P4_OPCODE_PACK(0x05, 0x02), + /* + * MSR_P4_TBPU_ESCR0: 4, 5 + * MSR_P4_TBPU_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE) = P4_OPCODE_PACK(0x04, 0x02), + /* + * MSR_P4_TBPU_ESCR0: 4, 5 + * MSR_P4_TBPU_ESCR1: 6, 7 + */ + + P4_OPCODE(P4_EVENT_RESOURCE_STALL) = P4_OPCODE_PACK(0x01, 0x01), + /* + * MSR_P4_ALF_ESCR0: 12, 13, 16 + * MSR_P4_ALF_ESCR1: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_WC_BUFFER) = P4_OPCODE_PACK(0x05, 0x05), + /* + * MSR_P4_DAC_ESCR0: 8, 9 + * MSR_P4_DAC_ESCR1: 10, 11 + */ + + P4_OPCODE(P4_EVENT_B2B_CYCLES) = P4_OPCODE_PACK(0x16, 0x03), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_BNR) = P4_OPCODE_PACK(0x08, 0x03), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_SNOOP) = P4_OPCODE_PACK(0x06, 0x03), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_RESPONSE) = P4_OPCODE_PACK(0x04, 0x03), + /* + * MSR_P4_FSB_ESCR0: 0, 1 + * MSR_P4_FSB_ESCR1: 2, 3 + */ + + P4_OPCODE(P4_EVENT_FRONT_END_EVENT) = P4_OPCODE_PACK(0x08, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_EXECUTION_EVENT) = P4_OPCODE_PACK(0x0c, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_REPLAY_EVENT) = P4_OPCODE_PACK(0x09, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_INSTR_RETIRED) = P4_OPCODE_PACK(0x02, 0x04), + /* + * MSR_P4_CRU_ESCR0: 12, 13, 16 + * MSR_P4_CRU_ESCR1: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_UOPS_RETIRED) = P4_OPCODE_PACK(0x01, 0x04), + /* + * MSR_P4_CRU_ESCR0: 12, 13, 16 + * MSR_P4_CRU_ESCR1: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_UOP_TYPE) = P4_OPCODE_PACK(0x02, 0x02), + /* + * MSR_P4_RAT_ESCR0: 12, 13, 16 + * MSR_P4_RAT_ESCR1: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_BRANCH_RETIRED) = P4_OPCODE_PACK(0x06, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED) = P4_OPCODE_PACK(0x03, 0x04), + /* + * MSR_P4_CRU_ESCR0: 12, 13, 16 + * MSR_P4_CRU_ESCR1: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_X87_ASSIST) = P4_OPCODE_PACK(0x03, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_MACHINE_CLEAR) = P4_OPCODE_PACK(0x02, 0x05), + /* + * MSR_P4_CRU_ESCR2: 12, 13, 16 + * MSR_P4_CRU_ESCR3: 14, 15, 17 + */ + + P4_OPCODE(P4_EVENT_INSTR_COMPLETED) = P4_OPCODE_PACK(0x07, 0x04), + /* + * MSR_P4_CRU_ESCR0: 12, 13, 16 + * MSR_P4_CRU_ESCR1: 14, 15, 17 + */ +}; + +/* + * a caller should use P4_ESCR_EMASK_NAME helper to + * pick the EventMask needed, for example + * + * P4_ESCR_EMASK_NAME(P4_EVENT_TC_DELIVER_MODE, DD) + */ +enum P4_ESCR_EMASKS { + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DD, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DB, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DI, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BD, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BB, 4), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BI, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, ID, 6), + + P4_GEN_ESCR_EMASK(P4_EVENT_BPU_FETCH_REQUEST, TCMISS, 0), + + P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, HIT, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, MISS, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, HIT_UK, 2), + + P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_CANCEL, 64K_CONF, 3), + + P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_COMPLETE, LSC, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_COMPLETE, SSC, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, NO_STA, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, NO_STD, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA, 4), + P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR, 5), + + P4_GEN_ESCR_EMASK(P4_EVENT_PAGE_WALK_TYPE, DTMISS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_PAGE_WALK_TYPE, ITMISS, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE, 4), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS, 8), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS, 9), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS, 10), + + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, DEFAULT, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, ALL_READ, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE, 6), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_UC, 7), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WC, 8), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WT, 9), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WP, 10), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WB, 11), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, OWN, 13), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, OTHER, 14), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, PREFETCH, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE, 6), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC, 7), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC, 8), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT, 9), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP, 10), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB, 11), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN, 13), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER, 14), + P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN, 4), + P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER, 5), + + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE, 6), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE, 7), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE, 8), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE, 9), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE, 10), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0, 11), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1, 12), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2, 13), + + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE, 6), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE, 7), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE, 8), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE, 9), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE, 10), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0, 11), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1, 12), + P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2, 13), + + P4_GEN_ESCR_EMASK(P4_EVENT_SSE_INPUT_ASSIST, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_PACKED_SP_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_PACKED_DP_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_SCALAR_SP_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_SCALAR_DP_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_64BIT_MMX_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_128BIT_MMX_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_X87_FP_UOP, ALL, 15), + + P4_GEN_ESCR_EMASK(P4_EVENT_TC_MISC, FLUSH, 4), + + P4_GEN_ESCR_EMASK(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING, 0), + + P4_GEN_ESCR_EMASK(P4_EVENT_TC_MS_XFER, CISC, 0), + + P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM, 2), + + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT, 4), + + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, CALL, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT, 4), + + P4_GEN_ESCR_EMASK(P4_EVENT_RESOURCE_STALL, SBFULL, 5), + + P4_GEN_ESCR_EMASK(P4_EVENT_WC_BUFFER, WCB_EVICTS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_FRONT_END_EVENT, NBOGUS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_FRONT_END_EVENT, BOGUS, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS0, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS1, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS2, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS3, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS0, 4), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS1, 5), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS2, 6), + P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS3, 7), + + P4_GEN_ESCR_EMASK(P4_EVENT_REPLAY_EVENT, NBOGUS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_REPLAY_EVENT, BOGUS, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, NBOGUSTAG, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, BOGUSNTAG, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, BOGUSTAG, 3), + + P4_GEN_ESCR_EMASK(P4_EVENT_UOPS_RETIRED, NBOGUS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_UOPS_RETIRED, BOGUS, 1), + + P4_GEN_ESCR_EMASK(P4_EVENT_UOP_TYPE, TAGLOADS, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_UOP_TYPE, TAGSTORES, 2), + + P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMNP, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMNM, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMTP, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMTM, 3), + + P4_GEN_ESCR_EMASK(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS, 0), + + P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, FPSU, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, FPSO, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, POAO, 2), + P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, POAU, 3), + P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, PREA, 4), + + P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, CLEAR, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, MOCLEAR, 1), + P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, SMCLEAR, 2), + + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_COMPLETED, NBOGUS, 0), + P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_COMPLETED, BOGUS, 1), +}; + +/* P4 PEBS: stale for a while */ +#define P4_PEBS_METRIC_MASK 0x00001fffU +#define P4_PEBS_UOB_TAG 0x01000000U +#define P4_PEBS_ENABLE 0x02000000U + +/* Replay metrics for MSR_IA32_PEBS_ENABLE and MSR_P4_PEBS_MATRIX_VERT */ +#define P4_PEBS__1stl_cache_load_miss_retired 0x3000001 +#define P4_PEBS__2ndl_cache_load_miss_retired 0x3000002 +#define P4_PEBS__dtlb_load_miss_retired 0x3000004 +#define P4_PEBS__dtlb_store_miss_retired 0x3000004 +#define P4_PEBS__dtlb_all_miss_retired 0x3000004 +#define P4_PEBS__tagged_mispred_branch 0x3018000 +#define P4_PEBS__mob_load_replay_retired 0x3000200 +#define P4_PEBS__split_load_retired 0x3000400 +#define P4_PEBS__split_store_retired 0x3000400 + +#define P4_VERT__1stl_cache_load_miss_retired 0x0000001 +#define P4_VERT__2ndl_cache_load_miss_retired 0x0000001 +#define P4_VERT__dtlb_load_miss_retired 0x0000001 +#define P4_VERT__dtlb_store_miss_retired 0x0000002 +#define P4_VERT__dtlb_all_miss_retired 0x0000003 +#define P4_VERT__tagged_mispred_branch 0x0000010 +#define P4_VERT__mob_load_replay_retired 0x0000001 +#define P4_VERT__split_load_retired 0x0000001 +#define P4_VERT__split_store_retired 0x0000002 + +enum P4_CACHE_EVENTS { + P4_CACHE__NONE, + + P4_CACHE__1stl_cache_load_miss_retired, + P4_CACHE__2ndl_cache_load_miss_retired, + P4_CACHE__dtlb_load_miss_retired, + P4_CACHE__dtlb_store_miss_retired, + P4_CACHE__itlb_reference_hit, + P4_CACHE__itlb_reference_miss, + + P4_CACHE__MAX +}; + +#endif /* PERF_EVENT_P4_H */ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index b753ea59703a..32428b410b55 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -21,7 +21,6 @@ struct mm_struct; #include <asm/msr.h> #include <asm/desc_defs.h> #include <asm/nops.h> -#include <asm/ds.h> #include <linux/personality.h> #include <linux/cpumask.h> @@ -29,6 +28,7 @@ struct mm_struct; #include <linux/threads.h> #include <linux/math64.h> #include <linux/init.h> +#include <linux/err.h> #define HBP_NUM 4 /* @@ -473,10 +473,6 @@ struct thread_struct { unsigned long iopl; /* Max allowed port in the bitmap, in bytes: */ unsigned io_bitmap_max; -/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set. */ - unsigned long debugctlmsr; - /* Debug Store context; see asm/ds.h */ - struct ds_context *ds_ctx; }; static inline unsigned long native_get_debugreg(int regno) @@ -803,7 +799,7 @@ extern void cpu_init(void); static inline unsigned long get_debugctlmsr(void) { - unsigned long debugctlmsr = 0; + unsigned long debugctlmsr = 0; #ifndef CONFIG_X86_DEBUGCTLMSR if (boot_cpu_data.x86 < 6) @@ -811,21 +807,6 @@ static inline unsigned long get_debugctlmsr(void) #endif rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr); - return debugctlmsr; -} - -static inline unsigned long get_debugctlmsr_on_cpu(int cpu) -{ - u64 debugctlmsr = 0; - u32 val1, val2; - -#ifndef CONFIG_X86_DEBUGCTLMSR - if (boot_cpu_data.x86 < 6) - return 0; -#endif - rdmsr_on_cpu(cpu, MSR_IA32_DEBUGCTLMSR, &val1, &val2); - debugctlmsr = val1 | ((u64)val2 << 32); - return debugctlmsr; } @@ -838,18 +819,6 @@ static inline void update_debugctlmsr(unsigned long debugctlmsr) wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr); } -static inline void update_debugctlmsr_on_cpu(int cpu, - unsigned long debugctlmsr) -{ -#ifndef CONFIG_X86_DEBUGCTLMSR - if (boot_cpu_data.x86 < 6) - return; -#endif - wrmsr_on_cpu(cpu, MSR_IA32_DEBUGCTLMSR, - (u32)((u64)debugctlmsr), - (u32)((u64)debugctlmsr >> 32)); -} - /* * from system description table in BIOS. Mostly for MCA use, but * others may find it useful: diff --git a/arch/x86/include/asm/ptrace-abi.h b/arch/x86/include/asm/ptrace-abi.h index 86723035a515..52b098a6eebb 100644 --- a/arch/x86/include/asm/ptrace-abi.h +++ b/arch/x86/include/asm/ptrace-abi.h @@ -82,61 +82,6 @@ #ifndef __ASSEMBLY__ #include <linux/types.h> - -/* configuration/status structure used in PTRACE_BTS_CONFIG and - PTRACE_BTS_STATUS commands. -*/ -struct ptrace_bts_config { - /* requested or actual size of BTS buffer in bytes */ - __u32 size; - /* bitmask of below flags */ - __u32 flags; - /* buffer overflow signal */ - __u32 signal; - /* actual size of bts_struct in bytes */ - __u32 bts_size; -}; -#endif /* __ASSEMBLY__ */ - -#define PTRACE_BTS_O_TRACE 0x1 /* branch trace */ -#define PTRACE_BTS_O_SCHED 0x2 /* scheduling events w/ jiffies */ -#define PTRACE_BTS_O_SIGNAL 0x4 /* send SIG<signal> on buffer overflow - instead of wrapping around */ -#define PTRACE_BTS_O_ALLOC 0x8 /* (re)allocate buffer */ - -#define PTRACE_BTS_CONFIG 40 -/* Configure branch trace recording. - ADDR points to a struct ptrace_bts_config. - DATA gives the size of that buffer. - A new buffer is allocated, if requested in the flags. - An overflow signal may only be requested for new buffers. - Returns the number of bytes read. -*/ -#define PTRACE_BTS_STATUS 41 -/* Return the current configuration in a struct ptrace_bts_config - pointed to by ADDR; DATA gives the size of that buffer. - Returns the number of bytes written. -*/ -#define PTRACE_BTS_SIZE 42 -/* Return the number of available BTS records for draining. - DATA and ADDR are ignored. -*/ -#define PTRACE_BTS_GET 43 -/* Get a single BTS record. - DATA defines the index into the BTS array, where 0 is the newest - entry, and higher indices refer to older entries. - ADDR is pointing to struct bts_struct (see asm/ds.h). -*/ -#define PTRACE_BTS_CLEAR 44 -/* Clear the BTS buffer. - DATA and ADDR are ignored. -*/ -#define PTRACE_BTS_DRAIN 45 -/* Read all available BTS records and clear the buffer. - ADDR points to an array of struct bts_struct. - DATA gives the size of that buffer. - BTS records are read from oldest to newest. - Returns number of BTS records drained. -*/ +#endif #endif /* _ASM_X86_PTRACE_ABI_H */ diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h index 69a686a7dff0..78cd1ea94500 100644 --- a/arch/x86/include/asm/ptrace.h +++ b/arch/x86/include/asm/ptrace.h @@ -289,12 +289,6 @@ extern int do_get_thread_area(struct task_struct *p, int idx, extern int do_set_thread_area(struct task_struct *p, int idx, struct user_desc __user *info, int can_allocate); -#ifdef CONFIG_X86_PTRACE_BTS -extern void ptrace_bts_untrace(struct task_struct *tsk); - -#define arch_ptrace_untrace(tsk) ptrace_bts_untrace(tsk) -#endif /* CONFIG_X86_PTRACE_BTS */ - #endif /* __KERNEL__ */ #endif /* !__ASSEMBLY__ */ diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index e0d28901e969..d017ed5502e2 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -92,8 +92,7 @@ struct thread_info { #define TIF_IO_BITMAP 22 /* uses I/O bitmap */ #define TIF_FREEZE 23 /* is freezing for suspend */ #define TIF_FORCED_TF 24 /* true if TF in eflags artificially */ -#define TIF_DEBUGCTLMSR 25 /* uses thread_struct.debugctlmsr */ -#define TIF_DS_AREA_MSR 26 /* uses thread_struct.ds_area_msr */ +#define TIF_BLOCKSTEP 25 /* set when we want DEBUGCTLMSR_BTF */ #define TIF_LAZY_MMU_UPDATES 27 /* task is updating the mmu lazily */ #define TIF_SYSCALL_TRACEPOINT 28 /* syscall tracepoint instrumentation */ @@ -115,8 +114,7 @@ struct thread_info { #define _TIF_IO_BITMAP (1 << TIF_IO_BITMAP) #define _TIF_FREEZE (1 << TIF_FREEZE) #define _TIF_FORCED_TF (1 << TIF_FORCED_TF) -#define _TIF_DEBUGCTLMSR (1 << TIF_DEBUGCTLMSR) -#define _TIF_DS_AREA_MSR (1 << TIF_DS_AREA_MSR) +#define _TIF_BLOCKSTEP (1 << TIF_BLOCKSTEP) #define _TIF_LAZY_MMU_UPDATES (1 << TIF_LAZY_MMU_UPDATES) #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) @@ -147,7 +145,7 @@ struct thread_info { /* flags to check in __switch_to() */ #define _TIF_WORK_CTXSW \ - (_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_NOTSC) + (_TIF_IO_BITMAP|_TIF_NOTSC|_TIF_BLOCKSTEP) #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY) #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG) diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h index b414d2b401f6..aa558ac0306e 100644 --- a/arch/x86/include/asm/uv/uv_bau.h +++ b/arch/x86/include/asm/uv/uv_bau.h @@ -27,13 +27,14 @@ * set 2 is at BASE + 2*512, set 3 at BASE + 3*512, and so on. * * We will use 31 sets, one for sending BAU messages from each of the 32 - * cpu's on the node. + * cpu's on the uvhub. * * TLB shootdown will use the first of the 8 descriptors of each set. * Each of the descriptors is 64 bytes in size (8*64 = 512 bytes in a set). */ #define UV_ITEMS_PER_DESCRIPTOR 8 +#define MAX_BAU_CONCURRENT 3 #define UV_CPUS_PER_ACT_STATUS 32 #define UV_ACT_STATUS_MASK 0x3 #define UV_ACT_STATUS_SIZE 2 @@ -45,6 +46,9 @@ #define UV_PAYLOADQ_PNODE_SHIFT 49 #define UV_PTC_BASENAME "sgi_uv/ptc_statistics" #define uv_physnodeaddr(x) ((__pa((unsigned long)(x)) & uv_mmask)) +#define UV_ENABLE_INTD_SOFT_ACK_MODE_SHIFT 15 +#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHIFT 16 +#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD 0x000000000bUL /* * bits in UVH_LB_BAU_SB_ACTIVATION_STATUS_0/1 @@ -55,15 +59,29 @@ #define DESC_STATUS_SOURCE_TIMEOUT 3 /* - * source side thresholds at which message retries print a warning + * source side threshholds at which message retries print a warning */ #define SOURCE_TIMEOUT_LIMIT 20 #define DESTINATION_TIMEOUT_LIMIT 20 /* + * misc. delays, in microseconds + */ +#define THROTTLE_DELAY 10 +#define TIMEOUT_DELAY 10 +#define BIOS_TO 1000 +/* BIOS is assumed to set the destination timeout to 1003520 nanoseconds */ + +/* + * threshholds at which to use IPI to free resources + */ +#define PLUGSB4RESET 100 +#define TIMEOUTSB4RESET 100 + +/* * number of entries in the destination side payload queue */ -#define DEST_Q_SIZE 17 +#define DEST_Q_SIZE 20 /* * number of destination side software ack resources */ @@ -72,9 +90,10 @@ /* * completion statuses for sending a TLB flush message */ -#define FLUSH_RETRY 1 -#define FLUSH_GIVEUP 2 -#define FLUSH_COMPLETE 3 +#define FLUSH_RETRY_PLUGGED 1 +#define FLUSH_RETRY_TIMEOUT 2 +#define FLUSH_GIVEUP 3 +#define FLUSH_COMPLETE 4 /* * Distribution: 32 bytes (256 bits) (bytes 0-0x1f of descriptor) @@ -86,14 +105,14 @@ * 'base_dest_nodeid' field of the header corresponds to the * destination nodeID associated with that specified bit. */ -struct bau_target_nodemask { - unsigned long bits[BITS_TO_LONGS(256)]; +struct bau_target_uvhubmask { + unsigned long bits[BITS_TO_LONGS(UV_DISTRIBUTION_SIZE)]; }; /* - * mask of cpu's on a node + * mask of cpu's on a uvhub * (during initialization we need to check that unsigned long has - * enough bits for max. cpu's per node) + * enough bits for max. cpu's per uvhub) */ struct bau_local_cpumask { unsigned long bits; @@ -135,8 +154,8 @@ struct bau_msg_payload { struct bau_msg_header { unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */ /* bits 5:0 */ - unsigned int base_dest_nodeid:15; /* nasid>>1 (pnode) of */ - /* bits 20:6 */ /* first bit in node_map */ + unsigned int base_dest_nodeid:15; /* nasid (pnode<<1) of */ + /* bits 20:6 */ /* first bit in uvhub map */ unsigned int command:8; /* message type */ /* bits 28:21 */ /* 0x38: SN3net EndPoint Message */ @@ -146,26 +165,38 @@ struct bau_msg_header { unsigned int rsvd_2:9; /* must be zero */ /* bits 40:32 */ /* Suppl_A is 56-41 */ - unsigned int payload_2a:8;/* becomes byte 16 of msg */ - /* bits 48:41 */ /* not currently using */ - unsigned int payload_2b:8;/* becomes byte 17 of msg */ - /* bits 56:49 */ /* not currently using */ + unsigned int sequence:16;/* message sequence number */ + /* bits 56:41 */ /* becomes bytes 16-17 of msg */ /* Address field (96:57) is never used as an address (these are address bits 42:3) */ + unsigned int rsvd_3:1; /* must be zero */ /* bit 57 */ /* address bits 27:4 are payload */ - /* these 24 bits become bytes 12-14 of msg */ + /* these next 24 (58-81) bits become bytes 12-14 of msg */ + + /* bits 65:58 land in byte 12 */ unsigned int replied_to:1;/* sent as 0 by the source to byte 12 */ /* bit 58 */ - - unsigned int payload_1a:5;/* not currently used */ - /* bits 63:59 */ - unsigned int payload_1b:8;/* not currently used */ - /* bits 71:64 */ - unsigned int payload_1c:8;/* not currently used */ - /* bits 79:72 */ - unsigned int payload_1d:2;/* not currently used */ + unsigned int msg_type:3; /* software type of the message*/ + /* bits 61:59 */ + unsigned int canceled:1; /* message canceled, resource to be freed*/ + /* bit 62 */ + unsigned int payload_1a:1;/* not currently used */ + /* bit 63 */ + unsigned int payload_1b:2;/* not currently used */ + /* bits 65:64 */ + + /* bits 73:66 land in byte 13 */ + unsigned int payload_1ca:6;/* not currently used */ + /* bits 71:66 */ + unsigned int payload_1c:2;/* not currently used */ + /* bits 73:72 */ + + /* bits 81:74 land in byte 14 */ + unsigned int payload_1d:6;/* not currently used */ + /* bits 79:74 */ + unsigned int payload_1e:2;/* not currently used */ /* bits 81:80 */ unsigned int rsvd_4:7; /* must be zero */ @@ -178,7 +209,7 @@ struct bau_msg_header { /* bits 95:90 */ unsigned int rsvd_6:5; /* must be zero */ /* bits 100:96 */ - unsigned int int_both:1;/* if 1, interrupt both sockets on the blade */ + unsigned int int_both:1;/* if 1, interrupt both sockets on the uvhub */ /* bit 101*/ unsigned int fairness:3;/* usually zero */ /* bits 104:102 */ @@ -191,13 +222,18 @@ struct bau_msg_header { /* bits 127:107 */ }; +/* see msg_type: */ +#define MSG_NOOP 0 +#define MSG_REGULAR 1 +#define MSG_RETRY 2 + /* * The activation descriptor: * The format of the message to send, plus all accompanying control * Should be 64 bytes */ struct bau_desc { - struct bau_target_nodemask distribution; + struct bau_target_uvhubmask distribution; /* * message template, consisting of header and payload: */ @@ -237,19 +273,25 @@ struct bau_payload_queue_entry { unsigned short acknowledge_count; /* filled in by destination */ /* 16 bits, bytes 10-11 */ - unsigned short replied_to:1; /* sent as 0 by the source */ - /* 1 bit */ - unsigned short unused1:7; /* not currently using */ - /* 7 bits: byte 12) */ + /* these next 3 bytes come from bits 58-81 of the message header */ + unsigned short replied_to:1; /* sent as 0 by the source */ + unsigned short msg_type:3; /* software message type */ + unsigned short canceled:1; /* sent as 0 by the source */ + unsigned short unused1:3; /* not currently using */ + /* byte 12 */ - unsigned char unused2[2]; /* not currently using */ - /* bytes 13-14 */ + unsigned char unused2a; /* not currently using */ + /* byte 13 */ + unsigned char unused2; /* not currently using */ + /* byte 14 */ unsigned char sw_ack_vector; /* filled in by the hardware */ /* byte 15 (bits 127:120) */ - unsigned char unused4[3]; /* not currently using bytes 17-19 */ - /* bytes 17-19 */ + unsigned short sequence; /* message sequence number */ + /* bytes 16-17 */ + unsigned char unused4[2]; /* not currently using bytes 18-19 */ + /* bytes 18-19 */ int number_of_cpus; /* filled in at destination */ /* 32 bits, bytes 20-23 (aligned) */ @@ -259,63 +301,93 @@ struct bau_payload_queue_entry { }; /* - * one for every slot in the destination payload queue - */ -struct bau_msg_status { - struct bau_local_cpumask seen_by; /* map of cpu's */ -}; - -/* - * one for every slot in the destination software ack resources - */ -struct bau_sw_ack_status { - struct bau_payload_queue_entry *msg; /* associated message */ - int watcher; /* cpu monitoring, or -1 */ -}; - -/* - * one on every node and per-cpu; to locate the software tables + * one per-cpu; to locate the software tables */ struct bau_control { struct bau_desc *descriptor_base; - struct bau_payload_queue_entry *bau_msg_head; struct bau_payload_queue_entry *va_queue_first; struct bau_payload_queue_entry *va_queue_last; - struct bau_msg_status *msg_statuses; - int *watching; /* pointer to array */ + struct bau_payload_queue_entry *bau_msg_head; + struct bau_control *uvhub_master; + struct bau_control *socket_master; + unsigned long timeout_interval; + atomic_t active_descriptor_count; + int max_concurrent; + int max_concurrent_constant; + int retry_message_scans; + int plugged_tries; + int timeout_tries; + int ipi_attempts; + int conseccompletes; + short cpu; + short uvhub_cpu; + short uvhub; + short cpus_in_socket; + short cpus_in_uvhub; + unsigned short message_number; + unsigned short uvhub_quiesce; + short socket_acknowledge_count[DEST_Q_SIZE]; + cycles_t send_message; + spinlock_t masks_lock; + spinlock_t uvhub_lock; + spinlock_t queue_lock; }; /* * This structure is allocated per_cpu for UV TLB shootdown statistics. */ struct ptc_stats { - unsigned long ptc_i; /* number of IPI-style flushes */ - unsigned long requestor; /* number of nodes this cpu sent to */ - unsigned long requestee; /* times cpu was remotely requested */ - unsigned long alltlb; /* times all tlb's on this cpu were flushed */ - unsigned long onetlb; /* times just one tlb on this cpu was flushed */ - unsigned long s_retry; /* retries on source side timeouts */ - unsigned long d_retry; /* retries on destination side timeouts */ - unsigned long sflush; /* cycles spent in uv_flush_tlb_others */ - unsigned long dflush; /* cycles spent on destination side */ - unsigned long retriesok; /* successes on retries */ - unsigned long nomsg; /* interrupts with no message */ - unsigned long multmsg; /* interrupts with multiple messages */ - unsigned long ntargeted;/* nodes targeted */ + /* sender statistics */ + unsigned long s_giveup; /* number of fall backs to IPI-style flushes */ + unsigned long s_requestor; /* number of shootdown requests */ + unsigned long s_stimeout; /* source side timeouts */ + unsigned long s_dtimeout; /* destination side timeouts */ + unsigned long s_time; /* time spent in sending side */ + unsigned long s_retriesok; /* successful retries */ + unsigned long s_ntargcpu; /* number of cpus targeted */ + unsigned long s_ntarguvhub; /* number of uvhubs targeted */ + unsigned long s_ntarguvhub16; /* number of times >= 16 target hubs */ + unsigned long s_ntarguvhub8; /* number of times >= 8 target hubs */ + unsigned long s_ntarguvhub4; /* number of times >= 4 target hubs */ + unsigned long s_ntarguvhub2; /* number of times >= 2 target hubs */ + unsigned long s_ntarguvhub1; /* number of times == 1 target hub */ + unsigned long s_resets_plug; /* ipi-style resets from plug state */ + unsigned long s_resets_timeout; /* ipi-style resets from timeouts */ + unsigned long s_busy; /* status stayed busy past s/w timer */ + unsigned long s_throttles; /* waits in throttle */ + unsigned long s_retry_messages; /* retry broadcasts */ + /* destination statistics */ + unsigned long d_alltlb; /* times all tlb's on this cpu were flushed */ + unsigned long d_onetlb; /* times just one tlb on this cpu was flushed */ + unsigned long d_multmsg; /* interrupts with multiple messages */ + unsigned long d_nomsg; /* interrupts with no message */ + unsigned long d_time; /* time spent on destination side */ + unsigned long d_requestee; /* number of messages processed */ + unsigned long d_retries; /* number of retry messages processed */ + unsigned long d_canceled; /* number of messages canceled by retries */ + unsigned long d_nocanceled; /* retries that found nothing to cancel */ + unsigned long d_resets; /* number of ipi-style requests processed */ + unsigned long d_rcanceled; /* number of messages canceled by resets */ }; -static inline int bau_node_isset(int node, struct bau_target_nodemask *dstp) +static inline int bau_uvhub_isset(int uvhub, struct bau_target_uvhubmask *dstp) { - return constant_test_bit(node, &dstp->bits[0]); + return constant_test_bit(uvhub, &dstp->bits[0]); } -static inline void bau_node_set(int node, struct bau_target_nodemask *dstp) +static inline void bau_uvhub_set(int uvhub, struct bau_target_uvhubmask *dstp) { - __set_bit(node, &dstp->bits[0]); + __set_bit(uvhub, &dstp->bits[0]); } -static inline void bau_nodes_clear(struct bau_target_nodemask *dstp, int nbits) +static inline void bau_uvhubs_clear(struct bau_target_uvhubmask *dstp, + int nbits) { bitmap_zero(&dstp->bits[0], nbits); } +static inline int bau_uvhub_weight(struct bau_target_uvhubmask *dstp) +{ + return bitmap_weight((unsigned long *)&dstp->bits[0], + UV_DISTRIBUTION_SIZE); +} static inline void bau_cpubits_clear(struct bau_local_cpumask *dstp, int nbits) { @@ -328,4 +400,35 @@ static inline void bau_cpubits_clear(struct bau_local_cpumask *dstp, int nbits) extern void uv_bau_message_intr1(void); extern void uv_bau_timeout_intr1(void); +struct atomic_short { + short counter; +}; + +/** + * atomic_read_short - read a short atomic variable + * @v: pointer of type atomic_short + * + * Atomically reads the value of @v. + */ +static inline int atomic_read_short(const struct atomic_short *v) +{ + return v->counter; +} + +/** + * atomic_add_short_return - add and return a short int + * @i: short value to add + * @v: pointer of type atomic_short + * + * Atomically adds @i to @v and returns @i + @v + */ +static inline int atomic_add_short_return(short i, struct atomic_short *v) +{ + short __i = i; + asm volatile(LOCK_PREFIX "xaddw %0, %1" + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +} + #endif /* _ASM_X86_UV_UV_BAU_H */ diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h index 14cc74ba5d23..bf6b88ef8eeb 100644 --- a/arch/x86/include/asm/uv/uv_hub.h +++ b/arch/x86/include/asm/uv/uv_hub.h @@ -307,7 +307,7 @@ static inline unsigned long uv_read_global_mmr32(int pnode, unsigned long offset * Access Global MMR space using the MMR space located at the top of physical * memory. */ -static inline unsigned long *uv_global_mmr64_address(int pnode, unsigned long offset) +static inline volatile void __iomem *uv_global_mmr64_address(int pnode, unsigned long offset) { return __va(UV_GLOBAL_MMR64_BASE | UV_GLOBAL_MMR64_PNODE_BITS(pnode) | offset); diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h index 2cae46c7c8a2..b2f2d2e05cec 100644 --- a/arch/x86/include/asm/uv/uv_mmrs.h +++ b/arch/x86/include/asm/uv/uv_mmrs.h @@ -1,4 +1,3 @@ - /* * This file is subject to the terms and conditions of the GNU General Public * License. See the file "COPYING" in the main directory of this archive @@ -15,13 +14,25 @@ #define UV_MMR_ENABLE (1UL << 63) /* ========================================================================= */ +/* UVH_BAU_DATA_BROADCAST */ +/* ========================================================================= */ +#define UVH_BAU_DATA_BROADCAST 0x61688UL +#define UVH_BAU_DATA_BROADCAST_32 0x0440 + +#define UVH_BAU_DATA_BROADCAST_ENABLE_SHFT 0 +#define UVH_BAU_DATA_BROADCAST_ENABLE_MASK 0x0000000000000001UL + +union uvh_bau_data_broadcast_u { + unsigned long v; + struct uvh_bau_data_broadcast_s { + unsigned long enable : 1; /* RW */ + unsigned long rsvd_1_63: 63; /* */ + } s; +}; + +/* ========================================================================= */ /* UVH_BAU_DATA_CONFIG */ /* ========================================================================= */ -#define UVH_LB_BAU_MISC_CONTROL 0x320170UL -#define UV_ENABLE_INTD_SOFT_ACK_MODE_SHIFT 15 -#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHIFT 16 -#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD 0x000000000bUL -/* 1011 timebase 7 (168millisec) * 3 ticks -> 500ms */ #define UVH_BAU_DATA_CONFIG 0x61680UL #define UVH_BAU_DATA_CONFIG_32 0x0438 @@ -604,6 +615,68 @@ union uvh_lb_bau_intd_software_acknowledge_u { #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS_32 0x0a70 /* ========================================================================= */ +/* UVH_LB_BAU_MISC_CONTROL */ +/* ========================================================================= */ +#define UVH_LB_BAU_MISC_CONTROL 0x320170UL +#define UVH_LB_BAU_MISC_CONTROL_32 0x00a10 + +#define UVH_LB_BAU_MISC_CONTROL_REJECTION_DELAY_SHFT 0 +#define UVH_LB_BAU_MISC_CONTROL_REJECTION_DELAY_MASK 0x00000000000000ffUL +#define UVH_LB_BAU_MISC_CONTROL_APIC_MODE_SHFT 8 +#define UVH_LB_BAU_MISC_CONTROL_APIC_MODE_MASK 0x0000000000000100UL +#define UVH_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_SHFT 9 +#define UVH_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_MASK 0x0000000000000200UL +#define UVH_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_SHFT 10 +#define UVH_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_MASK 0x0000000000000400UL +#define UVH_LB_BAU_MISC_CONTROL_CSI_AGENT_PRESENCE_VECTOR_SHFT 11 +#define UVH_LB_BAU_MISC_CONTROL_CSI_AGENT_PRESENCE_VECTOR_MASK 0x0000000000003800UL +#define UVH_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_SHFT 14 +#define UVH_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_MASK 0x0000000000004000UL +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT 15 +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_MASK 0x0000000000008000UL +#define UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT 16 +#define UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_MASK 0x00000000000f0000UL +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_SHFT 20 +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_MASK 0x0000000000100000UL +#define UVH_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_SHFT 21 +#define UVH_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_MASK 0x0000000000200000UL +#define UVH_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_SHFT 22 +#define UVH_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_MASK 0x0000000000400000UL +#define UVH_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_SHFT 23 +#define UVH_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_MASK 0x0000000000800000UL +#define UVH_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_SHFT 24 +#define UVH_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000007000000UL +#define UVH_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_SHFT 27 +#define UVH_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_MASK 0x0000000008000000UL +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_SHFT 28 +#define UVH_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000010000000UL +#define UVH_LB_BAU_MISC_CONTROL_FUN_SHFT 48 +#define UVH_LB_BAU_MISC_CONTROL_FUN_MASK 0xffff000000000000UL + +union uvh_lb_bau_misc_control_u { + unsigned long v; + struct uvh_lb_bau_misc_control_s { + unsigned long rejection_delay : 8; /* RW */ + unsigned long apic_mode : 1; /* RW */ + unsigned long force_broadcast : 1; /* RW */ + unsigned long force_lock_nop : 1; /* RW */ + unsigned long csi_agent_presence_vector : 3; /* RW */ + unsigned long descriptor_fetch_mode : 1; /* RW */ + unsigned long enable_intd_soft_ack_mode : 1; /* RW */ + unsigned long intd_soft_ack_timeout_period : 4; /* RW */ + unsigned long enable_dual_mapping_mode : 1; /* RW */ + unsigned long vga_io_port_decode_enable : 1; /* RW */ + unsigned long vga_io_port_16_bit_decode : 1; /* RW */ + unsigned long suppress_dest_registration : 1; /* RW */ + unsigned long programmed_initial_priority : 3; /* RW */ + unsigned long use_incoming_priority : 1; /* RW */ + unsigned long enable_programmed_initial_priority : 1; /* RW */ + unsigned long rsvd_29_47 : 19; /* */ + unsigned long fun : 16; /* RW */ + } s; +}; + +/* ========================================================================= */ /* UVH_LB_BAU_SB_ACTIVATION_CONTROL */ /* ========================================================================= */ #define UVH_LB_BAU_SB_ACTIVATION_CONTROL 0x320020UL @@ -681,334 +754,6 @@ union uvh_lb_bau_sb_descriptor_base_u { }; /* ========================================================================= */ -/* UVH_LB_MCAST_AOERR0_RPT_ENABLE */ -/* ========================================================================= */ -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE 0x50b20UL - -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_OBESE_MSG_SHFT 0 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_OBESE_MSG_MASK 0x0000000000000001UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_DATA_SB_ERR_SHFT 1 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_DATA_SB_ERR_MASK 0x0000000000000002UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_NACK_BUFF_PARITY_SHFT 2 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_NACK_BUFF_PARITY_MASK 0x0000000000000004UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_TIMEOUT_SHFT 3 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_TIMEOUT_MASK 0x0000000000000008UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_INACTIVE_REPLY_SHFT 4 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_INACTIVE_REPLY_MASK 0x0000000000000010UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_UPGRADE_ERROR_SHFT 5 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_UPGRADE_ERROR_MASK 0x0000000000000020UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_REG_COUNT_UNDERFLOW_SHFT 6 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_REG_COUNT_UNDERFLOW_MASK 0x0000000000000040UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_REP_OBESE_MSG_SHFT 7 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MCAST_REP_OBESE_MSG_MASK 0x0000000000000080UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_RUNT_MSG_SHFT 8 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_RUNT_MSG_MASK 0x0000000000000100UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_OBESE_MSG_SHFT 9 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_OBESE_MSG_MASK 0x0000000000000200UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_DATA_SB_ERR_SHFT 10 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REQ_DATA_SB_ERR_MASK 0x0000000000000400UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_RUNT_MSG_SHFT 11 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_RUNT_MSG_MASK 0x0000000000000800UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_OBESE_MSG_SHFT 12 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_OBESE_MSG_MASK 0x0000000000001000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_DATA_SB_ERR_SHFT 13 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_DATA_SB_ERR_MASK 0x0000000000002000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_COMMAND_ERR_SHFT 14 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_REP_COMMAND_ERR_MASK 0x0000000000004000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_PEND_TIMEOUT_SHFT 15 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_UCACHE_PEND_TIMEOUT_MASK 0x0000000000008000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_RUNT_MSG_SHFT 16 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_RUNT_MSG_MASK 0x0000000000010000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_OBESE_MSG_SHFT 17 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_OBESE_MSG_MASK 0x0000000000020000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_DATA_SB_ERR_SHFT 18 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REQ_DATA_SB_ERR_MASK 0x0000000000040000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_RUNT_MSG_SHFT 19 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_RUNT_MSG_MASK 0x0000000000080000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_OBESE_MSG_SHFT 20 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_OBESE_MSG_MASK 0x0000000000100000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_DATA_SB_ERR_SHFT 21 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_REP_DATA_SB_ERR_MASK 0x0000000000200000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_AMO_TIMEOUT_SHFT 22 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_AMO_TIMEOUT_MASK 0x0000000000400000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_PUT_TIMEOUT_SHFT 23 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_PUT_TIMEOUT_MASK 0x0000000000800000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_SPURIOUS_EVENT_SHFT 24 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_MACC_SPURIOUS_EVENT_MASK 0x0000000001000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_IOH_DESTINATION_TABLE_PARITY_SHFT 25 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_IOH_DESTINATION_TABLE_PARITY_MASK 0x0000000002000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_GET_HAD_ERROR_REPLY_SHFT 26 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_GET_HAD_ERROR_REPLY_MASK 0x0000000004000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_GET_TIMEOUT_SHFT 27 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_GET_TIMEOUT_MASK 0x0000000008000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_LOCK_MANAGER_HAD_ERROR_REPLY_SHFT 28 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_LOCK_MANAGER_HAD_ERROR_REPLY_MASK 0x0000000010000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_PUT_HAD_ERROR_REPLY_SHFT 29 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_PUT_HAD_ERROR_REPLY_MASK 0x0000000020000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_PUT_TIMEOUT_SHFT 30 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_PUT_TIMEOUT_MASK 0x0000000040000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_SB_ACTIVATION_OVERRUN_SHFT 31 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_SB_ACTIVATION_OVERRUN_MASK 0x0000000080000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_COMPLETED_GB_ACTIVATION_HAD_ERROR_REPLY_SHFT 32 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_COMPLETED_GB_ACTIVATION_HAD_ERROR_REPLY_MASK 0x0000000100000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_COMPLETED_GB_ACTIVATION_TIMEOUT_SHFT 33 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_COMPLETED_GB_ACTIVATION_TIMEOUT_MASK 0x0000000200000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_DESCRIPTOR_BUFFER_0_PARITY_SHFT 34 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_DESCRIPTOR_BUFFER_0_PARITY_MASK 0x0000000400000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_DESCRIPTOR_BUFFER_1_PARITY_SHFT 35 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_DESCRIPTOR_BUFFER_1_PARITY_MASK 0x0000000800000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_SOCKET_DESTINATION_TABLE_PARITY_SHFT 36 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_SOCKET_DESTINATION_TABLE_PARITY_MASK 0x0000001000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_BAU_REPLY_PAYLOAD_CORRUPTION_SHFT 37 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_BAU_REPLY_PAYLOAD_CORRUPTION_MASK 0x0000002000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_IO_PORT_DESTINATION_TABLE_PARITY_SHFT 38 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_IO_PORT_DESTINATION_TABLE_PARITY_MASK 0x0000004000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INTD_SOFT_ACK_TIMEOUT_SHFT 39 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INTD_SOFT_ACK_TIMEOUT_MASK 0x0000008000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_REP_OBESE_MSG_SHFT 40 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_REP_OBESE_MSG_MASK 0x0000010000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_REP_COMMAND_ERR_SHFT 41 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_REP_COMMAND_ERR_MASK 0x0000020000000000UL -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_TIMEOUT_SHFT 42 -#define UVH_LB_MCAST_AOERR0_RPT_ENABLE_INT_TIMEOUT_MASK 0x0000040000000000UL - -union uvh_lb_mcast_aoerr0_rpt_enable_u { - unsigned long v; - struct uvh_lb_mcast_aoerr0_rpt_enable_s { - unsigned long mcast_obese_msg : 1; /* RW */ - unsigned long mcast_data_sb_err : 1; /* RW */ - unsigned long mcast_nack_buff_parity : 1; /* RW */ - unsigned long mcast_timeout : 1; /* RW */ - unsigned long mcast_inactive_reply : 1; /* RW */ - unsigned long mcast_upgrade_error : 1; /* RW */ - unsigned long mcast_reg_count_underflow : 1; /* RW */ - unsigned long mcast_rep_obese_msg : 1; /* RW */ - unsigned long ucache_req_runt_msg : 1; /* RW */ - unsigned long ucache_req_obese_msg : 1; /* RW */ - unsigned long ucache_req_data_sb_err : 1; /* RW */ - unsigned long ucache_rep_runt_msg : 1; /* RW */ - unsigned long ucache_rep_obese_msg : 1; /* RW */ - unsigned long ucache_rep_data_sb_err : 1; /* RW */ - unsigned long ucache_rep_command_err : 1; /* RW */ - unsigned long ucache_pend_timeout : 1; /* RW */ - unsigned long macc_req_runt_msg : 1; /* RW */ - unsigned long macc_req_obese_msg : 1; /* RW */ - unsigned long macc_req_data_sb_err : 1; /* RW */ - unsigned long macc_rep_runt_msg : 1; /* RW */ - unsigned long macc_rep_obese_msg : 1; /* RW */ - unsigned long macc_rep_data_sb_err : 1; /* RW */ - unsigned long macc_amo_timeout : 1; /* RW */ - unsigned long macc_put_timeout : 1; /* RW */ - unsigned long macc_spurious_event : 1; /* RW */ - unsigned long ioh_destination_table_parity : 1; /* RW */ - unsigned long get_had_error_reply : 1; /* RW */ - unsigned long get_timeout : 1; /* RW */ - unsigned long lock_manager_had_error_reply : 1; /* RW */ - unsigned long put_had_error_reply : 1; /* RW */ - unsigned long put_timeout : 1; /* RW */ - unsigned long sb_activation_overrun : 1; /* RW */ - unsigned long completed_gb_activation_had_error_reply : 1; /* RW */ - unsigned long completed_gb_activation_timeout : 1; /* RW */ - unsigned long descriptor_buffer_0_parity : 1; /* RW */ - unsigned long descriptor_buffer_1_parity : 1; /* RW */ - unsigned long socket_destination_table_parity : 1; /* RW */ - unsigned long bau_reply_payload_corruption : 1; /* RW */ - unsigned long io_port_destination_table_parity : 1; /* RW */ - unsigned long intd_soft_ack_timeout : 1; /* RW */ - unsigned long int_rep_obese_msg : 1; /* RW */ - unsigned long int_rep_command_err : 1; /* RW */ - unsigned long int_timeout : 1; /* RW */ - unsigned long rsvd_43_63 : 21; /* */ - } s; -}; - -/* ========================================================================= */ -/* UVH_LOCAL_INT0_CONFIG */ -/* ========================================================================= */ -#define UVH_LOCAL_INT0_CONFIG 0x61000UL - -#define UVH_LOCAL_INT0_CONFIG_VECTOR_SHFT 0 -#define UVH_LOCAL_INT0_CONFIG_VECTOR_MASK 0x00000000000000ffUL -#define UVH_LOCAL_INT0_CONFIG_DM_SHFT 8 -#define UVH_LOCAL_INT0_CONFIG_DM_MASK 0x0000000000000700UL -#define UVH_LOCAL_INT0_CONFIG_DESTMODE_SHFT 11 -#define UVH_LOCAL_INT0_CONFIG_DESTMODE_MASK 0x0000000000000800UL -#define UVH_LOCAL_INT0_CONFIG_STATUS_SHFT 12 -#define UVH_LOCAL_INT0_CONFIG_STATUS_MASK 0x0000000000001000UL -#define UVH_LOCAL_INT0_CONFIG_P_SHFT 13 -#define UVH_LOCAL_INT0_CONFIG_P_MASK 0x0000000000002000UL -#define UVH_LOCAL_INT0_CONFIG_T_SHFT 15 -#define UVH_LOCAL_INT0_CONFIG_T_MASK 0x0000000000008000UL -#define UVH_LOCAL_INT0_CONFIG_M_SHFT 16 -#define UVH_LOCAL_INT0_CONFIG_M_MASK 0x0000000000010000UL -#define UVH_LOCAL_INT0_CONFIG_APIC_ID_SHFT 32 -#define UVH_LOCAL_INT0_CONFIG_APIC_ID_MASK 0xffffffff00000000UL - -union uvh_local_int0_config_u { - unsigned long v; - struct uvh_local_int0_config_s { - unsigned long vector_ : 8; /* RW */ - unsigned long dm : 3; /* RW */ - unsigned long destmode : 1; /* RW */ - unsigned long status : 1; /* RO */ - unsigned long p : 1; /* RO */ - unsigned long rsvd_14 : 1; /* */ - unsigned long t : 1; /* RO */ - unsigned long m : 1; /* RW */ - unsigned long rsvd_17_31: 15; /* */ - unsigned long apic_id : 32; /* RW */ - } s; -}; - -/* ========================================================================= */ -/* UVH_LOCAL_INT0_ENABLE */ -/* ========================================================================= */ -#define UVH_LOCAL_INT0_ENABLE 0x65000UL - -#define UVH_LOCAL_INT0_ENABLE_LB_HCERR_SHFT 0 -#define UVH_LOCAL_INT0_ENABLE_LB_HCERR_MASK 0x0000000000000001UL -#define UVH_LOCAL_INT0_ENABLE_GR0_HCERR_SHFT 1 -#define UVH_LOCAL_INT0_ENABLE_GR0_HCERR_MASK 0x0000000000000002UL -#define UVH_LOCAL_INT0_ENABLE_GR1_HCERR_SHFT 2 -#define UVH_LOCAL_INT0_ENABLE_GR1_HCERR_MASK 0x0000000000000004UL -#define UVH_LOCAL_INT0_ENABLE_LH_HCERR_SHFT 3 -#define UVH_LOCAL_INT0_ENABLE_LH_HCERR_MASK 0x0000000000000008UL -#define UVH_LOCAL_INT0_ENABLE_RH_HCERR_SHFT 4 -#define UVH_LOCAL_INT0_ENABLE_RH_HCERR_MASK 0x0000000000000010UL -#define UVH_LOCAL_INT0_ENABLE_XN_HCERR_SHFT 5 -#define UVH_LOCAL_INT0_ENABLE_XN_HCERR_MASK 0x0000000000000020UL -#define UVH_LOCAL_INT0_ENABLE_SI_HCERR_SHFT 6 -#define UVH_LOCAL_INT0_ENABLE_SI_HCERR_MASK 0x0000000000000040UL -#define UVH_LOCAL_INT0_ENABLE_LB_AOERR0_SHFT 7 -#define UVH_LOCAL_INT0_ENABLE_LB_AOERR0_MASK 0x0000000000000080UL -#define UVH_LOCAL_INT0_ENABLE_GR0_AOERR0_SHFT 8 -#define UVH_LOCAL_INT0_ENABLE_GR0_AOERR0_MASK 0x0000000000000100UL -#define UVH_LOCAL_INT0_ENABLE_GR1_AOERR0_SHFT 9 -#define UVH_LOCAL_INT0_ENABLE_GR1_AOERR0_MASK 0x0000000000000200UL -#define UVH_LOCAL_INT0_ENABLE_LH_AOERR0_SHFT 10 -#define UVH_LOCAL_INT0_ENABLE_LH_AOERR0_MASK 0x0000000000000400UL -#define UVH_LOCAL_INT0_ENABLE_RH_AOERR0_SHFT 11 -#define UVH_LOCAL_INT0_ENABLE_RH_AOERR0_MASK 0x0000000000000800UL -#define UVH_LOCAL_INT0_ENABLE_XN_AOERR0_SHFT 12 -#define UVH_LOCAL_INT0_ENABLE_XN_AOERR0_MASK 0x0000000000001000UL -#define UVH_LOCAL_INT0_ENABLE_SI_AOERR0_SHFT 13 -#define UVH_LOCAL_INT0_ENABLE_SI_AOERR0_MASK 0x0000000000002000UL -#define UVH_LOCAL_INT0_ENABLE_LB_AOERR1_SHFT 14 -#define UVH_LOCAL_INT0_ENABLE_LB_AOERR1_MASK 0x0000000000004000UL -#define UVH_LOCAL_INT0_ENABLE_GR0_AOERR1_SHFT 15 -#define UVH_LOCAL_INT0_ENABLE_GR0_AOERR1_MASK 0x0000000000008000UL -#define UVH_LOCAL_INT0_ENABLE_GR1_AOERR1_SHFT 16 -#define UVH_LOCAL_INT0_ENABLE_GR1_AOERR1_MASK 0x0000000000010000UL -#define UVH_LOCAL_INT0_ENABLE_LH_AOERR1_SHFT 17 -#define UVH_LOCAL_INT0_ENABLE_LH_AOERR1_MASK 0x0000000000020000UL -#define UVH_LOCAL_INT0_ENABLE_RH_AOERR1_SHFT 18 -#define UVH_LOCAL_INT0_ENABLE_RH_AOERR1_MASK 0x0000000000040000UL -#define UVH_LOCAL_INT0_ENABLE_XN_AOERR1_SHFT 19 -#define UVH_LOCAL_INT0_ENABLE_XN_AOERR1_MASK 0x0000000000080000UL -#define UVH_LOCAL_INT0_ENABLE_SI_AOERR1_SHFT 20 -#define UVH_LOCAL_INT0_ENABLE_SI_AOERR1_MASK 0x0000000000100000UL -#define UVH_LOCAL_INT0_ENABLE_RH_VPI_INT_SHFT 21 -#define UVH_LOCAL_INT0_ENABLE_RH_VPI_INT_MASK 0x0000000000200000UL -#define UVH_LOCAL_INT0_ENABLE_SYSTEM_SHUTDOWN_INT_SHFT 22 -#define UVH_LOCAL_INT0_ENABLE_SYSTEM_SHUTDOWN_INT_MASK 0x0000000000400000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_0_SHFT 23 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_0_MASK 0x0000000000800000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_1_SHFT 24 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_1_MASK 0x0000000001000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_2_SHFT 25 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_2_MASK 0x0000000002000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_3_SHFT 26 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_3_MASK 0x0000000004000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_4_SHFT 27 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_4_MASK 0x0000000008000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_5_SHFT 28 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_5_MASK 0x0000000010000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_6_SHFT 29 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_6_MASK 0x0000000020000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_7_SHFT 30 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_7_MASK 0x0000000040000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_8_SHFT 31 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_8_MASK 0x0000000080000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_9_SHFT 32 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_9_MASK 0x0000000100000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_10_SHFT 33 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_10_MASK 0x0000000200000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_11_SHFT 34 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_11_MASK 0x0000000400000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_12_SHFT 35 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_12_MASK 0x0000000800000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_13_SHFT 36 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_13_MASK 0x0000001000000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_14_SHFT 37 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_14_MASK 0x0000002000000000UL -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_15_SHFT 38 -#define UVH_LOCAL_INT0_ENABLE_LB_IRQ_INT_15_MASK 0x0000004000000000UL -#define UVH_LOCAL_INT0_ENABLE_L1_NMI_INT_SHFT 39 -#define UVH_LOCAL_INT0_ENABLE_L1_NMI_INT_MASK 0x0000008000000000UL -#define UVH_LOCAL_INT0_ENABLE_STOP_CLOCK_SHFT 40 -#define UVH_LOCAL_INT0_ENABLE_STOP_CLOCK_MASK 0x0000010000000000UL -#define UVH_LOCAL_INT0_ENABLE_ASIC_TO_L1_SHFT 41 -#define UVH_LOCAL_INT0_ENABLE_ASIC_TO_L1_MASK 0x0000020000000000UL -#define UVH_LOCAL_INT0_ENABLE_L1_TO_ASIC_SHFT 42 -#define UVH_LOCAL_INT0_ENABLE_L1_TO_ASIC_MASK 0x0000040000000000UL -#define UVH_LOCAL_INT0_ENABLE_LTC_INT_SHFT 43 -#define UVH_LOCAL_INT0_ENABLE_LTC_INT_MASK 0x0000080000000000UL -#define UVH_LOCAL_INT0_ENABLE_LA_SEQ_TRIGGER_SHFT 44 -#define UVH_LOCAL_INT0_ENABLE_LA_SEQ_TRIGGER_MASK 0x0000100000000000UL - -union uvh_local_int0_enable_u { - unsigned long v; - struct uvh_local_int0_enable_s { - unsigned long lb_hcerr : 1; /* RW */ - unsigned long gr0_hcerr : 1; /* RW */ - unsigned long gr1_hcerr : 1; /* RW */ - unsigned long lh_hcerr : 1; /* RW */ - unsigned long rh_hcerr : 1; /* RW */ - unsigned long xn_hcerr : 1; /* RW */ - unsigned long si_hcerr : 1; /* RW */ - unsigned long lb_aoerr0 : 1; /* RW */ - unsigned long gr0_aoerr0 : 1; /* RW */ - unsigned long gr1_aoerr0 : 1; /* RW */ - unsigned long lh_aoerr0 : 1; /* RW */ - unsigned long rh_aoerr0 : 1; /* RW */ - unsigned long xn_aoerr0 : 1; /* RW */ - unsigned long si_aoerr0 : 1; /* RW */ - unsigned long lb_aoerr1 : 1; /* RW */ - unsigned long gr0_aoerr1 : 1; /* RW */ - unsigned long gr1_aoerr1 : 1; /* RW */ - unsigned long lh_aoerr1 : 1; /* RW */ - unsigned long rh_aoerr1 : 1; /* RW */ - unsigned long xn_aoerr1 : 1; /* RW */ - unsigned long si_aoerr1 : 1; /* RW */ - unsigned long rh_vpi_int : 1; /* RW */ - unsigned long system_shutdown_int : 1; /* RW */ - unsigned long lb_irq_int_0 : 1; /* RW */ - unsigned long lb_irq_int_1 : 1; /* RW */ - unsigned long lb_irq_int_2 : 1; /* RW */ - unsigned long lb_irq_int_3 : 1; /* RW */ - unsigned long lb_irq_int_4 : 1; /* RW */ - unsigned long lb_irq_int_5 : 1; /* RW */ - unsigned long lb_irq_int_6 : 1; /* RW */ - unsigned long lb_irq_int_7 : 1; /* RW */ - unsigned long lb_irq_int_8 : 1; /* RW */ - unsigned long lb_irq_int_9 : 1; /* RW */ - unsigned long lb_irq_int_10 : 1; /* RW */ - unsigned long lb_irq_int_11 : 1; /* RW */ - unsigned long lb_irq_int_12 : 1; /* RW */ - unsigned long lb_irq_int_13 : 1; /* RW */ - unsigned long lb_irq_int_14 : 1; /* RW */ - unsigned long lb_irq_int_15 : 1; /* RW */ - unsigned long l1_nmi_int : 1; /* RW */ - unsigned long stop_clock : 1; /* RW */ - unsigned long asic_to_l1 : 1; /* RW */ - unsigned long l1_to_asic : 1; /* RW */ - unsigned long ltc_int : 1; /* RW */ - unsigned long la_seq_trigger : 1; /* RW */ - unsigned long rsvd_45_63 : 19; /* */ - } s; -}; - -/* ========================================================================= */ /* UVH_NODE_ID */ /* ========================================================================= */ #define UVH_NODE_ID 0x0UL @@ -1112,26 +857,6 @@ union uvh_rh_gam_alias210_redirect_config_2_mmr_u { }; /* ========================================================================= */ -/* UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR */ -/* ========================================================================= */ -#define UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR 0x1600020UL - -#define UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_BASE_SHFT 26 -#define UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffffc000000UL -#define UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 -#define UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL - -union uvh_rh_gam_cfg_overlay_config_mmr_u { - unsigned long v; - struct uvh_rh_gam_cfg_overlay_config_mmr_s { - unsigned long rsvd_0_25: 26; /* */ - unsigned long base : 20; /* RW */ - unsigned long rsvd_46_62: 17; /* */ - unsigned long enable : 1; /* RW */ - } s; -}; - -/* ========================================================================= */ /* UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR */ /* ========================================================================= */ #define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR 0x1600010UL @@ -1263,101 +988,6 @@ union uvh_rtc1_int_config_u { }; /* ========================================================================= */ -/* UVH_RTC2_INT_CONFIG */ -/* ========================================================================= */ -#define UVH_RTC2_INT_CONFIG 0x61600UL - -#define UVH_RTC2_INT_CONFIG_VECTOR_SHFT 0 -#define UVH_RTC2_INT_CONFIG_VECTOR_MASK 0x00000000000000ffUL -#define UVH_RTC2_INT_CONFIG_DM_SHFT 8 -#define UVH_RTC2_INT_CONFIG_DM_MASK 0x0000000000000700UL -#define UVH_RTC2_INT_CONFIG_DESTMODE_SHFT 11 -#define UVH_RTC2_INT_CONFIG_DESTMODE_MASK 0x0000000000000800UL -#define UVH_RTC2_INT_CONFIG_STATUS_SHFT 12 -#define UVH_RTC2_INT_CONFIG_STATUS_MASK 0x0000000000001000UL -#define UVH_RTC2_INT_CONFIG_P_SHFT 13 -#define UVH_RTC2_INT_CONFIG_P_MASK 0x0000000000002000UL -#define UVH_RTC2_INT_CONFIG_T_SHFT 15 -#define UVH_RTC2_INT_CONFIG_T_MASK 0x0000000000008000UL -#define UVH_RTC2_INT_CONFIG_M_SHFT 16 -#define UVH_RTC2_INT_CONFIG_M_MASK 0x0000000000010000UL -#define UVH_RTC2_INT_CONFIG_APIC_ID_SHFT 32 -#define UVH_RTC2_INT_CONFIG_APIC_ID_MASK 0xffffffff00000000UL - -union uvh_rtc2_int_config_u { - unsigned long v; - struct uvh_rtc2_int_config_s { - unsigned long vector_ : 8; /* RW */ - unsigned long dm : 3; /* RW */ - unsigned long destmode : 1; /* RW */ - unsigned long status : 1; /* RO */ - unsigned long p : 1; /* RO */ - unsigned long rsvd_14 : 1; /* */ - unsigned long t : 1; /* RO */ - unsigned long m : 1; /* RW */ - unsigned long rsvd_17_31: 15; /* */ - unsigned long apic_id : 32; /* RW */ - } s; -}; - -/* ========================================================================= */ -/* UVH_RTC3_INT_CONFIG */ -/* ========================================================================= */ -#define UVH_RTC3_INT_CONFIG 0x61640UL - -#define UVH_RTC3_INT_CONFIG_VECTOR_SHFT 0 -#define UVH_RTC3_INT_CONFIG_VECTOR_MASK 0x00000000000000ffUL -#define UVH_RTC3_INT_CONFIG_DM_SHFT 8 -#define UVH_RTC3_INT_CONFIG_DM_MASK 0x0000000000000700UL -#define UVH_RTC3_INT_CONFIG_DESTMODE_SHFT 11 -#define UVH_RTC3_INT_CONFIG_DESTMODE_MASK 0x0000000000000800UL -#define UVH_RTC3_INT_CONFIG_STATUS_SHFT 12 -#define UVH_RTC3_INT_CONFIG_STATUS_MASK 0x0000000000001000UL -#define UVH_RTC3_INT_CONFIG_P_SHFT 13 -#define UVH_RTC3_INT_CONFIG_P_MASK 0x0000000000002000UL -#define UVH_RTC3_INT_CONFIG_T_SHFT 15 -#define UVH_RTC3_INT_CONFIG_T_MASK 0x0000000000008000UL -#define UVH_RTC3_INT_CONFIG_M_SHFT 16 -#define UVH_RTC3_INT_CONFIG_M_MASK 0x0000000000010000UL -#define UVH_RTC3_INT_CONFIG_APIC_ID_SHFT 32 -#define UVH_RTC3_INT_CONFIG_APIC_ID_MASK 0xffffffff00000000UL - -union uvh_rtc3_int_config_u { - unsigned long v; - struct uvh_rtc3_int_config_s { - unsigned long vector_ : 8; /* RW */ - unsigned long dm : 3; /* RW */ - unsigned long destmode : 1; /* RW */ - unsigned long status : 1; /* RO */ - unsigned long p : 1; /* RO */ - unsigned long rsvd_14 : 1; /* */ - unsigned long t : 1; /* RO */ - unsigned long m : 1; /* RW */ - unsigned long rsvd_17_31: 15; /* */ - unsigned long apic_id : 32; /* RW */ - } s; -}; - -/* ========================================================================= */ -/* UVH_RTC_INC_RATIO */ -/* ========================================================================= */ -#define UVH_RTC_INC_RATIO 0x350000UL - -#define UVH_RTC_INC_RATIO_FRACTION_SHFT 0 -#define UVH_RTC_INC_RATIO_FRACTION_MASK 0x00000000000fffffUL -#define UVH_RTC_INC_RATIO_RATIO_SHFT 20 -#define UVH_RTC_INC_RATIO_RATIO_MASK 0x0000000000700000UL - -union uvh_rtc_inc_ratio_u { - unsigned long v; - struct uvh_rtc_inc_ratio_s { - unsigned long fraction : 20; /* RW */ - unsigned long ratio : 3; /* RW */ - unsigned long rsvd_23_63: 41; /* */ - } s; -}; - -/* ========================================================================= */ /* UVH_SI_ADDR_MAP_CONFIG */ /* ========================================================================= */ #define UVH_SI_ADDR_MAP_CONFIG 0xc80000UL diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 4c58352209e0..e77b22083721 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -47,8 +47,6 @@ obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o obj-y += i387.o xsave.o obj-y += ptrace.o -obj-$(CONFIG_X86_DS) += ds.o -obj-$(CONFIG_X86_DS_SELFTEST) += ds_selftest.o obj-$(CONFIG_X86_32) += tls.o obj-$(CONFIG_IA32_EMULATION) += tls.o obj-y += step.o diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index f3dadb571d9b..f854d89b7edf 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -118,7 +118,7 @@ static bool check_device(struct device *dev) return false; /* No device or no PCI device */ - if (!dev || dev->bus != &pci_bus_type) + if (dev->bus != &pci_bus_type) return false; devid = get_device_id(dev); @@ -392,6 +392,7 @@ static int __iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) u32 tail, head; u8 *target; + WARN_ON(iommu->cmd_buf_size & CMD_BUFFER_UNINITIALIZED); tail = readl(iommu->mmio_base + MMIO_CMD_TAIL_OFFSET); target = iommu->cmd_buf + tail; memcpy_toio(target, cmd, sizeof(*cmd)); @@ -2186,7 +2187,7 @@ static void prealloc_protection_domains(void) struct dma_ops_domain *dma_dom; u16 devid; - while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + for_each_pci_dev(dev) { /* Do we handle this device? */ if (!check_device(&dev->dev)) @@ -2298,7 +2299,7 @@ static void cleanup_domain(struct protection_domain *domain) list_for_each_entry_safe(dev_data, next, &domain->dev_list, list) { struct device *dev = dev_data->dev; - do_detach(dev); + __detach_device(dev); atomic_set(&dev_data->bind, 0); } @@ -2327,6 +2328,7 @@ static struct protection_domain *protection_domain_alloc(void) return NULL; spin_lock_init(&domain->lock); + mutex_init(&domain->api_lock); domain->id = domain_id_alloc(); if (!domain->id) goto out_err; @@ -2379,9 +2381,7 @@ static void amd_iommu_domain_destroy(struct iommu_domain *dom) free_pagetable(domain); - domain_id_free(domain->id); - - kfree(domain); + protection_domain_free(domain); dom->priv = NULL; } @@ -2456,6 +2456,8 @@ static int amd_iommu_map_range(struct iommu_domain *dom, iova &= PAGE_MASK; paddr &= PAGE_MASK; + mutex_lock(&domain->api_lock); + for (i = 0; i < npages; ++i) { ret = iommu_map_page(domain, iova, paddr, prot, PM_MAP_4k); if (ret) @@ -2465,6 +2467,8 @@ static int amd_iommu_map_range(struct iommu_domain *dom, paddr += PAGE_SIZE; } + mutex_unlock(&domain->api_lock); + return 0; } @@ -2477,12 +2481,16 @@ static void amd_iommu_unmap_range(struct iommu_domain *dom, iova &= PAGE_MASK; + mutex_lock(&domain->api_lock); + for (i = 0; i < npages; ++i) { iommu_unmap_page(domain, iova, PM_MAP_4k); iova += PAGE_SIZE; } iommu_flush_tlb_pde(domain); + + mutex_unlock(&domain->api_lock); } static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom, diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 42f5350b908f..6360abf993d4 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -138,9 +138,9 @@ int amd_iommus_present; bool amd_iommu_np_cache __read_mostly; /* - * Set to true if ACPI table parsing and hardware intialization went properly + * The ACPI table parsing functions set this variable on an error */ -static bool amd_iommu_initialized; +static int __initdata amd_iommu_init_err; /* * List of protection domains - used during resume @@ -391,9 +391,11 @@ static int __init find_last_devid_acpi(struct acpi_table_header *table) */ for (i = 0; i < table->length; ++i) checksum += p[i]; - if (checksum != 0) + if (checksum != 0) { /* ACPI table corrupt */ - return -ENODEV; + amd_iommu_init_err = -ENODEV; + return 0; + } p += IVRS_HEADER_LENGTH; @@ -436,7 +438,7 @@ static u8 * __init alloc_command_buffer(struct amd_iommu *iommu) if (cmd_buf == NULL) return NULL; - iommu->cmd_buf_size = CMD_BUFFER_SIZE; + iommu->cmd_buf_size = CMD_BUFFER_SIZE | CMD_BUFFER_UNINITIALIZED; return cmd_buf; } @@ -472,12 +474,13 @@ static void iommu_enable_command_buffer(struct amd_iommu *iommu) &entry, sizeof(entry)); amd_iommu_reset_cmd_buffer(iommu); + iommu->cmd_buf_size &= ~(CMD_BUFFER_UNINITIALIZED); } static void __init free_command_buffer(struct amd_iommu *iommu) { free_pages((unsigned long)iommu->cmd_buf, - get_order(iommu->cmd_buf_size)); + get_order(iommu->cmd_buf_size & ~(CMD_BUFFER_UNINITIALIZED))); } /* allocates the memory where the IOMMU will log its events to */ @@ -920,11 +923,16 @@ static int __init init_iommu_all(struct acpi_table_header *table) h->mmio_phys); iommu = kzalloc(sizeof(struct amd_iommu), GFP_KERNEL); - if (iommu == NULL) - return -ENOMEM; + if (iommu == NULL) { + amd_iommu_init_err = -ENOMEM; + return 0; + } + ret = init_iommu_one(iommu, h); - if (ret) - return ret; + if (ret) { + amd_iommu_init_err = ret; + return 0; + } break; default: break; @@ -934,8 +942,6 @@ static int __init init_iommu_all(struct acpi_table_header *table) } WARN_ON(p != end); - amd_iommu_initialized = true; - return 0; } @@ -1211,6 +1217,10 @@ static int __init amd_iommu_init(void) if (acpi_table_parse("IVRS", find_last_devid_acpi) != 0) return -ENODEV; + ret = amd_iommu_init_err; + if (ret) + goto out; + dev_table_size = tbl_size(DEV_TABLE_ENTRY_SIZE); alias_table_size = tbl_size(ALIAS_TABLE_ENTRY_SIZE); rlookup_table_size = tbl_size(RLOOKUP_TABLE_ENTRY_SIZE); @@ -1270,12 +1280,19 @@ static int __init amd_iommu_init(void) if (acpi_table_parse("IVRS", init_iommu_all) != 0) goto free; - if (!amd_iommu_initialized) + if (amd_iommu_init_err) { + ret = amd_iommu_init_err; goto free; + } if (acpi_table_parse("IVRS", init_memory_definitions) != 0) goto free; + if (amd_iommu_init_err) { + ret = amd_iommu_init_err; + goto free; + } + ret = sysdev_class_register(&amd_iommu_sysdev_class); if (ret) goto free; @@ -1288,6 +1305,8 @@ static int __init amd_iommu_init(void) if (ret) goto free; + enable_iommus(); + if (iommu_pass_through) ret = amd_iommu_init_passthrough(); else @@ -1300,8 +1319,6 @@ static int __init amd_iommu_init(void) amd_iommu_init_notifier(); - enable_iommus(); - if (iommu_pass_through) goto out; @@ -1315,6 +1332,7 @@ out: return ret; free: + disable_iommus(); amd_iommu_uninit_devices(); diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 3704997e8b25..b5d8b0bcf235 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -393,6 +393,7 @@ void __init gart_iommu_hole_init(void) for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) { int bus; int dev_base, dev_limit; + u32 ctl; bus = bus_dev_ranges[i].bus; dev_base = bus_dev_ranges[i].dev_base; @@ -406,7 +407,19 @@ void __init gart_iommu_hole_init(void) gart_iommu_aperture = 1; x86_init.iommu.iommu_init = gart_iommu_init; - aper_order = (read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL) >> 1) & 7; + ctl = read_pci_config(bus, slot, 3, + AMD64_GARTAPERTURECTL); + + /* + * Before we do anything else disable the GART. It may + * still be enabled if we boot into a crash-kernel here. + * Reconfiguring the GART while it is enabled could have + * unknown side-effects. + */ + ctl &= ~GARTEN; + write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl); + + aper_order = (ctl >> 1) & 7; aper_size = (32 * 1024 * 1024) << aper_order; aper_base = read_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE) & 0x7fff; aper_base <<= 25; diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile index 565c1bfc507d..1a4512e48d24 100644 --- a/arch/x86/kernel/apic/Makefile +++ b/arch/x86/kernel/apic/Makefile @@ -2,7 +2,12 @@ # Makefile for local APIC drivers and for the IO-APIC code # -obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o nmi.o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o +ifneq ($(CONFIG_NMI_WATCHDOG),y) +obj-$(CONFIG_X86_LOCAL_APIC) += nmi.o +endif +obj-$(CONFIG_NMI_WATCHDOG) += hw_nmi.o + obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_SMP) += ipi.o diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c new file mode 100644 index 000000000000..e8b78a0be5de --- /dev/null +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -0,0 +1,127 @@ +/* + * HW NMI watchdog support + * + * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc. + * + * Arch specific calls to support NMI watchdog + * + * Bits copied from original nmi.c file + * + */ + +#include <asm/apic.h> +#include <linux/smp.h> +#include <linux/cpumask.h> +#include <linux/sched.h> +#include <linux/percpu.h> +#include <linux/cpumask.h> +#include <linux/kernel_stat.h> +#include <asm/mce.h> + +#include <linux/nmi.h> +#include <linux/module.h> + +/* For reliability, we're prepared to waste bits here. */ +static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly; + +static DEFINE_PER_CPU(unsigned, last_irq_sum); + +/* + * Take the local apic timer and PIT/HPET into account. We don't + * know which one is active, when we have highres/dyntick on + */ +static inline unsigned int get_timer_irqs(int cpu) +{ + unsigned int irqs = per_cpu(irq_stat, cpu).irq0_irqs; + +#if defined(CONFIG_X86_LOCAL_APIC) + irqs += per_cpu(irq_stat, cpu).apic_timer_irqs; +#endif + + return irqs; +} + +static inline int mce_in_progress(void) +{ +#if defined(CONFIG_X86_MCE) + return atomic_read(&mce_entry) > 0; +#endif + return 0; +} + +int hw_nmi_is_cpu_stuck(struct pt_regs *regs) +{ + unsigned int sum; + int cpu = smp_processor_id(); + + /* FIXME: cheap hack for this check, probably should get its own + * die_notifier handler + */ + if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu); + show_regs(regs); + dump_stack(); + spin_unlock(&lock); + cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask)); + } + + /* if we are doing an mce, just assume the cpu is not stuck */ + /* Could check oops_in_progress here too, but it's safer not to */ + if (mce_in_progress()) + return 0; + + /* We determine if the cpu is stuck by checking whether any + * interrupts have happened since we last checked. Of course + * an nmi storm could create false positives, but the higher + * level logic should account for that + */ + sum = get_timer_irqs(cpu); + if (__get_cpu_var(last_irq_sum) == sum) { + return 1; + } else { + __get_cpu_var(last_irq_sum) = sum; + return 0; + } +} + +u64 hw_nmi_get_sample_period(void) +{ + return cpu_khz * 1000; +} + +#ifdef ARCH_HAS_NMI_WATCHDOG +void arch_trigger_all_cpu_backtrace(void) +{ + int i; + + cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask); + + printk(KERN_INFO "sending NMI to all CPUs:\n"); + apic->send_IPI_all(NMI_VECTOR); + + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpumask_empty(to_cpumask(backtrace_mask))) + break; + mdelay(1); + } +} +#endif + +/* STUB calls to mimic old nmi_watchdog behaviour */ +#if defined(CONFIG_X86_LOCAL_APIC) +unsigned int nmi_watchdog = NMI_NONE; +EXPORT_SYMBOL(nmi_watchdog); +void acpi_nmi_enable(void) { return; } +void acpi_nmi_disable(void) { return; } +#endif +atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ +EXPORT_SYMBOL(nmi_active); +int unknown_nmi_panic; +void cpu_nmi_set_wd_enabled(void) { return; } +void stop_apic_nmi_watchdog(void *unused) { return; } +void setup_apic_nmi_watchdog(void *unused) { return; } +int __init check_nmi_watchdog(void) { return 0; } diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index 1edaf15c0b8e..a43f71cb30f8 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -401,13 +401,6 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) int cpu = smp_processor_id(); int rc = 0; - /* check for other users first */ - if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) - == NOTIFY_STOP) { - rc = 1; - touched = 1; - } - sum = get_timer_irqs(cpu); if (__get_cpu_var(nmi_touch)) { diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index c085d52dbaf2..e46f98f36e31 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -735,9 +735,6 @@ void __init uv_system_init(void) uv_node_to_blade[nid] = blade; uv_cpu_to_blade[cpu] = blade; max_pnode = max(pnode, max_pnode); - - printk(KERN_DEBUG "UV: cpu %d, apicid 0x%x, pnode %d, nid %d, lcpu %d, blade %d\n", - cpu, apicid, pnode, nid, lcpu, blade); } /* Add blade/pnode info for nodes without cpus */ diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 031aa887b0eb..c4f9182ca3ac 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1224,7 +1224,7 @@ static void reinit_timer(void) #ifdef INIT_TIMER_AFTER_SUSPEND unsigned long flags; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); /* set the clock to HZ */ outb_pit(0x34, PIT_MODE); /* binary, mode 2, LSB/MSB, ch 0 */ udelay(10); @@ -1232,7 +1232,7 @@ static void reinit_timer(void) udelay(10); outb_pit(LATCH >> 8, PIT_CH0); /* MSB */ udelay(10); - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); #endif } diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c index 97ad79cdf688..fd1fc1902a47 100644 --- a/arch/x86/kernel/cpu/addon_cpuid_features.c +++ b/arch/x86/kernel/cpu/addon_cpuid_features.c @@ -30,8 +30,9 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) const struct cpuid_bit *cb; static const struct cpuid_bit __cpuinitconst cpuid_bits[] = { - { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 }, - { X86_FEATURE_ARAT, CR_EAX, 2, 0x00000006 }, + { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 }, + { X86_FEATURE_ARAT, CR_EAX, 2, 0x00000006 }, + { X86_FEATURE_CPB, CR_EDX, 9, 0x80000007 }, { X86_FEATURE_NPT, CR_EDX, 0, 0x8000000a }, { X86_FEATURE_LBRV, CR_EDX, 1, 0x8000000a }, { X86_FEATURE_SVML, CR_EDX, 2, 0x8000000a }, @@ -53,6 +54,14 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) if (regs[cb->reg] & (1 << cb->bit)) set_cpu_cap(c, cb->feature); } + + /* + * common AMD/Intel features + */ + if (c->cpuid_level >= 6) { + if (cpuid_ecx(6) & 0x1) + set_cpu_cap(c, X86_FEATURE_APERFMPERF); + } } /* leaf 0xb SMT level */ diff --git a/arch/x86/kernel/cpu/cpufreq/Makefile b/arch/x86/kernel/cpu/cpufreq/Makefile index 1840c0a5170b..bd54bf67e6fb 100644 --- a/arch/x86/kernel/cpu/cpufreq/Makefile +++ b/arch/x86/kernel/cpu/cpufreq/Makefile @@ -2,8 +2,8 @@ # K8 systems. ACPI is preferred to all other hardware-specific drivers. # speedstep-* is preferred over p4-clockmod. -obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o -obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o +obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o mperf.o +obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o mperf.o obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o obj-$(CONFIG_X86_POWERNOW_K7) += powernow-k7.o diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 459168083b77..1d3cddaa40ee 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -46,6 +46,7 @@ #include <asm/msr.h> #include <asm/processor.h> #include <asm/cpufeature.h> +#include "mperf.h" #define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ "acpi-cpufreq", msg) @@ -71,8 +72,6 @@ struct acpi_cpufreq_data { static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data); -static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf); - /* acpi_perf_data is a pointer to percpu data. */ static struct acpi_processor_performance *acpi_perf_data; @@ -240,45 +239,6 @@ static u32 get_cur_val(const struct cpumask *mask) return cmd.val; } -/* Called via smp_call_function_single(), on the target CPU */ -static void read_measured_perf_ctrs(void *_cur) -{ - struct aperfmperf *am = _cur; - - get_aperfmperf(am); -} - -/* - * Return the measured active (C0) frequency on this CPU since last call - * to this function. - * Input: cpu number - * Return: Average CPU frequency in terms of max frequency (zero on error) - * - * We use IA32_MPERF and IA32_APERF MSRs to get the measured performance - * over a period of time, while CPU is in C0 state. - * IA32_MPERF counts at the rate of max advertised frequency - * IA32_APERF counts at the rate of actual CPU frequency - * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and - * no meaning should be associated with absolute values of these MSRs. - */ -static unsigned int get_measured_perf(struct cpufreq_policy *policy, - unsigned int cpu) -{ - struct aperfmperf perf; - unsigned long ratio; - unsigned int retval; - - if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) - return 0; - - ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf); - per_cpu(acfreq_old_perf, cpu) = perf; - - retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; - - return retval; -} - static unsigned int get_cur_freq_on_cpu(unsigned int cpu) { struct acpi_cpufreq_data *data = per_cpu(acfreq_data, cpu); @@ -702,7 +662,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) /* Check for APERF/MPERF support in hardware */ if (cpu_has(c, X86_FEATURE_APERFMPERF)) - acpi_cpufreq_driver.getavg = get_measured_perf; + acpi_cpufreq_driver.getavg = cpufreq_get_measured_perf; dprintk("CPU%u - ACPI performance management activated.\n", cpu); for (i = 0; i < perf->state_count; i++) diff --git a/arch/x86/kernel/cpu/cpufreq/mperf.c b/arch/x86/kernel/cpu/cpufreq/mperf.c new file mode 100644 index 000000000000..911e193018ae --- /dev/null +++ b/arch/x86/kernel/cpu/cpufreq/mperf.c @@ -0,0 +1,51 @@ +#include <linux/kernel.h> +#include <linux/smp.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/cpufreq.h> +#include <linux/slab.h> + +#include "mperf.h" + +static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf); + +/* Called via smp_call_function_single(), on the target CPU */ +static void read_measured_perf_ctrs(void *_cur) +{ + struct aperfmperf *am = _cur; + + get_aperfmperf(am); +} + +/* + * Return the measured active (C0) frequency on this CPU since last call + * to this function. + * Input: cpu number + * Return: Average CPU frequency in terms of max frequency (zero on error) + * + * We use IA32_MPERF and IA32_APERF MSRs to get the measured performance + * over a period of time, while CPU is in C0 state. + * IA32_MPERF counts at the rate of max advertised frequency + * IA32_APERF counts at the rate of actual CPU frequency + * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and + * no meaning should be associated with absolute values of these MSRs. + */ +unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, + unsigned int cpu) +{ + struct aperfmperf perf; + unsigned long ratio; + unsigned int retval; + + if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) + return 0; + + ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf); + per_cpu(acfreq_old_perf, cpu) = perf; + + retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; + + return retval; +} +EXPORT_SYMBOL_GPL(cpufreq_get_measured_perf); +MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/mperf.h b/arch/x86/kernel/cpu/cpufreq/mperf.h new file mode 100644 index 000000000000..5dbf2950dc22 --- /dev/null +++ b/arch/x86/kernel/cpu/cpufreq/mperf.h @@ -0,0 +1,9 @@ +/* + * (c) 2010 Advanced Micro Devices, Inc. + * Your use of this code is subject to the terms and conditions of the + * GNU general public license version 2. See "COPYING" or + * http://www.gnu.org/licenses/gpl.html + */ + +unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, + unsigned int cpu); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index d360b56e9825..6f3dc8fbbfdc 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -1,6 +1,5 @@ - /* - * (c) 2003-2006 Advanced Micro Devices, Inc. + * (c) 2003-2010 Advanced Micro Devices, Inc. * Your use of this code is subject to the terms and conditions of the * GNU general public license version 2. See "COPYING" or * http://www.gnu.org/licenses/gpl.html @@ -46,6 +45,7 @@ #define PFX "powernow-k8: " #define VERSION "version 2.20.00" #include "powernow-k8.h" +#include "mperf.h" /* serialize freq changes */ static DEFINE_MUTEX(fidvid_mutex); @@ -54,6 +54,12 @@ static DEFINE_PER_CPU(struct powernow_k8_data *, powernow_data); static int cpu_family = CPU_OPTERON; +/* core performance boost */ +static bool cpb_capable, cpb_enabled; +static struct msr __percpu *msrs; + +static struct cpufreq_driver cpufreq_amd64_driver; + #ifndef CONFIG_SMP static inline const struct cpumask *cpu_core_mask(int cpu) { @@ -929,7 +935,8 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data, powernow_table[i].index = index; /* Frequency may be rounded for these */ - if (boot_cpu_data.x86 == 0x10 || boot_cpu_data.x86 == 0x11) { + if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10) + || boot_cpu_data.x86 == 0x11) { powernow_table[i].frequency = freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7); } else @@ -1248,6 +1255,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) struct powernow_k8_data *data; struct init_on_cpu init_on_cpu; int rc; + struct cpuinfo_x86 *c = &cpu_data(pol->cpu); if (!cpu_online(pol->cpu)) return -ENODEV; @@ -1322,6 +1330,10 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) return -EINVAL; } + /* Check for APERF/MPERF support in hardware */ + if (cpu_has(c, X86_FEATURE_APERFMPERF)) + cpufreq_amd64_driver.getavg = cpufreq_get_measured_perf; + cpufreq_frequency_table_get_attr(data->powernow_table, pol->cpu); if (cpu_family == CPU_HW_PSTATE) @@ -1393,8 +1405,77 @@ out: return khz; } +static void _cpb_toggle_msrs(bool t) +{ + int cpu; + + get_online_cpus(); + + rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); + + for_each_cpu(cpu, cpu_online_mask) { + struct msr *reg = per_cpu_ptr(msrs, cpu); + if (t) + reg->l &= ~BIT(25); + else + reg->l |= BIT(25); + } + wrmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); + + put_online_cpus(); +} + +/* + * Switch on/off core performance boosting. + * + * 0=disable + * 1=enable. + */ +static void cpb_toggle(bool t) +{ + if (!cpb_capable) + return; + + if (t && !cpb_enabled) { + cpb_enabled = true; + _cpb_toggle_msrs(t); + printk(KERN_INFO PFX "Core Boosting enabled.\n"); + } else if (!t && cpb_enabled) { + cpb_enabled = false; + _cpb_toggle_msrs(t); + printk(KERN_INFO PFX "Core Boosting disabled.\n"); + } +} + +static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf, + size_t count) +{ + int ret = -EINVAL; + unsigned long val = 0; + + ret = strict_strtoul(buf, 10, &val); + if (!ret && (val == 0 || val == 1) && cpb_capable) + cpb_toggle(val); + else + return -EINVAL; + + return count; +} + +static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf) +{ + return sprintf(buf, "%u\n", cpb_enabled); +} + +#define define_one_rw(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0644, show_##_name, store_##_name) + +define_one_rw(cpb); + static struct freq_attr *powernow_k8_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, + &cpb, NULL, }; @@ -1410,10 +1491,51 @@ static struct cpufreq_driver cpufreq_amd64_driver = { .attr = powernow_k8_attr, }; +/* + * Clear the boost-disable flag on the CPU_DOWN path so that this cpu + * cannot block the remaining ones from boosting. On the CPU_UP path we + * simply keep the boost-disable flag in sync with the current global + * state. + */ +static int __cpuinit cpb_notify(struct notifier_block *nb, unsigned long action, + void *hcpu) +{ + unsigned cpu = (long)hcpu; + u32 lo, hi; + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + + if (!cpb_enabled) { + rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); + lo |= BIT(25); + wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); + } + break; + + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); + lo &= ~BIT(25); + wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); + break; + + default: + break; + } + + return NOTIFY_OK; +} + +static struct notifier_block __cpuinitdata cpb_nb = { + .notifier_call = cpb_notify, +}; + /* driver entry point for init */ static int __cpuinit powernowk8_init(void) { - unsigned int i, supported_cpus = 0; + unsigned int i, supported_cpus = 0, cpu; for_each_online_cpu(i) { int rc; @@ -1422,15 +1544,36 @@ static int __cpuinit powernowk8_init(void) supported_cpus++; } - if (supported_cpus == num_online_cpus()) { - printk(KERN_INFO PFX "Found %d %s " - "processors (%d cpu cores) (" VERSION ")\n", - num_online_nodes(), - boot_cpu_data.x86_model_id, supported_cpus); - return cpufreq_register_driver(&cpufreq_amd64_driver); + if (supported_cpus != num_online_cpus()) + return -ENODEV; + + printk(KERN_INFO PFX "Found %d %s (%d cpu cores) (" VERSION ")\n", + num_online_nodes(), boot_cpu_data.x86_model_id, supported_cpus); + + if (boot_cpu_has(X86_FEATURE_CPB)) { + + cpb_capable = true; + + register_cpu_notifier(&cpb_nb); + + msrs = msrs_alloc(); + if (!msrs) { + printk(KERN_ERR "%s: Error allocating msrs!\n", __func__); + return -ENOMEM; + } + + rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); + + for_each_cpu(cpu, cpu_online_mask) { + struct msr *reg = per_cpu_ptr(msrs, cpu); + cpb_enabled |= !(!!(reg->l & BIT(25))); + } + + printk(KERN_INFO PFX "Core Performance Boosting: %s.\n", + (cpb_enabled ? "on" : "off")); } - return -ENODEV; + return cpufreq_register_driver(&cpufreq_amd64_driver); } /* driver entry point for term */ @@ -1438,6 +1581,13 @@ static void __exit powernowk8_exit(void) { dprintk("exit\n"); + if (boot_cpu_has(X86_FEATURE_CPB)) { + msrs_free(msrs); + msrs = NULL; + + unregister_cpu_notifier(&cpb_nb); + } + cpufreq_unregister_driver(&cpufreq_amd64_driver); } diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h index 02ce824073cb..df3529b1c02d 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h @@ -5,7 +5,6 @@ * http://www.gnu.org/licenses/gpl.html */ - enum pstate { HW_PSTATE_INVALID = 0xff, HW_PSTATE_0 = 0, @@ -55,7 +54,6 @@ struct powernow_k8_data { struct cpumask *available_cores; }; - /* processor's cpuid instruction support */ #define CPUID_PROCESSOR_SIGNATURE 1 /* function 1 */ #define CPUID_XFAM 0x0ff00000 /* extended family */ diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 7e1cca13af35..8b2265732ce8 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -12,7 +12,6 @@ #include <asm/processor.h> #include <asm/pgtable.h> #include <asm/msr.h> -#include <asm/ds.h> #include <asm/bugs.h> #include <asm/cpu.h> @@ -352,12 +351,6 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON); } - if (c->cpuid_level > 6) { - unsigned ecx = cpuid_ecx(6); - if (ecx & 0x01) - set_cpu_cap(c, X86_FEATURE_APERFMPERF); - } - if (cpu_has_xmm2) set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); if (cpu_has_ds) { @@ -367,7 +360,6 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_BTS); if (!(l1 & (1<<12))) set_cpu_cap(c, X86_FEATURE_PEBS); - ds_init_intel(c); } if (c->x86 == 6 && c->x86_model == 29 && cpu_has_clflush) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index db5bdc8addf8..626154a9f535 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -31,46 +31,51 @@ #include <asm/nmi.h> #include <asm/compat.h> -static u64 perf_event_mask __read_mostly; +#if 0 +#undef wrmsrl +#define wrmsrl(msr, val) \ +do { \ + trace_printk("wrmsrl(%lx, %lx)\n", (unsigned long)(msr),\ + (unsigned long)(val)); \ + native_write_msr((msr), (u32)((u64)(val)), \ + (u32)((u64)(val) >> 32)); \ +} while (0) +#endif -/* The maximal number of PEBS events: */ -#define MAX_PEBS_EVENTS 4 +/* + * best effort, GUP based copy_from_user() that assumes IRQ or NMI context + */ +static unsigned long +copy_from_user_nmi(void *to, const void __user *from, unsigned long n) +{ + unsigned long offset, addr = (unsigned long)from; + int type = in_nmi() ? KM_NMI : KM_IRQ0; + unsigned long size, len = 0; + struct page *page; + void *map; + int ret; -/* The size of a BTS record in bytes: */ -#define BTS_RECORD_SIZE 24 + do { + ret = __get_user_pages_fast(addr, 1, 0, &page); + if (!ret) + break; -/* The size of a per-cpu BTS buffer in bytes: */ -#define BTS_BUFFER_SIZE (BTS_RECORD_SIZE * 2048) + offset = addr & (PAGE_SIZE - 1); + size = min(PAGE_SIZE - offset, n - len); -/* The BTS overflow threshold in bytes from the end of the buffer: */ -#define BTS_OVFL_TH (BTS_RECORD_SIZE * 128) + map = kmap_atomic(page, type); + memcpy(to, map+offset, size); + kunmap_atomic(map, type); + put_page(page); + len += size; + to += size; + addr += size; -/* - * Bits in the debugctlmsr controlling branch tracing. - */ -#define X86_DEBUGCTL_TR (1 << 6) -#define X86_DEBUGCTL_BTS (1 << 7) -#define X86_DEBUGCTL_BTINT (1 << 8) -#define X86_DEBUGCTL_BTS_OFF_OS (1 << 9) -#define X86_DEBUGCTL_BTS_OFF_USR (1 << 10) + } while (len < n); -/* - * A debug store configuration. - * - * We only support architectures that use 64bit fields. - */ -struct debug_store { - u64 bts_buffer_base; - u64 bts_index; - u64 bts_absolute_maximum; - u64 bts_interrupt_threshold; - u64 pebs_buffer_base; - u64 pebs_index; - u64 pebs_absolute_maximum; - u64 pebs_interrupt_threshold; - u64 pebs_event_reset[MAX_PEBS_EVENTS]; -}; + return len; +} struct event_constraint { union { @@ -89,18 +94,39 @@ struct amd_nb { struct event_constraint event_constraints[X86_PMC_IDX_MAX]; }; +#define MAX_LBR_ENTRIES 16 + struct cpu_hw_events { + /* + * Generic x86 PMC bits + */ struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - unsigned long interrupts; int enabled; - struct debug_store *ds; int n_events; int n_added; int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ u64 tags[X86_PMC_IDX_MAX]; struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ + + /* + * Intel DebugStore bits + */ + struct debug_store *ds; + u64 pebs_enabled; + + /* + * Intel LBR bits + */ + int lbr_users; + void *lbr_context; + struct perf_branch_stack lbr_stack; + struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; + + /* + * AMD specific bits + */ struct amd_nb *amd_nb; }; @@ -114,11 +140,31 @@ struct cpu_hw_events { #define EVENT_CONSTRAINT(c, n, m) \ __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) +/* + * Constraint on the Event code. + */ #define INTEL_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVTSEL_MASK) + EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT) +/* + * Constraint on the Event code + UMask + fixed-mask + * + * filter mask to validate fixed counter events. + * the following filters disqualify for fixed counters: + * - inv + * - edge + * - cnt-mask + * The other filters are supported by fixed counters. + * The any-thread option is supported starting with v3. + */ #define FIXED_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, (1ULL << (32+n)), INTEL_ARCH_FIXED_MASK) + EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK) + +/* + * Constraint on the Event code + UMask + */ +#define PEBS_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK) #define EVENT_CONSTRAINT_END \ EVENT_CONSTRAINT(0, 0, 0) @@ -126,32 +172,43 @@ struct cpu_hw_events { #define for_each_event_constraint(e, c) \ for ((e) = (c); (e)->cmask; (e)++) +union perf_capabilities { + struct { + u64 lbr_format : 6; + u64 pebs_trap : 1; + u64 pebs_arch_reg : 1; + u64 pebs_format : 4; + u64 smm_freeze : 1; + }; + u64 capabilities; +}; + /* * struct x86_pmu - generic x86 pmu */ struct x86_pmu { + /* + * Generic x86 PMC bits + */ const char *name; int version; int (*handle_irq)(struct pt_regs *); void (*disable_all)(void); - void (*enable_all)(void); + void (*enable_all)(int added); void (*enable)(struct perf_event *); void (*disable)(struct perf_event *); + int (*hw_config)(struct perf_event *event); + int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); unsigned eventsel; unsigned perfctr; u64 (*event_map)(int); - u64 (*raw_event)(u64); int max_events; - int num_events; - int num_events_fixed; - int event_bits; - u64 event_mask; + int num_counters; + int num_counters_fixed; + int cntval_bits; + u64 cntval_mask; int apic; u64 max_period; - u64 intel_ctrl; - void (*enable_bts)(u64 config); - void (*disable_bts)(void); - struct event_constraint * (*get_event_constraints)(struct cpu_hw_events *cpuc, struct perf_event *event); @@ -159,11 +216,32 @@ struct x86_pmu { void (*put_event_constraints)(struct cpu_hw_events *cpuc, struct perf_event *event); struct event_constraint *event_constraints; + void (*quirks)(void); int (*cpu_prepare)(int cpu); void (*cpu_starting)(int cpu); void (*cpu_dying)(int cpu); void (*cpu_dead)(int cpu); + + /* + * Intel Arch Perfmon v2+ + */ + u64 intel_ctrl; + union perf_capabilities intel_cap; + + /* + * Intel DebugStore bits + */ + int bts, pebs; + int pebs_record_size; + void (*drain_pebs)(struct pt_regs *regs); + struct event_constraint *pebs_constraints; + + /* + * Intel LBR + */ + unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */ + int lbr_nr; /* hardware stack size */ }; static struct x86_pmu x86_pmu __read_mostly; @@ -198,7 +276,7 @@ static u64 x86_perf_event_update(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; - int shift = 64 - x86_pmu.event_bits; + int shift = 64 - x86_pmu.cntval_bits; u64 prev_raw_count, new_raw_count; int idx = hwc->idx; s64 delta; @@ -241,33 +319,32 @@ again: static atomic_t active_events; static DEFINE_MUTEX(pmc_reserve_mutex); +#ifdef CONFIG_X86_LOCAL_APIC + static bool reserve_pmc_hardware(void) { -#ifdef CONFIG_X86_LOCAL_APIC int i; if (nmi_watchdog == NMI_LOCAL_APIC) disable_lapic_nmi_watchdog(); - for (i = 0; i < x86_pmu.num_events; i++) { + for (i = 0; i < x86_pmu.num_counters; i++) { if (!reserve_perfctr_nmi(x86_pmu.perfctr + i)) goto perfctr_fail; } - for (i = 0; i < x86_pmu.num_events; i++) { + for (i = 0; i < x86_pmu.num_counters; i++) { if (!reserve_evntsel_nmi(x86_pmu.eventsel + i)) goto eventsel_fail; } -#endif return true; -#ifdef CONFIG_X86_LOCAL_APIC eventsel_fail: for (i--; i >= 0; i--) release_evntsel_nmi(x86_pmu.eventsel + i); - i = x86_pmu.num_events; + i = x86_pmu.num_counters; perfctr_fail: for (i--; i >= 0; i--) @@ -277,128 +354,36 @@ perfctr_fail: enable_lapic_nmi_watchdog(); return false; -#endif } static void release_pmc_hardware(void) { -#ifdef CONFIG_X86_LOCAL_APIC int i; - for (i = 0; i < x86_pmu.num_events; i++) { + for (i = 0; i < x86_pmu.num_counters; i++) { release_perfctr_nmi(x86_pmu.perfctr + i); release_evntsel_nmi(x86_pmu.eventsel + i); } if (nmi_watchdog == NMI_LOCAL_APIC) enable_lapic_nmi_watchdog(); -#endif -} - -static inline bool bts_available(void) -{ - return x86_pmu.enable_bts != NULL; -} - -static void init_debug_store_on_cpu(int cpu) -{ - struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; - - if (!ds) - return; - - wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, - (u32)((u64)(unsigned long)ds), - (u32)((u64)(unsigned long)ds >> 32)); -} - -static void fini_debug_store_on_cpu(int cpu) -{ - if (!per_cpu(cpu_hw_events, cpu).ds) - return; - - wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, 0, 0); -} - -static void release_bts_hardware(void) -{ - int cpu; - - if (!bts_available()) - return; - - get_online_cpus(); - - for_each_online_cpu(cpu) - fini_debug_store_on_cpu(cpu); - - for_each_possible_cpu(cpu) { - struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; - - if (!ds) - continue; - - per_cpu(cpu_hw_events, cpu).ds = NULL; - - kfree((void *)(unsigned long)ds->bts_buffer_base); - kfree(ds); - } - - put_online_cpus(); } -static int reserve_bts_hardware(void) -{ - int cpu, err = 0; - - if (!bts_available()) - return 0; - - get_online_cpus(); - - for_each_possible_cpu(cpu) { - struct debug_store *ds; - void *buffer; - - err = -ENOMEM; - buffer = kzalloc(BTS_BUFFER_SIZE, GFP_KERNEL); - if (unlikely(!buffer)) - break; - - ds = kzalloc(sizeof(*ds), GFP_KERNEL); - if (unlikely(!ds)) { - kfree(buffer); - break; - } - - ds->bts_buffer_base = (u64)(unsigned long)buffer; - ds->bts_index = ds->bts_buffer_base; - ds->bts_absolute_maximum = - ds->bts_buffer_base + BTS_BUFFER_SIZE; - ds->bts_interrupt_threshold = - ds->bts_absolute_maximum - BTS_OVFL_TH; +#else - per_cpu(cpu_hw_events, cpu).ds = ds; - err = 0; - } +static bool reserve_pmc_hardware(void) { return true; } +static void release_pmc_hardware(void) {} - if (err) - release_bts_hardware(); - else { - for_each_online_cpu(cpu) - init_debug_store_on_cpu(cpu); - } - - put_online_cpus(); +#endif - return err; -} +static int reserve_ds_buffers(void); +static void release_ds_buffers(void); static void hw_perf_event_destroy(struct perf_event *event) { if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) { release_pmc_hardware(); - release_bts_hardware(); + release_ds_buffers(); mutex_unlock(&pmc_reserve_mutex); } } @@ -441,6 +426,28 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event_attr *attr) return 0; } +static int x86_pmu_hw_config(struct perf_event *event) +{ + /* + * Generate PMC IRQs: + * (keep 'enabled' bit clear for now) + */ + event->hw.config = ARCH_PERFMON_EVENTSEL_INT; + + /* + * Count user and OS events unless requested not to + */ + if (!event->attr.exclude_user) + event->hw.config |= ARCH_PERFMON_EVENTSEL_USR; + if (!event->attr.exclude_kernel) + event->hw.config |= ARCH_PERFMON_EVENTSEL_OS; + + if (event->attr.type == PERF_TYPE_RAW) + event->hw.config |= event->attr.config & X86_RAW_EVENT_MASK; + + return 0; +} + /* * Setup the hardware configuration for a given attr_type */ @@ -460,8 +467,11 @@ static int __hw_perf_event_init(struct perf_event *event) if (atomic_read(&active_events) == 0) { if (!reserve_pmc_hardware()) err = -EBUSY; - else - err = reserve_bts_hardware(); + else { + err = reserve_ds_buffers(); + if (err) + release_pmc_hardware(); + } } if (!err) atomic_inc(&active_events); @@ -472,23 +482,14 @@ static int __hw_perf_event_init(struct perf_event *event) event->destroy = hw_perf_event_destroy; - /* - * Generate PMC IRQs: - * (keep 'enabled' bit clear for now) - */ - hwc->config = ARCH_PERFMON_EVENTSEL_INT; - hwc->idx = -1; hwc->last_cpu = -1; hwc->last_tag = ~0ULL; - /* - * Count user and OS events unless requested not to. - */ - if (!attr->exclude_user) - hwc->config |= ARCH_PERFMON_EVENTSEL_USR; - if (!attr->exclude_kernel) - hwc->config |= ARCH_PERFMON_EVENTSEL_OS; + /* Processor specifics */ + err = x86_pmu.hw_config(event); + if (err) + return err; if (!hwc->sample_period) { hwc->sample_period = x86_pmu.max_period; @@ -505,16 +506,8 @@ static int __hw_perf_event_init(struct perf_event *event) return -EOPNOTSUPP; } - /* - * Raw hw_event type provide the config in the hw_event structure - */ - if (attr->type == PERF_TYPE_RAW) { - hwc->config |= x86_pmu.raw_event(attr->config); - if ((hwc->config & ARCH_PERFMON_EVENTSEL_ANY) && - perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) - return -EACCES; + if (attr->type == PERF_TYPE_RAW) return 0; - } if (attr->type == PERF_TYPE_HW_CACHE) return set_ext_hw_attr(hwc, attr); @@ -539,11 +532,11 @@ static int __hw_perf_event_init(struct perf_event *event) if ((attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) && (hwc->sample_period == 1)) { /* BTS is not supported by this architecture. */ - if (!bts_available()) + if (!x86_pmu.bts) return -EOPNOTSUPP; /* BTS is currently only allowed for user-mode. */ - if (hwc->config & ARCH_PERFMON_EVENTSEL_OS) + if (!attr->exclude_kernel) return -EOPNOTSUPP; } @@ -557,7 +550,7 @@ static void x86_pmu_disable_all(void) struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; - for (idx = 0; idx < x86_pmu.num_events; idx++) { + for (idx = 0; idx < x86_pmu.num_counters; idx++) { u64 val; if (!test_bit(idx, cpuc->active_mask)) @@ -587,12 +580,12 @@ void hw_perf_disable(void) x86_pmu.disable_all(); } -static void x86_pmu_enable_all(void) +static void x86_pmu_enable_all(int added) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; - for (idx = 0; idx < x86_pmu.num_events; idx++) { + for (idx = 0; idx < x86_pmu.num_counters; idx++) { struct perf_event *event = cpuc->events[idx]; u64 val; @@ -667,14 +660,14 @@ static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) * assign events to counters starting with most * constrained events. */ - wmax = x86_pmu.num_events; + wmax = x86_pmu.num_counters; /* * when fixed event counters are present, * wmax is incremented by 1 to account * for one more choice */ - if (x86_pmu.num_events_fixed) + if (x86_pmu.num_counters_fixed) wmax++; for (w = 1, num = n; num && w <= wmax; w++) { @@ -724,7 +717,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, struct perf_event *event; int n, max_count; - max_count = x86_pmu.num_events + x86_pmu.num_events_fixed; + max_count = x86_pmu.num_counters + x86_pmu.num_counters_fixed; /* current number of events already accepted */ n = cpuc->n_events; @@ -795,7 +788,7 @@ void hw_perf_enable(void) struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct perf_event *event; struct hw_perf_event *hwc; - int i; + int i, added = cpuc->n_added; if (!x86_pmu_initialized()) return; @@ -847,19 +840,20 @@ void hw_perf_enable(void) cpuc->enabled = 1; barrier(); - x86_pmu.enable_all(); + x86_pmu.enable_all(added); } static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc) { - (void)checking_wrmsrl(hwc->config_base + hwc->idx, + wrmsrl(hwc->config_base + hwc->idx, hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE); } static inline void x86_pmu_disable_event(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; - (void)checking_wrmsrl(hwc->config_base + hwc->idx, hwc->config); + + wrmsrl(hwc->config_base + hwc->idx, hwc->config); } static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left); @@ -874,7 +868,7 @@ x86_perf_event_set_period(struct perf_event *event) struct hw_perf_event *hwc = &event->hw; s64 left = atomic64_read(&hwc->period_left); s64 period = hwc->sample_period; - int err, ret = 0, idx = hwc->idx; + int ret = 0, idx = hwc->idx; if (idx == X86_PMC_IDX_FIXED_BTS) return 0; @@ -912,8 +906,8 @@ x86_perf_event_set_period(struct perf_event *event) */ atomic64_set(&hwc->prev_count, (u64)-left); - err = checking_wrmsrl(hwc->event_base + idx, - (u64)(-left) & x86_pmu.event_mask); + wrmsrl(hwc->event_base + idx, + (u64)(-left) & x86_pmu.cntval_mask); perf_event_update_userpage(event); @@ -950,7 +944,7 @@ static int x86_pmu_enable(struct perf_event *event) if (n < 0) return n; - ret = x86_schedule_events(cpuc, n, assign); + ret = x86_pmu.schedule_events(cpuc, n, assign); if (ret) return ret; /* @@ -991,11 +985,12 @@ static void x86_pmu_unthrottle(struct perf_event *event) void perf_event_print_debug(void) { u64 ctrl, status, overflow, pmc_ctrl, pmc_count, prev_left, fixed; + u64 pebs; struct cpu_hw_events *cpuc; unsigned long flags; int cpu, idx; - if (!x86_pmu.num_events) + if (!x86_pmu.num_counters) return; local_irq_save(flags); @@ -1008,16 +1003,18 @@ void perf_event_print_debug(void) rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status); rdmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, overflow); rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR_CTRL, fixed); + rdmsrl(MSR_IA32_PEBS_ENABLE, pebs); pr_info("\n"); pr_info("CPU#%d: ctrl: %016llx\n", cpu, ctrl); pr_info("CPU#%d: status: %016llx\n", cpu, status); pr_info("CPU#%d: overflow: %016llx\n", cpu, overflow); pr_info("CPU#%d: fixed: %016llx\n", cpu, fixed); + pr_info("CPU#%d: pebs: %016llx\n", cpu, pebs); } - pr_info("CPU#%d: active: %016llx\n", cpu, *(u64 *)cpuc->active_mask); + pr_info("CPU#%d: active: %016llx\n", cpu, *(u64 *)cpuc->active_mask); - for (idx = 0; idx < x86_pmu.num_events; idx++) { + for (idx = 0; idx < x86_pmu.num_counters; idx++) { rdmsrl(x86_pmu.eventsel + idx, pmc_ctrl); rdmsrl(x86_pmu.perfctr + idx, pmc_count); @@ -1030,7 +1027,7 @@ void perf_event_print_debug(void) pr_info("CPU#%d: gen-PMC%d left: %016llx\n", cpu, idx, prev_left); } - for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) { + for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) { rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, pmc_count); pr_info("CPU#%d: fixed-PMC%d count: %016llx\n", @@ -1095,7 +1092,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); - for (idx = 0; idx < x86_pmu.num_events; idx++) { + for (idx = 0; idx < x86_pmu.num_counters; idx++) { if (!test_bit(idx, cpuc->active_mask)) continue; @@ -1103,7 +1100,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) hwc = &event->hw; val = x86_perf_event_update(event); - if (val & (1ULL << (x86_pmu.event_bits - 1))) + if (val & (1ULL << (x86_pmu.cntval_bits - 1))) continue; /* @@ -1146,7 +1143,6 @@ void set_perf_event_pending(void) void perf_events_lapic_init(void) { -#ifdef CONFIG_X86_LOCAL_APIC if (!x86_pmu.apic || !x86_pmu_initialized()) return; @@ -1154,7 +1150,6 @@ void perf_events_lapic_init(void) * Always use NMI for PMU */ apic_write(APIC_LVTPC, APIC_DM_NMI); -#endif } static int __kprobes @@ -1178,9 +1173,7 @@ perf_event_nmi_handler(struct notifier_block *self, regs = args->regs; -#ifdef CONFIG_X86_LOCAL_APIC apic_write(APIC_LVTPC, APIC_DM_NMI); -#endif /* * Can't rely on the handled return value to say it was our NMI, two * events could trigger 'simultaneously' raising two back-to-back NMIs. @@ -1274,12 +1267,15 @@ int hw_perf_group_sched_in(struct perf_event *leader, int assign[X86_PMC_IDX_MAX]; int n0, n1, ret; + if (!x86_pmu_initialized()) + return 0; + /* n0 = total number of events */ n0 = collect_events(cpuc, leader, true); if (n0 < 0) return n0; - ret = x86_schedule_events(cpuc, n0, assign); + ret = x86_pmu.schedule_events(cpuc, n0, assign); if (ret) return ret; @@ -1329,6 +1325,9 @@ undo: #include "perf_event_amd.c" #include "perf_event_p6.c" +#include "perf_event_p4.c" +#include "perf_event_intel_lbr.c" +#include "perf_event_intel_ds.c" #include "perf_event_intel.c" static int __cpuinit @@ -1402,48 +1401,50 @@ void __init init_hw_perf_events(void) pr_cont("%s PMU driver.\n", x86_pmu.name); - if (x86_pmu.num_events > X86_PMC_MAX_GENERIC) { + if (x86_pmu.quirks) + x86_pmu.quirks(); + + if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) { WARN(1, KERN_ERR "hw perf events %d > max(%d), clipping!", - x86_pmu.num_events, X86_PMC_MAX_GENERIC); - x86_pmu.num_events = X86_PMC_MAX_GENERIC; + x86_pmu.num_counters, X86_PMC_MAX_GENERIC); + x86_pmu.num_counters = X86_PMC_MAX_GENERIC; } - perf_event_mask = (1 << x86_pmu.num_events) - 1; - perf_max_events = x86_pmu.num_events; + x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1; + perf_max_events = x86_pmu.num_counters; - if (x86_pmu.num_events_fixed > X86_PMC_MAX_FIXED) { + if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) { WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!", - x86_pmu.num_events_fixed, X86_PMC_MAX_FIXED); - x86_pmu.num_events_fixed = X86_PMC_MAX_FIXED; + x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED); + x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED; } - perf_event_mask |= - ((1LL << x86_pmu.num_events_fixed)-1) << X86_PMC_IDX_FIXED; - x86_pmu.intel_ctrl = perf_event_mask; + x86_pmu.intel_ctrl |= + ((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED; perf_events_lapic_init(); register_die_notifier(&perf_event_nmi_notifier); unconstrained = (struct event_constraint) - __EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_events) - 1, - 0, x86_pmu.num_events); + __EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1, + 0, x86_pmu.num_counters); if (x86_pmu.event_constraints) { for_each_event_constraint(c, x86_pmu.event_constraints) { - if (c->cmask != INTEL_ARCH_FIXED_MASK) + if (c->cmask != X86_RAW_EVENT_MASK) continue; - c->idxmsk64 |= (1ULL << x86_pmu.num_events) - 1; - c->weight += x86_pmu.num_events; + c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1; + c->weight += x86_pmu.num_counters; } } pr_info("... version: %d\n", x86_pmu.version); - pr_info("... bit width: %d\n", x86_pmu.event_bits); - pr_info("... generic registers: %d\n", x86_pmu.num_events); - pr_info("... value mask: %016Lx\n", x86_pmu.event_mask); + pr_info("... bit width: %d\n", x86_pmu.cntval_bits); + pr_info("... generic registers: %d\n", x86_pmu.num_counters); + pr_info("... value mask: %016Lx\n", x86_pmu.cntval_mask); pr_info("... max period: %016Lx\n", x86_pmu.max_period); - pr_info("... fixed-purpose events: %d\n", x86_pmu.num_events_fixed); - pr_info("... event mask: %016Lx\n", perf_event_mask); + pr_info("... fixed-purpose events: %d\n", x86_pmu.num_counters_fixed); + pr_info("... event mask: %016Lx\n", x86_pmu.intel_ctrl); perf_cpu_notifier(x86_pmu_notifier); } @@ -1463,6 +1464,32 @@ static const struct pmu pmu = { }; /* + * validate that we can schedule this event + */ +static int validate_event(struct perf_event *event) +{ + struct cpu_hw_events *fake_cpuc; + struct event_constraint *c; + int ret = 0; + + fake_cpuc = kmalloc(sizeof(*fake_cpuc), GFP_KERNEL | __GFP_ZERO); + if (!fake_cpuc) + return -ENOMEM; + + c = x86_pmu.get_event_constraints(fake_cpuc, event); + + if (!c || !c->weight) + ret = -ENOSPC; + + if (x86_pmu.put_event_constraints) + x86_pmu.put_event_constraints(fake_cpuc, event); + + kfree(fake_cpuc); + + return ret; +} + +/* * validate a single event group * * validation include: @@ -1502,7 +1529,7 @@ static int validate_group(struct perf_event *event) fake_cpuc->n_events = n; - ret = x86_schedule_events(fake_cpuc, n, NULL); + ret = x86_pmu.schedule_events(fake_cpuc, n, NULL); out_free: kfree(fake_cpuc); @@ -1527,6 +1554,8 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) if (event->group_leader != event) err = validate_group(event); + else + err = validate_event(event); event->pmu = tmp; } @@ -1574,8 +1603,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) { struct perf_callchain_entry *entry = data; - if (reliable) - callchain_store(entry, addr); + callchain_store(entry, addr); } static const struct stacktrace_ops backtrace_ops = { @@ -1597,41 +1625,6 @@ perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); } -/* - * best effort, GUP based copy_from_user() that assumes IRQ or NMI context - */ -static unsigned long -copy_from_user_nmi(void *to, const void __user *from, unsigned long n) -{ - unsigned long offset, addr = (unsigned long)from; - int type = in_nmi() ? KM_NMI : KM_IRQ0; - unsigned long size, len = 0; - struct page *page; - void *map; - int ret; - - do { - ret = __get_user_pages_fast(addr, 1, 0, &page); - if (!ret) - break; - - offset = addr & (PAGE_SIZE - 1); - size = min(PAGE_SIZE - offset, n - len); - - map = kmap_atomic(page, type); - memcpy(to, map+offset, size); - kunmap_atomic(map, type); - put_page(page); - - len += size; - to += size; - addr += size; - - } while (len < n); - - return len; -} - #ifdef CONFIG_COMPAT static inline int perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry) diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index db6f7d4056e1..611df11ba15e 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -2,7 +2,7 @@ static DEFINE_RAW_SPINLOCK(amd_nb_lock); -static __initconst u64 amd_hw_cache_event_ids +static __initconst const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = @@ -111,22 +111,19 @@ static u64 amd_pmu_event_map(int hw_event) return amd_perfmon_event_map[hw_event]; } -static u64 amd_pmu_raw_event(u64 hw_event) +static int amd_pmu_hw_config(struct perf_event *event) { -#define K7_EVNTSEL_EVENT_MASK 0xF000000FFULL -#define K7_EVNTSEL_UNIT_MASK 0x00000FF00ULL -#define K7_EVNTSEL_EDGE_MASK 0x000040000ULL -#define K7_EVNTSEL_INV_MASK 0x000800000ULL -#define K7_EVNTSEL_REG_MASK 0x0FF000000ULL - -#define K7_EVNTSEL_MASK \ - (K7_EVNTSEL_EVENT_MASK | \ - K7_EVNTSEL_UNIT_MASK | \ - K7_EVNTSEL_EDGE_MASK | \ - K7_EVNTSEL_INV_MASK | \ - K7_EVNTSEL_REG_MASK) - - return hw_event & K7_EVNTSEL_MASK; + int ret = x86_pmu_hw_config(event); + + if (ret) + return ret; + + if (event->attr.type != PERF_TYPE_RAW) + return 0; + + event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK; + + return 0; } /* @@ -165,7 +162,7 @@ static void amd_put_event_constraints(struct cpu_hw_events *cpuc, * be removed on one CPU at a time AND PMU is disabled * when we come here */ - for (i = 0; i < x86_pmu.num_events; i++) { + for (i = 0; i < x86_pmu.num_counters; i++) { if (nb->owners[i] == event) { cmpxchg(nb->owners+i, event, NULL); break; @@ -215,7 +212,7 @@ amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) struct hw_perf_event *hwc = &event->hw; struct amd_nb *nb = cpuc->amd_nb; struct perf_event *old = NULL; - int max = x86_pmu.num_events; + int max = x86_pmu.num_counters; int i, j, k = -1; /* @@ -293,7 +290,7 @@ static struct amd_nb *amd_alloc_nb(int cpu, int nb_id) /* * initialize all possible NB constraints */ - for (i = 0; i < x86_pmu.num_events; i++) { + for (i = 0; i < x86_pmu.num_counters; i++) { __set_bit(i, nb->event_constraints[i].idxmsk); nb->event_constraints[i].weight = 1; } @@ -371,21 +368,22 @@ static void amd_pmu_cpu_dead(int cpu) raw_spin_unlock(&amd_nb_lock); } -static __initconst struct x86_pmu amd_pmu = { +static __initconst const struct x86_pmu amd_pmu = { .name = "AMD", .handle_irq = x86_pmu_handle_irq, .disable_all = x86_pmu_disable_all, .enable_all = x86_pmu_enable_all, .enable = x86_pmu_enable_event, .disable = x86_pmu_disable_event, + .hw_config = amd_pmu_hw_config, + .schedule_events = x86_schedule_events, .eventsel = MSR_K7_EVNTSEL0, .perfctr = MSR_K7_PERFCTR0, .event_map = amd_pmu_event_map, - .raw_event = amd_pmu_raw_event, .max_events = ARRAY_SIZE(amd_perfmon_event_map), - .num_events = 4, - .event_bits = 48, - .event_mask = (1ULL << 48) - 1, + .num_counters = 4, + .cntval_bits = 48, + .cntval_mask = (1ULL << 48) - 1, .apic = 1, /* use highest bit to detect overflow */ .max_period = (1ULL << 47) - 1, diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 9c794ac87837..a099df96f916 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -88,7 +88,7 @@ static u64 intel_pmu_event_map(int hw_event) return intel_perfmon_event_map[hw_event]; } -static __initconst u64 westmere_hw_cache_event_ids +static __initconst const u64 westmere_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = @@ -179,7 +179,7 @@ static __initconst u64 westmere_hw_cache_event_ids }, }; -static __initconst u64 nehalem_hw_cache_event_ids +static __initconst const u64 nehalem_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = @@ -270,7 +270,7 @@ static __initconst u64 nehalem_hw_cache_event_ids }, }; -static __initconst u64 core2_hw_cache_event_ids +static __initconst const u64 core2_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = @@ -361,7 +361,7 @@ static __initconst u64 core2_hw_cache_event_ids }, }; -static __initconst u64 atom_hw_cache_event_ids +static __initconst const u64 atom_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = @@ -452,60 +452,6 @@ static __initconst u64 atom_hw_cache_event_ids }, }; -static u64 intel_pmu_raw_event(u64 hw_event) -{ -#define CORE_EVNTSEL_EVENT_MASK 0x000000FFULL -#define CORE_EVNTSEL_UNIT_MASK 0x0000FF00ULL -#define CORE_EVNTSEL_EDGE_MASK 0x00040000ULL -#define CORE_EVNTSEL_INV_MASK 0x00800000ULL -#define CORE_EVNTSEL_REG_MASK 0xFF000000ULL - -#define CORE_EVNTSEL_MASK \ - (INTEL_ARCH_EVTSEL_MASK | \ - INTEL_ARCH_UNIT_MASK | \ - INTEL_ARCH_EDGE_MASK | \ - INTEL_ARCH_INV_MASK | \ - INTEL_ARCH_CNT_MASK) - - return hw_event & CORE_EVNTSEL_MASK; -} - -static void intel_pmu_enable_bts(u64 config) -{ - unsigned long debugctlmsr; - - debugctlmsr = get_debugctlmsr(); - - debugctlmsr |= X86_DEBUGCTL_TR; - debugctlmsr |= X86_DEBUGCTL_BTS; - debugctlmsr |= X86_DEBUGCTL_BTINT; - - if (!(config & ARCH_PERFMON_EVENTSEL_OS)) - debugctlmsr |= X86_DEBUGCTL_BTS_OFF_OS; - - if (!(config & ARCH_PERFMON_EVENTSEL_USR)) - debugctlmsr |= X86_DEBUGCTL_BTS_OFF_USR; - - update_debugctlmsr(debugctlmsr); -} - -static void intel_pmu_disable_bts(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - unsigned long debugctlmsr; - - if (!cpuc->ds) - return; - - debugctlmsr = get_debugctlmsr(); - - debugctlmsr &= - ~(X86_DEBUGCTL_TR | X86_DEBUGCTL_BTS | X86_DEBUGCTL_BTINT | - X86_DEBUGCTL_BTS_OFF_OS | X86_DEBUGCTL_BTS_OFF_USR); - - update_debugctlmsr(debugctlmsr); -} - static void intel_pmu_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -514,12 +460,17 @@ static void intel_pmu_disable_all(void) if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) intel_pmu_disable_bts(); + + intel_pmu_pebs_disable_all(); + intel_pmu_lbr_disable_all(); } -static void intel_pmu_enable_all(void) +static void intel_pmu_enable_all(int added) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + intel_pmu_pebs_enable_all(); + intel_pmu_lbr_enable_all(); wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) { @@ -533,6 +484,41 @@ static void intel_pmu_enable_all(void) } } +/* + * Workaround for: + * Intel Errata AAK100 (model 26) + * Intel Errata AAP53 (model 30) + * Intel Errata BD53 (model 44) + * + * These chips need to be 'reset' when adding counters by programming + * the magic three (non counting) events 0x4300D2, 0x4300B1 and 0x4300B5 + * either in sequence on the same PMC or on different PMCs. + */ +static void intel_pmu_nhm_enable_all(int added) +{ + if (added) { + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int i; + + wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 0, 0x4300D2); + wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 1, 0x4300B1); + wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 2, 0x4300B5); + + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x3); + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0); + + for (i = 0; i < 3; i++) { + struct perf_event *event = cpuc->events[i]; + + if (!event) + continue; + + __x86_pmu_enable_event(&event->hw); + } + } + intel_pmu_enable_all(added); +} + static inline u64 intel_pmu_get_status(void) { u64 status; @@ -547,8 +533,7 @@ static inline void intel_pmu_ack_status(u64 ack) wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack); } -static inline void -intel_pmu_disable_fixed(struct hw_perf_event *hwc) +static void intel_pmu_disable_fixed(struct hw_perf_event *hwc) { int idx = hwc->idx - X86_PMC_IDX_FIXED; u64 ctrl_val, mask; @@ -557,71 +542,10 @@ intel_pmu_disable_fixed(struct hw_perf_event *hwc) rdmsrl(hwc->config_base, ctrl_val); ctrl_val &= ~mask; - (void)checking_wrmsrl(hwc->config_base, ctrl_val); -} - -static void intel_pmu_drain_bts_buffer(void) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - struct debug_store *ds = cpuc->ds; - struct bts_record { - u64 from; - u64 to; - u64 flags; - }; - struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS]; - struct bts_record *at, *top; - struct perf_output_handle handle; - struct perf_event_header header; - struct perf_sample_data data; - struct pt_regs regs; - - if (!event) - return; - - if (!ds) - return; - - at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; - top = (struct bts_record *)(unsigned long)ds->bts_index; - - if (top <= at) - return; - - ds->bts_index = ds->bts_buffer_base; - - perf_sample_data_init(&data, 0); - - data.period = event->hw.last_period; - regs.ip = 0; - - /* - * Prepare a generic sample, i.e. fill in the invariant fields. - * We will overwrite the from and to address before we output - * the sample. - */ - perf_prepare_sample(&header, &data, event, ®s); - - if (perf_output_begin(&handle, event, - header.size * (top - at), 1, 1)) - return; - - for (; at < top; at++) { - data.ip = at->from; - data.addr = at->to; - - perf_output_sample(&handle, &header, &data, event); - } - - perf_output_end(&handle); - - /* There's new data available. */ - event->hw.interrupts++; - event->pending_kill = POLL_IN; + wrmsrl(hwc->config_base, ctrl_val); } -static inline void -intel_pmu_disable_event(struct perf_event *event) +static void intel_pmu_disable_event(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; @@ -637,14 +561,15 @@ intel_pmu_disable_event(struct perf_event *event) } x86_pmu_disable_event(event); + + if (unlikely(event->attr.precise)) + intel_pmu_pebs_disable(event); } -static inline void -intel_pmu_enable_fixed(struct hw_perf_event *hwc) +static void intel_pmu_enable_fixed(struct hw_perf_event *hwc) { int idx = hwc->idx - X86_PMC_IDX_FIXED; u64 ctrl_val, bits, mask; - int err; /* * Enable IRQ generation (0x8), @@ -669,7 +594,7 @@ intel_pmu_enable_fixed(struct hw_perf_event *hwc) rdmsrl(hwc->config_base, ctrl_val); ctrl_val &= ~mask; ctrl_val |= bits; - err = checking_wrmsrl(hwc->config_base, ctrl_val); + wrmsrl(hwc->config_base, ctrl_val); } static void intel_pmu_enable_event(struct perf_event *event) @@ -689,6 +614,9 @@ static void intel_pmu_enable_event(struct perf_event *event) return; } + if (unlikely(event->attr.precise)) + intel_pmu_pebs_enable(event); + __x86_pmu_enable_event(hwc); } @@ -708,20 +636,20 @@ static void intel_pmu_reset(void) unsigned long flags; int idx; - if (!x86_pmu.num_events) + if (!x86_pmu.num_counters) return; local_irq_save(flags); printk("clearing PMU state on CPU#%d\n", smp_processor_id()); - for (idx = 0; idx < x86_pmu.num_events; idx++) { + for (idx = 0; idx < x86_pmu.num_counters; idx++) { checking_wrmsrl(x86_pmu.eventsel + idx, 0ull); checking_wrmsrl(x86_pmu.perfctr + idx, 0ull); } - for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) { + for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull); - } + if (ds) ds->bts_index = ds->bts_buffer_base; @@ -747,7 +675,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) intel_pmu_drain_bts_buffer(); status = intel_pmu_get_status(); if (!status) { - intel_pmu_enable_all(); + intel_pmu_enable_all(0); return 0; } @@ -762,6 +690,15 @@ again: inc_irq_stat(apic_perf_irqs); ack = status; + + intel_pmu_lbr_read(); + + /* + * PEBS overflow sets bit 62 in the global status register + */ + if (__test_and_clear_bit(62, (unsigned long *)&status)) + x86_pmu.drain_pebs(regs); + for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { struct perf_event *event = cpuc->events[bit]; @@ -787,26 +724,22 @@ again: goto again; done: - intel_pmu_enable_all(); + intel_pmu_enable_all(0); return 1; } -static struct event_constraint bts_constraint = - EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0); - static struct event_constraint * -intel_special_constraints(struct perf_event *event) +intel_bts_constraints(struct perf_event *event) { - unsigned int hw_event; - - hw_event = event->hw.config & INTEL_ARCH_EVENT_MASK; + struct hw_perf_event *hwc = &event->hw; + unsigned int hw_event, bts_event; - if (unlikely((hw_event == - x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) && - (event->hw.sample_period == 1))) { + hw_event = hwc->config & INTEL_ARCH_EVENT_MASK; + bts_event = x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS); + if (unlikely(hw_event == bts_event && hwc->sample_period == 1)) return &bts_constraint; - } + return NULL; } @@ -815,24 +748,53 @@ intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event { struct event_constraint *c; - c = intel_special_constraints(event); + c = intel_bts_constraints(event); + if (c) + return c; + + c = intel_pebs_constraints(event); if (c) return c; return x86_get_event_constraints(cpuc, event); } -static __initconst struct x86_pmu core_pmu = { +static int intel_pmu_hw_config(struct perf_event *event) +{ + int ret = x86_pmu_hw_config(event); + + if (ret) + return ret; + + if (event->attr.type != PERF_TYPE_RAW) + return 0; + + if (!(event->attr.config & ARCH_PERFMON_EVENTSEL_ANY)) + return 0; + + if (x86_pmu.version < 3) + return -EINVAL; + + if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return -EACCES; + + event->hw.config |= ARCH_PERFMON_EVENTSEL_ANY; + + return 0; +} + +static __initconst const struct x86_pmu core_pmu = { .name = "core", .handle_irq = x86_pmu_handle_irq, .disable_all = x86_pmu_disable_all, .enable_all = x86_pmu_enable_all, .enable = x86_pmu_enable_event, .disable = x86_pmu_disable_event, + .hw_config = x86_pmu_hw_config, + .schedule_events = x86_schedule_events, .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, .perfctr = MSR_ARCH_PERFMON_PERFCTR0, .event_map = intel_pmu_event_map, - .raw_event = intel_pmu_raw_event, .max_events = ARRAY_SIZE(intel_perfmon_event_map), .apic = 1, /* @@ -845,17 +807,32 @@ static __initconst struct x86_pmu core_pmu = { .event_constraints = intel_core_event_constraints, }; -static __initconst struct x86_pmu intel_pmu = { +static void intel_pmu_cpu_starting(int cpu) +{ + init_debug_store_on_cpu(cpu); + /* + * Deal with CPUs that don't clear their LBRs on power-up. + */ + intel_pmu_lbr_reset(); +} + +static void intel_pmu_cpu_dying(int cpu) +{ + fini_debug_store_on_cpu(cpu); +} + +static __initconst const struct x86_pmu intel_pmu = { .name = "Intel", .handle_irq = intel_pmu_handle_irq, .disable_all = intel_pmu_disable_all, .enable_all = intel_pmu_enable_all, .enable = intel_pmu_enable_event, .disable = intel_pmu_disable_event, + .hw_config = intel_pmu_hw_config, + .schedule_events = x86_schedule_events, .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, .perfctr = MSR_ARCH_PERFMON_PERFCTR0, .event_map = intel_pmu_event_map, - .raw_event = intel_pmu_raw_event, .max_events = ARRAY_SIZE(intel_perfmon_event_map), .apic = 1, /* @@ -864,14 +841,38 @@ static __initconst struct x86_pmu intel_pmu = { * the generic event period: */ .max_period = (1ULL << 31) - 1, - .enable_bts = intel_pmu_enable_bts, - .disable_bts = intel_pmu_disable_bts, .get_event_constraints = intel_get_event_constraints, - .cpu_starting = init_debug_store_on_cpu, - .cpu_dying = fini_debug_store_on_cpu, + .cpu_starting = intel_pmu_cpu_starting, + .cpu_dying = intel_pmu_cpu_dying, }; +static void intel_clovertown_quirks(void) +{ + /* + * PEBS is unreliable due to: + * + * AJ67 - PEBS may experience CPL leaks + * AJ68 - PEBS PMI may be delayed by one event + * AJ69 - GLOBAL_STATUS[62] will only be set when DEBUGCTL[12] + * AJ106 - FREEZE_LBRS_ON_PMI doesn't work in combination with PEBS + * + * AJ67 could be worked around by restricting the OS/USR flags. + * AJ69 could be worked around by setting PMU_FREEZE_ON_PMI. + * + * AJ106 could possibly be worked around by not allowing LBR + * usage from PEBS, including the fixup. + * AJ68 could possibly be worked around by always programming + * a pebs_event_reset[0] value and coping with the lost events. + * + * But taken together it might just make sense to not enable PEBS on + * these chips. + */ + printk(KERN_WARNING "PEBS disabled due to CPU errata.\n"); + x86_pmu.pebs = 0; + x86_pmu.pebs_constraints = NULL; +} + static __init int intel_pmu_init(void) { union cpuid10_edx edx; @@ -881,12 +882,13 @@ static __init int intel_pmu_init(void) int version; if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { - /* check for P6 processor family */ - if (boot_cpu_data.x86 == 6) { - return p6_pmu_init(); - } else { + switch (boot_cpu_data.x86) { + case 0x6: + return p6_pmu_init(); + case 0xf: + return p4_pmu_init(); + } return -ENODEV; - } } /* @@ -904,16 +906,28 @@ static __init int intel_pmu_init(void) x86_pmu = intel_pmu; x86_pmu.version = version; - x86_pmu.num_events = eax.split.num_events; - x86_pmu.event_bits = eax.split.bit_width; - x86_pmu.event_mask = (1ULL << eax.split.bit_width) - 1; + x86_pmu.num_counters = eax.split.num_counters; + x86_pmu.cntval_bits = eax.split.bit_width; + x86_pmu.cntval_mask = (1ULL << eax.split.bit_width) - 1; /* * Quirk: v2 perfmon does not report fixed-purpose events, so * assume at least 3 events: */ if (version > 1) - x86_pmu.num_events_fixed = max((int)edx.split.num_events_fixed, 3); + x86_pmu.num_counters_fixed = max((int)edx.split.num_counters_fixed, 3); + + /* + * v2 and above have a perf capabilities MSR + */ + if (version > 1) { + u64 capabilities; + + rdmsrl(MSR_IA32_PERF_CAPABILITIES, capabilities); + x86_pmu.intel_cap.capabilities = capabilities; + } + + intel_ds_init(); /* * Install the hw-cache-events table: @@ -924,12 +938,15 @@ static __init int intel_pmu_init(void) break; case 15: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */ + x86_pmu.quirks = intel_clovertown_quirks; case 22: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */ case 23: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */ case 29: /* six-core 45 nm xeon "Dunnington" */ memcpy(hw_cache_event_ids, core2_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + intel_pmu_lbr_init_core(); + x86_pmu.event_constraints = intel_core2_event_constraints; pr_cont("Core2 events, "); break; @@ -940,13 +957,19 @@ static __init int intel_pmu_init(void) memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + intel_pmu_lbr_init_nhm(); + x86_pmu.event_constraints = intel_nehalem_event_constraints; - pr_cont("Nehalem/Corei7 events, "); + x86_pmu.enable_all = intel_pmu_nhm_enable_all; + pr_cont("Nehalem events, "); break; + case 28: /* Atom */ memcpy(hw_cache_event_ids, atom_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + intel_pmu_lbr_init_atom(); + x86_pmu.event_constraints = intel_gen_event_constraints; pr_cont("Atom events, "); break; @@ -956,7 +979,10 @@ static __init int intel_pmu_init(void) memcpy(hw_cache_event_ids, westmere_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + intel_pmu_lbr_init_nhm(); + x86_pmu.event_constraints = intel_westmere_event_constraints; + x86_pmu.enable_all = intel_pmu_nhm_enable_all; pr_cont("Westmere events, "); break; diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c new file mode 100644 index 000000000000..ec8b2e12e104 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -0,0 +1,664 @@ +#ifdef CONFIG_CPU_SUP_INTEL + +/* The maximal number of PEBS events: */ +#define MAX_PEBS_EVENTS 4 + +/* The size of a BTS record in bytes: */ +#define BTS_RECORD_SIZE 24 + +#define BTS_BUFFER_SIZE (PAGE_SIZE << 4) +#define PEBS_BUFFER_SIZE PAGE_SIZE + +/* + * pebs_record_32 for p4 and core not supported + +struct pebs_record_32 { + u32 flags, ip; + u32 ax, bc, cx, dx; + u32 si, di, bp, sp; +}; + + */ + +struct pebs_record_core { + u64 flags, ip; + u64 ax, bx, cx, dx; + u64 si, di, bp, sp; + u64 r8, r9, r10, r11; + u64 r12, r13, r14, r15; +}; + +struct pebs_record_nhm { + u64 flags, ip; + u64 ax, bx, cx, dx; + u64 si, di, bp, sp; + u64 r8, r9, r10, r11; + u64 r12, r13, r14, r15; + u64 status, dla, dse, lat; +}; + +/* + * A debug store configuration. + * + * We only support architectures that use 64bit fields. + */ +struct debug_store { + u64 bts_buffer_base; + u64 bts_index; + u64 bts_absolute_maximum; + u64 bts_interrupt_threshold; + u64 pebs_buffer_base; + u64 pebs_index; + u64 pebs_absolute_maximum; + u64 pebs_interrupt_threshold; + u64 pebs_event_reset[MAX_PEBS_EVENTS]; +}; + +static void init_debug_store_on_cpu(int cpu) +{ + struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; + + if (!ds) + return; + + wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, + (u32)((u64)(unsigned long)ds), + (u32)((u64)(unsigned long)ds >> 32)); +} + +static void fini_debug_store_on_cpu(int cpu) +{ + if (!per_cpu(cpu_hw_events, cpu).ds) + return; + + wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, 0, 0); +} + +static void release_ds_buffers(void) +{ + int cpu; + + if (!x86_pmu.bts && !x86_pmu.pebs) + return; + + get_online_cpus(); + + for_each_online_cpu(cpu) + fini_debug_store_on_cpu(cpu); + + for_each_possible_cpu(cpu) { + struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; + + if (!ds) + continue; + + per_cpu(cpu_hw_events, cpu).ds = NULL; + + kfree((void *)(unsigned long)ds->pebs_buffer_base); + kfree((void *)(unsigned long)ds->bts_buffer_base); + kfree(ds); + } + + put_online_cpus(); +} + +static int reserve_ds_buffers(void) +{ + int cpu, err = 0; + + if (!x86_pmu.bts && !x86_pmu.pebs) + return 0; + + get_online_cpus(); + + for_each_possible_cpu(cpu) { + struct debug_store *ds; + void *buffer; + int max, thresh; + + err = -ENOMEM; + ds = kzalloc(sizeof(*ds), GFP_KERNEL); + if (unlikely(!ds)) + break; + per_cpu(cpu_hw_events, cpu).ds = ds; + + if (x86_pmu.bts) { + buffer = kzalloc(BTS_BUFFER_SIZE, GFP_KERNEL); + if (unlikely(!buffer)) + break; + + max = BTS_BUFFER_SIZE / BTS_RECORD_SIZE; + thresh = max / 16; + + ds->bts_buffer_base = (u64)(unsigned long)buffer; + ds->bts_index = ds->bts_buffer_base; + ds->bts_absolute_maximum = ds->bts_buffer_base + + max * BTS_RECORD_SIZE; + ds->bts_interrupt_threshold = ds->bts_absolute_maximum - + thresh * BTS_RECORD_SIZE; + } + + if (x86_pmu.pebs) { + buffer = kzalloc(PEBS_BUFFER_SIZE, GFP_KERNEL); + if (unlikely(!buffer)) + break; + + max = PEBS_BUFFER_SIZE / x86_pmu.pebs_record_size; + + ds->pebs_buffer_base = (u64)(unsigned long)buffer; + ds->pebs_index = ds->pebs_buffer_base; + ds->pebs_absolute_maximum = ds->pebs_buffer_base + + max * x86_pmu.pebs_record_size; + /* + * Always use single record PEBS + */ + ds->pebs_interrupt_threshold = ds->pebs_buffer_base + + x86_pmu.pebs_record_size; + } + + err = 0; + } + + if (err) + release_ds_buffers(); + else { + for_each_online_cpu(cpu) + init_debug_store_on_cpu(cpu); + } + + put_online_cpus(); + + return err; +} + +/* + * BTS + */ + +static struct event_constraint bts_constraint = + EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0); + +static void intel_pmu_enable_bts(u64 config) +{ + unsigned long debugctlmsr; + + debugctlmsr = get_debugctlmsr(); + + debugctlmsr |= DEBUGCTLMSR_TR; + debugctlmsr |= DEBUGCTLMSR_BTS; + debugctlmsr |= DEBUGCTLMSR_BTINT; + + if (!(config & ARCH_PERFMON_EVENTSEL_OS)) + debugctlmsr |= DEBUGCTLMSR_BTS_OFF_OS; + + if (!(config & ARCH_PERFMON_EVENTSEL_USR)) + debugctlmsr |= DEBUGCTLMSR_BTS_OFF_USR; + + update_debugctlmsr(debugctlmsr); +} + +static void intel_pmu_disable_bts(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + unsigned long debugctlmsr; + + if (!cpuc->ds) + return; + + debugctlmsr = get_debugctlmsr(); + + debugctlmsr &= + ~(DEBUGCTLMSR_TR | DEBUGCTLMSR_BTS | DEBUGCTLMSR_BTINT | + DEBUGCTLMSR_BTS_OFF_OS | DEBUGCTLMSR_BTS_OFF_USR); + + update_debugctlmsr(debugctlmsr); +} + +static void intel_pmu_drain_bts_buffer(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct debug_store *ds = cpuc->ds; + struct bts_record { + u64 from; + u64 to; + u64 flags; + }; + struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS]; + struct bts_record *at, *top; + struct perf_output_handle handle; + struct perf_event_header header; + struct perf_sample_data data; + struct pt_regs regs; + + if (!event) + return; + + if (!ds) + return; + + at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; + top = (struct bts_record *)(unsigned long)ds->bts_index; + + if (top <= at) + return; + + ds->bts_index = ds->bts_buffer_base; + + perf_sample_data_init(&data, 0); + data.period = event->hw.last_period; + regs.ip = 0; + + /* + * Prepare a generic sample, i.e. fill in the invariant fields. + * We will overwrite the from and to address before we output + * the sample. + */ + perf_prepare_sample(&header, &data, event, ®s); + + if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1)) + return; + + for (; at < top; at++) { + data.ip = at->from; + data.addr = at->to; + + perf_output_sample(&handle, &header, &data, event); + } + + perf_output_end(&handle); + + /* There's new data available. */ + event->hw.interrupts++; + event->pending_kill = POLL_IN; +} + +/* + * PEBS + */ + +static struct event_constraint intel_core_pebs_events[] = { + PEBS_EVENT_CONSTRAINT(0x00c0, 0x1), /* INSTR_RETIRED.ANY */ + PEBS_EVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */ + PEBS_EVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */ + PEBS_EVENT_CONSTRAINT(0x1fc7, 0x1), /* SIMD_INST_RETURED.ANY */ + PEBS_EVENT_CONSTRAINT(0x01cb, 0x1), /* MEM_LOAD_RETIRED.L1D_MISS */ + PEBS_EVENT_CONSTRAINT(0x02cb, 0x1), /* MEM_LOAD_RETIRED.L1D_LINE_MISS */ + PEBS_EVENT_CONSTRAINT(0x04cb, 0x1), /* MEM_LOAD_RETIRED.L2_MISS */ + PEBS_EVENT_CONSTRAINT(0x08cb, 0x1), /* MEM_LOAD_RETIRED.L2_LINE_MISS */ + PEBS_EVENT_CONSTRAINT(0x10cb, 0x1), /* MEM_LOAD_RETIRED.DTLB_MISS */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint intel_nehalem_pebs_events[] = { + PEBS_EVENT_CONSTRAINT(0x00c0, 0xf), /* INSTR_RETIRED.ANY */ + PEBS_EVENT_CONSTRAINT(0xfec1, 0xf), /* X87_OPS_RETIRED.ANY */ + PEBS_EVENT_CONSTRAINT(0x00c5, 0xf), /* BR_INST_RETIRED.MISPRED */ + PEBS_EVENT_CONSTRAINT(0x1fc7, 0xf), /* SIMD_INST_RETURED.ANY */ + PEBS_EVENT_CONSTRAINT(0x01cb, 0xf), /* MEM_LOAD_RETIRED.L1D_MISS */ + PEBS_EVENT_CONSTRAINT(0x02cb, 0xf), /* MEM_LOAD_RETIRED.L1D_LINE_MISS */ + PEBS_EVENT_CONSTRAINT(0x04cb, 0xf), /* MEM_LOAD_RETIRED.L2_MISS */ + PEBS_EVENT_CONSTRAINT(0x08cb, 0xf), /* MEM_LOAD_RETIRED.L2_LINE_MISS */ + PEBS_EVENT_CONSTRAINT(0x10cb, 0xf), /* MEM_LOAD_RETIRED.DTLB_MISS */ + EVENT_CONSTRAINT_END +}; + +static struct event_constraint * +intel_pebs_constraints(struct perf_event *event) +{ + struct event_constraint *c; + + if (!event->attr.precise) + return NULL; + + if (x86_pmu.pebs_constraints) { + for_each_event_constraint(c, x86_pmu.pebs_constraints) { + if ((event->hw.config & c->cmask) == c->code) + return c; + } + } + + return &emptyconstraint; +} + +static void intel_pmu_pebs_enable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + + hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT; + + cpuc->pebs_enabled |= 1ULL << hwc->idx; + WARN_ON_ONCE(cpuc->enabled); + + if (x86_pmu.intel_cap.pebs_trap) + intel_pmu_lbr_enable(event); +} + +static void intel_pmu_pebs_disable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + + cpuc->pebs_enabled &= ~(1ULL << hwc->idx); + if (cpuc->enabled) + wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); + + hwc->config |= ARCH_PERFMON_EVENTSEL_INT; + + if (x86_pmu.intel_cap.pebs_trap) + intel_pmu_lbr_disable(event); +} + +static void intel_pmu_pebs_enable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (cpuc->pebs_enabled) + wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); +} + +static void intel_pmu_pebs_disable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (cpuc->pebs_enabled) + wrmsrl(MSR_IA32_PEBS_ENABLE, 0); +} + +#include <asm/insn.h> + +static inline bool kernel_ip(unsigned long ip) +{ +#ifdef CONFIG_X86_32 + return ip > PAGE_OFFSET; +#else + return (long)ip < 0; +#endif +} + +static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + unsigned long from = cpuc->lbr_entries[0].from; + unsigned long old_to, to = cpuc->lbr_entries[0].to; + unsigned long ip = regs->ip; + + /* + * We don't need to fixup if the PEBS assist is fault like + */ + if (!x86_pmu.intel_cap.pebs_trap) + return 1; + + /* + * No LBR entry, no basic block, no rewinding + */ + if (!cpuc->lbr_stack.nr || !from || !to) + return 0; + + /* + * Basic blocks should never cross user/kernel boundaries + */ + if (kernel_ip(ip) != kernel_ip(to)) + return 0; + + /* + * unsigned math, either ip is before the start (impossible) or + * the basic block is larger than 1 page (sanity) + */ + if ((ip - to) > PAGE_SIZE) + return 0; + + /* + * We sampled a branch insn, rewind using the LBR stack + */ + if (ip == to) { + regs->ip = from; + return 1; + } + + do { + struct insn insn; + u8 buf[MAX_INSN_SIZE]; + void *kaddr; + + old_to = to; + if (!kernel_ip(ip)) { + int bytes, size = MAX_INSN_SIZE; + + bytes = copy_from_user_nmi(buf, (void __user *)to, size); + if (bytes != size) + return 0; + + kaddr = buf; + } else + kaddr = (void *)to; + + kernel_insn_init(&insn, kaddr); + insn_get_length(&insn); + to += insn.length; + } while (to < ip); + + if (to == ip) { + regs->ip = old_to; + return 1; + } + + /* + * Even though we decoded the basic block, the instruction stream + * never matched the given IP, either the TO or the IP got corrupted. + */ + return 0; +} + +static int intel_pmu_save_and_restart(struct perf_event *event); + +static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct debug_store *ds = cpuc->ds; + struct perf_event *event = cpuc->events[0]; /* PMC0 only */ + struct pebs_record_core *at, *top; + struct perf_sample_data data; + struct perf_raw_record raw; + struct pt_regs regs; + int n; + + if (!ds || !x86_pmu.pebs) + return; + + at = (struct pebs_record_core *)(unsigned long)ds->pebs_buffer_base; + top = (struct pebs_record_core *)(unsigned long)ds->pebs_index; + + /* + * Whatever else happens, drain the thing + */ + ds->pebs_index = ds->pebs_buffer_base; + + if (!test_bit(0, cpuc->active_mask)) + return; + + WARN_ON_ONCE(!event); + + if (!event->attr.precise) + return; + + n = top - at; + if (n <= 0) + return; + + if (!intel_pmu_save_and_restart(event)) + return; + + /* + * Should not happen, we program the threshold at 1 and do not + * set a reset value. + */ + WARN_ON_ONCE(n > 1); + at += n - 1; + + perf_sample_data_init(&data, 0); + data.period = event->hw.last_period; + + if (event->attr.sample_type & PERF_SAMPLE_RAW) { + raw.size = x86_pmu.pebs_record_size; + raw.data = at; + data.raw = &raw; + } + + /* + * We use the interrupt regs as a base because the PEBS record + * does not contain a full regs set, specifically it seems to + * lack segment descriptors, which get used by things like + * user_mode(). + * + * In the simple case fix up only the IP and BP,SP regs, for + * PERF_SAMPLE_IP and PERF_SAMPLE_CALLCHAIN to function properly. + * A possible PERF_SAMPLE_REGS will have to transfer all regs. + */ + regs = *iregs; + regs.ip = at->ip; + regs.bp = at->bp; + regs.sp = at->sp; + + if (intel_pmu_pebs_fixup_ip(®s)) + regs.flags |= PERF_EFLAGS_EXACT; + else + regs.flags &= ~PERF_EFLAGS_EXACT; + + if (perf_event_overflow(event, 1, &data, ®s)) + x86_pmu_stop(event); +} + +static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct debug_store *ds = cpuc->ds; + struct pebs_record_nhm *at, *top; + struct perf_sample_data data; + struct perf_event *event = NULL; + struct perf_raw_record raw; + struct pt_regs regs; + u64 status = 0; + int bit, n; + + if (!ds || !x86_pmu.pebs) + return; + + at = (struct pebs_record_nhm *)(unsigned long)ds->pebs_buffer_base; + top = (struct pebs_record_nhm *)(unsigned long)ds->pebs_index; + + ds->pebs_index = ds->pebs_buffer_base; + + n = top - at; + if (n <= 0) + return; + + /* + * Should not happen, we program the threshold at 1 and do not + * set a reset value. + */ + WARN_ON_ONCE(n > MAX_PEBS_EVENTS); + + for ( ; at < top; at++) { + for_each_set_bit(bit, (unsigned long *)&at->status, MAX_PEBS_EVENTS) { + event = cpuc->events[bit]; + if (!test_bit(bit, cpuc->active_mask)) + continue; + + WARN_ON_ONCE(!event); + + if (!event->attr.precise) + continue; + + if (__test_and_set_bit(bit, (unsigned long *)&status)) + continue; + + break; + } + + if (!event || bit >= MAX_PEBS_EVENTS) + continue; + + if (!intel_pmu_save_and_restart(event)) + continue; + + perf_sample_data_init(&data, 0); + data.period = event->hw.last_period; + + if (event->attr.sample_type & PERF_SAMPLE_RAW) { + raw.size = x86_pmu.pebs_record_size; + raw.data = at; + data.raw = &raw; + } + + /* + * See the comment in intel_pmu_drain_pebs_core() + */ + regs = *iregs; + regs.ip = at->ip; + regs.bp = at->bp; + regs.sp = at->sp; + + if (intel_pmu_pebs_fixup_ip(®s)) + regs.flags |= PERF_EFLAGS_EXACT; + else + regs.flags &= ~PERF_EFLAGS_EXACT; + + if (perf_event_overflow(event, 1, &data, ®s)) + x86_pmu_stop(event); + } +} + +/* + * BTS, PEBS probe and setup + */ + +static void intel_ds_init(void) +{ + /* + * No support for 32bit formats + */ + if (!boot_cpu_has(X86_FEATURE_DTES64)) + return; + + x86_pmu.bts = boot_cpu_has(X86_FEATURE_BTS); + x86_pmu.pebs = boot_cpu_has(X86_FEATURE_PEBS); + if (x86_pmu.pebs) { + char pebs_type = x86_pmu.intel_cap.pebs_trap ? '+' : '-'; + int format = x86_pmu.intel_cap.pebs_format; + + switch (format) { + case 0: + printk(KERN_CONT "PEBS fmt0%c, ", pebs_type); + x86_pmu.pebs_record_size = sizeof(struct pebs_record_core); + x86_pmu.drain_pebs = intel_pmu_drain_pebs_core; + x86_pmu.pebs_constraints = intel_core_pebs_events; + break; + + case 1: + printk(KERN_CONT "PEBS fmt1%c, ", pebs_type); + x86_pmu.pebs_record_size = sizeof(struct pebs_record_nhm); + x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm; + x86_pmu.pebs_constraints = intel_nehalem_pebs_events; + break; + + default: + printk(KERN_CONT "no PEBS fmt%d%c, ", format, pebs_type); + x86_pmu.pebs = 0; + break; + } + } +} + +#else /* CONFIG_CPU_SUP_INTEL */ + +static int reserve_ds_buffers(void) +{ + return 0; +} + +static void release_ds_buffers(void) +{ +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c new file mode 100644 index 000000000000..d202c1bece1a --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c @@ -0,0 +1,218 @@ +#ifdef CONFIG_CPU_SUP_INTEL + +enum { + LBR_FORMAT_32 = 0x00, + LBR_FORMAT_LIP = 0x01, + LBR_FORMAT_EIP = 0x02, + LBR_FORMAT_EIP_FLAGS = 0x03, +}; + +/* + * We only support LBR implementations that have FREEZE_LBRS_ON_PMI + * otherwise it becomes near impossible to get a reliable stack. + */ + +static void __intel_pmu_lbr_enable(void) +{ + u64 debugctl; + + rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); + debugctl |= (DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI); + wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); +} + +static void __intel_pmu_lbr_disable(void) +{ + u64 debugctl; + + rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); + debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI); + wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); +} + +static void intel_pmu_lbr_reset_32(void) +{ + int i; + + for (i = 0; i < x86_pmu.lbr_nr; i++) + wrmsrl(x86_pmu.lbr_from + i, 0); +} + +static void intel_pmu_lbr_reset_64(void) +{ + int i; + + for (i = 0; i < x86_pmu.lbr_nr; i++) { + wrmsrl(x86_pmu.lbr_from + i, 0); + wrmsrl(x86_pmu.lbr_to + i, 0); + } +} + +static void intel_pmu_lbr_reset(void) +{ + if (!x86_pmu.lbr_nr) + return; + + if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_32) + intel_pmu_lbr_reset_32(); + else + intel_pmu_lbr_reset_64(); +} + +static void intel_pmu_lbr_enable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (!x86_pmu.lbr_nr) + return; + + WARN_ON_ONCE(cpuc->enabled); + + /* + * Reset the LBR stack if we changed task context to + * avoid data leaks. + */ + + if (event->ctx->task && cpuc->lbr_context != event->ctx) { + intel_pmu_lbr_reset(); + cpuc->lbr_context = event->ctx; + } + + cpuc->lbr_users++; +} + +static void intel_pmu_lbr_disable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (!x86_pmu.lbr_nr) + return; + + cpuc->lbr_users--; + WARN_ON_ONCE(cpuc->lbr_users < 0); + + if (cpuc->enabled && !cpuc->lbr_users) + __intel_pmu_lbr_disable(); +} + +static void intel_pmu_lbr_enable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (cpuc->lbr_users) + __intel_pmu_lbr_enable(); +} + +static void intel_pmu_lbr_disable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (cpuc->lbr_users) + __intel_pmu_lbr_disable(); +} + +static inline u64 intel_pmu_lbr_tos(void) +{ + u64 tos; + + rdmsrl(x86_pmu.lbr_tos, tos); + + return tos; +} + +static void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc) +{ + unsigned long mask = x86_pmu.lbr_nr - 1; + u64 tos = intel_pmu_lbr_tos(); + int i; + + for (i = 0; i < x86_pmu.lbr_nr; i++) { + unsigned long lbr_idx = (tos - i) & mask; + union { + struct { + u32 from; + u32 to; + }; + u64 lbr; + } msr_lastbranch; + + rdmsrl(x86_pmu.lbr_from + lbr_idx, msr_lastbranch.lbr); + + cpuc->lbr_entries[i].from = msr_lastbranch.from; + cpuc->lbr_entries[i].to = msr_lastbranch.to; + cpuc->lbr_entries[i].flags = 0; + } + cpuc->lbr_stack.nr = i; +} + +#define LBR_FROM_FLAG_MISPRED (1ULL << 63) + +/* + * Due to lack of segmentation in Linux the effective address (offset) + * is the same as the linear address, allowing us to merge the LIP and EIP + * LBR formats. + */ +static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc) +{ + unsigned long mask = x86_pmu.lbr_nr - 1; + int lbr_format = x86_pmu.intel_cap.lbr_format; + u64 tos = intel_pmu_lbr_tos(); + int i; + + for (i = 0; i < x86_pmu.lbr_nr; i++) { + unsigned long lbr_idx = (tos - i) & mask; + u64 from, to, flags = 0; + + rdmsrl(x86_pmu.lbr_from + lbr_idx, from); + rdmsrl(x86_pmu.lbr_to + lbr_idx, to); + + if (lbr_format == LBR_FORMAT_EIP_FLAGS) { + flags = !!(from & LBR_FROM_FLAG_MISPRED); + from = (u64)((((s64)from) << 1) >> 1); + } + + cpuc->lbr_entries[i].from = from; + cpuc->lbr_entries[i].to = to; + cpuc->lbr_entries[i].flags = flags; + } + cpuc->lbr_stack.nr = i; +} + +static void intel_pmu_lbr_read(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (!cpuc->lbr_users) + return; + + if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_32) + intel_pmu_lbr_read_32(cpuc); + else + intel_pmu_lbr_read_64(cpuc); +} + +static void intel_pmu_lbr_init_core(void) +{ + x86_pmu.lbr_nr = 4; + x86_pmu.lbr_tos = 0x01c9; + x86_pmu.lbr_from = 0x40; + x86_pmu.lbr_to = 0x60; +} + +static void intel_pmu_lbr_init_nhm(void) +{ + x86_pmu.lbr_nr = 16; + x86_pmu.lbr_tos = 0x01c9; + x86_pmu.lbr_from = 0x680; + x86_pmu.lbr_to = 0x6c0; +} + +static void intel_pmu_lbr_init_atom(void) +{ + x86_pmu.lbr_nr = 8; + x86_pmu.lbr_tos = 0x01c9; + x86_pmu.lbr_from = 0x40; + x86_pmu.lbr_to = 0x60; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c new file mode 100644 index 000000000000..15367cce66bd --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -0,0 +1,834 @@ +/* + * Netburst Perfomance Events (P4, old Xeon) + * + * Copyright (C) 2010 Parallels, Inc., Cyrill Gorcunov <gorcunov@openvz.org> + * Copyright (C) 2010 Intel Corporation, Lin Ming <ming.m.lin@intel.com> + * + * For licencing details see kernel-base/COPYING + */ + +#ifdef CONFIG_CPU_SUP_INTEL + +#include <asm/perf_event_p4.h> + +#define P4_CNTR_LIMIT 3 +/* + * array indices: 0,1 - HT threads, used with HT enabled cpu + */ +struct p4_event_bind { + unsigned int opcode; /* Event code and ESCR selector */ + unsigned int escr_msr[2]; /* ESCR MSR for this event */ + unsigned char cntr[2][P4_CNTR_LIMIT]; /* counter index (offset), -1 on abscence */ +}; + +struct p4_cache_event_bind { + unsigned int metric_pebs; + unsigned int metric_vert; +}; + +#define P4_GEN_CACHE_EVENT_BIND(name) \ + [P4_CACHE__##name] = { \ + .metric_pebs = P4_PEBS__##name, \ + .metric_vert = P4_VERT__##name, \ + } + +static struct p4_cache_event_bind p4_cache_event_bind_map[] = { + P4_GEN_CACHE_EVENT_BIND(1stl_cache_load_miss_retired), + P4_GEN_CACHE_EVENT_BIND(2ndl_cache_load_miss_retired), + P4_GEN_CACHE_EVENT_BIND(dtlb_load_miss_retired), + P4_GEN_CACHE_EVENT_BIND(dtlb_store_miss_retired), +}; + +/* + * Note that we don't use CCCR1 here, there is an + * exception for P4_BSQ_ALLOCATION but we just have + * no workaround + * + * consider this binding as resources which particular + * event may borrow, it doesn't contain EventMask, + * Tags and friends -- they are left to a caller + */ +static struct p4_event_bind p4_event_bind_map[] = { + [P4_EVENT_TC_DELIVER_MODE] = { + .opcode = P4_OPCODE(P4_EVENT_TC_DELIVER_MODE), + .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_BPU_FETCH_REQUEST] = { + .opcode = P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST), + .escr_msr = { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_ITLB_REFERENCE] = { + .opcode = P4_OPCODE(P4_EVENT_ITLB_REFERENCE), + .escr_msr = { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_MEMORY_CANCEL] = { + .opcode = P4_OPCODE(P4_EVENT_MEMORY_CANCEL), + .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_MEMORY_COMPLETE] = { + .opcode = P4_OPCODE(P4_EVENT_MEMORY_COMPLETE), + .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_LOAD_PORT_REPLAY] = { + .opcode = P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY), + .escr_msr = { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_STORE_PORT_REPLAY] = { + .opcode = P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY), + .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_MOB_LOAD_REPLAY] = { + .opcode = P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY), + .escr_msr = { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_PAGE_WALK_TYPE] = { + .opcode = P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE), + .escr_msr = { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_BSQ_CACHE_REFERENCE] = { + .opcode = P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE), + .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_IOQ_ALLOCATION] = { + .opcode = P4_OPCODE(P4_EVENT_IOQ_ALLOCATION), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_IOQ_ACTIVE_ENTRIES] = { /* shared ESCR */ + .opcode = P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES), + .escr_msr = { MSR_P4_FSB_ESCR1, MSR_P4_FSB_ESCR1 }, + .cntr = { {2, -1, -1}, {3, -1, -1} }, + }, + [P4_EVENT_FSB_DATA_ACTIVITY] = { + .opcode = P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_BSQ_ALLOCATION] = { /* shared ESCR, broken CCCR1 */ + .opcode = P4_OPCODE(P4_EVENT_BSQ_ALLOCATION), + .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 }, + .cntr = { {0, -1, -1}, {1, -1, -1} }, + }, + [P4_EVENT_BSQ_ACTIVE_ENTRIES] = { /* shared ESCR */ + .opcode = P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES), + .escr_msr = { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 }, + .cntr = { {2, -1, -1}, {3, -1, -1} }, + }, + [P4_EVENT_SSE_INPUT_ASSIST] = { + .opcode = P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_PACKED_SP_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_PACKED_SP_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_PACKED_DP_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_PACKED_DP_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_SCALAR_SP_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_SCALAR_SP_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_SCALAR_DP_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_SCALAR_DP_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_64BIT_MMX_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_64BIT_MMX_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_128BIT_MMX_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_128BIT_MMX_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_X87_FP_UOP] = { + .opcode = P4_OPCODE(P4_EVENT_X87_FP_UOP), + .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_TC_MISC] = { + .opcode = P4_OPCODE(P4_EVENT_TC_MISC), + .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_GLOBAL_POWER_EVENTS] = { + .opcode = P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_TC_MS_XFER] = { + .opcode = P4_OPCODE(P4_EVENT_TC_MS_XFER), + .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_UOP_QUEUE_WRITES] = { + .opcode = P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES), + .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = { + .opcode = P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE), + .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_RETIRED_BRANCH_TYPE] = { + .opcode = P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE), + .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 }, + .cntr = { {4, 5, -1}, {6, 7, -1} }, + }, + [P4_EVENT_RESOURCE_STALL] = { + .opcode = P4_OPCODE(P4_EVENT_RESOURCE_STALL), + .escr_msr = { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_WC_BUFFER] = { + .opcode = P4_OPCODE(P4_EVENT_WC_BUFFER), + .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .cntr = { {8, 9, -1}, {10, 11, -1} }, + }, + [P4_EVENT_B2B_CYCLES] = { + .opcode = P4_OPCODE(P4_EVENT_B2B_CYCLES), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_BNR] = { + .opcode = P4_OPCODE(P4_EVENT_BNR), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_SNOOP] = { + .opcode = P4_OPCODE(P4_EVENT_SNOOP), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_RESPONSE] = { + .opcode = P4_OPCODE(P4_EVENT_RESPONSE), + .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .cntr = { {0, -1, -1}, {2, -1, -1} }, + }, + [P4_EVENT_FRONT_END_EVENT] = { + .opcode = P4_OPCODE(P4_EVENT_FRONT_END_EVENT), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_EXECUTION_EVENT] = { + .opcode = P4_OPCODE(P4_EVENT_EXECUTION_EVENT), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_REPLAY_EVENT] = { + .opcode = P4_OPCODE(P4_EVENT_REPLAY_EVENT), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_INSTR_RETIRED] = { + .opcode = P4_OPCODE(P4_EVENT_INSTR_RETIRED), + .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_UOPS_RETIRED] = { + .opcode = P4_OPCODE(P4_EVENT_UOPS_RETIRED), + .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_UOP_TYPE] = { + .opcode = P4_OPCODE(P4_EVENT_UOP_TYPE), + .escr_msr = { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_BRANCH_RETIRED] = { + .opcode = P4_OPCODE(P4_EVENT_BRANCH_RETIRED), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_MISPRED_BRANCH_RETIRED] = { + .opcode = P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED), + .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_X87_ASSIST] = { + .opcode = P4_OPCODE(P4_EVENT_X87_ASSIST), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_MACHINE_CLEAR] = { + .opcode = P4_OPCODE(P4_EVENT_MACHINE_CLEAR), + .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, + [P4_EVENT_INSTR_COMPLETED] = { + .opcode = P4_OPCODE(P4_EVENT_INSTR_COMPLETED), + .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .cntr = { {12, 13, 16}, {14, 15, 17} }, + }, +}; + +#define P4_GEN_CACHE_EVENT(event, bit, cache_event) \ + p4_config_pack_escr(P4_ESCR_EVENT(event) | \ + P4_ESCR_EMASK_BIT(event, bit)) | \ + p4_config_pack_cccr(cache_event | \ + P4_CCCR_ESEL(P4_OPCODE_ESEL(P4_OPCODE(event)))) + +static __initconst const u64 p4_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, + P4_CACHE__1stl_cache_load_miss_retired), + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, + P4_CACHE__2ndl_cache_load_miss_retired), + }, +}, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, + P4_CACHE__dtlb_load_miss_retired), + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, + P4_CACHE__dtlb_store_miss_retired), + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, HIT, + P4_CACHE__itlb_reference_hit), + [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, MISS, + P4_CACHE__itlb_reference_miss), + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +static u64 p4_general_events[PERF_COUNT_HW_MAX] = { + /* non-halted CPU clocks */ + [PERF_COUNT_HW_CPU_CYCLES] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_GLOBAL_POWER_EVENTS) | + P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING)), + + /* + * retired instructions + * in a sake of simplicity we don't use the FSB tagging + */ + [PERF_COUNT_HW_INSTRUCTIONS] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_INSTR_RETIRED) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)), + + /* cache hits */ + [PERF_COUNT_HW_CACHE_REFERENCES] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)), + + /* cache misses */ + [PERF_COUNT_HW_CACHE_MISSES] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS)), + + /* branch instructions retired */ + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_RETIRED_BRANCH_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT)), + + /* mispredicted branches retired */ + [PERF_COUNT_HW_BRANCH_MISSES] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_MISPRED_BRANCH_RETIRED) | + P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS)), + + /* bus ready clocks (cpu is driving #DRDY_DRV\#DRDY_OWN): */ + [PERF_COUNT_HW_BUS_CYCLES] = + p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_FSB_DATA_ACTIVITY) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN)) | + p4_config_pack_cccr(P4_CCCR_EDGE | P4_CCCR_COMPARE), +}; + +static struct p4_event_bind *p4_config_get_bind(u64 config) +{ + unsigned int evnt = p4_config_unpack_event(config); + struct p4_event_bind *bind = NULL; + + if (evnt < ARRAY_SIZE(p4_event_bind_map)) + bind = &p4_event_bind_map[evnt]; + + return bind; +} + +static u64 p4_pmu_event_map(int hw_event) +{ + struct p4_event_bind *bind; + unsigned int esel; + u64 config; + + if (hw_event > ARRAY_SIZE(p4_general_events)) { + printk_once(KERN_ERR "P4 PMU: Bad index: %i\n", hw_event); + return 0; + } + + config = p4_general_events[hw_event]; + bind = p4_config_get_bind(config); + esel = P4_OPCODE_ESEL(bind->opcode); + config |= p4_config_pack_cccr(P4_CCCR_ESEL(esel)); + + return config; +} + +static int p4_hw_config(struct perf_event *event) +{ + int cpu = raw_smp_processor_id(); + u32 escr, cccr; + + /* + * the reason we use cpu that early is that: if we get scheduled + * first time on the same cpu -- we will not need swap thread + * specific flags in config (and will save some cpu cycles) + */ + + cccr = p4_default_cccr_conf(cpu); + escr = p4_default_escr_conf(cpu, event->attr.exclude_kernel, + event->attr.exclude_user); + event->hw.config = p4_config_pack_escr(escr) | + p4_config_pack_cccr(cccr); + + if (p4_ht_active() && p4_ht_thread(cpu)) + event->hw.config = p4_set_ht_bit(event->hw.config); + + if (event->attr.type != PERF_TYPE_RAW) + return 0; + + /* + * We don't control raw events so it's up to the caller + * to pass sane values (and we don't count the thread number + * on HT machine but allow HT-compatible specifics to be + * passed on) + * + * XXX: HT wide things should check perf_paranoid_cpu() && + * CAP_SYS_ADMIN + */ + event->hw.config |= event->attr.config & + (p4_config_pack_escr(P4_ESCR_MASK_HT) | + p4_config_pack_cccr(P4_CCCR_MASK_HT)); + + return 0; +} + +static inline void p4_pmu_clear_cccr_ovf(struct hw_perf_event *hwc) +{ + unsigned long dummy; + + rdmsrl(hwc->config_base + hwc->idx, dummy); + if (dummy & P4_CCCR_OVF) { + (void)checking_wrmsrl(hwc->config_base + hwc->idx, + ((u64)dummy) & ~P4_CCCR_OVF); + } +} + +static inline void p4_pmu_disable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + /* + * If event gets disabled while counter is in overflowed + * state we need to clear P4_CCCR_OVF, otherwise interrupt get + * asserted again and again + */ + (void)checking_wrmsrl(hwc->config_base + hwc->idx, + (u64)(p4_config_unpack_cccr(hwc->config)) & + ~P4_CCCR_ENABLE & ~P4_CCCR_OVF & ~P4_CCCR_RESERVED); +} + +static void p4_pmu_disable_all(void) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx; + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + struct perf_event *event = cpuc->events[idx]; + if (!test_bit(idx, cpuc->active_mask)) + continue; + p4_pmu_disable_event(event); + } +} + +static void p4_pmu_enable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + int thread = p4_ht_config_thread(hwc->config); + u64 escr_conf = p4_config_unpack_escr(p4_clear_ht_bit(hwc->config)); + unsigned int idx = p4_config_unpack_event(hwc->config); + unsigned int idx_cache = p4_config_unpack_cache_event(hwc->config); + struct p4_event_bind *bind; + struct p4_cache_event_bind *bind_cache; + u64 escr_addr, cccr; + + bind = &p4_event_bind_map[idx]; + escr_addr = (u64)bind->escr_msr[thread]; + + /* + * - we dont support cascaded counters yet + * - and counter 1 is broken (erratum) + */ + WARN_ON_ONCE(p4_is_event_cascaded(hwc->config)); + WARN_ON_ONCE(hwc->idx == 1); + + /* we need a real Event value */ + escr_conf &= ~P4_ESCR_EVENT_MASK; + escr_conf |= P4_ESCR_EVENT(P4_OPCODE_EVNT(bind->opcode)); + + cccr = p4_config_unpack_cccr(hwc->config); + + /* + * it could be Cache event so that we need to + * set metrics into additional MSRs + */ + BUILD_BUG_ON(P4_CACHE__MAX > P4_CCCR_CACHE_OPS_MASK); + if (idx_cache > P4_CACHE__NONE && + idx_cache < ARRAY_SIZE(p4_cache_event_bind_map)) { + bind_cache = &p4_cache_event_bind_map[idx_cache]; + (void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)bind_cache->metric_pebs); + (void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)bind_cache->metric_vert); + } + + (void)checking_wrmsrl(escr_addr, escr_conf); + (void)checking_wrmsrl(hwc->config_base + hwc->idx, + (cccr & ~P4_CCCR_RESERVED) | P4_CCCR_ENABLE); +} + +static void p4_pmu_enable_all(int added) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx; + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + struct perf_event *event = cpuc->events[idx]; + if (!test_bit(idx, cpuc->active_mask)) + continue; + p4_pmu_enable_event(event); + } +} + +static int p4_pmu_handle_irq(struct pt_regs *regs) +{ + struct perf_sample_data data; + struct cpu_hw_events *cpuc; + struct perf_event *event; + struct hw_perf_event *hwc; + int idx, handled = 0; + u64 val; + + data.addr = 0; + data.raw = NULL; + + cpuc = &__get_cpu_var(cpu_hw_events); + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + + if (!test_bit(idx, cpuc->active_mask)) + continue; + + event = cpuc->events[idx]; + hwc = &event->hw; + + WARN_ON_ONCE(hwc->idx != idx); + + /* + * FIXME: Redundant call, actually not needed + * but just to check if we're screwed + */ + p4_pmu_clear_cccr_ovf(hwc); + + val = x86_perf_event_update(event); + if (val & (1ULL << (x86_pmu.cntval_bits - 1))) + continue; + + /* + * event overflow + */ + handled = 1; + data.period = event->hw.last_period; + + if (!x86_perf_event_set_period(event)) + continue; + if (perf_event_overflow(event, 1, &data, regs)) + p4_pmu_disable_event(event); + } + + if (handled) { + /* p4 quirk: unmask it again */ + apic_write(APIC_LVTPC, apic_read(APIC_LVTPC) & ~APIC_LVT_MASKED); + inc_irq_stat(apic_perf_irqs); + } + + return handled; +} + +/* + * swap thread specific fields according to a thread + * we are going to run on + */ +static void p4_pmu_swap_config_ts(struct hw_perf_event *hwc, int cpu) +{ + u32 escr, cccr; + + /* + * we either lucky and continue on same cpu or no HT support + */ + if (!p4_should_swap_ts(hwc->config, cpu)) + return; + + /* + * the event is migrated from an another logical + * cpu, so we need to swap thread specific flags + */ + + escr = p4_config_unpack_escr(hwc->config); + cccr = p4_config_unpack_cccr(hwc->config); + + if (p4_ht_thread(cpu)) { + cccr &= ~P4_CCCR_OVF_PMI_T0; + cccr |= P4_CCCR_OVF_PMI_T1; + if (escr & P4_ESCR_T0_OS) { + escr &= ~P4_ESCR_T0_OS; + escr |= P4_ESCR_T1_OS; + } + if (escr & P4_ESCR_T0_USR) { + escr &= ~P4_ESCR_T0_USR; + escr |= P4_ESCR_T1_USR; + } + hwc->config = p4_config_pack_escr(escr); + hwc->config |= p4_config_pack_cccr(cccr); + hwc->config |= P4_CONFIG_HT; + } else { + cccr &= ~P4_CCCR_OVF_PMI_T1; + cccr |= P4_CCCR_OVF_PMI_T0; + if (escr & P4_ESCR_T1_OS) { + escr &= ~P4_ESCR_T1_OS; + escr |= P4_ESCR_T0_OS; + } + if (escr & P4_ESCR_T1_USR) { + escr &= ~P4_ESCR_T1_USR; + escr |= P4_ESCR_T0_USR; + } + hwc->config = p4_config_pack_escr(escr); + hwc->config |= p4_config_pack_cccr(cccr); + hwc->config &= ~P4_CONFIG_HT; + } +} + +/* ESCRs are not sequential in memory so we need a map */ +static const unsigned int p4_escr_map[ARCH_P4_TOTAL_ESCR] = { + MSR_P4_ALF_ESCR0, /* 0 */ + MSR_P4_ALF_ESCR1, /* 1 */ + MSR_P4_BPU_ESCR0, /* 2 */ + MSR_P4_BPU_ESCR1, /* 3 */ + MSR_P4_BSU_ESCR0, /* 4 */ + MSR_P4_BSU_ESCR1, /* 5 */ + MSR_P4_CRU_ESCR0, /* 6 */ + MSR_P4_CRU_ESCR1, /* 7 */ + MSR_P4_CRU_ESCR2, /* 8 */ + MSR_P4_CRU_ESCR3, /* 9 */ + MSR_P4_CRU_ESCR4, /* 10 */ + MSR_P4_CRU_ESCR5, /* 11 */ + MSR_P4_DAC_ESCR0, /* 12 */ + MSR_P4_DAC_ESCR1, /* 13 */ + MSR_P4_FIRM_ESCR0, /* 14 */ + MSR_P4_FIRM_ESCR1, /* 15 */ + MSR_P4_FLAME_ESCR0, /* 16 */ + MSR_P4_FLAME_ESCR1, /* 17 */ + MSR_P4_FSB_ESCR0, /* 18 */ + MSR_P4_FSB_ESCR1, /* 19 */ + MSR_P4_IQ_ESCR0, /* 20 */ + MSR_P4_IQ_ESCR1, /* 21 */ + MSR_P4_IS_ESCR0, /* 22 */ + MSR_P4_IS_ESCR1, /* 23 */ + MSR_P4_ITLB_ESCR0, /* 24 */ + MSR_P4_ITLB_ESCR1, /* 25 */ + MSR_P4_IX_ESCR0, /* 26 */ + MSR_P4_IX_ESCR1, /* 27 */ + MSR_P4_MOB_ESCR0, /* 28 */ + MSR_P4_MOB_ESCR1, /* 29 */ + MSR_P4_MS_ESCR0, /* 30 */ + MSR_P4_MS_ESCR1, /* 31 */ + MSR_P4_PMH_ESCR0, /* 32 */ + MSR_P4_PMH_ESCR1, /* 33 */ + MSR_P4_RAT_ESCR0, /* 34 */ + MSR_P4_RAT_ESCR1, /* 35 */ + MSR_P4_SAAT_ESCR0, /* 36 */ + MSR_P4_SAAT_ESCR1, /* 37 */ + MSR_P4_SSU_ESCR0, /* 38 */ + MSR_P4_SSU_ESCR1, /* 39 */ + MSR_P4_TBPU_ESCR0, /* 40 */ + MSR_P4_TBPU_ESCR1, /* 41 */ + MSR_P4_TC_ESCR0, /* 42 */ + MSR_P4_TC_ESCR1, /* 43 */ + MSR_P4_U2L_ESCR0, /* 44 */ + MSR_P4_U2L_ESCR1, /* 45 */ +}; + +static int p4_get_escr_idx(unsigned int addr) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(p4_escr_map); i++) { + if (addr == p4_escr_map[i]) + return i; + } + + return -1; +} + +static int p4_next_cntr(int thread, unsigned long *used_mask, + struct p4_event_bind *bind) +{ + int i = 0, j; + + for (i = 0; i < P4_CNTR_LIMIT; i++) { + j = bind->cntr[thread][i++]; + if (j == -1 || !test_bit(j, used_mask)) + return j; + } + + return -1; +} + +static int p4_pmu_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) +{ + unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + unsigned long escr_mask[BITS_TO_LONGS(ARCH_P4_TOTAL_ESCR)]; + int cpu = raw_smp_processor_id(); + struct hw_perf_event *hwc; + struct p4_event_bind *bind; + unsigned int i, thread, num; + int cntr_idx, escr_idx; + + bitmap_zero(used_mask, X86_PMC_IDX_MAX); + bitmap_zero(escr_mask, ARCH_P4_TOTAL_ESCR); + + for (i = 0, num = n; i < n; i++, num--) { + + hwc = &cpuc->event_list[i]->hw; + thread = p4_ht_thread(cpu); + bind = p4_config_get_bind(hwc->config); + escr_idx = p4_get_escr_idx(bind->escr_msr[thread]); + + if (hwc->idx != -1 && !p4_should_swap_ts(hwc->config, cpu)) { + cntr_idx = hwc->idx; + if (assign) + assign[i] = hwc->idx; + goto reserve; + } + + cntr_idx = p4_next_cntr(thread, used_mask, bind); + if (cntr_idx == -1 || test_bit(escr_idx, escr_mask)) + goto done; + + p4_pmu_swap_config_ts(hwc, cpu); + if (assign) + assign[i] = cntr_idx; +reserve: + set_bit(cntr_idx, used_mask); + set_bit(escr_idx, escr_mask); + } + +done: + return num ? -ENOSPC : 0; +} + +static __initconst const struct x86_pmu p4_pmu = { + .name = "Netburst P4/Xeon", + .handle_irq = p4_pmu_handle_irq, + .disable_all = p4_pmu_disable_all, + .enable_all = p4_pmu_enable_all, + .enable = p4_pmu_enable_event, + .disable = p4_pmu_disable_event, + .eventsel = MSR_P4_BPU_CCCR0, + .perfctr = MSR_P4_BPU_PERFCTR0, + .event_map = p4_pmu_event_map, + .max_events = ARRAY_SIZE(p4_general_events), + .get_event_constraints = x86_get_event_constraints, + /* + * IF HT disabled we may need to use all + * ARCH_P4_MAX_CCCR counters simulaneously + * though leave it restricted at moment assuming + * HT is on + */ + .num_counters = ARCH_P4_MAX_CCCR, + .apic = 1, + .cntval_bits = 40, + .cntval_mask = (1ULL << 40) - 1, + .max_period = (1ULL << 39) - 1, + .hw_config = p4_hw_config, + .schedule_events = p4_pmu_schedule_events, +}; + +static __init int p4_pmu_init(void) +{ + unsigned int low, high; + + /* If we get stripped -- indexig fails */ + BUILD_BUG_ON(ARCH_P4_MAX_CCCR > X86_PMC_MAX_GENERIC); + + rdmsr(MSR_IA32_MISC_ENABLE, low, high); + if (!(low & (1 << 7))) { + pr_cont("unsupported Netburst CPU model %d ", + boot_cpu_data.x86_model); + return -ENODEV; + } + + memcpy(hw_cache_event_ids, p4_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); + + pr_cont("Netburst events, "); + + x86_pmu = p4_pmu; + + return 0; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c index a330485d14da..34ba07be2cda 100644 --- a/arch/x86/kernel/cpu/perf_event_p6.c +++ b/arch/x86/kernel/cpu/perf_event_p6.c @@ -27,24 +27,6 @@ static u64 p6_pmu_event_map(int hw_event) */ #define P6_NOP_EVENT 0x0000002EULL -static u64 p6_pmu_raw_event(u64 hw_event) -{ -#define P6_EVNTSEL_EVENT_MASK 0x000000FFULL -#define P6_EVNTSEL_UNIT_MASK 0x0000FF00ULL -#define P6_EVNTSEL_EDGE_MASK 0x00040000ULL -#define P6_EVNTSEL_INV_MASK 0x00800000ULL -#define P6_EVNTSEL_REG_MASK 0xFF000000ULL - -#define P6_EVNTSEL_MASK \ - (P6_EVNTSEL_EVENT_MASK | \ - P6_EVNTSEL_UNIT_MASK | \ - P6_EVNTSEL_EDGE_MASK | \ - P6_EVNTSEL_INV_MASK | \ - P6_EVNTSEL_REG_MASK) - - return hw_event & P6_EVNTSEL_MASK; -} - static struct event_constraint p6_event_constraints[] = { INTEL_EVENT_CONSTRAINT(0xc1, 0x1), /* FLOPS */ @@ -66,7 +48,7 @@ static void p6_pmu_disable_all(void) wrmsrl(MSR_P6_EVNTSEL0, val); } -static void p6_pmu_enable_all(void) +static void p6_pmu_enable_all(int added) { unsigned long val; @@ -102,22 +84,23 @@ static void p6_pmu_enable_event(struct perf_event *event) (void)checking_wrmsrl(hwc->config_base + hwc->idx, val); } -static __initconst struct x86_pmu p6_pmu = { +static __initconst const struct x86_pmu p6_pmu = { .name = "p6", .handle_irq = x86_pmu_handle_irq, .disable_all = p6_pmu_disable_all, .enable_all = p6_pmu_enable_all, .enable = p6_pmu_enable_event, .disable = p6_pmu_disable_event, + .hw_config = x86_pmu_hw_config, + .schedule_events = x86_schedule_events, .eventsel = MSR_P6_EVNTSEL0, .perfctr = MSR_P6_PERFCTR0, .event_map = p6_pmu_event_map, - .raw_event = p6_pmu_raw_event, .max_events = ARRAY_SIZE(p6_perfmon_event_map), .apic = 1, .max_period = (1ULL << 31) - 1, .version = 0, - .num_events = 2, + .num_counters = 2, /* * Events have 40 bits implemented. However they are designed such * that bits [32-39] are sign extensions of bit 31. As such the @@ -125,8 +108,8 @@ static __initconst struct x86_pmu p6_pmu = { * * See IA-32 Intel Architecture Software developer manual Vol 3B */ - .event_bits = 32, - .event_mask = (1ULL << 32) - 1, + .cntval_bits = 32, + .cntval_mask = (1ULL << 32) - 1, .get_event_constraints = x86_get_event_constraints, .event_constraints = p6_event_constraints, }; diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index a4849c10a77e..ebd4c51d096a 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -27,7 +27,6 @@ #include <asm/cpu.h> #include <asm/reboot.h> #include <asm/virtext.h> -#include <asm/x86_init.h> #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) @@ -103,10 +102,5 @@ void native_machine_crash_shutdown(struct pt_regs *regs) #ifdef CONFIG_HPET_TIMER hpet_disable(); #endif - -#ifdef CONFIG_X86_64 - x86_platform.iommu_shutdown(); -#endif - crash_save_cpu(regs, safe_smp_processor_id()); } diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c deleted file mode 100644 index 1c47390dd0e5..000000000000 --- a/arch/x86/kernel/ds.c +++ /dev/null @@ -1,1437 +0,0 @@ -/* - * Debug Store support - * - * This provides a low-level interface to the hardware's Debug Store - * feature that is used for branch trace store (BTS) and - * precise-event based sampling (PEBS). - * - * It manages: - * - DS and BTS hardware configuration - * - buffer overflow handling (to be done) - * - buffer access - * - * It does not do: - * - security checking (is the caller allowed to trace the task) - * - buffer allocation (memory accounting) - * - * - * Copyright (C) 2007-2009 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, 2007-2009 - */ - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/slab.h> -#include <linux/mm.h> -#include <linux/trace_clock.h> - -#include <asm/ds.h> - -#include "ds_selftest.h" - -/* - * The configuration for a particular DS hardware implementation: - */ -struct ds_configuration { - /* The name of the configuration: */ - const char *name; - - /* The size of pointer-typed fields in DS, BTS, and PEBS: */ - unsigned char sizeof_ptr_field; - - /* The size of a BTS/PEBS record in bytes: */ - unsigned char sizeof_rec[2]; - - /* The number of pebs counter reset values in the DS structure. */ - unsigned char nr_counter_reset; - - /* Control bit-masks indexed by enum ds_feature: */ - unsigned long ctl[dsf_ctl_max]; -}; -static struct ds_configuration ds_cfg __read_mostly; - - -/* Maximal size of a DS configuration: */ -#define MAX_SIZEOF_DS 0x80 - -/* Maximal size of a BTS record: */ -#define MAX_SIZEOF_BTS (3 * 8) - -/* BTS and PEBS buffer alignment: */ -#define DS_ALIGNMENT (1 << 3) - -/* Number of buffer pointers in DS: */ -#define NUM_DS_PTR_FIELDS 8 - -/* Size of a pebs reset value in DS: */ -#define PEBS_RESET_FIELD_SIZE 8 - -/* Mask of control bits in the DS MSR register: */ -#define BTS_CONTROL \ - ( ds_cfg.ctl[dsf_bts] | \ - ds_cfg.ctl[dsf_bts_kernel] | \ - ds_cfg.ctl[dsf_bts_user] | \ - ds_cfg.ctl[dsf_bts_overflow] ) - -/* - * A BTS or PEBS tracer. - * - * This holds the configuration of the tracer and serves as a handle - * to identify tracers. - */ -struct ds_tracer { - /* The DS context (partially) owned by this tracer. */ - struct ds_context *context; - /* The buffer provided on ds_request() and its size in bytes. */ - void *buffer; - size_t size; -}; - -struct bts_tracer { - /* The common DS part: */ - struct ds_tracer ds; - - /* The trace including the DS configuration: */ - struct bts_trace trace; - - /* Buffer overflow notification function: */ - bts_ovfl_callback_t ovfl; - - /* Active flags affecting trace collection. */ - unsigned int flags; -}; - -struct pebs_tracer { - /* The common DS part: */ - struct ds_tracer ds; - - /* The trace including the DS configuration: */ - struct pebs_trace trace; - - /* Buffer overflow notification function: */ - pebs_ovfl_callback_t ovfl; -}; - -/* - * Debug Store (DS) save area configuration (see Intel64 and IA32 - * Architectures Software Developer's Manual, section 18.5) - * - * The DS configuration consists of the following fields; different - * architetures vary in the size of those fields. - * - * - double-word aligned base linear address of the BTS buffer - * - write pointer into the BTS buffer - * - end linear address of the BTS buffer (one byte beyond the end of - * the buffer) - * - interrupt pointer into BTS buffer - * (interrupt occurs when write pointer passes interrupt pointer) - * - double-word aligned base linear address of the PEBS buffer - * - write pointer into the PEBS buffer - * - end linear address of the PEBS buffer (one byte beyond the end of - * the buffer) - * - interrupt pointer into PEBS buffer - * (interrupt occurs when write pointer passes interrupt pointer) - * - value to which counter is reset following counter overflow - * - * Later architectures use 64bit pointers throughout, whereas earlier - * architectures use 32bit pointers in 32bit mode. - * - * - * We compute the base address for the first 8 fields based on: - * - the field size stored in the DS configuration - * - the relative field position - * - an offset giving the start of the respective region - * - * This offset is further used to index various arrays holding - * information for BTS and PEBS at the respective index. - * - * On later 32bit processors, we only access the lower 32bit of the - * 64bit pointer fields. The upper halves will be zeroed out. - */ - -enum ds_field { - ds_buffer_base = 0, - ds_index, - ds_absolute_maximum, - ds_interrupt_threshold, -}; - -enum ds_qualifier { - ds_bts = 0, - ds_pebs -}; - -static inline unsigned long -ds_get(const unsigned char *base, enum ds_qualifier qual, enum ds_field field) -{ - base += (ds_cfg.sizeof_ptr_field * (field + (4 * qual))); - return *(unsigned long *)base; -} - -static inline void -ds_set(unsigned char *base, enum ds_qualifier qual, enum ds_field field, - unsigned long value) -{ - base += (ds_cfg.sizeof_ptr_field * (field + (4 * qual))); - (*(unsigned long *)base) = value; -} - - -/* - * Locking is done only for allocating BTS or PEBS resources. - */ -static DEFINE_SPINLOCK(ds_lock); - -/* - * We either support (system-wide) per-cpu or per-thread allocation. - * We distinguish the two based on the task_struct pointer, where a - * NULL pointer indicates per-cpu allocation for the current cpu. - * - * Allocations are use-counted. As soon as resources are allocated, - * further allocations must be of the same type (per-cpu or - * per-thread). We model this by counting allocations (i.e. the number - * of tracers of a certain type) for one type negatively: - * =0 no tracers - * >0 number of per-thread tracers - * <0 number of per-cpu tracers - * - * Tracers essentially gives the number of ds contexts for a certain - * type of allocation. - */ -static atomic_t tracers = ATOMIC_INIT(0); - -static inline int get_tracer(struct task_struct *task) -{ - int error; - - spin_lock_irq(&ds_lock); - - if (task) { - error = -EPERM; - if (atomic_read(&tracers) < 0) - goto out; - atomic_inc(&tracers); - } else { - error = -EPERM; - if (atomic_read(&tracers) > 0) - goto out; - atomic_dec(&tracers); - } - - error = 0; -out: - spin_unlock_irq(&ds_lock); - return error; -} - -static inline void put_tracer(struct task_struct *task) -{ - if (task) - atomic_dec(&tracers); - else - atomic_inc(&tracers); -} - -/* - * The DS context is either attached to a thread or to a cpu: - * - in the former case, the thread_struct contains a pointer to the - * attached context. - * - in the latter case, we use a static array of per-cpu context - * pointers. - * - * Contexts are use-counted. They are allocated on first access and - * deallocated when the last user puts the context. - */ -struct ds_context { - /* The DS configuration; goes into MSR_IA32_DS_AREA: */ - unsigned char ds[MAX_SIZEOF_DS]; - - /* The owner of the BTS and PEBS configuration, respectively: */ - struct bts_tracer *bts_master; - struct pebs_tracer *pebs_master; - - /* Use count: */ - unsigned long count; - - /* Pointer to the context pointer field: */ - struct ds_context **this; - - /* The traced task; NULL for cpu tracing: */ - struct task_struct *task; - - /* The traced cpu; only valid if task is NULL: */ - int cpu; -}; - -static DEFINE_PER_CPU(struct ds_context *, cpu_ds_context); - - -static struct ds_context *ds_get_context(struct task_struct *task, int cpu) -{ - struct ds_context **p_context = - (task ? &task->thread.ds_ctx : &per_cpu(cpu_ds_context, cpu)); - struct ds_context *context = NULL; - struct ds_context *new_context = NULL; - - /* Chances are small that we already have a context. */ - new_context = kzalloc(sizeof(*new_context), GFP_KERNEL); - if (!new_context) - return NULL; - - spin_lock_irq(&ds_lock); - - context = *p_context; - if (likely(!context)) { - context = new_context; - - context->this = p_context; - context->task = task; - context->cpu = cpu; - context->count = 0; - - *p_context = context; - } - - context->count++; - - spin_unlock_irq(&ds_lock); - - if (context != new_context) - kfree(new_context); - - return context; -} - -static void ds_put_context(struct ds_context *context) -{ - struct task_struct *task; - unsigned long irq; - - if (!context) - return; - - spin_lock_irqsave(&ds_lock, irq); - - if (--context->count) { - spin_unlock_irqrestore(&ds_lock, irq); - return; - } - - *(context->this) = NULL; - - task = context->task; - - if (task) - clear_tsk_thread_flag(task, TIF_DS_AREA_MSR); - - /* - * We leave the (now dangling) pointer to the DS configuration in - * the DS_AREA msr. This is as good or as bad as replacing it with - * NULL - the hardware would crash if we enabled tracing. - * - * This saves us some problems with having to write an msr on a - * different cpu while preventing others from doing the same for the - * next context for that same cpu. - */ - - spin_unlock_irqrestore(&ds_lock, irq); - - /* The context might still be in use for context switching. */ - if (task && (task != current)) - wait_task_context_switch(task); - - kfree(context); -} - -static void ds_install_ds_area(struct ds_context *context) -{ - unsigned long ds; - - ds = (unsigned long)context->ds; - - /* - * There is a race between the bts master and the pebs master. - * - * The thread/cpu access is synchronized via get/put_cpu() for - * task tracing and via wrmsr_on_cpu for cpu tracing. - * - * If bts and pebs are collected for the same task or same cpu, - * the same confiuration is written twice. - */ - if (context->task) { - get_cpu(); - if (context->task == current) - wrmsrl(MSR_IA32_DS_AREA, ds); - set_tsk_thread_flag(context->task, TIF_DS_AREA_MSR); - put_cpu(); - } else - wrmsr_on_cpu(context->cpu, MSR_IA32_DS_AREA, - (u32)((u64)ds), (u32)((u64)ds >> 32)); -} - -/* - * Call the tracer's callback on a buffer overflow. - * - * context: the ds context - * qual: the buffer type - */ -static void ds_overflow(struct ds_context *context, enum ds_qualifier qual) -{ - switch (qual) { - case ds_bts: - if (context->bts_master && - context->bts_master->ovfl) - context->bts_master->ovfl(context->bts_master); - break; - case ds_pebs: - if (context->pebs_master && - context->pebs_master->ovfl) - context->pebs_master->ovfl(context->pebs_master); - break; - } -} - - -/* - * Write raw data into the BTS or PEBS buffer. - * - * The remainder of any partially written record is zeroed out. - * - * context: the DS context - * qual: the buffer type - * record: the data to write - * size: the size of the data - */ -static int ds_write(struct ds_context *context, enum ds_qualifier qual, - const void *record, size_t size) -{ - int bytes_written = 0; - - if (!record) - return -EINVAL; - - while (size) { - unsigned long base, index, end, write_end, int_th; - unsigned long write_size, adj_write_size; - - /* - * Write as much as possible without producing an - * overflow interrupt. - * - * Interrupt_threshold must either be - * - bigger than absolute_maximum or - * - point to a record between buffer_base and absolute_maximum - * - * Index points to a valid record. - */ - base = ds_get(context->ds, qual, ds_buffer_base); - index = ds_get(context->ds, qual, ds_index); - end = ds_get(context->ds, qual, ds_absolute_maximum); - int_th = ds_get(context->ds, qual, ds_interrupt_threshold); - - write_end = min(end, int_th); - - /* - * If we are already beyond the interrupt threshold, - * we fill the entire buffer. - */ - if (write_end <= index) - write_end = end; - - if (write_end <= index) - break; - - write_size = min((unsigned long) size, write_end - index); - memcpy((void *)index, record, write_size); - - record = (const char *)record + write_size; - size -= write_size; - bytes_written += write_size; - - adj_write_size = write_size / ds_cfg.sizeof_rec[qual]; - adj_write_size *= ds_cfg.sizeof_rec[qual]; - - /* Zero out trailing bytes. */ - memset((char *)index + write_size, 0, - adj_write_size - write_size); - index += adj_write_size; - - if (index >= end) - index = base; - ds_set(context->ds, qual, ds_index, index); - - if (index >= int_th) - ds_overflow(context, qual); - } - - return bytes_written; -} - - -/* - * Branch Trace Store (BTS) uses the following format. Different - * architectures vary in the size of those fields. - * - source linear address - * - destination linear address - * - flags - * - * Later architectures use 64bit pointers throughout, whereas earlier - * architectures use 32bit pointers in 32bit mode. - * - * We compute the base address for the fields based on: - * - the field size stored in the DS configuration - * - the relative field position - * - * In order to store additional information in the BTS buffer, we use - * a special source address to indicate that the record requires - * special interpretation. - * - * Netburst indicated via a bit in the flags field whether the branch - * was predicted; this is ignored. - * - * We use two levels of abstraction: - * - the raw data level defined here - * - an arch-independent level defined in ds.h - */ - -enum bts_field { - bts_from, - bts_to, - bts_flags, - - bts_qual = bts_from, - bts_clock = bts_to, - bts_pid = bts_flags, - - bts_qual_mask = (bts_qual_max - 1), - bts_escape = ((unsigned long)-1 & ~bts_qual_mask) -}; - -static inline unsigned long bts_get(const char *base, unsigned long field) -{ - base += (ds_cfg.sizeof_ptr_field * field); - return *(unsigned long *)base; -} - -static inline void bts_set(char *base, unsigned long field, unsigned long val) -{ - base += (ds_cfg.sizeof_ptr_field * field); - (*(unsigned long *)base) = val; -} - - -/* - * The raw BTS data is architecture dependent. - * - * For higher-level users, we give an arch-independent view. - * - ds.h defines struct bts_struct - * - bts_read translates one raw bts record into a bts_struct - * - bts_write translates one bts_struct into the raw format and - * writes it into the top of the parameter tracer's buffer. - * - * return: bytes read/written on success; -Eerrno, otherwise - */ -static int -bts_read(struct bts_tracer *tracer, const void *at, struct bts_struct *out) -{ - if (!tracer) - return -EINVAL; - - if (at < tracer->trace.ds.begin) - return -EINVAL; - - if (tracer->trace.ds.end < (at + tracer->trace.ds.size)) - return -EINVAL; - - memset(out, 0, sizeof(*out)); - if ((bts_get(at, bts_qual) & ~bts_qual_mask) == bts_escape) { - out->qualifier = (bts_get(at, bts_qual) & bts_qual_mask); - out->variant.event.clock = bts_get(at, bts_clock); - out->variant.event.pid = bts_get(at, bts_pid); - } else { - out->qualifier = bts_branch; - out->variant.lbr.from = bts_get(at, bts_from); - out->variant.lbr.to = bts_get(at, bts_to); - - if (!out->variant.lbr.from && !out->variant.lbr.to) - out->qualifier = bts_invalid; - } - - return ds_cfg.sizeof_rec[ds_bts]; -} - -static int bts_write(struct bts_tracer *tracer, const struct bts_struct *in) -{ - unsigned char raw[MAX_SIZEOF_BTS]; - - if (!tracer) - return -EINVAL; - - if (MAX_SIZEOF_BTS < ds_cfg.sizeof_rec[ds_bts]) - return -EOVERFLOW; - - switch (in->qualifier) { - case bts_invalid: - bts_set(raw, bts_from, 0); - bts_set(raw, bts_to, 0); - bts_set(raw, bts_flags, 0); - break; - case bts_branch: - bts_set(raw, bts_from, in->variant.lbr.from); - bts_set(raw, bts_to, in->variant.lbr.to); - bts_set(raw, bts_flags, 0); - break; - case bts_task_arrives: - case bts_task_departs: - bts_set(raw, bts_qual, (bts_escape | in->qualifier)); - bts_set(raw, bts_clock, in->variant.event.clock); - bts_set(raw, bts_pid, in->variant.event.pid); - break; - default: - return -EINVAL; - } - - return ds_write(tracer->ds.context, ds_bts, raw, - ds_cfg.sizeof_rec[ds_bts]); -} - - -static void ds_write_config(struct ds_context *context, - struct ds_trace *cfg, enum ds_qualifier qual) -{ - unsigned char *ds = context->ds; - - ds_set(ds, qual, ds_buffer_base, (unsigned long)cfg->begin); - ds_set(ds, qual, ds_index, (unsigned long)cfg->top); - ds_set(ds, qual, ds_absolute_maximum, (unsigned long)cfg->end); - ds_set(ds, qual, ds_interrupt_threshold, (unsigned long)cfg->ith); -} - -static void ds_read_config(struct ds_context *context, - struct ds_trace *cfg, enum ds_qualifier qual) -{ - unsigned char *ds = context->ds; - - cfg->begin = (void *)ds_get(ds, qual, ds_buffer_base); - cfg->top = (void *)ds_get(ds, qual, ds_index); - cfg->end = (void *)ds_get(ds, qual, ds_absolute_maximum); - cfg->ith = (void *)ds_get(ds, qual, ds_interrupt_threshold); -} - -static void ds_init_ds_trace(struct ds_trace *trace, enum ds_qualifier qual, - void *base, size_t size, size_t ith, - unsigned int flags) { - unsigned long buffer, adj; - - /* - * Adjust the buffer address and size to meet alignment - * constraints: - * - buffer is double-word aligned - * - size is multiple of record size - * - * We checked the size at the very beginning; we have enough - * space to do the adjustment. - */ - buffer = (unsigned long)base; - - adj = ALIGN(buffer, DS_ALIGNMENT) - buffer; - buffer += adj; - size -= adj; - - trace->n = size / ds_cfg.sizeof_rec[qual]; - trace->size = ds_cfg.sizeof_rec[qual]; - - size = (trace->n * trace->size); - - trace->begin = (void *)buffer; - trace->top = trace->begin; - trace->end = (void *)(buffer + size); - /* - * The value for 'no threshold' is -1, which will set the - * threshold outside of the buffer, just like we want it. - */ - ith *= ds_cfg.sizeof_rec[qual]; - trace->ith = (void *)(buffer + size - ith); - - trace->flags = flags; -} - - -static int ds_request(struct ds_tracer *tracer, struct ds_trace *trace, - enum ds_qualifier qual, struct task_struct *task, - int cpu, void *base, size_t size, size_t th) -{ - struct ds_context *context; - int error; - size_t req_size; - - error = -EOPNOTSUPP; - if (!ds_cfg.sizeof_rec[qual]) - goto out; - - error = -EINVAL; - if (!base) - goto out; - - req_size = ds_cfg.sizeof_rec[qual]; - /* We might need space for alignment adjustments. */ - if (!IS_ALIGNED((unsigned long)base, DS_ALIGNMENT)) - req_size += DS_ALIGNMENT; - - error = -EINVAL; - if (size < req_size) - goto out; - - if (th != (size_t)-1) { - th *= ds_cfg.sizeof_rec[qual]; - - error = -EINVAL; - if (size <= th) - goto out; - } - - tracer->buffer = base; - tracer->size = size; - - error = -ENOMEM; - context = ds_get_context(task, cpu); - if (!context) - goto out; - tracer->context = context; - - /* - * Defer any tracer-specific initialization work for the context until - * context ownership has been clarified. - */ - - error = 0; - out: - return error; -} - -static struct bts_tracer *ds_request_bts(struct task_struct *task, int cpu, - void *base, size_t size, - bts_ovfl_callback_t ovfl, size_t th, - unsigned int flags) -{ - struct bts_tracer *tracer; - int error; - - /* Buffer overflow notification is not yet implemented. */ - error = -EOPNOTSUPP; - if (ovfl) - goto out; - - error = get_tracer(task); - if (error < 0) - goto out; - - error = -ENOMEM; - tracer = kzalloc(sizeof(*tracer), GFP_KERNEL); - if (!tracer) - goto out_put_tracer; - tracer->ovfl = ovfl; - - /* Do some more error checking and acquire a tracing context. */ - error = ds_request(&tracer->ds, &tracer->trace.ds, - ds_bts, task, cpu, base, size, th); - if (error < 0) - goto out_tracer; - - /* Claim the bts part of the tracing context we acquired above. */ - spin_lock_irq(&ds_lock); - - error = -EPERM; - if (tracer->ds.context->bts_master) - goto out_unlock; - tracer->ds.context->bts_master = tracer; - - spin_unlock_irq(&ds_lock); - - /* - * Now that we own the bts part of the context, let's complete the - * initialization for that part. - */ - ds_init_ds_trace(&tracer->trace.ds, ds_bts, base, size, th, flags); - ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts); - ds_install_ds_area(tracer->ds.context); - - tracer->trace.read = bts_read; - tracer->trace.write = bts_write; - - /* Start tracing. */ - ds_resume_bts(tracer); - - return tracer; - - out_unlock: - spin_unlock_irq(&ds_lock); - ds_put_context(tracer->ds.context); - out_tracer: - kfree(tracer); - out_put_tracer: - put_tracer(task); - out: - return ERR_PTR(error); -} - -struct bts_tracer *ds_request_bts_task(struct task_struct *task, - void *base, size_t size, - bts_ovfl_callback_t ovfl, - size_t th, unsigned int flags) -{ - return ds_request_bts(task, 0, base, size, ovfl, th, flags); -} - -struct bts_tracer *ds_request_bts_cpu(int cpu, void *base, size_t size, - bts_ovfl_callback_t ovfl, - size_t th, unsigned int flags) -{ - return ds_request_bts(NULL, cpu, base, size, ovfl, th, flags); -} - -static struct pebs_tracer *ds_request_pebs(struct task_struct *task, int cpu, - void *base, size_t size, - pebs_ovfl_callback_t ovfl, size_t th, - unsigned int flags) -{ - struct pebs_tracer *tracer; - int error; - - /* Buffer overflow notification is not yet implemented. */ - error = -EOPNOTSUPP; - if (ovfl) - goto out; - - error = get_tracer(task); - if (error < 0) - goto out; - - error = -ENOMEM; - tracer = kzalloc(sizeof(*tracer), GFP_KERNEL); - if (!tracer) - goto out_put_tracer; - tracer->ovfl = ovfl; - - /* Do some more error checking and acquire a tracing context. */ - error = ds_request(&tracer->ds, &tracer->trace.ds, - ds_pebs, task, cpu, base, size, th); - if (error < 0) - goto out_tracer; - - /* Claim the pebs part of the tracing context we acquired above. */ - spin_lock_irq(&ds_lock); - - error = -EPERM; - if (tracer->ds.context->pebs_master) - goto out_unlock; - tracer->ds.context->pebs_master = tracer; - - spin_unlock_irq(&ds_lock); - - /* - * Now that we own the pebs part of the context, let's complete the - * initialization for that part. - */ - ds_init_ds_trace(&tracer->trace.ds, ds_pebs, base, size, th, flags); - ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_pebs); - ds_install_ds_area(tracer->ds.context); - - /* Start tracing. */ - ds_resume_pebs(tracer); - - return tracer; - - out_unlock: - spin_unlock_irq(&ds_lock); - ds_put_context(tracer->ds.context); - out_tracer: - kfree(tracer); - out_put_tracer: - put_tracer(task); - out: - return ERR_PTR(error); -} - -struct pebs_tracer *ds_request_pebs_task(struct task_struct *task, - void *base, size_t size, - pebs_ovfl_callback_t ovfl, - size_t th, unsigned int flags) -{ - return ds_request_pebs(task, 0, base, size, ovfl, th, flags); -} - -struct pebs_tracer *ds_request_pebs_cpu(int cpu, void *base, size_t size, - pebs_ovfl_callback_t ovfl, - size_t th, unsigned int flags) -{ - return ds_request_pebs(NULL, cpu, base, size, ovfl, th, flags); -} - -static void ds_free_bts(struct bts_tracer *tracer) -{ - struct task_struct *task; - - task = tracer->ds.context->task; - - WARN_ON_ONCE(tracer->ds.context->bts_master != tracer); - tracer->ds.context->bts_master = NULL; - - /* Make sure tracing stopped and the tracer is not in use. */ - if (task && (task != current)) - wait_task_context_switch(task); - - ds_put_context(tracer->ds.context); - put_tracer(task); - - kfree(tracer); -} - -void ds_release_bts(struct bts_tracer *tracer) -{ - might_sleep(); - - if (!tracer) - return; - - ds_suspend_bts(tracer); - ds_free_bts(tracer); -} - -int ds_release_bts_noirq(struct bts_tracer *tracer) -{ - struct task_struct *task; - unsigned long irq; - int error; - - if (!tracer) - return 0; - - task = tracer->ds.context->task; - - local_irq_save(irq); - - error = -EPERM; - if (!task && - (tracer->ds.context->cpu != smp_processor_id())) - goto out; - - error = -EPERM; - if (task && (task != current)) - goto out; - - ds_suspend_bts_noirq(tracer); - ds_free_bts(tracer); - - error = 0; - out: - local_irq_restore(irq); - return error; -} - -static void update_task_debugctlmsr(struct task_struct *task, - unsigned long debugctlmsr) -{ - task->thread.debugctlmsr = debugctlmsr; - - get_cpu(); - if (task == current) - update_debugctlmsr(debugctlmsr); - put_cpu(); -} - -void ds_suspend_bts(struct bts_tracer *tracer) -{ - struct task_struct *task; - unsigned long debugctlmsr; - int cpu; - - if (!tracer) - return; - - tracer->flags = 0; - - task = tracer->ds.context->task; - cpu = tracer->ds.context->cpu; - - WARN_ON(!task && irqs_disabled()); - - debugctlmsr = (task ? - task->thread.debugctlmsr : - get_debugctlmsr_on_cpu(cpu)); - debugctlmsr &= ~BTS_CONTROL; - - if (task) - update_task_debugctlmsr(task, debugctlmsr); - else - update_debugctlmsr_on_cpu(cpu, debugctlmsr); -} - -int ds_suspend_bts_noirq(struct bts_tracer *tracer) -{ - struct task_struct *task; - unsigned long debugctlmsr, irq; - int cpu, error = 0; - - if (!tracer) - return 0; - - tracer->flags = 0; - - task = tracer->ds.context->task; - cpu = tracer->ds.context->cpu; - - local_irq_save(irq); - - error = -EPERM; - if (!task && (cpu != smp_processor_id())) - goto out; - - debugctlmsr = (task ? - task->thread.debugctlmsr : - get_debugctlmsr()); - debugctlmsr &= ~BTS_CONTROL; - - if (task) - update_task_debugctlmsr(task, debugctlmsr); - else - update_debugctlmsr(debugctlmsr); - - error = 0; - out: - local_irq_restore(irq); - return error; -} - -static unsigned long ds_bts_control(struct bts_tracer *tracer) -{ - unsigned long control; - - control = ds_cfg.ctl[dsf_bts]; - if (!(tracer->trace.ds.flags & BTS_KERNEL)) - control |= ds_cfg.ctl[dsf_bts_kernel]; - if (!(tracer->trace.ds.flags & BTS_USER)) - control |= ds_cfg.ctl[dsf_bts_user]; - - return control; -} - -void ds_resume_bts(struct bts_tracer *tracer) -{ - struct task_struct *task; - unsigned long debugctlmsr; - int cpu; - - if (!tracer) - return; - - tracer->flags = tracer->trace.ds.flags; - - task = tracer->ds.context->task; - cpu = tracer->ds.context->cpu; - - WARN_ON(!task && irqs_disabled()); - - debugctlmsr = (task ? - task->thread.debugctlmsr : - get_debugctlmsr_on_cpu(cpu)); - debugctlmsr |= ds_bts_control(tracer); - - if (task) - update_task_debugctlmsr(task, debugctlmsr); - else - update_debugctlmsr_on_cpu(cpu, debugctlmsr); -} - -int ds_resume_bts_noirq(struct bts_tracer *tracer) -{ - struct task_struct *task; - unsigned long debugctlmsr, irq; - int cpu, error = 0; - - if (!tracer) - return 0; - - tracer->flags = tracer->trace.ds.flags; - - task = tracer->ds.context->task; - cpu = tracer->ds.context->cpu; - - local_irq_save(irq); - - error = -EPERM; - if (!task && (cpu != smp_processor_id())) - goto out; - - debugctlmsr = (task ? - task->thread.debugctlmsr : - get_debugctlmsr()); - debugctlmsr |= ds_bts_control(tracer); - - if (task) - update_task_debugctlmsr(task, debugctlmsr); - else - update_debugctlmsr(debugctlmsr); - - error = 0; - out: - local_irq_restore(irq); - return error; -} - -static void ds_free_pebs(struct pebs_tracer *tracer) -{ - struct task_struct *task; - - task = tracer->ds.context->task; - - WARN_ON_ONCE(tracer->ds.context->pebs_master != tracer); - tracer->ds.context->pebs_master = NULL; - - ds_put_context(tracer->ds.context); - put_tracer(task); - - kfree(tracer); -} - -void ds_release_pebs(struct pebs_tracer *tracer) -{ - might_sleep(); - - if (!tracer) - return; - - ds_suspend_pebs(tracer); - ds_free_pebs(tracer); -} - -int ds_release_pebs_noirq(struct pebs_tracer *tracer) -{ - struct task_struct *task; - unsigned long irq; - int error; - - if (!tracer) - return 0; - - task = tracer->ds.context->task; - - local_irq_save(irq); - - error = -EPERM; - if (!task && - (tracer->ds.context->cpu != smp_processor_id())) - goto out; - - error = -EPERM; - if (task && (task != current)) - goto out; - - ds_suspend_pebs_noirq(tracer); - ds_free_pebs(tracer); - - error = 0; - out: - local_irq_restore(irq); - return error; -} - -void ds_suspend_pebs(struct pebs_tracer *tracer) -{ - -} - -int ds_suspend_pebs_noirq(struct pebs_tracer *tracer) -{ - return 0; -} - -void ds_resume_pebs(struct pebs_tracer *tracer) -{ - -} - -int ds_resume_pebs_noirq(struct pebs_tracer *tracer) -{ - return 0; -} - -const struct bts_trace *ds_read_bts(struct bts_tracer *tracer) -{ - if (!tracer) - return NULL; - - ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_bts); - return &tracer->trace; -} - -const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer) -{ - if (!tracer) - return NULL; - - ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_pebs); - - tracer->trace.counters = ds_cfg.nr_counter_reset; - memcpy(tracer->trace.counter_reset, - tracer->ds.context->ds + - (NUM_DS_PTR_FIELDS * ds_cfg.sizeof_ptr_field), - ds_cfg.nr_counter_reset * PEBS_RESET_FIELD_SIZE); - - return &tracer->trace; -} - -int ds_reset_bts(struct bts_tracer *tracer) -{ - if (!tracer) - return -EINVAL; - - tracer->trace.ds.top = tracer->trace.ds.begin; - - ds_set(tracer->ds.context->ds, ds_bts, ds_index, - (unsigned long)tracer->trace.ds.top); - - return 0; -} - -int ds_reset_pebs(struct pebs_tracer *tracer) -{ - if (!tracer) - return -EINVAL; - - tracer->trace.ds.top = tracer->trace.ds.begin; - - ds_set(tracer->ds.context->ds, ds_pebs, ds_index, - (unsigned long)tracer->trace.ds.top); - - return 0; -} - -int ds_set_pebs_reset(struct pebs_tracer *tracer, - unsigned int counter, u64 value) -{ - if (!tracer) - return -EINVAL; - - if (ds_cfg.nr_counter_reset < counter) - return -EINVAL; - - *(u64 *)(tracer->ds.context->ds + - (NUM_DS_PTR_FIELDS * ds_cfg.sizeof_ptr_field) + - (counter * PEBS_RESET_FIELD_SIZE)) = value; - - return 0; -} - -static const struct ds_configuration ds_cfg_netburst = { - .name = "Netburst", - .ctl[dsf_bts] = (1 << 2) | (1 << 3), - .ctl[dsf_bts_kernel] = (1 << 5), - .ctl[dsf_bts_user] = (1 << 6), - .nr_counter_reset = 1, -}; -static const struct ds_configuration ds_cfg_pentium_m = { - .name = "Pentium M", - .ctl[dsf_bts] = (1 << 6) | (1 << 7), - .nr_counter_reset = 1, -}; -static const struct ds_configuration ds_cfg_core2_atom = { - .name = "Core 2/Atom", - .ctl[dsf_bts] = (1 << 6) | (1 << 7), - .ctl[dsf_bts_kernel] = (1 << 9), - .ctl[dsf_bts_user] = (1 << 10), - .nr_counter_reset = 1, -}; -static const struct ds_configuration ds_cfg_core_i7 = { - .name = "Core i7", - .ctl[dsf_bts] = (1 << 6) | (1 << 7), - .ctl[dsf_bts_kernel] = (1 << 9), - .ctl[dsf_bts_user] = (1 << 10), - .nr_counter_reset = 4, -}; - -static void -ds_configure(const struct ds_configuration *cfg, - struct cpuinfo_x86 *cpu) -{ - unsigned long nr_pebs_fields = 0; - - printk(KERN_INFO "[ds] using %s configuration\n", cfg->name); - -#ifdef __i386__ - nr_pebs_fields = 10; -#else - nr_pebs_fields = 18; -#endif - - /* - * Starting with version 2, architectural performance - * monitoring supports a format specifier. - */ - if ((cpuid_eax(0xa) & 0xff) > 1) { - unsigned long perf_capabilities, format; - - rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_capabilities); - - format = (perf_capabilities >> 8) & 0xf; - - switch (format) { - case 0: - nr_pebs_fields = 18; - break; - case 1: - nr_pebs_fields = 22; - break; - default: - printk(KERN_INFO - "[ds] unknown PEBS format: %lu\n", format); - nr_pebs_fields = 0; - break; - } - } - - memset(&ds_cfg, 0, sizeof(ds_cfg)); - ds_cfg = *cfg; - - ds_cfg.sizeof_ptr_field = - (cpu_has(cpu, X86_FEATURE_DTES64) ? 8 : 4); - - ds_cfg.sizeof_rec[ds_bts] = ds_cfg.sizeof_ptr_field * 3; - ds_cfg.sizeof_rec[ds_pebs] = ds_cfg.sizeof_ptr_field * nr_pebs_fields; - - if (!cpu_has(cpu, X86_FEATURE_BTS)) { - ds_cfg.sizeof_rec[ds_bts] = 0; - printk(KERN_INFO "[ds] bts not available\n"); - } - if (!cpu_has(cpu, X86_FEATURE_PEBS)) { - ds_cfg.sizeof_rec[ds_pebs] = 0; - printk(KERN_INFO "[ds] pebs not available\n"); - } - - printk(KERN_INFO "[ds] sizes: address: %u bit, ", - 8 * ds_cfg.sizeof_ptr_field); - printk("bts/pebs record: %u/%u bytes\n", - ds_cfg.sizeof_rec[ds_bts], ds_cfg.sizeof_rec[ds_pebs]); - - WARN_ON_ONCE(MAX_PEBS_COUNTERS < ds_cfg.nr_counter_reset); -} - -void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) -{ - /* Only configure the first cpu. Others are identical. */ - if (ds_cfg.name) - return; - - switch (c->x86) { - case 0x6: - switch (c->x86_model) { - case 0x9: - case 0xd: /* Pentium M */ - ds_configure(&ds_cfg_pentium_m, c); - break; - case 0xf: - case 0x17: /* Core2 */ - case 0x1c: /* Atom */ - ds_configure(&ds_cfg_core2_atom, c); - break; - case 0x1a: /* Core i7 */ - ds_configure(&ds_cfg_core_i7, c); - break; - default: - /* Sorry, don't know about them. */ - break; - } - break; - case 0xf: - switch (c->x86_model) { - case 0x0: - case 0x1: - case 0x2: /* Netburst */ - ds_configure(&ds_cfg_netburst, c); - break; - default: - /* Sorry, don't know about them. */ - break; - } - break; - default: - /* Sorry, don't know about them. */ - break; - } -} - -static inline void ds_take_timestamp(struct ds_context *context, - enum bts_qualifier qualifier, - struct task_struct *task) -{ - struct bts_tracer *tracer = context->bts_master; - struct bts_struct ts; - - /* Prevent compilers from reading the tracer pointer twice. */ - barrier(); - - if (!tracer || !(tracer->flags & BTS_TIMESTAMPS)) - return; - - memset(&ts, 0, sizeof(ts)); - ts.qualifier = qualifier; - ts.variant.event.clock = trace_clock_global(); - ts.variant.event.pid = task->pid; - - bts_write(tracer, &ts); -} - -/* - * Change the DS configuration from tracing prev to tracing next. - */ -void ds_switch_to(struct task_struct *prev, struct task_struct *next) -{ - struct ds_context *prev_ctx = prev->thread.ds_ctx; - struct ds_context *next_ctx = next->thread.ds_ctx; - unsigned long debugctlmsr = next->thread.debugctlmsr; - - /* Make sure all data is read before we start. */ - barrier(); - - if (prev_ctx) { - update_debugctlmsr(0); - - ds_take_timestamp(prev_ctx, bts_task_departs, prev); - } - - if (next_ctx) { - ds_take_timestamp(next_ctx, bts_task_arrives, next); - - wrmsrl(MSR_IA32_DS_AREA, (unsigned long)next_ctx->ds); - } - - update_debugctlmsr(debugctlmsr); -} - -static __init int ds_selftest(void) -{ - if (ds_cfg.sizeof_rec[ds_bts]) { - int error; - - error = ds_selftest_bts(); - if (error) { - WARN(1, "[ds] selftest failed. disabling bts.\n"); - ds_cfg.sizeof_rec[ds_bts] = 0; - } - } - - if (ds_cfg.sizeof_rec[ds_pebs]) { - int error; - - error = ds_selftest_pebs(); - if (error) { - WARN(1, "[ds] selftest failed. disabling pebs.\n"); - ds_cfg.sizeof_rec[ds_pebs] = 0; - } - } - - return 0; -} -device_initcall(ds_selftest); diff --git a/arch/x86/kernel/ds_selftest.c b/arch/x86/kernel/ds_selftest.c deleted file mode 100644 index 6bc7c199ab99..000000000000 --- a/arch/x86/kernel/ds_selftest.c +++ /dev/null @@ -1,408 +0,0 @@ -/* - * Debug Store support - selftest - * - * - * Copyright (C) 2009 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, 2009 - */ - -#include "ds_selftest.h" - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/smp.h> -#include <linux/cpu.h> - -#include <asm/ds.h> - - -#define BUFFER_SIZE 521 /* Intentionally chose an odd size. */ -#define SMALL_BUFFER_SIZE 24 /* A single bts entry. */ - -struct ds_selftest_bts_conf { - struct bts_tracer *tracer; - int error; - int (*suspend)(struct bts_tracer *); - int (*resume)(struct bts_tracer *); -}; - -static int ds_selftest_bts_consistency(const struct bts_trace *trace) -{ - int error = 0; - - if (!trace) { - printk(KERN_CONT "failed to access trace..."); - /* Bail out. Other tests are pointless. */ - return -1; - } - - if (!trace->read) { - printk(KERN_CONT "bts read not available..."); - error = -1; - } - - /* Do some sanity checks on the trace configuration. */ - if (!trace->ds.n) { - printk(KERN_CONT "empty bts buffer..."); - error = -1; - } - if (!trace->ds.size) { - printk(KERN_CONT "bad bts trace setup..."); - error = -1; - } - if (trace->ds.end != - (char *)trace->ds.begin + (trace->ds.n * trace->ds.size)) { - printk(KERN_CONT "bad bts buffer setup..."); - error = -1; - } - /* - * We allow top in [begin; end], since its not clear when the - * overflow adjustment happens: after the increment or before the - * write. - */ - if ((trace->ds.top < trace->ds.begin) || - (trace->ds.end < trace->ds.top)) { - printk(KERN_CONT "bts top out of bounds..."); - error = -1; - } - - return error; -} - -static int ds_selftest_bts_read(struct bts_tracer *tracer, - const struct bts_trace *trace, - const void *from, const void *to) -{ - const unsigned char *at; - - /* - * Check a few things which do not belong to this test. - * They should be covered by other tests. - */ - if (!trace) - return -1; - - if (!trace->read) - return -1; - - if (to < from) - return -1; - - if (from < trace->ds.begin) - return -1; - - if (trace->ds.end < to) - return -1; - - if (!trace->ds.size) - return -1; - - /* Now to the test itself. */ - for (at = from; (void *)at < to; at += trace->ds.size) { - struct bts_struct bts; - unsigned long index; - int error; - - if (((void *)at - trace->ds.begin) % trace->ds.size) { - printk(KERN_CONT - "read from non-integer index..."); - return -1; - } - index = ((void *)at - trace->ds.begin) / trace->ds.size; - - memset(&bts, 0, sizeof(bts)); - error = trace->read(tracer, at, &bts); - if (error < 0) { - printk(KERN_CONT - "error reading bts trace at [%lu] (0x%p)...", - index, at); - return error; - } - - switch (bts.qualifier) { - case BTS_BRANCH: - break; - default: - printk(KERN_CONT - "unexpected bts entry %llu at [%lu] (0x%p)...", - bts.qualifier, index, at); - return -1; - } - } - - return 0; -} - -static void ds_selftest_bts_cpu(void *arg) -{ - struct ds_selftest_bts_conf *conf = arg; - const struct bts_trace *trace; - void *top; - - if (IS_ERR(conf->tracer)) { - conf->error = PTR_ERR(conf->tracer); - conf->tracer = NULL; - - printk(KERN_CONT - "initialization failed (err: %d)...", conf->error); - return; - } - - /* We should meanwhile have enough trace. */ - conf->error = conf->suspend(conf->tracer); - if (conf->error < 0) - return; - - /* Let's see if we can access the trace. */ - trace = ds_read_bts(conf->tracer); - - conf->error = ds_selftest_bts_consistency(trace); - if (conf->error < 0) - return; - - /* If everything went well, we should have a few trace entries. */ - if (trace->ds.top == trace->ds.begin) { - /* - * It is possible but highly unlikely that we got a - * buffer overflow and end up at exactly the same - * position we started from. - * Let's issue a warning, but continue. - */ - printk(KERN_CONT "no trace/overflow..."); - } - - /* Let's try to read the trace we collected. */ - conf->error = - ds_selftest_bts_read(conf->tracer, trace, - trace->ds.begin, trace->ds.top); - if (conf->error < 0) - return; - - /* - * Let's read the trace again. - * Since we suspended tracing, we should get the same result. - */ - top = trace->ds.top; - - trace = ds_read_bts(conf->tracer); - conf->error = ds_selftest_bts_consistency(trace); - if (conf->error < 0) - return; - - if (top != trace->ds.top) { - printk(KERN_CONT "suspend not working..."); - conf->error = -1; - return; - } - - /* Let's collect some more trace - see if resume is working. */ - conf->error = conf->resume(conf->tracer); - if (conf->error < 0) - return; - - conf->error = conf->suspend(conf->tracer); - if (conf->error < 0) - return; - - trace = ds_read_bts(conf->tracer); - - conf->error = ds_selftest_bts_consistency(trace); - if (conf->error < 0) - return; - - if (trace->ds.top == top) { - /* - * It is possible but highly unlikely that we got a - * buffer overflow and end up at exactly the same - * position we started from. - * Let's issue a warning and check the full trace. - */ - printk(KERN_CONT - "no resume progress/overflow..."); - - conf->error = - ds_selftest_bts_read(conf->tracer, trace, - trace->ds.begin, trace->ds.end); - } else if (trace->ds.top < top) { - /* - * We had a buffer overflow - the entire buffer should - * contain trace records. - */ - conf->error = - ds_selftest_bts_read(conf->tracer, trace, - trace->ds.begin, trace->ds.end); - } else { - /* - * It is quite likely that the buffer did not overflow. - * Let's just check the delta trace. - */ - conf->error = - ds_selftest_bts_read(conf->tracer, trace, top, - trace->ds.top); - } - if (conf->error < 0) - return; - - conf->error = 0; -} - -static int ds_suspend_bts_wrap(struct bts_tracer *tracer) -{ - ds_suspend_bts(tracer); - return 0; -} - -static int ds_resume_bts_wrap(struct bts_tracer *tracer) -{ - ds_resume_bts(tracer); - return 0; -} - -static void ds_release_bts_noirq_wrap(void *tracer) -{ - (void)ds_release_bts_noirq(tracer); -} - -static int ds_selftest_bts_bad_release_noirq(int cpu, - struct bts_tracer *tracer) -{ - int error = -EPERM; - - /* Try to release the tracer on the wrong cpu. */ - get_cpu(); - if (cpu != smp_processor_id()) { - error = ds_release_bts_noirq(tracer); - if (error != -EPERM) - printk(KERN_CONT "release on wrong cpu..."); - } - put_cpu(); - - return error ? 0 : -1; -} - -static int ds_selftest_bts_bad_request_cpu(int cpu, void *buffer) -{ - struct bts_tracer *tracer; - int error; - - /* Try to request cpu tracing while task tracing is active. */ - tracer = ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE, NULL, - (size_t)-1, BTS_KERNEL); - error = PTR_ERR(tracer); - if (!IS_ERR(tracer)) { - ds_release_bts(tracer); - error = 0; - } - - if (error != -EPERM) - printk(KERN_CONT "cpu/task tracing overlap..."); - - return error ? 0 : -1; -} - -static int ds_selftest_bts_bad_request_task(void *buffer) -{ - struct bts_tracer *tracer; - int error; - - /* Try to request cpu tracing while task tracing is active. */ - tracer = ds_request_bts_task(current, buffer, BUFFER_SIZE, NULL, - (size_t)-1, BTS_KERNEL); - error = PTR_ERR(tracer); - if (!IS_ERR(tracer)) { - error = 0; - ds_release_bts(tracer); - } - - if (error != -EPERM) - printk(KERN_CONT "task/cpu tracing overlap..."); - - return error ? 0 : -1; -} - -int ds_selftest_bts(void) -{ - struct ds_selftest_bts_conf conf; - unsigned char buffer[BUFFER_SIZE], *small_buffer; - unsigned long irq; - int cpu; - - printk(KERN_INFO "[ds] bts selftest..."); - conf.error = 0; - - small_buffer = (unsigned char *)ALIGN((unsigned long)buffer, 8) + 8; - - get_online_cpus(); - for_each_online_cpu(cpu) { - conf.suspend = ds_suspend_bts_wrap; - conf.resume = ds_resume_bts_wrap; - conf.tracer = - ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE, - NULL, (size_t)-1, BTS_KERNEL); - ds_selftest_bts_cpu(&conf); - if (conf.error >= 0) - conf.error = ds_selftest_bts_bad_request_task(buffer); - ds_release_bts(conf.tracer); - if (conf.error < 0) - goto out; - - conf.suspend = ds_suspend_bts_noirq; - conf.resume = ds_resume_bts_noirq; - conf.tracer = - ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE, - NULL, (size_t)-1, BTS_KERNEL); - smp_call_function_single(cpu, ds_selftest_bts_cpu, &conf, 1); - if (conf.error >= 0) { - conf.error = - ds_selftest_bts_bad_release_noirq(cpu, - conf.tracer); - /* We must not release the tracer twice. */ - if (conf.error < 0) - conf.tracer = NULL; - } - if (conf.error >= 0) - conf.error = ds_selftest_bts_bad_request_task(buffer); - smp_call_function_single(cpu, ds_release_bts_noirq_wrap, - conf.tracer, 1); - if (conf.error < 0) - goto out; - } - - conf.suspend = ds_suspend_bts_wrap; - conf.resume = ds_resume_bts_wrap; - conf.tracer = - ds_request_bts_task(current, buffer, BUFFER_SIZE, - NULL, (size_t)-1, BTS_KERNEL); - ds_selftest_bts_cpu(&conf); - if (conf.error >= 0) - conf.error = ds_selftest_bts_bad_request_cpu(0, buffer); - ds_release_bts(conf.tracer); - if (conf.error < 0) - goto out; - - conf.suspend = ds_suspend_bts_noirq; - conf.resume = ds_resume_bts_noirq; - conf.tracer = - ds_request_bts_task(current, small_buffer, SMALL_BUFFER_SIZE, - NULL, (size_t)-1, BTS_KERNEL); - local_irq_save(irq); - ds_selftest_bts_cpu(&conf); - if (conf.error >= 0) - conf.error = ds_selftest_bts_bad_request_cpu(0, buffer); - ds_release_bts_noirq(conf.tracer); - local_irq_restore(irq); - if (conf.error < 0) - goto out; - - conf.error = 0; - out: - put_online_cpus(); - printk(KERN_CONT "%s.\n", (conf.error ? "failed" : "passed")); - - return conf.error; -} - -int ds_selftest_pebs(void) -{ - return 0; -} diff --git a/arch/x86/kernel/ds_selftest.h b/arch/x86/kernel/ds_selftest.h deleted file mode 100644 index 2ba8745c6663..000000000000 --- a/arch/x86/kernel/ds_selftest.h +++ /dev/null @@ -1,15 +0,0 @@ -/* - * Debug Store support - selftest - * - * - * Copyright (C) 2009 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, 2009 - */ - -#ifdef CONFIG_X86_DS_SELFTEST -extern int ds_selftest_bts(void); -extern int ds_selftest_pebs(void); -#else -static inline int ds_selftest_bts(void) { return 0; } -static inline int ds_selftest_pebs(void) { return 0; } -#endif diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 6d817554780a..c89a386930b7 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -224,11 +224,6 @@ unsigned __kprobes long oops_begin(void) int cpu; unsigned long flags; - /* notify the hw-branch tracer so it may disable tracing and - add the last trace to the trace buffer - - the earlier this happens, the more useful the trace. */ - trace_hw_branch_oops(); - oops_enter(); /* racy, but better than risking deadlock. */ diff --git a/arch/x86/kernel/dumpstack.h b/arch/x86/kernel/dumpstack.h index e39e77168a37..e1a93be4fd44 100644 --- a/arch/x86/kernel/dumpstack.h +++ b/arch/x86/kernel/dumpstack.h @@ -14,6 +14,8 @@ #define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) #endif +#include <linux/uaccess.h> + extern void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, char *log_lvl); @@ -42,8 +44,10 @@ static inline unsigned long rewind_frame_pointer(int n) get_bp(frame); #ifdef CONFIG_FRAME_POINTER - while (n--) - frame = frame->next_frame; + while (n--) { + if (probe_kernel_address(&frame->next_frame, frame)) + break; + } #endif return (unsigned long)frame; diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c index 23c167925a5c..2dfd31597443 100644 --- a/arch/x86/kernel/i8253.c +++ b/arch/x86/kernel/i8253.c @@ -16,7 +16,7 @@ #include <asm/hpet.h> #include <asm/smp.h> -DEFINE_SPINLOCK(i8253_lock); +DEFINE_RAW_SPINLOCK(i8253_lock); EXPORT_SYMBOL(i8253_lock); /* @@ -33,7 +33,7 @@ struct clock_event_device *global_clock_event; static void init_pit_timer(enum clock_event_mode mode, struct clock_event_device *evt) { - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); switch (mode) { case CLOCK_EVT_MODE_PERIODIC: @@ -62,7 +62,7 @@ static void init_pit_timer(enum clock_event_mode mode, /* Nothing to do here */ break; } - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); } /* @@ -72,10 +72,10 @@ static void init_pit_timer(enum clock_event_mode mode, */ static int pit_next_event(unsigned long delta, struct clock_event_device *evt) { - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); outb_pit(delta & 0xff , PIT_CH0); /* LSB */ outb_pit(delta >> 8 , PIT_CH0); /* MSB */ - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); return 0; } @@ -130,7 +130,7 @@ static cycle_t pit_read(struct clocksource *cs) int count; u32 jifs; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); /* * Although our caller may have the read side of xtime_lock, * this is now a seqlock, and we are cheating in this routine @@ -176,7 +176,7 @@ static cycle_t pit_read(struct clocksource *cs) old_count = count; old_jifs = jifs; - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); count = (LATCH - 1) - count; diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index b43bbaebe2c0..f2f56c0967b6 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -422,14 +422,22 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs, static void __kprobes clear_btf(void) { - if (test_thread_flag(TIF_DEBUGCTLMSR)) - update_debugctlmsr(0); + if (test_thread_flag(TIF_BLOCKSTEP)) { + unsigned long debugctl = get_debugctlmsr(); + + debugctl &= ~DEBUGCTLMSR_BTF; + update_debugctlmsr(debugctl); + } } static void __kprobes restore_btf(void) { - if (test_thread_flag(TIF_DEBUGCTLMSR)) - update_debugctlmsr(current->thread.debugctlmsr); + if (test_thread_flag(TIF_BLOCKSTEP)) { + unsigned long debugctl = get_debugctlmsr(); + + debugctl |= DEBUGCTLMSR_BTF; + update_debugctlmsr(debugctl); + } } void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index cceb5bc3c3c2..2cd8c544e41a 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -201,9 +201,9 @@ static int do_microcode_update(const void __user *buf, size_t size) return error; } -static int microcode_open(struct inode *unused1, struct file *unused2) +static int microcode_open(struct inode *inode, struct file *file) { - return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; + return capable(CAP_SYS_RAWIO) ? nonseekable_open(inode, file) : -EPERM; } static ssize_t microcode_write(struct file *file, const char __user *buf, diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c index 85a343e28937..356170262a93 100644 --- a/arch/x86/kernel/microcode_intel.c +++ b/arch/x86/kernel/microcode_intel.c @@ -343,10 +343,11 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, int (*get_ucode_data)(void *, const void *, size_t)) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - u8 *ucode_ptr = data, *new_mc = NULL, *mc; + u8 *ucode_ptr = data, *new_mc = NULL, *mc = NULL; int new_rev = uci->cpu_sig.rev; unsigned int leftover = size; enum ucode_state state = UCODE_OK; + unsigned int curr_mc_size = 0; while (leftover) { struct microcode_header_intel mc_header; @@ -361,9 +362,15 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, break; } - mc = vmalloc(mc_size); - if (!mc) - break; + /* For performance reasons, reuse mc area when possible */ + if (!mc || mc_size > curr_mc_size) { + if (mc) + vfree(mc); + mc = vmalloc(mc_size); + if (!mc) + break; + curr_mc_size = mc_size; + } if (get_ucode_data(mc, ucode_ptr, mc_size) || microcode_sanity_check(mc) < 0) { @@ -376,13 +383,16 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, vfree(new_mc); new_rev = mc_header.rev; new_mc = mc; - } else - vfree(mc); + mc = NULL; /* trigger new vmalloc */ + } ucode_ptr += mc_size; leftover -= mc_size; } + if (mc) + vfree(mc); + if (leftover) { if (new_mc) vfree(new_mc); diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 68cd24f9deae..0f7f130caa67 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -565,6 +565,9 @@ static void enable_gart_translations(void) enable_gart_translation(dev, __pa(agp_gatt_table)); } + + /* Flush the GART-TLB to remove stale entries */ + k8_flush_garts(); } /* diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 28ad9f4d8b94..eccdb57094e3 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -20,7 +20,6 @@ #include <asm/idle.h> #include <asm/uaccess.h> #include <asm/i387.h> -#include <asm/ds.h> #include <asm/debugreg.h> unsigned long idle_halt; @@ -50,8 +49,6 @@ void free_thread_xstate(struct task_struct *tsk) kmem_cache_free(task_xstate_cachep, tsk->thread.xstate); tsk->thread.xstate = NULL; } - - WARN(tsk->thread.ds_ctx, "leaking DS context\n"); } void free_thread_info(struct thread_info *ti) @@ -198,11 +195,16 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, prev = &prev_p->thread; next = &next_p->thread; - if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) || - test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR)) - ds_switch_to(prev_p, next_p); - else if (next->debugctlmsr != prev->debugctlmsr) - update_debugctlmsr(next->debugctlmsr); + if (test_tsk_thread_flag(prev_p, TIF_BLOCKSTEP) ^ + test_tsk_thread_flag(next_p, TIF_BLOCKSTEP)) { + unsigned long debugctl = get_debugctlmsr(); + + debugctl &= ~DEBUGCTLMSR_BTF; + if (test_tsk_thread_flag(next_p, TIF_BLOCKSTEP)) + debugctl |= DEBUGCTLMSR_BTF; + + update_debugctlmsr(debugctl); + } if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^ test_tsk_thread_flag(next_p, TIF_NOTSC)) { diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index f6c62667e30c..75090c589b7a 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -55,7 +55,6 @@ #include <asm/cpu.h> #include <asm/idle.h> #include <asm/syscalls.h> -#include <asm/ds.h> #include <asm/debugreg.h> asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -238,13 +237,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, kfree(p->thread.io_bitmap_ptr); p->thread.io_bitmap_max = 0; } - - clear_tsk_thread_flag(p, TIF_DS_AREA_MSR); - p->thread.ds_ctx = NULL; - - clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); - p->thread.debugctlmsr = 0; - return err; } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index dc9690b4c4cc..cc4258f2beb5 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -49,7 +49,6 @@ #include <asm/ia32.h> #include <asm/idle.h> #include <asm/syscalls.h> -#include <asm/ds.h> #include <asm/debugreg.h> asmlinkage extern void ret_from_fork(void); @@ -313,13 +312,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, if (err) goto out; } - - clear_tsk_thread_flag(p, TIF_DS_AREA_MSR); - p->thread.ds_ctx = NULL; - - clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); - p->thread.debugctlmsr = 0; - err = 0; out: if (err && p->thread.io_bitmap_ptr) { diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 2e9b55027b7e..055be0afd330 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -2,9 +2,6 @@ /* * Pentium III FXSR, SSE support * Gareth Hughes <gareth@valinux.com>, May 2000 - * - * BTS tracing - * Markus Metzger <markus.t.metzger@intel.com>, Dec 2007 */ #include <linux/kernel.h> @@ -22,7 +19,6 @@ #include <linux/audit.h> #include <linux/seccomp.h> #include <linux/signal.h> -#include <linux/workqueue.h> #include <linux/perf_event.h> #include <linux/hw_breakpoint.h> @@ -36,7 +32,6 @@ #include <asm/desc.h> #include <asm/prctl.h> #include <asm/proto.h> -#include <asm/ds.h> #include <asm/hw_breakpoint.h> #include "tls.h" @@ -789,342 +784,6 @@ static int ioperm_get(struct task_struct *target, 0, IO_BITMAP_BYTES); } -#ifdef CONFIG_X86_PTRACE_BTS -/* - * A branch trace store context. - * - * Contexts may only be installed by ptrace_bts_config() and only for - * ptraced tasks. - * - * Contexts are destroyed when the tracee is detached from the tracer. - * The actual destruction work requires interrupts enabled, so the - * work is deferred and will be scheduled during __ptrace_unlink(). - * - * Contexts hold an additional task_struct reference on the traced - * task, as well as a reference on the tracer's mm. - * - * Ptrace already holds a task_struct for the duration of ptrace operations, - * but since destruction is deferred, it may be executed after both - * tracer and tracee exited. - */ -struct bts_context { - /* The branch trace handle. */ - struct bts_tracer *tracer; - - /* The buffer used to store the branch trace and its size. */ - void *buffer; - unsigned int size; - - /* The mm that paid for the above buffer. */ - struct mm_struct *mm; - - /* The task this context belongs to. */ - struct task_struct *task; - - /* The signal to send on a bts buffer overflow. */ - unsigned int bts_ovfl_signal; - - /* The work struct to destroy a context. */ - struct work_struct work; -}; - -static int alloc_bts_buffer(struct bts_context *context, unsigned int size) -{ - void *buffer = NULL; - int err = -ENOMEM; - - err = account_locked_memory(current->mm, current->signal->rlim, size); - if (err < 0) - return err; - - buffer = kzalloc(size, GFP_KERNEL); - if (!buffer) - goto out_refund; - - context->buffer = buffer; - context->size = size; - context->mm = get_task_mm(current); - - return 0; - - out_refund: - refund_locked_memory(current->mm, size); - return err; -} - -static inline void free_bts_buffer(struct bts_context *context) -{ - if (!context->buffer) - return; - - kfree(context->buffer); - context->buffer = NULL; - - refund_locked_memory(context->mm, context->size); - context->size = 0; - - mmput(context->mm); - context->mm = NULL; -} - -static void free_bts_context_work(struct work_struct *w) -{ - struct bts_context *context; - - context = container_of(w, struct bts_context, work); - - ds_release_bts(context->tracer); - put_task_struct(context->task); - free_bts_buffer(context); - kfree(context); -} - -static inline void free_bts_context(struct bts_context *context) -{ - INIT_WORK(&context->work, free_bts_context_work); - schedule_work(&context->work); -} - -static inline struct bts_context *alloc_bts_context(struct task_struct *task) -{ - struct bts_context *context = kzalloc(sizeof(*context), GFP_KERNEL); - if (context) { - context->task = task; - task->bts = context; - - get_task_struct(task); - } - - return context; -} - -static int ptrace_bts_read_record(struct task_struct *child, size_t index, - struct bts_struct __user *out) -{ - struct bts_context *context; - const struct bts_trace *trace; - struct bts_struct bts; - const unsigned char *at; - int error; - - context = child->bts; - if (!context) - return -ESRCH; - - trace = ds_read_bts(context->tracer); - if (!trace) - return -ESRCH; - - at = trace->ds.top - ((index + 1) * trace->ds.size); - if ((void *)at < trace->ds.begin) - at += (trace->ds.n * trace->ds.size); - - if (!trace->read) - return -EOPNOTSUPP; - - error = trace->read(context->tracer, at, &bts); - if (error < 0) - return error; - - if (copy_to_user(out, &bts, sizeof(bts))) - return -EFAULT; - - return sizeof(bts); -} - -static int ptrace_bts_drain(struct task_struct *child, - long size, - struct bts_struct __user *out) -{ - struct bts_context *context; - const struct bts_trace *trace; - const unsigned char *at; - int error, drained = 0; - - context = child->bts; - if (!context) - return -ESRCH; - - trace = ds_read_bts(context->tracer); - if (!trace) - return -ESRCH; - - if (!trace->read) - return -EOPNOTSUPP; - - if (size < (trace->ds.top - trace->ds.begin)) - return -EIO; - - for (at = trace->ds.begin; (void *)at < trace->ds.top; - out++, drained++, at += trace->ds.size) { - struct bts_struct bts; - - error = trace->read(context->tracer, at, &bts); - if (error < 0) - return error; - - if (copy_to_user(out, &bts, sizeof(bts))) - return -EFAULT; - } - - memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size); - - error = ds_reset_bts(context->tracer); - if (error < 0) - return error; - - return drained; -} - -static int ptrace_bts_config(struct task_struct *child, - long cfg_size, - const struct ptrace_bts_config __user *ucfg) -{ - struct bts_context *context; - struct ptrace_bts_config cfg; - unsigned int flags = 0; - - if (cfg_size < sizeof(cfg)) - return -EIO; - - if (copy_from_user(&cfg, ucfg, sizeof(cfg))) - return -EFAULT; - - context = child->bts; - if (!context) - context = alloc_bts_context(child); - if (!context) - return -ENOMEM; - - if (cfg.flags & PTRACE_BTS_O_SIGNAL) { - if (!cfg.signal) - return -EINVAL; - - return -EOPNOTSUPP; - context->bts_ovfl_signal = cfg.signal; - } - - ds_release_bts(context->tracer); - context->tracer = NULL; - - if ((cfg.flags & PTRACE_BTS_O_ALLOC) && (cfg.size != context->size)) { - int err; - - free_bts_buffer(context); - if (!cfg.size) - return 0; - - err = alloc_bts_buffer(context, cfg.size); - if (err < 0) - return err; - } - - if (cfg.flags & PTRACE_BTS_O_TRACE) - flags |= BTS_USER; - - if (cfg.flags & PTRACE_BTS_O_SCHED) - flags |= BTS_TIMESTAMPS; - - context->tracer = - ds_request_bts_task(child, context->buffer, context->size, - NULL, (size_t)-1, flags); - if (unlikely(IS_ERR(context->tracer))) { - int error = PTR_ERR(context->tracer); - - free_bts_buffer(context); - context->tracer = NULL; - return error; - } - - return sizeof(cfg); -} - -static int ptrace_bts_status(struct task_struct *child, - long cfg_size, - struct ptrace_bts_config __user *ucfg) -{ - struct bts_context *context; - const struct bts_trace *trace; - struct ptrace_bts_config cfg; - - context = child->bts; - if (!context) - return -ESRCH; - - if (cfg_size < sizeof(cfg)) - return -EIO; - - trace = ds_read_bts(context->tracer); - if (!trace) - return -ESRCH; - - memset(&cfg, 0, sizeof(cfg)); - cfg.size = trace->ds.end - trace->ds.begin; - cfg.signal = context->bts_ovfl_signal; - cfg.bts_size = sizeof(struct bts_struct); - - if (cfg.signal) - cfg.flags |= PTRACE_BTS_O_SIGNAL; - - if (trace->ds.flags & BTS_USER) - cfg.flags |= PTRACE_BTS_O_TRACE; - - if (trace->ds.flags & BTS_TIMESTAMPS) - cfg.flags |= PTRACE_BTS_O_SCHED; - - if (copy_to_user(ucfg, &cfg, sizeof(cfg))) - return -EFAULT; - - return sizeof(cfg); -} - -static int ptrace_bts_clear(struct task_struct *child) -{ - struct bts_context *context; - const struct bts_trace *trace; - - context = child->bts; - if (!context) - return -ESRCH; - - trace = ds_read_bts(context->tracer); - if (!trace) - return -ESRCH; - - memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size); - - return ds_reset_bts(context->tracer); -} - -static int ptrace_bts_size(struct task_struct *child) -{ - struct bts_context *context; - const struct bts_trace *trace; - - context = child->bts; - if (!context) - return -ESRCH; - - trace = ds_read_bts(context->tracer); - if (!trace) - return -ESRCH; - - return (trace->ds.top - trace->ds.begin) / trace->ds.size; -} - -/* - * Called from __ptrace_unlink() after the child has been moved back - * to its original parent. - */ -void ptrace_bts_untrace(struct task_struct *child) -{ - if (unlikely(child->bts)) { - free_bts_context(child->bts); - child->bts = NULL; - } -} -#endif /* CONFIG_X86_PTRACE_BTS */ - /* * Called by kernel/ptrace.c when detaching.. * @@ -1252,39 +911,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; #endif - /* - * These bits need more cooking - not enabled yet: - */ -#ifdef CONFIG_X86_PTRACE_BTS - case PTRACE_BTS_CONFIG: - ret = ptrace_bts_config - (child, data, (struct ptrace_bts_config __user *)addr); - break; - - case PTRACE_BTS_STATUS: - ret = ptrace_bts_status - (child, data, (struct ptrace_bts_config __user *)addr); - break; - - case PTRACE_BTS_SIZE: - ret = ptrace_bts_size(child); - break; - - case PTRACE_BTS_GET: - ret = ptrace_bts_read_record - (child, data, (struct bts_struct __user *) addr); - break; - - case PTRACE_BTS_CLEAR: - ret = ptrace_bts_clear(child); - break; - - case PTRACE_BTS_DRAIN: - ret = ptrace_bts_drain - (child, data, (struct bts_struct __user *) addr); - break; -#endif /* CONFIG_X86_PTRACE_BTS */ - default: ret = ptrace_request(child, request, addr, data); break; @@ -1544,14 +1170,6 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, case PTRACE_GET_THREAD_AREA: case PTRACE_SET_THREAD_AREA: -#ifdef CONFIG_X86_PTRACE_BTS - case PTRACE_BTS_CONFIG: - case PTRACE_BTS_STATUS: - case PTRACE_BTS_SIZE: - case PTRACE_BTS_GET: - case PTRACE_BTS_CLEAR: - case PTRACE_BTS_DRAIN: -#endif /* CONFIG_X86_PTRACE_BTS */ return arch_ptrace(child, request, addr, data); default: diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c index 3149032ff107..58de45ee08b6 100644 --- a/arch/x86/kernel/step.c +++ b/arch/x86/kernel/step.c @@ -158,22 +158,6 @@ static int enable_single_step(struct task_struct *child) } /* - * Install this value in MSR_IA32_DEBUGCTLMSR whenever child is running. - */ -static void write_debugctlmsr(struct task_struct *child, unsigned long val) -{ - if (child->thread.debugctlmsr == val) - return; - - child->thread.debugctlmsr = val; - - if (child != current) - return; - - update_debugctlmsr(val); -} - -/* * Enable single or block step. */ static void enable_step(struct task_struct *child, bool block) @@ -186,15 +170,17 @@ static void enable_step(struct task_struct *child, bool block) * that uses user-mode single stepping itself. */ if (enable_single_step(child) && block) { - set_tsk_thread_flag(child, TIF_DEBUGCTLMSR); - write_debugctlmsr(child, - child->thread.debugctlmsr | DEBUGCTLMSR_BTF); - } else { - write_debugctlmsr(child, - child->thread.debugctlmsr & ~DEBUGCTLMSR_BTF); - - if (!child->thread.debugctlmsr) - clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); + unsigned long debugctl = get_debugctlmsr(); + + debugctl |= DEBUGCTLMSR_BTF; + update_debugctlmsr(debugctl); + set_tsk_thread_flag(child, TIF_BLOCKSTEP); + } else if (test_tsk_thread_flag(child, TIF_BLOCKSTEP)) { + unsigned long debugctl = get_debugctlmsr(); + + debugctl &= ~DEBUGCTLMSR_BTF; + update_debugctlmsr(debugctl); + clear_tsk_thread_flag(child, TIF_BLOCKSTEP); } } @@ -213,11 +199,13 @@ void user_disable_single_step(struct task_struct *child) /* * Make sure block stepping (BTF) is disabled. */ - write_debugctlmsr(child, - child->thread.debugctlmsr & ~DEBUGCTLMSR_BTF); + if (test_tsk_thread_flag(child, TIF_BLOCKSTEP)) { + unsigned long debugctl = get_debugctlmsr(); - if (!child->thread.debugctlmsr) - clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR); + debugctl &= ~DEBUGCTLMSR_BTF; + update_debugctlmsr(debugctl); + clear_tsk_thread_flag(child, TIF_BLOCKSTEP); + } /* Always clear TIF_SINGLESTEP... */ clear_tsk_thread_flag(child, TIF_SINGLESTEP); diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index 86c9f91b48ae..cc2c60474fd0 100644 --- a/arch/x86/kernel/tboot.c +++ b/arch/x86/kernel/tboot.c @@ -175,6 +175,9 @@ static void add_mac_region(phys_addr_t start, unsigned long size) struct tboot_mac_region *mr; phys_addr_t end = start + size; + if (tboot->num_mac_regions >= MAX_TB_MAC_REGIONS) + panic("tboot: Too many MAC regions\n"); + if (start && size) { mr = &tboot->mac_regions[tboot->num_mac_regions++]; mr->start = round_down(start, PAGE_SIZE); @@ -184,18 +187,17 @@ static void add_mac_region(phys_addr_t start, unsigned long size) static int tboot_setup_sleep(void) { + int i; + tboot->num_mac_regions = 0; - /* S3 resume code */ - add_mac_region(acpi_wakeup_address, WAKEUP_SIZE); + for (i = 0; i < e820.nr_map; i++) { + if ((e820.map[i].type != E820_RAM) + && (e820.map[i].type != E820_RESERVED_KERN)) + continue; -#ifdef CONFIG_X86_TRAMPOLINE - /* AP trampoline code */ - add_mac_region(virt_to_phys(trampoline_base), TRAMPOLINE_SIZE); -#endif - - /* kernel code + data + bss */ - add_mac_region(virt_to_phys(_text), _end - _text); + add_mac_region(e820.map[i].addr, e820.map[i].size); + } tboot->acpi_sinfo.kernel_s3_resume_vector = acpi_wakeup_address; diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 17b03dd3a6b5..7fea555929e2 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -1,7 +1,7 @@ /* * SGI UltraViolet TLB flush routines. * - * (c) 2008 Cliff Wickman <cpw@sgi.com>, SGI. + * (c) 2008-2010 Cliff Wickman <cpw@sgi.com>, SGI. * * This code is released under the GNU General Public License version 2 or * later. @@ -20,42 +20,67 @@ #include <asm/idle.h> #include <asm/tsc.h> #include <asm/irq_vectors.h> +#include <asm/timer.h> -static struct bau_control **uv_bau_table_bases __read_mostly; -static int uv_bau_retry_limit __read_mostly; +struct msg_desc { + struct bau_payload_queue_entry *msg; + int msg_slot; + int sw_ack_slot; + struct bau_payload_queue_entry *va_queue_first; + struct bau_payload_queue_entry *va_queue_last; +}; -/* base pnode in this partition */ -static int uv_partition_base_pnode __read_mostly; +#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD 0x000000000bUL + +static int uv_bau_max_concurrent __read_mostly; + +static int nobau; +static int __init setup_nobau(char *arg) +{ + nobau = 1; + return 0; +} +early_param("nobau", setup_nobau); -static unsigned long uv_mmask __read_mostly; +/* base pnode in this partition */ +static int uv_partition_base_pnode __read_mostly; +/* position of pnode (which is nasid>>1): */ +static int uv_nshift __read_mostly; +static unsigned long uv_mmask __read_mostly; static DEFINE_PER_CPU(struct ptc_stats, ptcstats); static DEFINE_PER_CPU(struct bau_control, bau_control); +static DEFINE_PER_CPU(cpumask_var_t, uv_flush_tlb_mask); + +struct reset_args { + int sender; +}; /* - * Determine the first node on a blade. + * Determine the first node on a uvhub. 'Nodes' are used for kernel + * memory allocation. */ -static int __init blade_to_first_node(int blade) +static int __init uvhub_to_first_node(int uvhub) { int node, b; for_each_online_node(node) { b = uv_node_to_blade_id(node); - if (blade == b) + if (uvhub == b) return node; } - return -1; /* shouldn't happen */ + return -1; } /* - * Determine the apicid of the first cpu on a blade. + * Determine the apicid of the first cpu on a uvhub. */ -static int __init blade_to_first_apicid(int blade) +static int __init uvhub_to_first_apicid(int uvhub) { int cpu; for_each_present_cpu(cpu) - if (blade == uv_cpu_to_blade_id(cpu)) + if (uvhub == uv_cpu_to_blade_id(cpu)) return per_cpu(x86_cpu_to_apicid, cpu); return -1; } @@ -68,195 +93,459 @@ static int __init blade_to_first_apicid(int blade) * clear of the Timeout bit (as well) will free the resource. No reply will * be sent (the hardware will only do one reply per message). */ -static void uv_reply_to_message(int resource, - struct bau_payload_queue_entry *msg, - struct bau_msg_status *msp) +static inline void uv_reply_to_message(struct msg_desc *mdp, + struct bau_control *bcp) { unsigned long dw; + struct bau_payload_queue_entry *msg; - dw = (1 << (resource + UV_SW_ACK_NPENDING)) | (1 << resource); + msg = mdp->msg; + if (!msg->canceled) { + dw = (msg->sw_ack_vector << UV_SW_ACK_NPENDING) | + msg->sw_ack_vector; + uv_write_local_mmr( + UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, dw); + } msg->replied_to = 1; msg->sw_ack_vector = 0; - if (msp) - msp->seen_by.bits = 0; - uv_write_local_mmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, dw); } /* - * Do all the things a cpu should do for a TLB shootdown message. - * Other cpu's may come here at the same time for this message. + * Process the receipt of a RETRY message */ -static void uv_bau_process_message(struct bau_payload_queue_entry *msg, - int msg_slot, int sw_ack_slot) +static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, + struct bau_control *bcp) { - unsigned long this_cpu_mask; - struct bau_msg_status *msp; - int cpu; + int i; + int cancel_count = 0; + int slot2; + unsigned long msg_res; + unsigned long mmr = 0; + struct bau_payload_queue_entry *msg; + struct bau_payload_queue_entry *msg2; + struct ptc_stats *stat; - msp = __get_cpu_var(bau_control).msg_statuses + msg_slot; - cpu = uv_blade_processor_id(); - msg->number_of_cpus = - uv_blade_nr_online_cpus(uv_node_to_blade_id(numa_node_id())); - this_cpu_mask = 1UL << cpu; - if (msp->seen_by.bits & this_cpu_mask) - return; - atomic_or_long(&msp->seen_by.bits, this_cpu_mask); + msg = mdp->msg; + stat = &per_cpu(ptcstats, bcp->cpu); + stat->d_retries++; + /* + * cancel any message from msg+1 to the retry itself + */ + for (msg2 = msg+1, i = 0; i < DEST_Q_SIZE; msg2++, i++) { + if (msg2 > mdp->va_queue_last) + msg2 = mdp->va_queue_first; + if (msg2 == msg) + break; + + /* same conditions for cancellation as uv_do_reset */ + if ((msg2->replied_to == 0) && (msg2->canceled == 0) && + (msg2->sw_ack_vector) && ((msg2->sw_ack_vector & + msg->sw_ack_vector) == 0) && + (msg2->sending_cpu == msg->sending_cpu) && + (msg2->msg_type != MSG_NOOP)) { + slot2 = msg2 - mdp->va_queue_first; + mmr = uv_read_local_mmr + (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); + msg_res = ((msg2->sw_ack_vector << 8) | + msg2->sw_ack_vector); + /* + * This is a message retry; clear the resources held + * by the previous message only if they timed out. + * If it has not timed out we have an unexpected + * situation to report. + */ + if (mmr & (msg_res << 8)) { + /* + * is the resource timed out? + * make everyone ignore the cancelled message. + */ + msg2->canceled = 1; + stat->d_canceled++; + cancel_count++; + uv_write_local_mmr( + UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, + (msg_res << 8) | msg_res); + } else + printk(KERN_INFO "note bau retry: no effect\n"); + } + } + if (!cancel_count) + stat->d_nocanceled++; +} - if (msg->replied_to == 1) - return; +/* + * Do all the things a cpu should do for a TLB shootdown message. + * Other cpu's may come here at the same time for this message. + */ +static void uv_bau_process_message(struct msg_desc *mdp, + struct bau_control *bcp) +{ + int msg_ack_count; + short socket_ack_count = 0; + struct ptc_stats *stat; + struct bau_payload_queue_entry *msg; + struct bau_control *smaster = bcp->socket_master; + /* + * This must be a normal message, or retry of a normal message + */ + msg = mdp->msg; + stat = &per_cpu(ptcstats, bcp->cpu); if (msg->address == TLB_FLUSH_ALL) { local_flush_tlb(); - __get_cpu_var(ptcstats).alltlb++; + stat->d_alltlb++; } else { __flush_tlb_one(msg->address); - __get_cpu_var(ptcstats).onetlb++; + stat->d_onetlb++; } + stat->d_requestee++; + + /* + * One cpu on each uvhub has the additional job on a RETRY + * of releasing the resource held by the message that is + * being retried. That message is identified by sending + * cpu number. + */ + if (msg->msg_type == MSG_RETRY && bcp == bcp->uvhub_master) + uv_bau_process_retry_msg(mdp, bcp); - __get_cpu_var(ptcstats).requestee++; + /* + * This is a sw_ack message, so we have to reply to it. + * Count each responding cpu on the socket. This avoids + * pinging the count's cache line back and forth between + * the sockets. + */ + socket_ack_count = atomic_add_short_return(1, (struct atomic_short *) + &smaster->socket_acknowledge_count[mdp->msg_slot]); + if (socket_ack_count == bcp->cpus_in_socket) { + /* + * Both sockets dump their completed count total into + * the message's count. + */ + smaster->socket_acknowledge_count[mdp->msg_slot] = 0; + msg_ack_count = atomic_add_short_return(socket_ack_count, + (struct atomic_short *)&msg->acknowledge_count); + + if (msg_ack_count == bcp->cpus_in_uvhub) { + /* + * All cpus in uvhub saw it; reply + */ + uv_reply_to_message(mdp, bcp); + } + } - atomic_inc_short(&msg->acknowledge_count); - if (msg->number_of_cpus == msg->acknowledge_count) - uv_reply_to_message(sw_ack_slot, msg, msp); + return; } /* - * Examine the payload queue on one distribution node to see - * which messages have not been seen, and which cpu(s) have not seen them. + * Determine the first cpu on a uvhub. + */ +static int uvhub_to_first_cpu(int uvhub) +{ + int cpu; + for_each_present_cpu(cpu) + if (uvhub == uv_cpu_to_blade_id(cpu)) + return cpu; + return -1; +} + +/* + * Last resort when we get a large number of destination timeouts is + * to clear resources held by a given cpu. + * Do this with IPI so that all messages in the BAU message queue + * can be identified by their nonzero sw_ack_vector field. * - * Returns the number of cpu's that have not responded. + * This is entered for a single cpu on the uvhub. + * The sender want's this uvhub to free a specific message's + * sw_ack resources. */ -static int uv_examine_destination(struct bau_control *bau_tablesp, int sender) +static void +uv_do_reset(void *ptr) { - struct bau_payload_queue_entry *msg; - struct bau_msg_status *msp; - int count = 0; int i; - int j; + int slot; + int count = 0; + unsigned long mmr; + unsigned long msg_res; + struct bau_control *bcp; + struct reset_args *rap; + struct bau_payload_queue_entry *msg; + struct ptc_stats *stat; - for (msg = bau_tablesp->va_queue_first, i = 0; i < DEST_Q_SIZE; - msg++, i++) { - if ((msg->sending_cpu == sender) && (!msg->replied_to)) { - msp = bau_tablesp->msg_statuses + i; - printk(KERN_DEBUG - "blade %d: address:%#lx %d of %d, not cpu(s): ", - i, msg->address, msg->acknowledge_count, - msg->number_of_cpus); - for (j = 0; j < msg->number_of_cpus; j++) { - if (!((1L << j) & msp->seen_by.bits)) { - count++; - printk("%d ", j); - } + bcp = &per_cpu(bau_control, smp_processor_id()); + rap = (struct reset_args *)ptr; + stat = &per_cpu(ptcstats, bcp->cpu); + stat->d_resets++; + + /* + * We're looking for the given sender, and + * will free its sw_ack resource. + * If all cpu's finally responded after the timeout, its + * message 'replied_to' was set. + */ + for (msg = bcp->va_queue_first, i = 0; i < DEST_Q_SIZE; msg++, i++) { + /* uv_do_reset: same conditions for cancellation as + uv_bau_process_retry_msg() */ + if ((msg->replied_to == 0) && + (msg->canceled == 0) && + (msg->sending_cpu == rap->sender) && + (msg->sw_ack_vector) && + (msg->msg_type != MSG_NOOP)) { + /* + * make everyone else ignore this message + */ + msg->canceled = 1; + slot = msg - bcp->va_queue_first; + count++; + /* + * only reset the resource if it is still pending + */ + mmr = uv_read_local_mmr + (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); + msg_res = ((msg->sw_ack_vector << 8) | + msg->sw_ack_vector); + if (mmr & msg_res) { + stat->d_rcanceled++; + uv_write_local_mmr( + UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, + msg_res); } - printk("\n"); } } - return count; + return; } /* - * Examine the payload queue on all the distribution nodes to see - * which messages have not been seen, and which cpu(s) have not seen them. - * - * Returns the number of cpu's that have not responded. + * Use IPI to get all target uvhubs to release resources held by + * a given sending cpu number. */ -static int uv_examine_destinations(struct bau_target_nodemask *distribution) +static void uv_reset_with_ipi(struct bau_target_uvhubmask *distribution, + int sender) { - int sender; - int i; - int count = 0; + int uvhub; + int cpu; + cpumask_t mask; + struct reset_args reset_args; + + reset_args.sender = sender; - sender = smp_processor_id(); - for (i = 0; i < sizeof(struct bau_target_nodemask) * BITSPERBYTE; i++) { - if (!bau_node_isset(i, distribution)) + cpus_clear(mask); + /* find a single cpu for each uvhub in this distribution mask */ + for (uvhub = 0; + uvhub < sizeof(struct bau_target_uvhubmask) * BITSPERBYTE; + uvhub++) { + if (!bau_uvhub_isset(uvhub, distribution)) continue; - count += uv_examine_destination(uv_bau_table_bases[i], sender); + /* find a cpu for this uvhub */ + cpu = uvhub_to_first_cpu(uvhub); + cpu_set(cpu, mask); } - return count; + /* IPI all cpus; Preemption is already disabled */ + smp_call_function_many(&mask, uv_do_reset, (void *)&reset_args, 1); + return; +} + +static inline unsigned long +cycles_2_us(unsigned long long cyc) +{ + unsigned long long ns; + unsigned long us; + ns = (cyc * per_cpu(cyc2ns, smp_processor_id())) + >> CYC2NS_SCALE_FACTOR; + us = ns / 1000; + return us; } /* - * wait for completion of a broadcast message - * - * return COMPLETE, RETRY or GIVEUP + * wait for all cpus on this hub to finish their sends and go quiet + * leaves uvhub_quiesce set so that no new broadcasts are started by + * bau_flush_send_and_wait() + */ +static inline void +quiesce_local_uvhub(struct bau_control *hmaster) +{ + atomic_add_short_return(1, (struct atomic_short *) + &hmaster->uvhub_quiesce); +} + +/* + * mark this quiet-requestor as done + */ +static inline void +end_uvhub_quiesce(struct bau_control *hmaster) +{ + atomic_add_short_return(-1, (struct atomic_short *) + &hmaster->uvhub_quiesce); +} + +/* + * Wait for completion of a broadcast software ack message + * return COMPLETE, RETRY(PLUGGED or TIMEOUT) or GIVEUP */ static int uv_wait_completion(struct bau_desc *bau_desc, - unsigned long mmr_offset, int right_shift) + unsigned long mmr_offset, int right_shift, int this_cpu, + struct bau_control *bcp, struct bau_control *smaster, long try) { - int exams = 0; - long destination_timeouts = 0; - long source_timeouts = 0; + int relaxes = 0; unsigned long descriptor_status; + unsigned long mmr; + unsigned long mask; + cycles_t ttime; + cycles_t timeout_time; + struct ptc_stats *stat = &per_cpu(ptcstats, this_cpu); + struct bau_control *hmaster; + + hmaster = bcp->uvhub_master; + timeout_time = get_cycles() + bcp->timeout_interval; + /* spin on the status MMR, waiting for it to go idle */ while ((descriptor_status = (((unsigned long) uv_read_local_mmr(mmr_offset) >> right_shift) & UV_ACT_STATUS_MASK)) != DESC_STATUS_IDLE) { - if (descriptor_status == DESC_STATUS_SOURCE_TIMEOUT) { - source_timeouts++; - if (source_timeouts > SOURCE_TIMEOUT_LIMIT) - source_timeouts = 0; - __get_cpu_var(ptcstats).s_retry++; - return FLUSH_RETRY; - } /* - * spin here looking for progress at the destinations + * Our software ack messages may be blocked because there are + * no swack resources available. As long as none of them + * has timed out hardware will NACK our message and its + * state will stay IDLE. */ - if (descriptor_status == DESC_STATUS_DESTINATION_TIMEOUT) { - destination_timeouts++; - if (destination_timeouts > DESTINATION_TIMEOUT_LIMIT) { - /* - * returns number of cpus not responding - */ - if (uv_examine_destinations - (&bau_desc->distribution) == 0) { - __get_cpu_var(ptcstats).d_retry++; - return FLUSH_RETRY; - } - exams++; - if (exams >= uv_bau_retry_limit) { - printk(KERN_DEBUG - "uv_flush_tlb_others"); - printk("giving up on cpu %d\n", - smp_processor_id()); + if (descriptor_status == DESC_STATUS_SOURCE_TIMEOUT) { + stat->s_stimeout++; + return FLUSH_GIVEUP; + } else if (descriptor_status == + DESC_STATUS_DESTINATION_TIMEOUT) { + stat->s_dtimeout++; + ttime = get_cycles(); + + /* + * Our retries may be blocked by all destination + * swack resources being consumed, and a timeout + * pending. In that case hardware returns the + * ERROR that looks like a destination timeout. + */ + if (cycles_2_us(ttime - bcp->send_message) < BIOS_TO) { + bcp->conseccompletes = 0; + return FLUSH_RETRY_PLUGGED; + } + + bcp->conseccompletes = 0; + return FLUSH_RETRY_TIMEOUT; + } else { + /* + * descriptor_status is still BUSY + */ + cpu_relax(); + relaxes++; + if (relaxes >= 10000) { + relaxes = 0; + if (get_cycles() > timeout_time) { + quiesce_local_uvhub(hmaster); + + /* single-thread the register change */ + spin_lock(&hmaster->masks_lock); + mmr = uv_read_local_mmr(mmr_offset); + mask = 0UL; + mask |= (3UL < right_shift); + mask = ~mask; + mmr &= mask; + uv_write_local_mmr(mmr_offset, mmr); + spin_unlock(&hmaster->masks_lock); + end_uvhub_quiesce(hmaster); + stat->s_busy++; return FLUSH_GIVEUP; } - /* - * delays can hang the simulator - udelay(1000); - */ - destination_timeouts = 0; } } - cpu_relax(); } + bcp->conseccompletes++; return FLUSH_COMPLETE; } +static inline cycles_t +sec_2_cycles(unsigned long sec) +{ + unsigned long ns; + cycles_t cyc; + + ns = sec * 1000000000; + cyc = (ns << CYC2NS_SCALE_FACTOR)/(per_cpu(cyc2ns, smp_processor_id())); + return cyc; +} + +/* + * conditionally add 1 to *v, unless *v is >= u + * return 0 if we cannot add 1 to *v because it is >= u + * return 1 if we can add 1 to *v because it is < u + * the add is atomic + * + * This is close to atomic_add_unless(), but this allows the 'u' value + * to be lowered below the current 'v'. atomic_add_unless can only stop + * on equal. + */ +static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u) +{ + spin_lock(lock); + if (atomic_read(v) >= u) { + spin_unlock(lock); + return 0; + } + atomic_inc(v); + spin_unlock(lock); + return 1; +} + /** * uv_flush_send_and_wait * - * Send a broadcast and wait for a broadcast message to complete. + * Send a broadcast and wait for it to complete. * - * The flush_mask contains the cpus the broadcast was sent to. + * The flush_mask contains the cpus the broadcast is to be sent to, plus + * cpus that are on the local uvhub. * - * Returns NULL if all remote flushing was done. The mask is zeroed. + * Returns NULL if all flushing represented in the mask was done. The mask + * is zeroed. * Returns @flush_mask if some remote flushing remains to be done. The - * mask will have some bits still set. + * mask will have some bits still set, representing any cpus on the local + * uvhub (not current cpu) and any on remote uvhubs if the broadcast failed. */ -const struct cpumask *uv_flush_send_and_wait(int cpu, int this_pnode, - struct bau_desc *bau_desc, - struct cpumask *flush_mask) +const struct cpumask *uv_flush_send_and_wait(struct bau_desc *bau_desc, + struct cpumask *flush_mask, + struct bau_control *bcp) { - int completion_status = 0; int right_shift; - int tries = 0; - int pnode; + int uvhub; int bit; + int completion_status = 0; + int seq_number = 0; + long try = 0; + int cpu = bcp->uvhub_cpu; + int this_cpu = bcp->cpu; + int this_uvhub = bcp->uvhub; unsigned long mmr_offset; unsigned long index; cycles_t time1; cycles_t time2; + struct ptc_stats *stat = &per_cpu(ptcstats, bcp->cpu); + struct bau_control *smaster = bcp->socket_master; + struct bau_control *hmaster = bcp->uvhub_master; + + /* + * Spin here while there are hmaster->max_concurrent or more active + * descriptors. This is the per-uvhub 'throttle'. + */ + if (!atomic_inc_unless_ge(&hmaster->uvhub_lock, + &hmaster->active_descriptor_count, + hmaster->max_concurrent)) { + stat->s_throttles++; + do { + cpu_relax(); + } while (!atomic_inc_unless_ge(&hmaster->uvhub_lock, + &hmaster->active_descriptor_count, + hmaster->max_concurrent)); + } + + while (hmaster->uvhub_quiesce) + cpu_relax(); if (cpu < UV_CPUS_PER_ACT_STATUS) { mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_0; @@ -268,24 +557,108 @@ const struct cpumask *uv_flush_send_and_wait(int cpu, int this_pnode, } time1 = get_cycles(); do { - tries++; + /* + * Every message from any given cpu gets a unique message + * sequence number. But retries use that same number. + * Our message may have timed out at the destination because + * all sw-ack resources are in use and there is a timeout + * pending there. In that case, our last send never got + * placed into the queue and we need to persist until it + * does. + * + * Make any retry a type MSG_RETRY so that the destination will + * free any resource held by a previous message from this cpu. + */ + if (try == 0) { + /* use message type set by the caller the first time */ + seq_number = bcp->message_number++; + } else { + /* use RETRY type on all the rest; same sequence */ + bau_desc->header.msg_type = MSG_RETRY; + stat->s_retry_messages++; + } + bau_desc->header.sequence = seq_number; index = (1UL << UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT) | - cpu; + bcp->uvhub_cpu; + bcp->send_message = get_cycles(); + uv_write_local_mmr(UVH_LB_BAU_SB_ACTIVATION_CONTROL, index); + + try++; completion_status = uv_wait_completion(bau_desc, mmr_offset, - right_shift); - } while (completion_status == FLUSH_RETRY); + right_shift, this_cpu, bcp, smaster, try); + + if (completion_status == FLUSH_RETRY_PLUGGED) { + /* + * Our retries may be blocked by all destination swack + * resources being consumed, and a timeout pending. In + * that case hardware immediately returns the ERROR + * that looks like a destination timeout. + */ + udelay(TIMEOUT_DELAY); + bcp->plugged_tries++; + if (bcp->plugged_tries >= PLUGSB4RESET) { + bcp->plugged_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); + uv_reset_with_ipi(&bau_desc->distribution, + this_cpu); + spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; + stat->s_resets_plug++; + } + } else if (completion_status == FLUSH_RETRY_TIMEOUT) { + hmaster->max_concurrent = 1; + bcp->timeout_tries++; + udelay(TIMEOUT_DELAY); + if (bcp->timeout_tries >= TIMEOUTSB4RESET) { + bcp->timeout_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); + uv_reset_with_ipi(&bau_desc->distribution, + this_cpu); + spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; + stat->s_resets_timeout++; + } + } + if (bcp->ipi_attempts >= 3) { + bcp->ipi_attempts = 0; + completion_status = FLUSH_GIVEUP; + break; + } + cpu_relax(); + } while ((completion_status == FLUSH_RETRY_PLUGGED) || + (completion_status == FLUSH_RETRY_TIMEOUT)); time2 = get_cycles(); - __get_cpu_var(ptcstats).sflush += (time2 - time1); - if (tries > 1) - __get_cpu_var(ptcstats).retriesok++; - if (completion_status == FLUSH_GIVEUP) { + if ((completion_status == FLUSH_COMPLETE) && (bcp->conseccompletes > 5) + && (hmaster->max_concurrent < hmaster->max_concurrent_constant)) + hmaster->max_concurrent++; + + /* + * hold any cpu not timing out here; no other cpu currently held by + * the 'throttle' should enter the activation code + */ + while (hmaster->uvhub_quiesce) + cpu_relax(); + atomic_dec(&hmaster->active_descriptor_count); + + /* guard against cycles wrap */ + if (time2 > time1) + stat->s_time += (time2 - time1); + else + stat->s_requestor--; /* don't count this one */ + if (completion_status == FLUSH_COMPLETE && try > 1) + stat->s_retriesok++; + else if (completion_status == FLUSH_GIVEUP) { /* * Cause the caller to do an IPI-style TLB shootdown on - * the cpu's, all of which are still in the mask. + * the target cpu's, all of which are still in the mask. */ - __get_cpu_var(ptcstats).ptc_i++; + stat->s_giveup++; return flush_mask; } @@ -294,18 +667,17 @@ const struct cpumask *uv_flush_send_and_wait(int cpu, int this_pnode, * use the IPI method of shootdown on them. */ for_each_cpu(bit, flush_mask) { - pnode = uv_cpu_to_pnode(bit); - if (pnode == this_pnode) + uvhub = uv_cpu_to_blade_id(bit); + if (uvhub == this_uvhub) continue; cpumask_clear_cpu(bit, flush_mask); } if (!cpumask_empty(flush_mask)) return flush_mask; + return NULL; } -static DEFINE_PER_CPU(cpumask_var_t, uv_flush_tlb_mask); - /** * uv_flush_tlb_others - globally purge translation cache of a virtual * address or all TLB's @@ -322,8 +694,8 @@ static DEFINE_PER_CPU(cpumask_var_t, uv_flush_tlb_mask); * The caller has derived the cpumask from the mm_struct. This function * is called only if there are bits set in the mask. (e.g. flush_tlb_page()) * - * The cpumask is converted into a nodemask of the nodes containing - * the cpus. + * The cpumask is converted into a uvhubmask of the uvhubs containing + * those cpus. * * Note that this function should be called with preemption disabled. * @@ -335,52 +707,82 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm, unsigned long va, unsigned int cpu) { - struct cpumask *flush_mask = __get_cpu_var(uv_flush_tlb_mask); - int i; - int bit; - int pnode; - int uv_cpu; - int this_pnode; + int remotes; + int tcpu; + int uvhub; int locals = 0; struct bau_desc *bau_desc; + struct cpumask *flush_mask; + struct ptc_stats *stat; + struct bau_control *bcp; - cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu)); + if (nobau) + return cpumask; - uv_cpu = uv_blade_processor_id(); - this_pnode = uv_hub_info->pnode; - bau_desc = __get_cpu_var(bau_control).descriptor_base; - bau_desc += UV_ITEMS_PER_DESCRIPTOR * uv_cpu; + bcp = &per_cpu(bau_control, cpu); + /* + * Each sending cpu has a per-cpu mask which it fills from the caller's + * cpu mask. Only remote cpus are converted to uvhubs and copied. + */ + flush_mask = (struct cpumask *)per_cpu(uv_flush_tlb_mask, cpu); + /* + * copy cpumask to flush_mask, removing current cpu + * (current cpu should already have been flushed by the caller and + * should never be returned if we return flush_mask) + */ + cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu)); + if (cpu_isset(cpu, *cpumask)) + locals++; /* current cpu was targeted */ - bau_nodes_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); + bau_desc = bcp->descriptor_base; + bau_desc += UV_ITEMS_PER_DESCRIPTOR * bcp->uvhub_cpu; - i = 0; - for_each_cpu(bit, flush_mask) { - pnode = uv_cpu_to_pnode(bit); - BUG_ON(pnode > (UV_DISTRIBUTION_SIZE - 1)); - if (pnode == this_pnode) { + bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); + remotes = 0; + for_each_cpu(tcpu, flush_mask) { + uvhub = uv_cpu_to_blade_id(tcpu); + if (uvhub == bcp->uvhub) { locals++; continue; } - bau_node_set(pnode - uv_partition_base_pnode, - &bau_desc->distribution); - i++; + bau_uvhub_set(uvhub, &bau_desc->distribution); + remotes++; } - if (i == 0) { + if (remotes == 0) { /* - * no off_node flushing; return status for local node + * No off_hub flushing; return status for local hub. + * Return the caller's mask if all were local (the current + * cpu may be in that mask). */ if (locals) - return flush_mask; + return cpumask; else return NULL; } - __get_cpu_var(ptcstats).requestor++; - __get_cpu_var(ptcstats).ntargeted += i; + stat = &per_cpu(ptcstats, cpu); + stat->s_requestor++; + stat->s_ntargcpu += remotes; + remotes = bau_uvhub_weight(&bau_desc->distribution); + stat->s_ntarguvhub += remotes; + if (remotes >= 16) + stat->s_ntarguvhub16++; + else if (remotes >= 8) + stat->s_ntarguvhub8++; + else if (remotes >= 4) + stat->s_ntarguvhub4++; + else if (remotes >= 2) + stat->s_ntarguvhub2++; + else + stat->s_ntarguvhub1++; bau_desc->payload.address = va; bau_desc->payload.sending_cpu = cpu; - return uv_flush_send_and_wait(uv_cpu, this_pnode, bau_desc, flush_mask); + /* + * uv_flush_send_and_wait returns null if all cpu's were messaged, or + * the adjusted flush_mask if any cpu's were not messaged. + */ + return uv_flush_send_and_wait(bau_desc, flush_mask, bcp); } /* @@ -389,87 +791,70 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, * * We received a broadcast assist message. * - * Interrupts may have been disabled; this interrupt could represent + * Interrupts are disabled; this interrupt could represent * the receipt of several messages. * - * All cores/threads on this node get this interrupt. - * The last one to see it does the s/w ack. + * All cores/threads on this hub get this interrupt. + * The last one to see it does the software ack. * (the resource will not be freed until noninterruptable cpus see this - * interrupt; hardware will timeout the s/w ack and reply ERROR) + * interrupt; hardware may timeout the s/w ack and reply ERROR) */ void uv_bau_message_interrupt(struct pt_regs *regs) { - struct bau_payload_queue_entry *va_queue_first; - struct bau_payload_queue_entry *va_queue_last; - struct bau_payload_queue_entry *msg; - struct pt_regs *old_regs = set_irq_regs(regs); - cycles_t time1; - cycles_t time2; - int msg_slot; - int sw_ack_slot; - int fw; int count = 0; - unsigned long local_pnode; - - ack_APIC_irq(); - exit_idle(); - irq_enter(); - - time1 = get_cycles(); - - local_pnode = uv_blade_to_pnode(uv_numa_blade_id()); - - va_queue_first = __get_cpu_var(bau_control).va_queue_first; - va_queue_last = __get_cpu_var(bau_control).va_queue_last; - - msg = __get_cpu_var(bau_control).bau_msg_head; + cycles_t time_start; + struct bau_payload_queue_entry *msg; + struct bau_control *bcp; + struct ptc_stats *stat; + struct msg_desc msgdesc; + + time_start = get_cycles(); + bcp = &per_cpu(bau_control, smp_processor_id()); + stat = &per_cpu(ptcstats, smp_processor_id()); + msgdesc.va_queue_first = bcp->va_queue_first; + msgdesc.va_queue_last = bcp->va_queue_last; + msg = bcp->bau_msg_head; while (msg->sw_ack_vector) { count++; - fw = msg->sw_ack_vector; - msg_slot = msg - va_queue_first; - sw_ack_slot = ffs(fw) - 1; - - uv_bau_process_message(msg, msg_slot, sw_ack_slot); - + msgdesc.msg_slot = msg - msgdesc.va_queue_first; + msgdesc.sw_ack_slot = ffs(msg->sw_ack_vector) - 1; + msgdesc.msg = msg; + uv_bau_process_message(&msgdesc, bcp); msg++; - if (msg > va_queue_last) - msg = va_queue_first; - __get_cpu_var(bau_control).bau_msg_head = msg; + if (msg > msgdesc.va_queue_last) + msg = msgdesc.va_queue_first; + bcp->bau_msg_head = msg; } + stat->d_time += (get_cycles() - time_start); if (!count) - __get_cpu_var(ptcstats).nomsg++; + stat->d_nomsg++; else if (count > 1) - __get_cpu_var(ptcstats).multmsg++; - - time2 = get_cycles(); - __get_cpu_var(ptcstats).dflush += (time2 - time1); - - irq_exit(); - set_irq_regs(old_regs); + stat->d_multmsg++; + ack_APIC_irq(); } /* * uv_enable_timeouts * - * Each target blade (i.e. blades that have cpu's) needs to have + * Each target uvhub (i.e. a uvhub that has no cpu's) needs to have * shootdown message timeouts enabled. The timeout does not cause * an interrupt, but causes an error message to be returned to * the sender. */ static void uv_enable_timeouts(void) { - int blade; - int nblades; + int uvhub; + int nuvhubs; int pnode; unsigned long mmr_image; - nblades = uv_num_possible_blades(); + nuvhubs = uv_num_possible_blades(); - for (blade = 0; blade < nblades; blade++) { - if (!uv_blade_nr_possible_cpus(blade)) + for (uvhub = 0; uvhub < nuvhubs; uvhub++) { + if (!uv_blade_nr_possible_cpus(uvhub)) continue; - pnode = uv_blade_to_pnode(blade); + pnode = uv_blade_to_pnode(uvhub); mmr_image = uv_read_global_mmr64(pnode, UVH_LB_BAU_MISC_CONTROL); /* @@ -479,16 +864,16 @@ static void uv_enable_timeouts(void) * To program the period, the SOFT_ACK_MODE must be off. */ mmr_image &= ~((unsigned long)1 << - UV_ENABLE_INTD_SOFT_ACK_MODE_SHIFT); + UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT); uv_write_global_mmr64 (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); /* * Set the 4-bit period. */ mmr_image &= ~((unsigned long)0xf << - UV_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHIFT); + UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT); mmr_image |= (UV_INTD_SOFT_ACK_TIMEOUT_PERIOD << - UV_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHIFT); + UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT); uv_write_global_mmr64 (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); /* @@ -497,7 +882,7 @@ static void uv_enable_timeouts(void) * indicated in bits 2:0 (7 causes all of them to timeout). */ mmr_image |= ((unsigned long)1 << - UV_ENABLE_INTD_SOFT_ACK_MODE_SHIFT); + UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT); uv_write_global_mmr64 (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); } @@ -522,9 +907,20 @@ static void uv_ptc_seq_stop(struct seq_file *file, void *data) { } +static inline unsigned long long +millisec_2_cycles(unsigned long millisec) +{ + unsigned long ns; + unsigned long long cyc; + + ns = millisec * 1000; + cyc = (ns << CYC2NS_SCALE_FACTOR)/(per_cpu(cyc2ns, smp_processor_id())); + return cyc; +} + /* - * Display the statistics thru /proc - * data points to the cpu number + * Display the statistics thru /proc. + * 'data' points to the cpu number */ static int uv_ptc_seq_show(struct seq_file *file, void *data) { @@ -535,78 +931,155 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) if (!cpu) { seq_printf(file, - "# cpu requestor requestee one all sretry dretry ptc_i "); + "# cpu sent stime numuvhubs numuvhubs16 numuvhubs8 "); seq_printf(file, - "sw_ack sflush dflush sok dnomsg dmult starget\n"); + "numuvhubs4 numuvhubs2 numuvhubs1 numcpus dto "); + seq_printf(file, + "retries rok resetp resett giveup sto bz throt "); + seq_printf(file, + "sw_ack recv rtime all "); + seq_printf(file, + "one mult none retry canc nocan reset rcan\n"); } if (cpu < num_possible_cpus() && cpu_online(cpu)) { stat = &per_cpu(ptcstats, cpu); - seq_printf(file, "cpu %d %ld %ld %ld %ld %ld %ld %ld ", - cpu, stat->requestor, - stat->requestee, stat->onetlb, stat->alltlb, - stat->s_retry, stat->d_retry, stat->ptc_i); - seq_printf(file, "%lx %ld %ld %ld %ld %ld %ld\n", + /* source side statistics */ + seq_printf(file, + "cpu %d %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ", + cpu, stat->s_requestor, cycles_2_us(stat->s_time), + stat->s_ntarguvhub, stat->s_ntarguvhub16, + stat->s_ntarguvhub8, stat->s_ntarguvhub4, + stat->s_ntarguvhub2, stat->s_ntarguvhub1, + stat->s_ntargcpu, stat->s_dtimeout); + seq_printf(file, "%ld %ld %ld %ld %ld %ld %ld %ld ", + stat->s_retry_messages, stat->s_retriesok, + stat->s_resets_plug, stat->s_resets_timeout, + stat->s_giveup, stat->s_stimeout, + stat->s_busy, stat->s_throttles); + /* destination side statistics */ + seq_printf(file, + "%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld\n", uv_read_global_mmr64(uv_cpu_to_pnode(cpu), UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE), - stat->sflush, stat->dflush, - stat->retriesok, stat->nomsg, - stat->multmsg, stat->ntargeted); + stat->d_requestee, cycles_2_us(stat->d_time), + stat->d_alltlb, stat->d_onetlb, stat->d_multmsg, + stat->d_nomsg, stat->d_retries, stat->d_canceled, + stat->d_nocanceled, stat->d_resets, + stat->d_rcanceled); } return 0; } /* + * -1: resetf the statistics * 0: display meaning of the statistics - * >0: retry limit + * >0: maximum concurrent active descriptors per uvhub (throttle) */ static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, size_t count, loff_t *data) { - long newmode; + int cpu; + long input_arg; char optstr[64]; + struct ptc_stats *stat; + struct bau_control *bcp; if (count == 0 || count > sizeof(optstr)) return -EINVAL; if (copy_from_user(optstr, user, count)) return -EFAULT; optstr[count - 1] = '\0'; - if (strict_strtoul(optstr, 10, &newmode) < 0) { + if (strict_strtol(optstr, 10, &input_arg) < 0) { printk(KERN_DEBUG "%s is invalid\n", optstr); return -EINVAL; } - if (newmode == 0) { + if (input_arg == 0) { printk(KERN_DEBUG "# cpu: cpu number\n"); + printk(KERN_DEBUG "Sender statistics:\n"); + printk(KERN_DEBUG + "sent: number of shootdown messages sent\n"); + printk(KERN_DEBUG + "stime: time spent sending messages\n"); + printk(KERN_DEBUG + "numuvhubs: number of hubs targeted with shootdown\n"); + printk(KERN_DEBUG + "numuvhubs16: number times 16 or more hubs targeted\n"); + printk(KERN_DEBUG + "numuvhubs8: number times 8 or more hubs targeted\n"); + printk(KERN_DEBUG + "numuvhubs4: number times 4 or more hubs targeted\n"); + printk(KERN_DEBUG + "numuvhubs2: number times 2 or more hubs targeted\n"); + printk(KERN_DEBUG + "numuvhubs1: number times 1 hub targeted\n"); + printk(KERN_DEBUG + "numcpus: number of cpus targeted with shootdown\n"); + printk(KERN_DEBUG + "dto: number of destination timeouts\n"); + printk(KERN_DEBUG + "retries: destination timeout retries sent\n"); + printk(KERN_DEBUG + "rok: : destination timeouts successfully retried\n"); + printk(KERN_DEBUG + "resetp: ipi-style resource resets for plugs\n"); + printk(KERN_DEBUG + "resett: ipi-style resource resets for timeouts\n"); + printk(KERN_DEBUG + "giveup: fall-backs to ipi-style shootdowns\n"); + printk(KERN_DEBUG + "sto: number of source timeouts\n"); + printk(KERN_DEBUG + "bz: number of stay-busy's\n"); + printk(KERN_DEBUG + "throt: number times spun in throttle\n"); + printk(KERN_DEBUG "Destination side statistics:\n"); printk(KERN_DEBUG - "requestor: times this cpu was the flush requestor\n"); + "sw_ack: image of UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE\n"); printk(KERN_DEBUG - "requestee: times this cpu was requested to flush its TLBs\n"); + "recv: shootdown messages received\n"); printk(KERN_DEBUG - "one: times requested to flush a single address\n"); + "rtime: time spent processing messages\n"); printk(KERN_DEBUG - "all: times requested to flush all TLB's\n"); + "all: shootdown all-tlb messages\n"); printk(KERN_DEBUG - "sretry: number of retries of source-side timeouts\n"); + "one: shootdown one-tlb messages\n"); printk(KERN_DEBUG - "dretry: number of retries of destination-side timeouts\n"); + "mult: interrupts that found multiple messages\n"); printk(KERN_DEBUG - "ptc_i: times UV fell through to IPI-style flushes\n"); + "none: interrupts that found no messages\n"); printk(KERN_DEBUG - "sw_ack: image of UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE\n"); + "retry: number of retry messages processed\n"); printk(KERN_DEBUG - "sflush_us: cycles spent in uv_flush_tlb_others()\n"); + "canc: number messages canceled by retries\n"); printk(KERN_DEBUG - "dflush_us: cycles spent in handling flush requests\n"); - printk(KERN_DEBUG "sok: successes on retry\n"); - printk(KERN_DEBUG "dnomsg: interrupts with no message\n"); + "nocan: number retries that found nothing to cancel\n"); printk(KERN_DEBUG - "dmult: interrupts with multiple messages\n"); - printk(KERN_DEBUG "starget: nodes targeted\n"); + "reset: number of ipi-style reset requests processed\n"); + printk(KERN_DEBUG + "rcan: number messages canceled by reset requests\n"); + } else if (input_arg == -1) { + for_each_present_cpu(cpu) { + stat = &per_cpu(ptcstats, cpu); + memset(stat, 0, sizeof(struct ptc_stats)); + } } else { - uv_bau_retry_limit = newmode; - printk(KERN_DEBUG "timeout retry limit:%d\n", - uv_bau_retry_limit); + uv_bau_max_concurrent = input_arg; + bcp = &per_cpu(bau_control, smp_processor_id()); + if (uv_bau_max_concurrent < 1 || + uv_bau_max_concurrent > bcp->cpus_in_uvhub) { + printk(KERN_DEBUG + "Error: BAU max concurrent %d; %d is invalid\n", + bcp->max_concurrent, uv_bau_max_concurrent); + return -EINVAL; + } + printk(KERN_DEBUG "Set BAU max concurrent:%d\n", + uv_bau_max_concurrent); + for_each_present_cpu(cpu) { + bcp = &per_cpu(bau_control, cpu); + bcp->max_concurrent = uv_bau_max_concurrent; + } } return count; @@ -650,79 +1123,30 @@ static int __init uv_ptc_init(void) } /* - * begin the initialization of the per-blade control structures - */ -static struct bau_control * __init uv_table_bases_init(int blade, int node) -{ - int i; - struct bau_msg_status *msp; - struct bau_control *bau_tabp; - - bau_tabp = - kmalloc_node(sizeof(struct bau_control), GFP_KERNEL, node); - BUG_ON(!bau_tabp); - - bau_tabp->msg_statuses = - kmalloc_node(sizeof(struct bau_msg_status) * - DEST_Q_SIZE, GFP_KERNEL, node); - BUG_ON(!bau_tabp->msg_statuses); - - for (i = 0, msp = bau_tabp->msg_statuses; i < DEST_Q_SIZE; i++, msp++) - bau_cpubits_clear(&msp->seen_by, (int) - uv_blade_nr_possible_cpus(blade)); - - uv_bau_table_bases[blade] = bau_tabp; - - return bau_tabp; -} - -/* - * finish the initialization of the per-blade control structures - */ -static void __init -uv_table_bases_finish(int blade, - struct bau_control *bau_tablesp, - struct bau_desc *adp) -{ - struct bau_control *bcp; - int cpu; - - for_each_present_cpu(cpu) { - if (blade != uv_cpu_to_blade_id(cpu)) - continue; - - bcp = (struct bau_control *)&per_cpu(bau_control, cpu); - bcp->bau_msg_head = bau_tablesp->va_queue_first; - bcp->va_queue_first = bau_tablesp->va_queue_first; - bcp->va_queue_last = bau_tablesp->va_queue_last; - bcp->msg_statuses = bau_tablesp->msg_statuses; - bcp->descriptor_base = adp; - } -} - -/* * initialize the sending side's sending buffers */ -static struct bau_desc * __init +static void uv_activation_descriptor_init(int node, int pnode) { int i; + int cpu; unsigned long pa; unsigned long m; unsigned long n; - struct bau_desc *adp; - struct bau_desc *ad2; + struct bau_desc *bau_desc; + struct bau_desc *bd2; + struct bau_control *bcp; /* * each bau_desc is 64 bytes; there are 8 (UV_ITEMS_PER_DESCRIPTOR) - * per cpu; and up to 32 (UV_ADP_SIZE) cpu's per blade + * per cpu; and up to 32 (UV_ADP_SIZE) cpu's per uvhub */ - adp = (struct bau_desc *)kmalloc_node(sizeof(struct bau_desc)* + bau_desc = (struct bau_desc *)kmalloc_node(sizeof(struct bau_desc)* UV_ADP_SIZE*UV_ITEMS_PER_DESCRIPTOR, GFP_KERNEL, node); - BUG_ON(!adp); + BUG_ON(!bau_desc); - pa = uv_gpa(adp); /* need the real nasid*/ - n = uv_gpa_to_pnode(pa); + pa = uv_gpa(bau_desc); /* need the real nasid*/ + n = pa >> uv_nshift; m = pa & uv_mmask; uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, @@ -731,96 +1155,188 @@ uv_activation_descriptor_init(int node, int pnode) /* * initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each * cpu even though we only use the first one; one descriptor can - * describe a broadcast to 256 nodes. + * describe a broadcast to 256 uv hubs. */ - for (i = 0, ad2 = adp; i < (UV_ADP_SIZE*UV_ITEMS_PER_DESCRIPTOR); - i++, ad2++) { - memset(ad2, 0, sizeof(struct bau_desc)); - ad2->header.sw_ack_flag = 1; + for (i = 0, bd2 = bau_desc; i < (UV_ADP_SIZE*UV_ITEMS_PER_DESCRIPTOR); + i++, bd2++) { + memset(bd2, 0, sizeof(struct bau_desc)); + bd2->header.sw_ack_flag = 1; /* - * base_dest_nodeid is the first node in the partition, so - * the bit map will indicate partition-relative node numbers. - * note that base_dest_nodeid is actually a nasid. + * base_dest_nodeid is the nasid (pnode<<1) of the first uvhub + * in the partition. The bit map will indicate uvhub numbers, + * which are 0-N in a partition. Pnodes are unique system-wide. */ - ad2->header.base_dest_nodeid = uv_partition_base_pnode << 1; - ad2->header.dest_subnodeid = 0x10; /* the LB */ - ad2->header.command = UV_NET_ENDPOINT_INTD; - ad2->header.int_both = 1; + bd2->header.base_dest_nodeid = uv_partition_base_pnode << 1; + bd2->header.dest_subnodeid = 0x10; /* the LB */ + bd2->header.command = UV_NET_ENDPOINT_INTD; + bd2->header.int_both = 1; /* * all others need to be set to zero: * fairness chaining multilevel count replied_to */ } - return adp; + for_each_present_cpu(cpu) { + if (pnode != uv_blade_to_pnode(uv_cpu_to_blade_id(cpu))) + continue; + bcp = &per_cpu(bau_control, cpu); + bcp->descriptor_base = bau_desc; + } } /* * initialize the destination side's receiving buffers + * entered for each uvhub in the partition + * - node is first node (kernel memory notion) on the uvhub + * - pnode is the uvhub's physical identifier */ -static struct bau_payload_queue_entry * __init -uv_payload_queue_init(int node, int pnode, struct bau_control *bau_tablesp) +static void +uv_payload_queue_init(int node, int pnode) { - struct bau_payload_queue_entry *pqp; - unsigned long pa; int pn; + int cpu; char *cp; + unsigned long pa; + struct bau_payload_queue_entry *pqp; + struct bau_payload_queue_entry *pqp_malloc; + struct bau_control *bcp; pqp = (struct bau_payload_queue_entry *) kmalloc_node( (DEST_Q_SIZE + 1) * sizeof(struct bau_payload_queue_entry), GFP_KERNEL, node); BUG_ON(!pqp); + pqp_malloc = pqp; cp = (char *)pqp + 31; pqp = (struct bau_payload_queue_entry *)(((unsigned long)cp >> 5) << 5); - bau_tablesp->va_queue_first = pqp; + + for_each_present_cpu(cpu) { + if (pnode != uv_cpu_to_pnode(cpu)) + continue; + /* for every cpu on this pnode: */ + bcp = &per_cpu(bau_control, cpu); + bcp->va_queue_first = pqp; + bcp->bau_msg_head = pqp; + bcp->va_queue_last = pqp + (DEST_Q_SIZE - 1); + } /* * need the pnode of where the memory was really allocated */ pa = uv_gpa(pqp); - pn = uv_gpa_to_pnode(pa); + pn = pa >> uv_nshift; uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, ((unsigned long)pn << UV_PAYLOADQ_PNODE_SHIFT) | uv_physnodeaddr(pqp)); uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL, uv_physnodeaddr(pqp)); - bau_tablesp->va_queue_last = pqp + (DEST_Q_SIZE - 1); uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST, (unsigned long) - uv_physnodeaddr(bau_tablesp->va_queue_last)); + uv_physnodeaddr(pqp + (DEST_Q_SIZE - 1))); + /* in effect, all msg_type's are set to MSG_NOOP */ memset(pqp, 0, sizeof(struct bau_payload_queue_entry) * DEST_Q_SIZE); - - return pqp; } /* - * Initialization of each UV blade's structures + * Initialization of each UV hub's structures */ -static int __init uv_init_blade(int blade) +static void __init uv_init_uvhub(int uvhub, int vector) { int node; int pnode; - unsigned long pa; unsigned long apicid; - struct bau_desc *adp; - struct bau_payload_queue_entry *pqp; - struct bau_control *bau_tablesp; - - node = blade_to_first_node(blade); - bau_tablesp = uv_table_bases_init(blade, node); - pnode = uv_blade_to_pnode(blade); - adp = uv_activation_descriptor_init(node, pnode); - pqp = uv_payload_queue_init(node, pnode, bau_tablesp); - uv_table_bases_finish(blade, bau_tablesp, adp); + + node = uvhub_to_first_node(uvhub); + pnode = uv_blade_to_pnode(uvhub); + uv_activation_descriptor_init(node, pnode); + uv_payload_queue_init(node, pnode); /* * the below initialization can't be in firmware because the * messaging IRQ will be determined by the OS */ - apicid = blade_to_first_apicid(blade); - pa = uv_read_global_mmr64(pnode, UVH_BAU_DATA_CONFIG); + apicid = uvhub_to_first_apicid(uvhub); uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, - ((apicid << 32) | UV_BAU_MESSAGE)); - return 0; + ((apicid << 32) | vector)); +} + +/* + * initialize the bau_control structure for each cpu + */ +static void uv_init_per_cpu(int nuvhubs) +{ + int i, j, k; + int cpu; + int pnode; + int uvhub; + short socket = 0; + struct bau_control *bcp; + struct uvhub_desc *bdp; + struct socket_desc *sdp; + struct bau_control *hmaster = NULL; + struct bau_control *smaster = NULL; + struct socket_desc { + short num_cpus; + short cpu_number[16]; + }; + struct uvhub_desc { + short num_sockets; + short num_cpus; + short uvhub; + short pnode; + struct socket_desc socket[2]; + }; + struct uvhub_desc *uvhub_descs; + + uvhub_descs = (struct uvhub_desc *) + kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); + memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc)); + for_each_present_cpu(cpu) { + bcp = &per_cpu(bau_control, cpu); + memset(bcp, 0, sizeof(struct bau_control)); + spin_lock_init(&bcp->masks_lock); + bcp->max_concurrent = uv_bau_max_concurrent; + pnode = uv_cpu_hub_info(cpu)->pnode; + uvhub = uv_cpu_hub_info(cpu)->numa_blade_id; + bdp = &uvhub_descs[uvhub]; + bdp->num_cpus++; + bdp->uvhub = uvhub; + bdp->pnode = pnode; + /* time interval to catch a hardware stay-busy bug */ + bcp->timeout_interval = millisec_2_cycles(3); + /* kludge: assume uv_hub.h is constant */ + socket = (cpu_physical_id(cpu)>>5)&1; + if (socket >= bdp->num_sockets) + bdp->num_sockets = socket+1; + sdp = &bdp->socket[socket]; + sdp->cpu_number[sdp->num_cpus] = cpu; + sdp->num_cpus++; + } + socket = 0; + for_each_possible_blade(uvhub) { + bdp = &uvhub_descs[uvhub]; + for (i = 0; i < bdp->num_sockets; i++) { + sdp = &bdp->socket[i]; + for (j = 0; j < sdp->num_cpus; j++) { + cpu = sdp->cpu_number[j]; + bcp = &per_cpu(bau_control, cpu); + bcp->cpu = cpu; + if (j == 0) { + smaster = bcp; + if (i == 0) + hmaster = bcp; + } + bcp->cpus_in_uvhub = bdp->num_cpus; + bcp->cpus_in_socket = sdp->num_cpus; + bcp->socket_master = smaster; + bcp->uvhub_master = hmaster; + for (k = 0; k < DEST_Q_SIZE; k++) + bcp->socket_acknowledge_count[k] = 0; + bcp->uvhub_cpu = + uv_cpu_hub_info(cpu)->blade_processor_id; + } + socket++; + } + } + kfree(uvhub_descs); } /* @@ -828,38 +1344,54 @@ static int __init uv_init_blade(int blade) */ static int __init uv_bau_init(void) { - int blade; - int nblades; + int uvhub; + int pnode; + int nuvhubs; int cur_cpu; + int vector; + unsigned long mmr; if (!is_uv_system()) return 0; + if (nobau) + return 0; + for_each_possible_cpu(cur_cpu) zalloc_cpumask_var_node(&per_cpu(uv_flush_tlb_mask, cur_cpu), GFP_KERNEL, cpu_to_node(cur_cpu)); - uv_bau_retry_limit = 1; + uv_bau_max_concurrent = MAX_BAU_CONCURRENT; + uv_nshift = uv_hub_info->m_val; uv_mmask = (1UL << uv_hub_info->m_val) - 1; - nblades = uv_num_possible_blades(); + nuvhubs = uv_num_possible_blades(); - uv_bau_table_bases = (struct bau_control **) - kmalloc(nblades * sizeof(struct bau_control *), GFP_KERNEL); - BUG_ON(!uv_bau_table_bases); + uv_init_per_cpu(nuvhubs); uv_partition_base_pnode = 0x7fffffff; - for (blade = 0; blade < nblades; blade++) - if (uv_blade_nr_possible_cpus(blade) && - (uv_blade_to_pnode(blade) < uv_partition_base_pnode)) - uv_partition_base_pnode = uv_blade_to_pnode(blade); - for (blade = 0; blade < nblades; blade++) - if (uv_blade_nr_possible_cpus(blade)) - uv_init_blade(blade); - - alloc_intr_gate(UV_BAU_MESSAGE, uv_bau_message_intr1); + for (uvhub = 0; uvhub < nuvhubs; uvhub++) + if (uv_blade_nr_possible_cpus(uvhub) && + (uv_blade_to_pnode(uvhub) < uv_partition_base_pnode)) + uv_partition_base_pnode = uv_blade_to_pnode(uvhub); + + vector = UV_BAU_MESSAGE; + for_each_possible_blade(uvhub) + if (uv_blade_nr_possible_cpus(uvhub)) + uv_init_uvhub(uvhub, vector); + uv_enable_timeouts(); + alloc_intr_gate(vector, uv_bau_message_intr1); + + for_each_possible_blade(uvhub) { + pnode = uv_blade_to_pnode(uvhub); + /* INIT the bau */ + uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_ACTIVATION_CONTROL, + ((unsigned long)1 << 63)); + mmr = 1; /* should be 1 to broadcast to both sockets */ + uv_write_global_mmr64(pnode, UVH_BAU_DATA_BROADCAST, mmr); + } return 0; } -__initcall(uv_bau_init); -__initcall(uv_ptc_init); +core_initcall(uv_bau_init); +core_initcall(uv_ptc_init); diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 1168e4454188..88504990561a 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -400,7 +400,13 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; + #ifdef CONFIG_X86_LOCAL_APIC + if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) + == NOTIFY_STOP) + return; + +#ifndef CONFIG_NMI_WATCHDOG /* * Ok, so this is none of the documented NMI sources, * so it must be the NMI watchdog. @@ -408,6 +414,7 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) if (nmi_watchdog_tick(regs, reason)) return; if (!do_nmi_callback(regs, cpu)) +#endif /* !CONFIG_NMI_WATCHDOG */ unknown_nmi_error(reason, regs); #else unknown_nmi_error(reason, regs); @@ -543,11 +550,11 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) /* DR6 may or may not be cleared by the CPU */ set_debugreg(0, 6); + /* * The processor cleared BTF, so don't mark that we need it set. */ - clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR); - tsk->thread.debugctlmsr = 0; + clear_tsk_thread_flag(tsk, TIF_BLOCKSTEP); /* Store the virtualized DR6 value */ tsk->thread.debugreg6 = dr6; diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index 1d40336b030a..1132129db792 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -44,7 +44,7 @@ static void uv_ack_apic(unsigned int irq) ack_APIC_irq(); } -struct irq_chip uv_irq_chip = { +static struct irq_chip uv_irq_chip = { .name = "UV-CORE", .startup = uv_noop_ret, .shutdown = uv_noop, @@ -141,7 +141,7 @@ int uv_irq_2_mmr_info(int irq, unsigned long *offset, int *pnode) */ static int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, - unsigned long mmr_offset, int restrict) + unsigned long mmr_offset, int limit) { const struct cpumask *eligible_cpu = cpumask_of(cpu); struct irq_desc *desc = irq_to_desc(irq); @@ -160,7 +160,7 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, if (err != 0) return err; - if (restrict == UV_AFFINITY_CPU) + if (limit == UV_AFFINITY_CPU) desc->status |= IRQ_NO_BALANCING; else desc->status |= IRQ_MOVE_PCNTXT; @@ -214,7 +214,7 @@ static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) unsigned long mmr_value; struct uv_IO_APIC_route_entry *entry; unsigned long mmr_offset; - unsigned mmr_pnode; + int mmr_pnode; if (set_desc_affinity(desc, mask, &dest)) return -1; @@ -248,7 +248,7 @@ static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) * interrupt is raised. */ int uv_setup_irq(char *irq_name, int cpu, int mmr_blade, - unsigned long mmr_offset, int restrict) + unsigned long mmr_offset, int limit) { int irq, ret; @@ -258,7 +258,7 @@ int uv_setup_irq(char *irq_name, int cpu, int mmr_blade, return -EBUSY; ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset, - restrict); + limit); if (ret == irq) uv_set_irq_2_mmr_info(irq, mmr_offset, mmr_blade); else diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 7e59dc1d3fc2..2bdf628066bd 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -115,7 +115,7 @@ static void async_hcall(unsigned long call, unsigned long arg1, local_irq_save(flags); if (lguest_data.hcall_status[next_call] != 0xFF) { /* Table full, so do normal hcall which will flush table. */ - kvm_hypercall4(call, arg1, arg2, arg3, arg4); + hcall(call, arg1, arg2, arg3, arg4); } else { lguest_data.hcalls[next_call].arg0 = call; lguest_data.hcalls[next_call].arg1 = arg1; @@ -145,46 +145,45 @@ static void async_hcall(unsigned long call, unsigned long arg1, * So, when we're in lazy mode, we call async_hcall() to store the call for * future processing: */ -static void lazy_hcall1(unsigned long call, - unsigned long arg1) +static void lazy_hcall1(unsigned long call, unsigned long arg1) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) - kvm_hypercall1(call, arg1); + hcall(call, arg1, 0, 0, 0); else async_hcall(call, arg1, 0, 0, 0); } /* You can imagine what lazy_hcall2, 3 and 4 look like. :*/ static void lazy_hcall2(unsigned long call, - unsigned long arg1, - unsigned long arg2) + unsigned long arg1, + unsigned long arg2) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) - kvm_hypercall2(call, arg1, arg2); + hcall(call, arg1, arg2, 0, 0); else async_hcall(call, arg1, arg2, 0, 0); } static void lazy_hcall3(unsigned long call, - unsigned long arg1, - unsigned long arg2, - unsigned long arg3) + unsigned long arg1, + unsigned long arg2, + unsigned long arg3) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) - kvm_hypercall3(call, arg1, arg2, arg3); + hcall(call, arg1, arg2, arg3, 0); else async_hcall(call, arg1, arg2, arg3, 0); } #ifdef CONFIG_X86_PAE static void lazy_hcall4(unsigned long call, - unsigned long arg1, - unsigned long arg2, - unsigned long arg3, - unsigned long arg4) + unsigned long arg1, + unsigned long arg2, + unsigned long arg3, + unsigned long arg4) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) - kvm_hypercall4(call, arg1, arg2, arg3, arg4); + hcall(call, arg1, arg2, arg3, arg4); else async_hcall(call, arg1, arg2, arg3, arg4); } @@ -196,13 +195,13 @@ static void lazy_hcall4(unsigned long call, :*/ static void lguest_leave_lazy_mmu_mode(void) { - kvm_hypercall0(LHCALL_FLUSH_ASYNC); + hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0, 0); paravirt_leave_lazy_mmu(); } static void lguest_end_context_switch(struct task_struct *next) { - kvm_hypercall0(LHCALL_FLUSH_ASYNC); + hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0, 0); paravirt_end_context_switch(next); } @@ -286,7 +285,7 @@ static void lguest_write_idt_entry(gate_desc *dt, /* Keep the local copy up to date. */ native_write_idt_entry(dt, entrynum, g); /* Tell Host about this new entry. */ - kvm_hypercall3(LHCALL_LOAD_IDT_ENTRY, entrynum, desc[0], desc[1]); + hcall(LHCALL_LOAD_IDT_ENTRY, entrynum, desc[0], desc[1], 0); } /* @@ -300,7 +299,7 @@ static void lguest_load_idt(const struct desc_ptr *desc) struct desc_struct *idt = (void *)desc->address; for (i = 0; i < (desc->size+1)/8; i++) - kvm_hypercall3(LHCALL_LOAD_IDT_ENTRY, i, idt[i].a, idt[i].b); + hcall(LHCALL_LOAD_IDT_ENTRY, i, idt[i].a, idt[i].b, 0); } /* @@ -321,7 +320,7 @@ static void lguest_load_gdt(const struct desc_ptr *desc) struct desc_struct *gdt = (void *)desc->address; for (i = 0; i < (desc->size+1)/8; i++) - kvm_hypercall3(LHCALL_LOAD_GDT_ENTRY, i, gdt[i].a, gdt[i].b); + hcall(LHCALL_LOAD_GDT_ENTRY, i, gdt[i].a, gdt[i].b, 0); } /* @@ -334,8 +333,8 @@ static void lguest_write_gdt_entry(struct desc_struct *dt, int entrynum, { native_write_gdt_entry(dt, entrynum, desc, type); /* Tell Host about this new entry. */ - kvm_hypercall3(LHCALL_LOAD_GDT_ENTRY, entrynum, - dt[entrynum].a, dt[entrynum].b); + hcall(LHCALL_LOAD_GDT_ENTRY, entrynum, + dt[entrynum].a, dt[entrynum].b, 0); } /* @@ -931,7 +930,7 @@ static int lguest_clockevent_set_next_event(unsigned long delta, } /* Please wake us this far in the future. */ - kvm_hypercall1(LHCALL_SET_CLOCKEVENT, delta); + hcall(LHCALL_SET_CLOCKEVENT, delta, 0, 0, 0); return 0; } @@ -942,7 +941,7 @@ static void lguest_clockevent_set_mode(enum clock_event_mode mode, case CLOCK_EVT_MODE_UNUSED: case CLOCK_EVT_MODE_SHUTDOWN: /* A 0 argument shuts the clock down. */ - kvm_hypercall0(LHCALL_SET_CLOCKEVENT); + hcall(LHCALL_SET_CLOCKEVENT, 0, 0, 0, 0); break; case CLOCK_EVT_MODE_ONESHOT: /* This is what we expect. */ @@ -1100,7 +1099,7 @@ static void set_lguest_basic_apic_ops(void) /* STOP! Until an interrupt comes in. */ static void lguest_safe_halt(void) { - kvm_hypercall0(LHCALL_HALT); + hcall(LHCALL_HALT, 0, 0, 0, 0); } /* @@ -1112,8 +1111,8 @@ static void lguest_safe_halt(void) */ static void lguest_power_off(void) { - kvm_hypercall2(LHCALL_SHUTDOWN, __pa("Power down"), - LGUEST_SHUTDOWN_POWEROFF); + hcall(LHCALL_SHUTDOWN, __pa("Power down"), + LGUEST_SHUTDOWN_POWEROFF, 0, 0); } /* @@ -1123,7 +1122,7 @@ static void lguest_power_off(void) */ static int lguest_panic(struct notifier_block *nb, unsigned long l, void *p) { - kvm_hypercall2(LHCALL_SHUTDOWN, __pa(p), LGUEST_SHUTDOWN_POWEROFF); + hcall(LHCALL_SHUTDOWN, __pa(p), LGUEST_SHUTDOWN_POWEROFF, 0, 0); /* The hcall won't return, but to keep gcc happy, we're "done". */ return NOTIFY_DONE; } @@ -1162,7 +1161,7 @@ static __init int early_put_chars(u32 vtermno, const char *buf, int count) len = sizeof(scratch) - 1; scratch[len] = '\0'; memcpy(scratch, buf, len); - kvm_hypercall1(LHCALL_NOTIFY, __pa(scratch)); + hcall(LHCALL_NOTIFY, __pa(scratch), 0, 0, 0); /* This routine returns the number of bytes actually written. */ return len; @@ -1174,7 +1173,7 @@ static __init int early_put_chars(u32 vtermno, const char *buf, int count) */ static void lguest_restart(char *reason) { - kvm_hypercall2(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART); + hcall(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART, 0, 0); } /*G:050 diff --git a/arch/x86/lguest/i386_head.S b/arch/x86/lguest/i386_head.S index 27eac0faee48..4f420c2f2d55 100644 --- a/arch/x86/lguest/i386_head.S +++ b/arch/x86/lguest/i386_head.S @@ -32,7 +32,7 @@ ENTRY(lguest_entry) */ movl $LHCALL_LGUEST_INIT, %eax movl $lguest_data - __PAGE_OFFSET, %ebx - .byte 0x0f,0x01,0xc1 /* KVM_HYPERCALL */ + int $LGUEST_TRAP_ENTRY /* Set up the initial stack so we can run C code. */ movl $(init_thread_union+THREAD_SIZE),%esp diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile index 419386c24b82..cbaf8f2b83df 100644 --- a/arch/x86/lib/Makefile +++ b/arch/x86/lib/Makefile @@ -20,7 +20,7 @@ lib-y := delay.o lib-y += thunk_$(BITS).o lib-y += usercopy_$(BITS).o getuser.o putuser.o lib-y += memcpy_$(BITS).o -lib-$(CONFIG_KPROBES) += insn.o inat.o +lib-$(CONFIG_INSTRUCTION_DECODER) += insn.o inat.o obj-y += msr.o msr-reg.o msr-reg-export.o diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 06630d26e56d..a4c768397baa 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -6,6 +6,7 @@ nostackp := $(call cc-option, -fno-stack-protector) CFLAGS_physaddr.o := $(nostackp) CFLAGS_setup_nx.o := $(nostackp) +obj-$(CONFIG_X86_PAT) += pat_rbtree.o obj-$(CONFIG_SMP) += tlb.o obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index edc8b95afc1a..931e5b753e77 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -30,6 +30,8 @@ #include <asm/pat.h> #include <asm/io.h> +#include "pat_internal.h" + #ifdef CONFIG_X86_PAT int __read_mostly pat_enabled = 1; @@ -53,19 +55,15 @@ static inline void pat_disable(const char *reason) #endif -static int debug_enable; +int pat_debug_enable; static int __init pat_debug_setup(char *str) { - debug_enable = 1; + pat_debug_enable = 1; return 0; } __setup("debugpat", pat_debug_setup); -#define dprintk(fmt, arg...) \ - do { if (debug_enable) printk(KERN_INFO fmt, ##arg); } while (0) - - static u64 __read_mostly boot_pat_state; enum { @@ -132,84 +130,7 @@ void pat_init(void) #undef PAT -static char *cattr_name(unsigned long flags) -{ - switch (flags & _PAGE_CACHE_MASK) { - case _PAGE_CACHE_UC: return "uncached"; - case _PAGE_CACHE_UC_MINUS: return "uncached-minus"; - case _PAGE_CACHE_WB: return "write-back"; - case _PAGE_CACHE_WC: return "write-combining"; - default: return "broken"; - } -} - -/* - * The global memtype list keeps track of memory type for specific - * physical memory areas. Conflicting memory types in different - * mappings can cause CPU cache corruption. To avoid this we keep track. - * - * The list is sorted based on starting address and can contain multiple - * entries for each address (this allows reference counting for overlapping - * areas). All the aliases have the same cache attributes of course. - * Zero attributes are represented as holes. - * - * The data structure is a list that is also organized as an rbtree - * sorted on the start address of memtype range. - * - * memtype_lock protects both the linear list and rbtree. - */ - -struct memtype { - u64 start; - u64 end; - unsigned long type; - struct list_head nd; - struct rb_node rb; -}; - -static struct rb_root memtype_rbroot = RB_ROOT; -static LIST_HEAD(memtype_list); -static DEFINE_SPINLOCK(memtype_lock); /* protects memtype list */ - -static struct memtype *memtype_rb_search(struct rb_root *root, u64 start) -{ - struct rb_node *node = root->rb_node; - struct memtype *last_lower = NULL; - - while (node) { - struct memtype *data = container_of(node, struct memtype, rb); - - if (data->start < start) { - last_lower = data; - node = node->rb_right; - } else if (data->start > start) { - node = node->rb_left; - } else - return data; - } - - /* Will return NULL if there is no entry with its start <= start */ - return last_lower; -} - -static void memtype_rb_insert(struct rb_root *root, struct memtype *data) -{ - struct rb_node **new = &(root->rb_node); - struct rb_node *parent = NULL; - - while (*new) { - struct memtype *this = container_of(*new, struct memtype, rb); - - parent = *new; - if (data->start <= this->start) - new = &((*new)->rb_left); - else if (data->start > this->start) - new = &((*new)->rb_right); - } - - rb_link_node(&data->rb, parent, new); - rb_insert_color(&data->rb, root); -} +static DEFINE_SPINLOCK(memtype_lock); /* protects memtype accesses */ /* * Does intersection of PAT memory type and MTRR memory type and returns @@ -237,33 +158,6 @@ static unsigned long pat_x_mtrr_type(u64 start, u64 end, unsigned long req_type) return req_type; } -static int -chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type) -{ - if (new->type != entry->type) { - if (type) { - new->type = entry->type; - *type = entry->type; - } else - goto conflict; - } - - /* check overlaps with more than one entry in the list */ - list_for_each_entry_continue(entry, &memtype_list, nd) { - if (new->end <= entry->start) - break; - else if (new->type != entry->type) - goto conflict; - } - return 0; - - conflict: - printk(KERN_INFO "%s:%d conflicting memory types " - "%Lx-%Lx %s<->%s\n", current->comm, current->pid, new->start, - new->end, cattr_name(new->type), cattr_name(entry->type)); - return -EBUSY; -} - static int pat_pagerange_is_ram(unsigned long start, unsigned long end) { int ram_page = 0, not_rampage = 0; @@ -364,9 +258,8 @@ static int free_ram_pages_type(u64 start, u64 end) int reserve_memtype(u64 start, u64 end, unsigned long req_type, unsigned long *new_type) { - struct memtype *new, *entry; + struct memtype *new; unsigned long actual_type; - struct list_head *where; int is_range_ram; int err = 0; @@ -423,42 +316,7 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, spin_lock(&memtype_lock); - /* Search for existing mapping that overlaps the current range */ - where = NULL; - list_for_each_entry(entry, &memtype_list, nd) { - if (end <= entry->start) { - where = entry->nd.prev; - break; - } else if (start <= entry->start) { /* end > entry->start */ - err = chk_conflict(new, entry, new_type); - if (!err) { - dprintk("Overlap at 0x%Lx-0x%Lx\n", - entry->start, entry->end); - where = entry->nd.prev; - } - break; - } else if (start < entry->end) { /* start > entry->start */ - err = chk_conflict(new, entry, new_type); - if (!err) { - dprintk("Overlap at 0x%Lx-0x%Lx\n", - entry->start, entry->end); - - /* - * Move to right position in the linked - * list to add this new entry - */ - list_for_each_entry_continue(entry, - &memtype_list, nd) { - if (start <= entry->start) { - where = entry->nd.prev; - break; - } - } - } - break; - } - } - + err = rbt_memtype_check_insert(new, new_type); if (err) { printk(KERN_INFO "reserve_memtype failed 0x%Lx-0x%Lx, " "track %s, req %s\n", @@ -469,13 +327,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, return err; } - if (where) - list_add(&new->nd, where); - else - list_add_tail(&new->nd, &memtype_list); - - memtype_rb_insert(&memtype_rbroot, new); - spin_unlock(&memtype_lock); dprintk("reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s, ret %s\n", @@ -487,7 +338,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, int free_memtype(u64 start, u64 end) { - struct memtype *entry, *saved_entry; int err = -EINVAL; int is_range_ram; @@ -511,46 +361,7 @@ int free_memtype(u64 start, u64 end) } spin_lock(&memtype_lock); - - entry = memtype_rb_search(&memtype_rbroot, start); - if (unlikely(entry == NULL)) - goto unlock_ret; - - /* - * Saved entry points to an entry with start same or less than what - * we searched for. Now go through the list in both directions to look - * for the entry that matches with both start and end, with list stored - * in sorted start address - */ - saved_entry = entry; - list_for_each_entry_from(entry, &memtype_list, nd) { - if (entry->start == start && entry->end == end) { - rb_erase(&entry->rb, &memtype_rbroot); - list_del(&entry->nd); - kfree(entry); - err = 0; - break; - } else if (entry->start > start) { - break; - } - } - - if (!err) - goto unlock_ret; - - entry = saved_entry; - list_for_each_entry_reverse(entry, &memtype_list, nd) { - if (entry->start == start && entry->end == end) { - rb_erase(&entry->rb, &memtype_rbroot); - list_del(&entry->nd); - kfree(entry); - err = 0; - break; - } else if (entry->start < start) { - break; - } - } -unlock_ret: + err = rbt_memtype_erase(start, end); spin_unlock(&memtype_lock); if (err) { @@ -599,7 +410,7 @@ static unsigned long lookup_memtype(u64 paddr) spin_lock(&memtype_lock); - entry = memtype_rb_search(&memtype_rbroot, paddr); + entry = rbt_memtype_lookup(paddr); if (entry != NULL) rettype = entry->type; else @@ -936,29 +747,25 @@ EXPORT_SYMBOL_GPL(pgprot_writecombine); #if defined(CONFIG_DEBUG_FS) && defined(CONFIG_X86_PAT) -/* get Nth element of the linked list */ static struct memtype *memtype_get_idx(loff_t pos) { - struct memtype *list_node, *print_entry; - int i = 1; + struct memtype *print_entry; + int ret; - print_entry = kmalloc(sizeof(struct memtype), GFP_KERNEL); + print_entry = kzalloc(sizeof(struct memtype), GFP_KERNEL); if (!print_entry) return NULL; spin_lock(&memtype_lock); - list_for_each_entry(list_node, &memtype_list, nd) { - if (pos == i) { - *print_entry = *list_node; - spin_unlock(&memtype_lock); - return print_entry; - } - ++i; - } + ret = rbt_memtype_copy_nth_element(print_entry, pos); spin_unlock(&memtype_lock); - kfree(print_entry); - return NULL; + if (!ret) { + return print_entry; + } else { + kfree(print_entry); + return NULL; + } } static void *memtype_seq_start(struct seq_file *seq, loff_t *pos) diff --git a/arch/x86/mm/pat_internal.h b/arch/x86/mm/pat_internal.h new file mode 100644 index 000000000000..4f39eefa3e61 --- /dev/null +++ b/arch/x86/mm/pat_internal.h @@ -0,0 +1,46 @@ +#ifndef __PAT_INTERNAL_H_ +#define __PAT_INTERNAL_H_ + +extern int pat_debug_enable; + +#define dprintk(fmt, arg...) \ + do { if (pat_debug_enable) printk(KERN_INFO fmt, ##arg); } while (0) + +struct memtype { + u64 start; + u64 end; + u64 subtree_max_end; + unsigned long type; + struct rb_node rb; +}; + +static inline char *cattr_name(unsigned long flags) +{ + switch (flags & _PAGE_CACHE_MASK) { + case _PAGE_CACHE_UC: return "uncached"; + case _PAGE_CACHE_UC_MINUS: return "uncached-minus"; + case _PAGE_CACHE_WB: return "write-back"; + case _PAGE_CACHE_WC: return "write-combining"; + default: return "broken"; + } +} + +#ifdef CONFIG_X86_PAT +extern int rbt_memtype_check_insert(struct memtype *new, + unsigned long *new_type); +extern int rbt_memtype_erase(u64 start, u64 end); +extern struct memtype *rbt_memtype_lookup(u64 addr); +extern int rbt_memtype_copy_nth_element(struct memtype *out, loff_t pos); +#else +static inline int rbt_memtype_check_insert(struct memtype *new, + unsigned long *new_type) +{ return 0; } +static inline int rbt_memtype_erase(u64 start, u64 end) +{ return 0; } +static inline struct memtype *rbt_memtype_lookup(u64 addr) +{ return NULL; } +static inline int rbt_memtype_copy_nth_element(struct memtype *out, loff_t pos) +{ return 0; } +#endif + +#endif /* __PAT_INTERNAL_H_ */ diff --git a/arch/x86/mm/pat_rbtree.c b/arch/x86/mm/pat_rbtree.c new file mode 100644 index 000000000000..07de4cb8cc30 --- /dev/null +++ b/arch/x86/mm/pat_rbtree.c @@ -0,0 +1,273 @@ +/* + * Handle caching attributes in page tables (PAT) + * + * Authors: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> + * Suresh B Siddha <suresh.b.siddha@intel.com> + * + * Interval tree (augmented rbtree) used to store the PAT memory type + * reservations. + */ + +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/rbtree.h> +#include <linux/sched.h> +#include <linux/gfp.h> + +#include <asm/pgtable.h> +#include <asm/pat.h> + +#include "pat_internal.h" + +/* + * The memtype tree keeps track of memory type for specific + * physical memory areas. Without proper tracking, conflicting memory + * types in different mappings can cause CPU cache corruption. + * + * The tree is an interval tree (augmented rbtree) with tree ordered + * on starting address. Tree can contain multiple entries for + * different regions which overlap. All the aliases have the same + * cache attributes of course. + * + * memtype_lock protects the rbtree. + */ + +static void memtype_rb_augment_cb(struct rb_node *node); +static struct rb_root memtype_rbroot = RB_AUGMENT_ROOT(&memtype_rb_augment_cb); + +static int is_node_overlap(struct memtype *node, u64 start, u64 end) +{ + if (node->start >= end || node->end <= start) + return 0; + + return 1; +} + +static u64 get_subtree_max_end(struct rb_node *node) +{ + u64 ret = 0; + if (node) { + struct memtype *data = container_of(node, struct memtype, rb); + ret = data->subtree_max_end; + } + return ret; +} + +/* Update 'subtree_max_end' for a node, based on node and its children */ +static void update_node_max_end(struct rb_node *node) +{ + struct memtype *data; + u64 max_end, child_max_end; + + if (!node) + return; + + data = container_of(node, struct memtype, rb); + max_end = data->end; + + child_max_end = get_subtree_max_end(node->rb_right); + if (child_max_end > max_end) + max_end = child_max_end; + + child_max_end = get_subtree_max_end(node->rb_left); + if (child_max_end > max_end) + max_end = child_max_end; + + data->subtree_max_end = max_end; +} + +/* Update 'subtree_max_end' for a node and all its ancestors */ +static void update_path_max_end(struct rb_node *node) +{ + u64 old_max_end, new_max_end; + + while (node) { + struct memtype *data = container_of(node, struct memtype, rb); + + old_max_end = data->subtree_max_end; + update_node_max_end(node); + new_max_end = data->subtree_max_end; + + if (new_max_end == old_max_end) + break; + + node = rb_parent(node); + } +} + +/* Find the first (lowest start addr) overlapping range from rb tree */ +static struct memtype *memtype_rb_lowest_match(struct rb_root *root, + u64 start, u64 end) +{ + struct rb_node *node = root->rb_node; + struct memtype *last_lower = NULL; + + while (node) { + struct memtype *data = container_of(node, struct memtype, rb); + + if (get_subtree_max_end(node->rb_left) > start) { + /* Lowest overlap if any must be on left side */ + node = node->rb_left; + } else if (is_node_overlap(data, start, end)) { + last_lower = data; + break; + } else if (start >= data->start) { + /* Lowest overlap if any must be on right side */ + node = node->rb_right; + } else { + break; + } + } + return last_lower; /* Returns NULL if there is no overlap */ +} + +static struct memtype *memtype_rb_exact_match(struct rb_root *root, + u64 start, u64 end) +{ + struct memtype *match; + + match = memtype_rb_lowest_match(root, start, end); + while (match != NULL && match->start < end) { + struct rb_node *node; + + if (match->start == start && match->end == end) + return match; + + node = rb_next(&match->rb); + if (node) + match = container_of(node, struct memtype, rb); + else + match = NULL; + } + + return NULL; /* Returns NULL if there is no exact match */ +} + +static int memtype_rb_check_conflict(struct rb_root *root, + u64 start, u64 end, + unsigned long reqtype, unsigned long *newtype) +{ + struct rb_node *node; + struct memtype *match; + int found_type = reqtype; + + match = memtype_rb_lowest_match(&memtype_rbroot, start, end); + if (match == NULL) + goto success; + + if (match->type != found_type && newtype == NULL) + goto failure; + + dprintk("Overlap at 0x%Lx-0x%Lx\n", match->start, match->end); + found_type = match->type; + + node = rb_next(&match->rb); + while (node) { + match = container_of(node, struct memtype, rb); + + if (match->start >= end) /* Checked all possible matches */ + goto success; + + if (is_node_overlap(match, start, end) && + match->type != found_type) { + goto failure; + } + + node = rb_next(&match->rb); + } +success: + if (newtype) + *newtype = found_type; + + return 0; + +failure: + printk(KERN_INFO "%s:%d conflicting memory types " + "%Lx-%Lx %s<->%s\n", current->comm, current->pid, start, + end, cattr_name(found_type), cattr_name(match->type)); + return -EBUSY; +} + +static void memtype_rb_augment_cb(struct rb_node *node) +{ + if (node) + update_path_max_end(node); +} + +static void memtype_rb_insert(struct rb_root *root, struct memtype *newdata) +{ + struct rb_node **node = &(root->rb_node); + struct rb_node *parent = NULL; + + while (*node) { + struct memtype *data = container_of(*node, struct memtype, rb); + + parent = *node; + if (newdata->start <= data->start) + node = &((*node)->rb_left); + else if (newdata->start > data->start) + node = &((*node)->rb_right); + } + + rb_link_node(&newdata->rb, parent, node); + rb_insert_color(&newdata->rb, root); +} + +int rbt_memtype_check_insert(struct memtype *new, unsigned long *ret_type) +{ + int err = 0; + + err = memtype_rb_check_conflict(&memtype_rbroot, new->start, new->end, + new->type, ret_type); + + if (!err) { + if (ret_type) + new->type = *ret_type; + + memtype_rb_insert(&memtype_rbroot, new); + } + return err; +} + +int rbt_memtype_erase(u64 start, u64 end) +{ + struct memtype *data; + + data = memtype_rb_exact_match(&memtype_rbroot, start, end); + if (!data) + return -EINVAL; + + rb_erase(&data->rb, &memtype_rbroot); + return 0; +} + +struct memtype *rbt_memtype_lookup(u64 addr) +{ + struct memtype *data; + data = memtype_rb_lowest_match(&memtype_rbroot, addr, addr + PAGE_SIZE); + return data; +} + +#if defined(CONFIG_DEBUG_FS) +int rbt_memtype_copy_nth_element(struct memtype *out, loff_t pos) +{ + struct rb_node *node; + int i = 1; + + node = rb_first(&memtype_rbroot); + while (node && pos != i) { + node = rb_next(node); + i++; + } + + if (node) { /* pos == i */ + struct memtype *this = container_of(node, struct memtype, rb); + *out = *this; + return 0; + } else { + return 1; + } +} +#endif diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 2bf90fafa7b5..c8abc4d1bf35 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -239,11 +239,11 @@ static void arch_perfmon_setup_counters(void) if (eax.split.version_id == 0 && current_cpu_data.x86 == 6 && current_cpu_data.x86_model == 15) { eax.split.version_id = 2; - eax.split.num_events = 2; + eax.split.num_counters = 2; eax.split.bit_width = 40; } - num_counters = eax.split.num_events; + num_counters = eax.split.num_counters; op_arch_perfmon_spec.num_counters = num_counters; op_arch_perfmon_spec.num_controls = num_counters; diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 32764b8880b5..b3c6c59ed302 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -476,6 +476,7 @@ void xen_timer_resume(void) __init void xen_time_init(void) { int cpu = smp_processor_id(); + struct timespec tp; clocksource_register(&xen_clocksource); @@ -487,9 +488,8 @@ __init void xen_time_init(void) } /* Set initial system time with full resolution */ - xen_read_wallclock(&xtime); - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); + xen_read_wallclock(&tp); + do_settimeofday(&tp); setup_force_cpu_cap(X86_FEATURE_TSC); diff --git a/arch/xtensa/kernel/time.c b/arch/xtensa/kernel/time.c index 19f7df30937f..19df764f6399 100644 --- a/arch/xtensa/kernel/time.c +++ b/arch/xtensa/kernel/time.c @@ -60,11 +60,6 @@ static struct irqaction timer_irqaction = { void __init time_init(void) { - /* FIXME: xtime&wall_to_monotonic are set in timekeeping_init. */ - read_persistent_clock(&xtime); - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - #ifdef CONFIG_XTENSA_CALIBRATE_CCOUNT printk("Calibrating CPU frequency "); platform_calibrate_ccount(); diff --git a/drivers/acpi/acpica/exprep.c b/drivers/acpi/acpica/exprep.c index a610ebe18edd..2fbfe51fb141 100644 --- a/drivers/acpi/acpica/exprep.c +++ b/drivers/acpi/acpica/exprep.c @@ -471,13 +471,18 @@ acpi_status acpi_ex_prep_field_value(struct acpi_create_field_info *info) /* allow full data read from EC address space */ if (obj_desc->field.region_obj->region.space_id == ACPI_ADR_SPACE_EC) { - if (obj_desc->common_field.bit_length > 8) - obj_desc->common_field.access_bit_width = - ACPI_ROUND_UP(obj_desc->common_field. - bit_length, 8); + if (obj_desc->common_field.bit_length > 8) { + unsigned width = + ACPI_ROUND_BITS_UP_TO_BYTES( + obj_desc->common_field.bit_length); + // access_bit_width is u8, don't overflow it + if (width > 8) + width = 8; obj_desc->common_field.access_byte_width = - ACPI_DIV_8(obj_desc->common_field. - access_bit_width); + width; + obj_desc->common_field.access_bit_width = + 8 * width; + } } ACPI_DEBUG_PRINT((ACPI_DB_BFIELD, diff --git a/drivers/block/hd.c b/drivers/block/hd.c index 034e6dfc878c..81c78b3ce2df 100644 --- a/drivers/block/hd.c +++ b/drivers/block/hd.c @@ -164,12 +164,12 @@ unsigned long read_timer(void) unsigned long t, flags; int i; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); t = jiffies * 11932; outb_p(0, 0x43); i = inb_p(0x40); i |= inb(0x40) << 8; - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); return(t - i); } #endif diff --git a/drivers/clocksource/cs5535-clockevt.c b/drivers/clocksource/cs5535-clockevt.c index b314a999aabe..d7be69f13154 100644 --- a/drivers/clocksource/cs5535-clockevt.c +++ b/drivers/clocksource/cs5535-clockevt.c @@ -154,14 +154,14 @@ static int __init cs5535_mfgpt_init(void) if (cs5535_mfgpt_setup_irq(timer, MFGPT_CMP2, &timer_irq)) { printk(KERN_ERR DRV_NAME ": Could not set up IRQ %d\n", timer_irq); - return -EIO; + goto err_timer; } /* And register it with the kernel */ ret = setup_irq(timer_irq, &mfgptirq); if (ret) { printk(KERN_ERR DRV_NAME ": Unable to set up the interrupt.\n"); - goto err; + goto err_irq; } /* Set the clock scale and enable the event mode for CMP2 */ @@ -184,8 +184,10 @@ static int __init cs5535_mfgpt_init(void) return 0; -err: +err_irq: cs5535_mfgpt_release_irq(cs5535_event_clock, MFGPT_CMP2, &timer_irq); +err_timer: + cs5535_mfgpt_free_timer(cs5535_event_clock); printk(KERN_ERR DRV_NAME ": Unable to set up the MFGPT clock source\n"); return -EIO; } diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 2d5d575e889d..e02e4174c2c8 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -662,32 +662,20 @@ static ssize_t show_bios_limit(struct cpufreq_policy *policy, char *buf) return sprintf(buf, "%u\n", policy->cpuinfo.max_freq); } -#define define_one_ro(_name) \ -static struct freq_attr _name = \ -__ATTR(_name, 0444, show_##_name, NULL) - -#define define_one_ro0400(_name) \ -static struct freq_attr _name = \ -__ATTR(_name, 0400, show_##_name, NULL) - -#define define_one_rw(_name) \ -static struct freq_attr _name = \ -__ATTR(_name, 0644, show_##_name, store_##_name) - -define_one_ro0400(cpuinfo_cur_freq); -define_one_ro(cpuinfo_min_freq); -define_one_ro(cpuinfo_max_freq); -define_one_ro(cpuinfo_transition_latency); -define_one_ro(scaling_available_governors); -define_one_ro(scaling_driver); -define_one_ro(scaling_cur_freq); -define_one_ro(bios_limit); -define_one_ro(related_cpus); -define_one_ro(affected_cpus); -define_one_rw(scaling_min_freq); -define_one_rw(scaling_max_freq); -define_one_rw(scaling_governor); -define_one_rw(scaling_setspeed); +cpufreq_freq_attr_ro_perm(cpuinfo_cur_freq, 0400); +cpufreq_freq_attr_ro(cpuinfo_min_freq); +cpufreq_freq_attr_ro(cpuinfo_max_freq); +cpufreq_freq_attr_ro(cpuinfo_transition_latency); +cpufreq_freq_attr_ro(scaling_available_governors); +cpufreq_freq_attr_ro(scaling_driver); +cpufreq_freq_attr_ro(scaling_cur_freq); +cpufreq_freq_attr_ro(bios_limit); +cpufreq_freq_attr_ro(related_cpus); +cpufreq_freq_attr_ro(affected_cpus); +cpufreq_freq_attr_rw(scaling_min_freq); +cpufreq_freq_attr_rw(scaling_max_freq); +cpufreq_freq_attr_rw(scaling_governor); +cpufreq_freq_attr_rw(scaling_setspeed); static struct attribute *default_attrs[] = { &cpuinfo_min_freq.attr, diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index 599a40b25cb0..ce5248e04218 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -178,12 +178,8 @@ static ssize_t show_sampling_rate_min(struct kobject *kobj, return sprintf(buf, "%u\n", min_sampling_rate); } -#define define_one_ro(_name) \ -static struct global_attr _name = \ -__ATTR(_name, 0444, show_##_name, NULL) - -define_one_ro(sampling_rate_max); -define_one_ro(sampling_rate_min); +define_one_global_ro(sampling_rate_max); +define_one_global_ro(sampling_rate_min); /* cpufreq_conservative Governor Tunables */ #define show_one(file_name, object) \ @@ -221,12 +217,8 @@ show_one_old(freq_step); show_one_old(sampling_rate_min); show_one_old(sampling_rate_max); -#define define_one_ro_old(object, _name) \ -static struct freq_attr object = \ -__ATTR(_name, 0444, show_##_name##_old, NULL) - -define_one_ro_old(sampling_rate_min_old, sampling_rate_min); -define_one_ro_old(sampling_rate_max_old, sampling_rate_max); +cpufreq_freq_attr_ro_old(sampling_rate_min); +cpufreq_freq_attr_ro_old(sampling_rate_max); /*** delete after deprecation time ***/ @@ -364,16 +356,12 @@ static ssize_t store_freq_step(struct kobject *a, struct attribute *b, return count; } -#define define_one_rw(_name) \ -static struct global_attr _name = \ -__ATTR(_name, 0644, show_##_name, store_##_name) - -define_one_rw(sampling_rate); -define_one_rw(sampling_down_factor); -define_one_rw(up_threshold); -define_one_rw(down_threshold); -define_one_rw(ignore_nice_load); -define_one_rw(freq_step); +define_one_global_rw(sampling_rate); +define_one_global_rw(sampling_down_factor); +define_one_global_rw(up_threshold); +define_one_global_rw(down_threshold); +define_one_global_rw(ignore_nice_load); +define_one_global_rw(freq_step); static struct attribute *dbs_attributes[] = { &sampling_rate_max.attr, @@ -409,16 +397,12 @@ write_one_old(down_threshold); write_one_old(ignore_nice_load); write_one_old(freq_step); -#define define_one_rw_old(object, _name) \ -static struct freq_attr object = \ -__ATTR(_name, 0644, show_##_name##_old, store_##_name##_old) - -define_one_rw_old(sampling_rate_old, sampling_rate); -define_one_rw_old(sampling_down_factor_old, sampling_down_factor); -define_one_rw_old(up_threshold_old, up_threshold); -define_one_rw_old(down_threshold_old, down_threshold); -define_one_rw_old(ignore_nice_load_old, ignore_nice_load); -define_one_rw_old(freq_step_old, freq_step); +cpufreq_freq_attr_rw_old(sampling_rate); +cpufreq_freq_attr_rw_old(sampling_down_factor); +cpufreq_freq_attr_rw_old(up_threshold); +cpufreq_freq_attr_rw_old(down_threshold); +cpufreq_freq_attr_rw_old(ignore_nice_load); +cpufreq_freq_attr_rw_old(freq_step); static struct attribute *dbs_attributes_old[] = { &sampling_rate_max_old.attr, diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index bd444dc93cf2..c00b25f4d243 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -234,12 +234,8 @@ static ssize_t show_sampling_rate_min(struct kobject *kobj, return sprintf(buf, "%u\n", min_sampling_rate); } -#define define_one_ro(_name) \ -static struct global_attr _name = \ -__ATTR(_name, 0444, show_##_name, NULL) - -define_one_ro(sampling_rate_max); -define_one_ro(sampling_rate_min); +define_one_global_ro(sampling_rate_max); +define_one_global_ro(sampling_rate_min); /* cpufreq_ondemand Governor Tunables */ #define show_one(file_name, object) \ @@ -274,12 +270,8 @@ show_one_old(powersave_bias); show_one_old(sampling_rate_min); show_one_old(sampling_rate_max); -#define define_one_ro_old(object, _name) \ -static struct freq_attr object = \ -__ATTR(_name, 0444, show_##_name##_old, NULL) - -define_one_ro_old(sampling_rate_min_old, sampling_rate_min); -define_one_ro_old(sampling_rate_max_old, sampling_rate_max); +cpufreq_freq_attr_ro_old(sampling_rate_min); +cpufreq_freq_attr_ro_old(sampling_rate_max); /*** delete after deprecation time ***/ @@ -376,14 +368,10 @@ static ssize_t store_powersave_bias(struct kobject *a, struct attribute *b, return count; } -#define define_one_rw(_name) \ -static struct global_attr _name = \ -__ATTR(_name, 0644, show_##_name, store_##_name) - -define_one_rw(sampling_rate); -define_one_rw(up_threshold); -define_one_rw(ignore_nice_load); -define_one_rw(powersave_bias); +define_one_global_rw(sampling_rate); +define_one_global_rw(up_threshold); +define_one_global_rw(ignore_nice_load); +define_one_global_rw(powersave_bias); static struct attribute *dbs_attributes[] = { &sampling_rate_max.attr, @@ -415,14 +403,10 @@ write_one_old(up_threshold); write_one_old(ignore_nice_load); write_one_old(powersave_bias); -#define define_one_rw_old(object, _name) \ -static struct freq_attr object = \ -__ATTR(_name, 0644, show_##_name##_old, store_##_name##_old) - -define_one_rw_old(sampling_rate_old, sampling_rate); -define_one_rw_old(up_threshold_old, up_threshold); -define_one_rw_old(ignore_nice_load_old, ignore_nice_load); -define_one_rw_old(powersave_bias_old, powersave_bias); +cpufreq_freq_attr_rw_old(sampling_rate); +cpufreq_freq_attr_rw_old(up_threshold); +cpufreq_freq_attr_rw_old(ignore_nice_load); +cpufreq_freq_attr_rw_old(powersave_bias); static struct attribute *dbs_attributes_old[] = { &sampling_rate_max_old.attr, diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c index 702dcc98c074..14a34d99eea2 100644 --- a/drivers/firewire/core-cdev.c +++ b/drivers/firewire/core-cdev.c @@ -960,6 +960,8 @@ static int ioctl_queue_iso(struct client *client, union ioctl_arg *arg) u.packet.header_length = GET_HEADER_LENGTH(control); if (ctx->type == FW_ISO_CONTEXT_TRANSMIT) { + if (u.packet.header_length % 4 != 0) + return -EINVAL; header_length = u.packet.header_length; } else { /* @@ -969,7 +971,8 @@ static int ioctl_queue_iso(struct client *client, union ioctl_arg *arg) if (ctx->header_size == 0) { if (u.packet.header_length > 0) return -EINVAL; - } else if (u.packet.header_length % ctx->header_size != 0) { + } else if (u.packet.header_length == 0 || + u.packet.header_length % ctx->header_size != 0) { return -EINVAL; } header_length = 0; @@ -1354,24 +1357,24 @@ static int dispatch_ioctl(struct client *client, return -ENODEV; if (_IOC_TYPE(cmd) != '#' || - _IOC_NR(cmd) >= ARRAY_SIZE(ioctl_handlers)) + _IOC_NR(cmd) >= ARRAY_SIZE(ioctl_handlers) || + _IOC_SIZE(cmd) > sizeof(buffer)) return -EINVAL; - if (_IOC_DIR(cmd) & _IOC_WRITE) { - if (_IOC_SIZE(cmd) > sizeof(buffer) || - copy_from_user(&buffer, arg, _IOC_SIZE(cmd))) + if (_IOC_DIR(cmd) == _IOC_READ) + memset(&buffer, 0, _IOC_SIZE(cmd)); + + if (_IOC_DIR(cmd) & _IOC_WRITE) + if (copy_from_user(&buffer, arg, _IOC_SIZE(cmd))) return -EFAULT; - } ret = ioctl_handlers[_IOC_NR(cmd)](client, &buffer); if (ret < 0) return ret; - if (_IOC_DIR(cmd) & _IOC_READ) { - if (_IOC_SIZE(cmd) > sizeof(buffer) || - copy_to_user(arg, &buffer, _IOC_SIZE(cmd))) + if (_IOC_DIR(cmd) & _IOC_READ) + if (copy_to_user(arg, &buffer, _IOC_SIZE(cmd))) return -EFAULT; - } return ret; } diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index c1605b528e8f..0f28d91f29d8 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -142,6 +142,12 @@ static const char *temperature_sensors_sets[][41] = { "TM1S", "TM2P", "TM2S", "TM3S", "TM8P", "TM8S", "TM9P", "TM9S", "TN0C", "TN0D", "TN0H", "TS0C", "Tp0C", "Tp1C", "Tv0S", "Tv1S", NULL }, +/* Set 17: iMac 9,1 */ + { "TA0P", "TC0D", "TC0H", "TC0P", "TG0D", "TG0H", "TH0P", "TL0P", + "TN0D", "TN0H", "TN0P", "TO0P", "Tm0P", "Tp0P", NULL }, +/* Set 18: MacBook Pro 2,2 */ + { "TB0T", "TC0D", "TC0P", "TG0H", "TG0P", "TG0T", "TM0P", "TTF0", + "Th0H", "Th1H", "Tm0P", "Ts0P", NULL }, }; /* List of keys used to read/write fan speeds */ @@ -1350,6 +1356,10 @@ static __initdata struct dmi_match_data applesmc_dmi_data[] = { { .accelerometer = 1, .light = 1, .temperature_set = 15 }, /* MacPro3,1: temperature set 16 */ { .accelerometer = 0, .light = 0, .temperature_set = 16 }, +/* iMac 9,1: light sensor only, temperature set 17 */ + { .accelerometer = 0, .light = 0, .temperature_set = 17 }, +/* MacBook Pro 2,2: accelerometer, backlight and temperature set 18 */ + { .accelerometer = 1, .light = 1, .temperature_set = 18 }, }; /* Note that DMI_MATCH(...,"MacBook") will match "MacBookPro1,1". @@ -1375,6 +1385,10 @@ static __initdata struct dmi_system_id applesmc_whitelist[] = { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro3") }, &applesmc_dmi_data[9]}, + { applesmc_dmi_match, "Apple MacBook Pro 2,2", { + DMI_MATCH(DMI_BOARD_VENDOR, "Apple Computer, Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro2,2") }, + &applesmc_dmi_data[18]}, { applesmc_dmi_match, "Apple MacBook Pro", { DMI_MATCH(DMI_BOARD_VENDOR,"Apple"), DMI_MATCH(DMI_PRODUCT_NAME,"MacBookPro") }, @@ -1415,6 +1429,10 @@ static __initdata struct dmi_system_id applesmc_whitelist[] = { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "MacPro") }, &applesmc_dmi_data[4]}, + { applesmc_dmi_match, "Apple iMac 9,1", { + DMI_MATCH(DMI_BOARD_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "iMac9,1") }, + &applesmc_dmi_data[17]}, { applesmc_dmi_match, "Apple iMac 8", { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "iMac8") }, diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c index 1002befd87d5..5be09c048c5f 100644 --- a/drivers/hwmon/it87.c +++ b/drivers/hwmon/it87.c @@ -539,14 +539,14 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, struct it87_data *data = dev_get_drvdata(dev); long val; + u8 reg; if (strict_strtol(buf, 10, &val) < 0) return -EINVAL; - mutex_lock(&data->update_lock); - - data->sensor &= ~(1 << nr); - data->sensor &= ~(8 << nr); + reg = it87_read_value(data, IT87_REG_TEMP_ENABLE); + reg &= ~(1 << nr); + reg &= ~(8 << nr); if (val == 2) { /* backwards compatibility */ dev_warn(dev, "Sensor type 2 is deprecated, please use 4 " "instead\n"); @@ -554,14 +554,16 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, } /* 3 = thermal diode; 4 = thermistor; 0 = disabled */ if (val == 3) - data->sensor |= 1 << nr; + reg |= 1 << nr; else if (val == 4) - data->sensor |= 8 << nr; - else if (val != 0) { - mutex_unlock(&data->update_lock); + reg |= 8 << nr; + else if (val != 0) return -EINVAL; - } + + mutex_lock(&data->update_lock); + data->sensor = reg; it87_write_value(data, IT87_REG_TEMP_ENABLE, data->sensor); + data->valid = 0; /* Force cache refresh */ mutex_unlock(&data->update_lock); return count; } @@ -1841,14 +1843,10 @@ static void __devinit it87_init_device(struct platform_device *pdev) it87_write_value(data, IT87_REG_TEMP_HIGH(i), 127); } - /* Check if temperature channels are reset manually or by some reason */ - tmp = it87_read_value(data, IT87_REG_TEMP_ENABLE); - if ((tmp & 0x3f) == 0) { - /* Temp1,Temp3=thermistor; Temp2=thermal diode */ - tmp = (tmp & 0xc0) | 0x2a; - it87_write_value(data, IT87_REG_TEMP_ENABLE, tmp); - } - data->sensor = tmp; + /* Temperature channels are not forcibly enabled, as they can be + * set to two different sensor types and we can't guess which one + * is correct for a given system. These channels can be enabled at + * run-time through the temp{1-3}_type sysfs accessors if needed. */ /* Check if voltage monitors are reset manually or by some reason */ tmp = it87_read_value(data, IT87_REG_VIN_ENABLE); diff --git a/drivers/hwmon/sht15.c b/drivers/hwmon/sht15.c index 6b2d8ae64fe1..a610e7880fb3 100644 --- a/drivers/hwmon/sht15.c +++ b/drivers/hwmon/sht15.c @@ -303,13 +303,13 @@ error_ret: **/ static inline int sht15_calc_temp(struct sht15_data *data) { - int d1 = 0; + int d1 = temppoints[0].d1; int i; - for (i = 1; i < ARRAY_SIZE(temppoints); i++) + for (i = ARRAY_SIZE(temppoints) - 1; i > 0; i--) /* Find pointer to interpolate */ if (data->supply_uV > temppoints[i - 1].vdd) { - d1 = (data->supply_uV/1000 - temppoints[i - 1].vdd) + d1 = (data->supply_uV - temppoints[i - 1].vdd) * (temppoints[i].d1 - temppoints[i - 1].d1) / (temppoints[i].vdd - temppoints[i - 1].vdd) + temppoints[i - 1].d1; @@ -542,7 +542,12 @@ static int __devinit sht15_probe(struct platform_device *pdev) /* If a regulator is available, query what the supply voltage actually is!*/ data->reg = regulator_get(data->dev, "vcc"); if (!IS_ERR(data->reg)) { - data->supply_uV = regulator_get_voltage(data->reg); + int voltage; + + voltage = regulator_get_voltage(data->reg); + if (voltage) + data->supply_uV = voltage; + regulator_enable(data->reg); /* setup a notifier block to update this if another device * causes the voltage to change */ diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c index 7e18bcf05a66..46239e47a260 100644 --- a/drivers/input/gameport/gameport.c +++ b/drivers/input/gameport/gameport.c @@ -59,11 +59,11 @@ static unsigned int get_time_pit(void) unsigned long flags; unsigned int count; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); outb_p(0x00, 0x43); count = inb_p(0x40); count |= inb_p(0x40) << 8; - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); return count; } diff --git a/drivers/input/input.c b/drivers/input/input.c index afd4e2b7658c..9c79bd56b51a 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -660,7 +660,14 @@ static int input_default_setkeycode(struct input_dev *dev, int input_get_keycode(struct input_dev *dev, unsigned int scancode, unsigned int *keycode) { - return dev->getkeycode(dev, scancode, keycode); + unsigned long flags; + int retval; + + spin_lock_irqsave(&dev->event_lock, flags); + retval = dev->getkeycode(dev, scancode, keycode); + spin_unlock_irqrestore(&dev->event_lock, flags); + + return retval; } EXPORT_SYMBOL(input_get_keycode); diff --git a/drivers/input/joystick/analog.c b/drivers/input/joystick/analog.c index 1c0b529c06aa..4afe0a3b4884 100644 --- a/drivers/input/joystick/analog.c +++ b/drivers/input/joystick/analog.c @@ -146,11 +146,11 @@ static unsigned int get_time_pit(void) unsigned long flags; unsigned int count; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); outb_p(0x00, 0x43); count = inb_p(0x40); count |= inb_p(0x40) << 8; - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); return count; } diff --git a/drivers/input/keyboard/matrix_keypad.c b/drivers/input/keyboard/matrix_keypad.c index ffc25cfcef7a..b443e088fd3c 100644 --- a/drivers/input/keyboard/matrix_keypad.c +++ b/drivers/input/keyboard/matrix_keypad.c @@ -374,7 +374,9 @@ static int __devinit matrix_keypad_probe(struct platform_device *pdev) input_dev->name = pdev->name; input_dev->id.bustype = BUS_HOST; input_dev->dev.parent = &pdev->dev; - input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_REP); + input_dev->evbit[0] = BIT_MASK(EV_KEY); + if (!pdata->no_autorepeat) + input_dev->evbit[0] |= BIT_MASK(EV_REP); input_dev->open = matrix_keypad_start; input_dev->close = matrix_keypad_stop; diff --git a/drivers/input/misc/pcspkr.c b/drivers/input/misc/pcspkr.c index ea4e1fd12651..f080dd31499b 100644 --- a/drivers/input/misc/pcspkr.c +++ b/drivers/input/misc/pcspkr.c @@ -30,7 +30,7 @@ MODULE_ALIAS("platform:pcspkr"); #include <asm/i8253.h> #else #include <asm/8253pit.h> -static DEFINE_SPINLOCK(i8253_lock); +static DEFINE_RAW_SPINLOCK(i8253_lock); #endif static int pcspkr_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) @@ -50,7 +50,7 @@ static int pcspkr_event(struct input_dev *dev, unsigned int type, unsigned int c if (value > 20 && value < 32767) count = PIT_TICK_RATE / value; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); if (count) { /* set command for counter 2, 2 byte write */ @@ -65,7 +65,7 @@ static int pcspkr_event(struct input_dev *dev, unsigned int type, unsigned int c outb(inb_p(0x61) & 0xFC, 0x61); } - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); return 0; } diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c index 99d58764ef03..0d22cb9ce42e 100644 --- a/drivers/input/mouse/alps.c +++ b/drivers/input/mouse/alps.c @@ -64,6 +64,7 @@ static const struct alps_model_info alps_model_data[] = { { { 0x62, 0x02, 0x14 }, 0xcf, 0xcf, ALPS_PASS | ALPS_DUALPOINT | ALPS_PS2_INTERLEAVED }, { { 0x73, 0x02, 0x50 }, 0xcf, 0xcf, ALPS_FOUR_BUTTONS }, /* Dell Vostro 1400 */ + { { 0x73, 0x02, 0x64 }, 0xf8, 0xf8, 0 }, /* HP Pavilion dm3 */ { { 0x52, 0x01, 0x14 }, 0xff, 0xff, ALPS_PASS | ALPS_DUALPOINT | ALPS_PS2_INTERLEAVED }, /* Toshiba Tecra A11-11L */ }; diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c index 4f8fe0886b2a..b89879bd860f 100644 --- a/drivers/input/mouse/bcm5974.c +++ b/drivers/input/mouse/bcm5974.c @@ -803,7 +803,6 @@ static struct usb_driver bcm5974_driver = { .disconnect = bcm5974_disconnect, .suspend = bcm5974_suspend, .resume = bcm5974_resume, - .reset_resume = bcm5974_resume, .id_table = bcm5974_table, .supports_autosuspend = 1, }; diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 577688b5b951..6440a8f55686 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -39,7 +39,7 @@ MODULE_PARM_DESC(noaux, "Do not probe or use AUX (mouse) port."); static bool i8042_nomux; module_param_named(nomux, i8042_nomux, bool, 0); -MODULE_PARM_DESC(nomux, "Do not check whether an active multiplexing conrtoller is present."); +MODULE_PARM_DESC(nomux, "Do not check whether an active multiplexing controller is present."); static bool i8042_unlock; module_param_named(unlock, i8042_unlock, bool, 0); diff --git a/drivers/input/sparse-keymap.c b/drivers/input/sparse-keymap.c index 82ae18d29685..014248344763 100644 --- a/drivers/input/sparse-keymap.c +++ b/drivers/input/sparse-keymap.c @@ -68,12 +68,14 @@ static int sparse_keymap_getkeycode(struct input_dev *dev, unsigned int scancode, unsigned int *keycode) { - const struct key_entry *key = - sparse_keymap_entry_from_scancode(dev, scancode); + const struct key_entry *key; - if (key && key->type == KE_KEY) { - *keycode = key->keycode; - return 0; + if (dev->keycode) { + key = sparse_keymap_entry_from_scancode(dev, scancode); + if (key && key->type == KE_KEY) { + *keycode = key->keycode; + return 0; + } } return -EINVAL; @@ -86,17 +88,16 @@ static int sparse_keymap_setkeycode(struct input_dev *dev, struct key_entry *key; int old_keycode; - if (keycode < 0 || keycode > KEY_MAX) - return -EINVAL; - - key = sparse_keymap_entry_from_scancode(dev, scancode); - if (key && key->type == KE_KEY) { - old_keycode = key->keycode; - key->keycode = keycode; - set_bit(keycode, dev->keybit); - if (!sparse_keymap_entry_from_keycode(dev, old_keycode)) - clear_bit(old_keycode, dev->keybit); - return 0; + if (dev->keycode) { + key = sparse_keymap_entry_from_scancode(dev, scancode); + if (key && key->type == KE_KEY) { + old_keycode = key->keycode; + key->keycode = keycode; + set_bit(keycode, dev->keybit); + if (!sparse_keymap_entry_from_keycode(dev, old_keycode)) + clear_bit(old_keycode, dev->keybit); + return 0; + } } return -EINVAL; @@ -164,7 +165,7 @@ int sparse_keymap_setup(struct input_dev *dev, return 0; err_out: - kfree(keymap); + kfree(map); return error; } @@ -176,14 +177,27 @@ EXPORT_SYMBOL(sparse_keymap_setup); * * This function is used to free memory allocated by sparse keymap * in an input device that was set up by sparse_keymap_setup(). + * NOTE: It is safe to cal this function while input device is + * still registered (however the drivers should care not to try to + * use freed keymap and thus have to shut off interrups/polling + * before freeing the keymap). */ void sparse_keymap_free(struct input_dev *dev) { + unsigned long flags; + + /* + * Take event lock to prevent racing with input_get_keycode() + * and input_set_keycode() if we are called while input device + * is still registered. + */ + spin_lock_irqsave(&dev->event_lock, flags); + kfree(dev->keycode); dev->keycode = NULL; dev->keycodemax = 0; - dev->getkeycode = NULL; - dev->setkeycode = NULL; + + spin_unlock_irqrestore(&dev->event_lock, flags); } EXPORT_SYMBOL(sparse_keymap_free); diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c index 8b5d2873f0c4..f46502589e4e 100644 --- a/drivers/input/tablet/wacom_sys.c +++ b/drivers/input/tablet/wacom_sys.c @@ -673,13 +673,15 @@ static int wacom_resume(struct usb_interface *intf) int rv; mutex_lock(&wacom->lock); - if (wacom->open) { + + /* switch to wacom mode first */ + wacom_query_tablet_data(intf, features); + + if (wacom->open) rv = usb_submit_urb(wacom->irq, GFP_NOIO); - /* switch to wacom mode if needed */ - if (!wacom_retrieve_hid_descriptor(intf, features)) - wacom_query_tablet_data(intf, features); - } else + else rv = 0; + mutex_unlock(&wacom->lock); return rv; diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index b3ba3437a2eb..4a852d815c68 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -155,19 +155,19 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) { struct wacom_features *features = &wacom->features; unsigned char *data = wacom->data; - int x, y, prox; - int rw = 0; - int retval = 0; + int x, y, rw; + static int penData = 0; if (data[0] != WACOM_REPORT_PENABLED) { dbg("wacom_graphire_irq: received unknown report #%d", data[0]); - goto exit; + return 0; } - prox = data[1] & 0x80; - if (prox || wacom->id[0]) { - if (prox) { - switch ((data[1] >> 5) & 3) { + if (data[1] & 0x80) { + /* in prox and not a pad data */ + penData = 1; + + switch ((data[1] >> 5) & 3) { case 0: /* Pen */ wacom->tool[0] = BTN_TOOL_PEN; @@ -181,13 +181,23 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) case 2: /* Mouse with wheel */ wacom_report_key(wcombo, BTN_MIDDLE, data[1] & 0x04); + if (features->type == WACOM_G4 || features->type == WACOM_MO) { + rw = data[7] & 0x04 ? (data[7] & 0x03)-4 : (data[7] & 0x03); + wacom_report_rel(wcombo, REL_WHEEL, -rw); + } else + wacom_report_rel(wcombo, REL_WHEEL, -(signed char) data[6]); /* fall through */ case 3: /* Mouse without wheel */ wacom->tool[0] = BTN_TOOL_MOUSE; wacom->id[0] = CURSOR_DEVICE_ID; + wacom_report_key(wcombo, BTN_LEFT, data[1] & 0x01); + wacom_report_key(wcombo, BTN_RIGHT, data[1] & 0x02); + if (features->type == WACOM_G4 || features->type == WACOM_MO) + wacom_report_abs(wcombo, ABS_DISTANCE, data[6] & 0x3f); + else + wacom_report_abs(wcombo, ABS_DISTANCE, data[7] & 0x3f); break; - } } x = wacom_le16_to_cpu(&data[2]); y = wacom_le16_to_cpu(&data[4]); @@ -198,32 +208,36 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x01); wacom_report_key(wcombo, BTN_STYLUS, data[1] & 0x02); wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 0x04); - } else { - wacom_report_key(wcombo, BTN_LEFT, data[1] & 0x01); - wacom_report_key(wcombo, BTN_RIGHT, data[1] & 0x02); - if (features->type == WACOM_G4 || - features->type == WACOM_MO) { - wacom_report_abs(wcombo, ABS_DISTANCE, data[6] & 0x3f); - rw = (signed)(data[7] & 0x04) - (data[7] & 0x03); - } else { - wacom_report_abs(wcombo, ABS_DISTANCE, data[7] & 0x3f); - rw = -(signed)data[6]; - } - wacom_report_rel(wcombo, REL_WHEEL, rw); } - - if (!prox) - wacom->id[0] = 0; wacom_report_abs(wcombo, ABS_MISC, wacom->id[0]); /* report tool id */ - wacom_report_key(wcombo, wacom->tool[0], prox); - wacom_input_sync(wcombo); /* sync last event */ + wacom_report_key(wcombo, wacom->tool[0], 1); + } else if (wacom->id[0]) { + wacom_report_abs(wcombo, ABS_X, 0); + wacom_report_abs(wcombo, ABS_Y, 0); + if (wacom->tool[0] == BTN_TOOL_MOUSE) { + wacom_report_key(wcombo, BTN_LEFT, 0); + wacom_report_key(wcombo, BTN_RIGHT, 0); + wacom_report_abs(wcombo, ABS_DISTANCE, 0); + } else { + wacom_report_abs(wcombo, ABS_PRESSURE, 0); + wacom_report_key(wcombo, BTN_TOUCH, 0); + wacom_report_key(wcombo, BTN_STYLUS, 0); + wacom_report_key(wcombo, BTN_STYLUS2, 0); + } + wacom->id[0] = 0; + wacom_report_abs(wcombo, ABS_MISC, 0); /* reset tool id */ + wacom_report_key(wcombo, wacom->tool[0], 0); } /* send pad data */ switch (features->type) { case WACOM_G4: - prox = data[7] & 0xf8; - if (prox || wacom->id[1]) { + if (data[7] & 0xf8) { + if (penData) { + wacom_input_sync(wcombo); /* sync last event */ + if (!wacom->id[0]) + penData = 0; + } wacom->id[1] = PAD_DEVICE_ID; wacom_report_key(wcombo, BTN_0, (data[7] & 0x40)); wacom_report_key(wcombo, BTN_4, (data[7] & 0x80)); @@ -231,16 +245,29 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) wacom_report_rel(wcombo, REL_WHEEL, rw); wacom_report_key(wcombo, BTN_TOOL_FINGER, 0xf0); wacom_report_abs(wcombo, ABS_MISC, wacom->id[1]); - if (!prox) - wacom->id[1] = 0; - wacom_report_abs(wcombo, ABS_MISC, wacom->id[1]); + wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0); + } else if (wacom->id[1]) { + if (penData) { + wacom_input_sync(wcombo); /* sync last event */ + if (!wacom->id[0]) + penData = 0; + } + wacom->id[1] = 0; + wacom_report_key(wcombo, BTN_0, (data[7] & 0x40)); + wacom_report_key(wcombo, BTN_4, (data[7] & 0x80)); + wacom_report_rel(wcombo, REL_WHEEL, 0); + wacom_report_key(wcombo, BTN_TOOL_FINGER, 0); + wacom_report_abs(wcombo, ABS_MISC, 0); wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0); } - retval = 1; break; case WACOM_MO: - prox = (data[7] & 0xf8) || data[8]; - if (prox || wacom->id[1]) { + if ((data[7] & 0xf8) || (data[8] & 0xff)) { + if (penData) { + wacom_input_sync(wcombo); /* sync last event */ + if (!wacom->id[0]) + penData = 0; + } wacom->id[1] = PAD_DEVICE_ID; wacom_report_key(wcombo, BTN_0, (data[7] & 0x08)); wacom_report_key(wcombo, BTN_1, (data[7] & 0x20)); @@ -248,16 +275,27 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) wacom_report_key(wcombo, BTN_5, (data[7] & 0x40)); wacom_report_abs(wcombo, ABS_WHEEL, (data[8] & 0x7f)); wacom_report_key(wcombo, BTN_TOOL_FINGER, 0xf0); - if (!prox) - wacom->id[1] = 0; wacom_report_abs(wcombo, ABS_MISC, wacom->id[1]); wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0); + } else if (wacom->id[1]) { + if (penData) { + wacom_input_sync(wcombo); /* sync last event */ + if (!wacom->id[0]) + penData = 0; + } + wacom->id[1] = 0; + wacom_report_key(wcombo, BTN_0, (data[7] & 0x08)); + wacom_report_key(wcombo, BTN_1, (data[7] & 0x20)); + wacom_report_key(wcombo, BTN_4, (data[7] & 0x10)); + wacom_report_key(wcombo, BTN_5, (data[7] & 0x40)); + wacom_report_abs(wcombo, ABS_WHEEL, (data[8] & 0x7f)); + wacom_report_key(wcombo, BTN_TOOL_FINGER, 0); + wacom_report_abs(wcombo, ABS_MISC, 0); + wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0); } - retval = 1; break; } -exit: - return retval; + return 1; } static int wacom_intuos_inout(struct wacom_wac *wacom, void *wcombo) @@ -598,9 +636,9 @@ static int wacom_intuos_irq(struct wacom_wac *wacom, void *wcombo) static void wacom_tpc_finger_in(struct wacom_wac *wacom, void *wcombo, char *data, int idx) { wacom_report_abs(wcombo, ABS_X, - data[2 + idx * 2] | ((data[3 + idx * 2] & 0x7f) << 8)); + (data[2 + idx * 2] & 0xff) | ((data[3 + idx * 2] & 0x7f) << 8)); wacom_report_abs(wcombo, ABS_Y, - data[6 + idx * 2] | ((data[7 + idx * 2] & 0x7f) << 8)); + (data[6 + idx * 2] & 0xff) | ((data[7 + idx * 2] & 0x7f) << 8)); wacom_report_abs(wcombo, ABS_MISC, wacom->id[0]); wacom_report_key(wcombo, wacom->tool[idx], 1); if (idx) @@ -744,24 +782,31 @@ static int wacom_tpc_irq(struct wacom_wac *wacom, void *wcombo) touchInProx = 0; - if (!wacom->id[0]) { /* first in prox */ - /* Going into proximity select tool */ - wacom->tool[0] = (data[1] & 0x0c) ? BTN_TOOL_RUBBER : BTN_TOOL_PEN; - if (wacom->tool[0] == BTN_TOOL_PEN) - wacom->id[0] = STYLUS_DEVICE_ID; - else - wacom->id[0] = ERASER_DEVICE_ID; - } - wacom_report_key(wcombo, BTN_STYLUS, data[1] & 0x02); - wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 0x10); - wacom_report_abs(wcombo, ABS_X, wacom_le16_to_cpu(&data[2])); - wacom_report_abs(wcombo, ABS_Y, wacom_le16_to_cpu(&data[4])); - pressure = ((data[7] & 0x01) << 8) | data[6]; - if (pressure < 0) - pressure = features->pressure_max + pressure + 1; - wacom_report_abs(wcombo, ABS_PRESSURE, pressure); - wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x05); - if (!prox) { /* out-prox */ + if (prox) { /* in prox */ + if (!wacom->id[0]) { + /* Going into proximity select tool */ + wacom->tool[0] = (data[1] & 0x0c) ? BTN_TOOL_RUBBER : BTN_TOOL_PEN; + if (wacom->tool[0] == BTN_TOOL_PEN) + wacom->id[0] = STYLUS_DEVICE_ID; + else + wacom->id[0] = ERASER_DEVICE_ID; + } + wacom_report_key(wcombo, BTN_STYLUS, data[1] & 0x02); + wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 0x10); + wacom_report_abs(wcombo, ABS_X, wacom_le16_to_cpu(&data[2])); + wacom_report_abs(wcombo, ABS_Y, wacom_le16_to_cpu(&data[4])); + pressure = ((data[7] & 0x01) << 8) | data[6]; + if (pressure < 0) + pressure = features->pressure_max + pressure + 1; + wacom_report_abs(wcombo, ABS_PRESSURE, pressure); + wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x05); + } else { + wacom_report_abs(wcombo, ABS_X, 0); + wacom_report_abs(wcombo, ABS_Y, 0); + wacom_report_abs(wcombo, ABS_PRESSURE, 0); + wacom_report_key(wcombo, BTN_STYLUS, 0); + wacom_report_key(wcombo, BTN_STYLUS2, 0); + wacom_report_key(wcombo, BTN_TOUCH, 0); wacom->id[0] = 0; /* pen is out so touch can be enabled now */ touchInProx = 1; diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c index 07090f379c63..69c84a1d88ea 100644 --- a/drivers/lguest/lguest_device.c +++ b/drivers/lguest/lguest_device.c @@ -178,7 +178,7 @@ static void set_status(struct virtio_device *vdev, u8 status) /* We set the status. */ to_lgdev(vdev)->desc->status = status; - kvm_hypercall1(LHCALL_NOTIFY, (max_pfn << PAGE_SHIFT) + offset); + hcall(LHCALL_NOTIFY, (max_pfn << PAGE_SHIFT) + offset, 0, 0, 0); } static void lg_set_status(struct virtio_device *vdev, u8 status) @@ -229,7 +229,7 @@ static void lg_notify(struct virtqueue *vq) */ struct lguest_vq_info *lvq = vq->priv; - kvm_hypercall1(LHCALL_NOTIFY, lvq->config.pfn << PAGE_SHIFT); + hcall(LHCALL_NOTIFY, lvq->config.pfn << PAGE_SHIFT, 0, 0, 0); } /* An extern declaration inside a C file is bad form. Don't do it. */ diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c index fb2b7ef7868e..b4eb675a807e 100644 --- a/drivers/lguest/x86/core.c +++ b/drivers/lguest/x86/core.c @@ -288,6 +288,18 @@ static int emulate_insn(struct lg_cpu *cpu) insn = lgread(cpu, physaddr, u8); /* + * Around 2.6.33, the kernel started using an emulation for the + * cmpxchg8b instruction in early boot on many configurations. This + * code isn't paravirtualized, and it tries to disable interrupts. + * Ignore it, which will Mostly Work. + */ + if (insn == 0xfa) { + /* "cli", or Clear Interrupt Enable instruction. Skip it. */ + cpu->regs->eip++; + return 1; + } + + /* * 0x66 is an "operand prefix". It means it's using the upper 16 bits * of the eax register. */ diff --git a/drivers/net/cnic.c b/drivers/net/cnic.c index 9781942992e9..4b451a7c03e9 100644 --- a/drivers/net/cnic.c +++ b/drivers/net/cnic.c @@ -2334,13 +2334,13 @@ static int cnic_service_bnx2x(void *data, void *status_blk) struct cnic_local *cp = dev->cnic_priv; u16 prod = cp->kcq_prod_idx & MAX_KCQ_IDX; - prefetch(cp->status_blk.bnx2x); - prefetch(&cp->kcq[KCQ_PG(prod)][KCQ_IDX(prod)]); + if (likely(test_bit(CNIC_F_CNIC_UP, &dev->flags))) { + prefetch(cp->status_blk.bnx2x); + prefetch(&cp->kcq[KCQ_PG(prod)][KCQ_IDX(prod)]); - if (likely(test_bit(CNIC_F_CNIC_UP, &dev->flags))) tasklet_schedule(&cp->cnic_irq_task); - - cnic_chk_pkt_rings(cp); + cnic_chk_pkt_rings(cp); + } return 0; } diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index cfd09cea7214..73d43c53015a 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -661,6 +661,8 @@ static bool e1000_clean_tx_irq(struct e1000_adapter *adapter) i = 0; } + if (i == tx_ring->next_to_use) + break; eop = tx_ring->buffer_info[i].next_to_watch; eop_desc = E1000_TX_DESC(*tx_ring, eop); } diff --git a/drivers/net/igb/igb_ethtool.c b/drivers/net/igb/igb_ethtool.c index d313fae992da..743038490104 100644 --- a/drivers/net/igb/igb_ethtool.c +++ b/drivers/net/igb/igb_ethtool.c @@ -1814,6 +1814,7 @@ static int igb_wol_exclusion(struct igb_adapter *adapter, retval = 0; break; case E1000_DEV_ID_82576_QUAD_COPPER: + case E1000_DEV_ID_82576_QUAD_COPPER_ET2: /* quad port adapters only support WoL on port A */ if (!(adapter->flags & IGB_FLAG_QUAD_PORT_A)) { wol->supported = 0; diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index 9b3c51ab1758..c9baa2aa98cd 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -1612,6 +1612,7 @@ static int __devinit igb_probe(struct pci_dev *pdev, adapter->eeprom_wol = 0; break; case E1000_DEV_ID_82576_QUAD_COPPER: + case E1000_DEV_ID_82576_QUAD_COPPER_ET2: /* if quad port adapter, disable WoL on all but port A */ if (global_quad_port_a != 0) adapter->eeprom_wol = 0; diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 471887742b02..ecde0876a785 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -1690,7 +1690,7 @@ myri10ge_set_pauseparam(struct net_device *netdev, if (pause->tx_pause != mgp->pause) return myri10ge_change_pause(mgp, pause->tx_pause); if (pause->rx_pause != mgp->pause) - return myri10ge_change_pause(mgp, pause->tx_pause); + return myri10ge_change_pause(mgp, pause->rx_pause); if (pause->autoneg != 0) return -EINVAL; return 0; diff --git a/drivers/net/pcmcia/smc91c92_cs.c b/drivers/net/pcmcia/smc91c92_cs.c index ff7eb9116b6a..fd9d6e34fda4 100644 --- a/drivers/net/pcmcia/smc91c92_cs.c +++ b/drivers/net/pcmcia/smc91c92_cs.c @@ -1608,9 +1608,12 @@ static void set_rx_mode(struct net_device *dev) { unsigned int ioaddr = dev->base_addr; struct smc_private *smc = netdev_priv(dev); - u_int multicast_table[ 2 ] = { 0, }; + unsigned char multicast_table[8]; unsigned long flags; u_short rx_cfg_setting; + int i; + + memset(multicast_table, 0, sizeof(multicast_table)); if (dev->flags & IFF_PROMISC) { rx_cfg_setting = RxStripCRC | RxEnable | RxPromisc | RxAllMulti; @@ -1622,10 +1625,6 @@ static void set_rx_mode(struct net_device *dev) netdev_for_each_mc_addr(mc_addr, dev) { u_int position = ether_crc(6, mc_addr->dmi_addr); -#ifndef final_version /* Verify multicast address. */ - if ((mc_addr->dmi_addr[0] & 1) == 0) - continue; -#endif multicast_table[position >> 29] |= 1 << ((position >> 26) & 7); } } @@ -1635,8 +1634,8 @@ static void set_rx_mode(struct net_device *dev) /* Load MC table and Rx setting into the chip without interrupts. */ spin_lock_irqsave(&smc->lock, flags); SMC_SELECT_BANK(3); - outl(multicast_table[0], ioaddr + MULTICAST0); - outl(multicast_table[1], ioaddr + MULTICAST4); + for (i = 0; i < 8; i++) + outb(multicast_table[i], ioaddr + MULTICAST0 + i); SMC_SELECT_BANK(0); outw(rx_cfg_setting, ioaddr + RCR); SMC_SELECT_BANK(2); diff --git a/drivers/net/qlcnic/qlcnic_hw.c b/drivers/net/qlcnic/qlcnic_hw.c index a6ef266a2fe2..e73ba455aa20 100644 --- a/drivers/net/qlcnic/qlcnic_hw.c +++ b/drivers/net/qlcnic/qlcnic_hw.c @@ -431,6 +431,9 @@ void qlcnic_set_multi(struct net_device *netdev) u8 bcast_addr[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }; u32 mode = VPORT_MISS_MODE_DROP; + if (adapter->is_up != QLCNIC_ADAPTER_UP_MAGIC) + return; + qlcnic_nic_add_mac(adapter, adapter->mac_addr); qlcnic_nic_add_mac(adapter, bcast_addr); diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c index 43afdb6b25e6..0298d8c1dcb6 100644 --- a/drivers/net/r6040.c +++ b/drivers/net/r6040.c @@ -134,7 +134,7 @@ #define RX_DESC_SIZE (RX_DCNT * sizeof(struct r6040_descriptor)) #define TX_DESC_SIZE (TX_DCNT * sizeof(struct r6040_descriptor)) #define MBCR_DEFAULT 0x012A /* MAC Bus Control Register */ -#define MCAST_MAX 4 /* Max number multicast addresses to filter */ +#define MCAST_MAX 3 /* Max number multicast addresses to filter */ /* Descriptor status */ #define DSC_OWNER_MAC 0x8000 /* MAC is the owner of this descriptor */ @@ -982,9 +982,6 @@ static void r6040_multicast_list(struct net_device *dev) crc >>= 26; hash_table[crc >> 4] |= 1 << (15 - (crc & 0xf)); } - /* Write the index of the hash table */ - for (i = 0; i < 4; i++) - iowrite16(hash_table[i] << 14, ioaddr + MCR1); /* Fill the MAC hash tables with their values */ iowrite16(hash_table[0], ioaddr + MAR0); iowrite16(hash_table[1], ioaddr + MAR1); @@ -1000,9 +997,9 @@ static void r6040_multicast_list(struct net_device *dev) iowrite16(adrp[1], ioaddr + MID_1M + 8 * i); iowrite16(adrp[2], ioaddr + MID_1H + 8 * i); } else { - iowrite16(0xffff, ioaddr + MID_0L + 8 * i); - iowrite16(0xffff, ioaddr + MID_0M + 8 * i); - iowrite16(0xffff, ioaddr + MID_0H + 8 * i); + iowrite16(0xffff, ioaddr + MID_1L + 8 * i); + iowrite16(0xffff, ioaddr + MID_1M + 8 * i); + iowrite16(0xffff, ioaddr + MID_1H + 8 * i); } i++; } diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c index a214a1627e8b..4111a85ec80e 100644 --- a/drivers/net/stmmac/stmmac_main.c +++ b/drivers/net/stmmac/stmmac_main.c @@ -1686,7 +1686,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev) } pr_info("done!\n"); - if (!request_mem_region(res->start, (res->end - res->start), + if (!request_mem_region(res->start, resource_size(res), pdev->name)) { pr_err("%s: ERROR: memory allocation failed" "cannot get the I/O addr 0x%x\n", @@ -1695,9 +1695,9 @@ static int stmmac_dvr_probe(struct platform_device *pdev) goto out; } - addr = ioremap(res->start, (res->end - res->start)); + addr = ioremap(res->start, resource_size(res)); if (!addr) { - pr_err("%s: ERROR: memory mapping failed \n", __func__); + pr_err("%s: ERROR: memory mapping failed\n", __func__); ret = -ENOMEM; goto out; } @@ -1775,7 +1775,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev) out: if (ret < 0) { platform_set_drvdata(pdev, NULL); - release_mem_region(res->start, (res->end - res->start)); + release_mem_region(res->start, resource_size(res)); if (addr != NULL) iounmap(addr); } @@ -1813,7 +1813,7 @@ static int stmmac_dvr_remove(struct platform_device *pdev) iounmap((void *)ndev->base_addr); res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - release_mem_region(res->start, (res->end - res->start)); + release_mem_region(res->start, resource_size(res)); free_netdev(ndev); diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 6fb783ce20b9..b0577dd1a42d 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -327,6 +327,7 @@ static int add_recvbuf_small(struct virtnet_info *vi, gfp_t gfp) struct scatterlist sg[2]; int err; + sg_init_table(sg, 2); skb = netdev_alloc_skb_ip_align(vi->dev, MAX_PACKET_LEN); if (unlikely(!skb)) return -ENOMEM; @@ -352,6 +353,7 @@ static int add_recvbuf_big(struct virtnet_info *vi, gfp_t gfp) char *p; int i, err, offset; + sg_init_table(sg, MAX_SKB_FRAGS + 2); /* page in sg[MAX_SKB_FRAGS + 1] is list tail */ for (i = MAX_SKB_FRAGS + 1; i > 1; --i) { first = get_a_page(vi, gfp); diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 67ca4e5a6017..115e1aeedb59 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -1532,8 +1532,7 @@ static int ath9k_config(struct ieee80211_hw *hw, u32 changed) all_wiphys_idle = ath9k_all_wiphys_idle(sc); ath9k_set_wiphy_idle(aphy, idle); - if (!idle && all_wiphys_idle) - enable_radio = true; + enable_radio = (!idle && all_wiphys_idle); /* * After we unlock here its possible another wiphy diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c index 83c52a682622..8972166386cb 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965.c +++ b/drivers/net/wireless/iwlwifi/iwl-4965.c @@ -2015,7 +2015,9 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv, IWL_DEBUG_TX_REPLY(priv, "Retry scheduler reclaim scd_ssn " "%d index %d\n", scd_ssn , index); freed = iwl_tx_queue_reclaim(priv, txq_id, index); - iwl_free_tfds_in_queue(priv, sta_id, tid, freed); + if (qc) + iwl_free_tfds_in_queue(priv, sta_id, + tid, freed); if (priv->mac80211_registered && (iwl_queue_space(&txq->q) > txq->q.low_mark) && @@ -2041,14 +2043,17 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv, tx_resp->failure_frame); freed = iwl_tx_queue_reclaim(priv, txq_id, index); - iwl_free_tfds_in_queue(priv, sta_id, tid, freed); + if (qc && likely(sta_id != IWL_INVALID_STATION)) + iwl_free_tfds_in_queue(priv, sta_id, tid, freed); + else if (sta_id == IWL_INVALID_STATION) + IWL_DEBUG_TX_REPLY(priv, "Station not known\n"); if (priv->mac80211_registered && (iwl_queue_space(&txq->q) > txq->q.low_mark)) iwl_wake_queue(priv, txq_id); } - - iwl_txq_check_empty(priv, sta_id, tid, txq_id); + if (qc && likely(sta_id != IWL_INVALID_STATION)) + iwl_txq_check_empty(priv, sta_id, tid, txq_id); if (iwl_check_bits(status, TX_ABORT_REQUIRED_MSK)) IWL_ERR(priv, "TODO: Implement Tx ABORT REQUIRED!!!\n"); diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-rs.c b/drivers/net/wireless/iwlwifi/iwl-agn-rs.c index 35f819ac87a3..1460116d329f 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn-rs.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn-rs.c @@ -346,6 +346,17 @@ static inline int get_num_of_ant_from_rate(u32 rate_n_flags) !!(rate_n_flags & RATE_MCS_ANT_C_MSK); } +/* + * Static function to get the expected throughput from an iwl_scale_tbl_info + * that wraps a NULL pointer check + */ +static s32 get_expected_tpt(struct iwl_scale_tbl_info *tbl, int rs_index) +{ + if (tbl->expected_tpt) + return tbl->expected_tpt[rs_index]; + return 0; +} + /** * rs_collect_tx_data - Update the success/failure sliding window * @@ -353,19 +364,21 @@ static inline int get_num_of_ant_from_rate(u32 rate_n_flags) * at this rate. window->data contains the bitmask of successful * packets. */ -static int rs_collect_tx_data(struct iwl_rate_scale_data *windows, - int scale_index, s32 tpt, int attempts, - int successes) +static int rs_collect_tx_data(struct iwl_scale_tbl_info *tbl, + int scale_index, int attempts, int successes) { struct iwl_rate_scale_data *window = NULL; static const u64 mask = (((u64)1) << (IWL_RATE_MAX_WINDOW - 1)); - s32 fail_count; + s32 fail_count, tpt; if (scale_index < 0 || scale_index >= IWL_RATE_COUNT) return -EINVAL; /* Select window for current tx bit rate */ - window = &(windows[scale_index]); + window = &(tbl->win[scale_index]); + + /* Get expected throughput */ + tpt = get_expected_tpt(tbl, scale_index); /* * Keep track of only the latest 62 tx frame attempts in this rate's @@ -739,16 +752,6 @@ static bool table_type_matches(struct iwl_scale_tbl_info *a, return (a->lq_type == b->lq_type) && (a->ant_type == b->ant_type) && (a->is_SGI == b->is_SGI); } -/* - * Static function to get the expected throughput from an iwl_scale_tbl_info - * that wraps a NULL pointer check - */ -static s32 get_expected_tpt(struct iwl_scale_tbl_info *tbl, int rs_index) -{ - if (tbl->expected_tpt) - return tbl->expected_tpt[rs_index]; - return 0; -} /* * mac80211 sends us Tx status @@ -765,12 +768,10 @@ static void rs_tx_status(void *priv_r, struct ieee80211_supported_band *sband, struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; struct iwl_priv *priv = (struct iwl_priv *)priv_r; struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - struct iwl_rate_scale_data *window = NULL; enum mac80211_rate_control_flags mac_flags; u32 tx_rate; struct iwl_scale_tbl_info tbl_type; - struct iwl_scale_tbl_info *curr_tbl, *other_tbl; - s32 tpt = 0; + struct iwl_scale_tbl_info *curr_tbl, *other_tbl, *tmp_tbl; IWL_DEBUG_RATE_LIMIT(priv, "get frame ack response, update rate scale window\n"); @@ -853,7 +854,6 @@ static void rs_tx_status(void *priv_r, struct ieee80211_supported_band *sband, IWL_DEBUG_RATE(priv, "Neither active nor search matches tx rate\n"); return; } - window = (struct iwl_rate_scale_data *)&(curr_tbl->win[0]); /* * Updating the frame history depends on whether packets were @@ -866,8 +866,7 @@ static void rs_tx_status(void *priv_r, struct ieee80211_supported_band *sband, tx_rate = le32_to_cpu(table->rs_table[0].rate_n_flags); rs_get_tbl_info_from_mcs(tx_rate, priv->band, &tbl_type, &rs_index); - tpt = get_expected_tpt(curr_tbl, rs_index); - rs_collect_tx_data(window, rs_index, tpt, + rs_collect_tx_data(curr_tbl, rs_index, info->status.ampdu_ack_len, info->status.ampdu_ack_map); @@ -897,19 +896,13 @@ static void rs_tx_status(void *priv_r, struct ieee80211_supported_band *sband, * table as active/search. */ if (table_type_matches(&tbl_type, curr_tbl)) - tpt = get_expected_tpt(curr_tbl, rs_index); + tmp_tbl = curr_tbl; else if (table_type_matches(&tbl_type, other_tbl)) - tpt = get_expected_tpt(other_tbl, rs_index); + tmp_tbl = other_tbl; else continue; - - /* Constants mean 1 transmission, 0 successes */ - if (i < retries) - rs_collect_tx_data(window, rs_index, tpt, 1, - 0); - else - rs_collect_tx_data(window, rs_index, tpt, 1, - legacy_success); + rs_collect_tx_data(tmp_tbl, rs_index, 1, + i < retries ? 0 : legacy_success); } /* Update success/fail counts if not searching for new mode */ diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c index db050b811232..3352f7086632 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.c +++ b/drivers/net/wireless/iwlwifi/iwl-core.c @@ -308,10 +308,13 @@ int iwl_hw_nic_init(struct iwl_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); - /* Allocate and init all Tx and Command queues */ - ret = iwl_txq_ctx_reset(priv); - if (ret) - return ret; + /* Allocate or reset and init all Tx and Command queues */ + if (!priv->txq) { + ret = iwl_txq_ctx_alloc(priv); + if (ret) + return ret; + } else + iwl_txq_ctx_reset(priv); set_bit(STATUS_INIT, &priv->status); diff --git a/drivers/net/wireless/iwlwifi/iwl-core.h b/drivers/net/wireless/iwlwifi/iwl-core.h index 4ef7739f9e8e..732590f5fe30 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.h +++ b/drivers/net/wireless/iwlwifi/iwl-core.h @@ -442,7 +442,8 @@ void iwl_rx_csa(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb); /***************************************************** * TX ******************************************************/ -int iwl_txq_ctx_reset(struct iwl_priv *priv); +int iwl_txq_ctx_alloc(struct iwl_priv *priv); +void iwl_txq_ctx_reset(struct iwl_priv *priv); void iwl_hw_txq_free_tfd(struct iwl_priv *priv, struct iwl_tx_queue *txq); int iwl_hw_txq_attach_buf_to_tfd(struct iwl_priv *priv, struct iwl_tx_queue *txq, @@ -456,6 +457,8 @@ void iwl_free_tfds_in_queue(struct iwl_priv *priv, void iwl_txq_update_write_ptr(struct iwl_priv *priv, struct iwl_tx_queue *txq); int iwl_tx_queue_init(struct iwl_priv *priv, struct iwl_tx_queue *txq, int slots_num, u32 txq_id); +void iwl_tx_queue_reset(struct iwl_priv *priv, struct iwl_tx_queue *txq, + int slots_num, u32 txq_id); void iwl_tx_queue_free(struct iwl_priv *priv, int txq_id); int iwl_tx_agg_start(struct iwl_priv *priv, const u8 *ra, u16 tid, u16 *ssn); int iwl_tx_agg_stop(struct iwl_priv *priv , const u8 *ra, u16 tid); diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c index f0b7e6cfbe4f..8dd0c036d547 100644 --- a/drivers/net/wireless/iwlwifi/iwl-tx.c +++ b/drivers/net/wireless/iwlwifi/iwl-tx.c @@ -194,10 +194,34 @@ void iwl_cmd_queue_free(struct iwl_priv *priv) struct iwl_queue *q = &txq->q; struct device *dev = &priv->pci_dev->dev; int i; + bool huge = false; if (q->n_bd == 0) return; + for (; q->read_ptr != q->write_ptr; + q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) { + /* we have no way to tell if it is a huge cmd ATM */ + i = get_cmd_index(q, q->read_ptr, 0); + + if (txq->meta[i].flags & CMD_SIZE_HUGE) { + huge = true; + continue; + } + + pci_unmap_single(priv->pci_dev, + pci_unmap_addr(&txq->meta[i], mapping), + pci_unmap_len(&txq->meta[i], len), + PCI_DMA_BIDIRECTIONAL); + } + if (huge) { + i = q->n_window; + pci_unmap_single(priv->pci_dev, + pci_unmap_addr(&txq->meta[i], mapping), + pci_unmap_len(&txq->meta[i], len), + PCI_DMA_BIDIRECTIONAL); + } + /* De-alloc array of command/tx buffers */ for (i = 0; i <= TFD_CMD_SLOTS; i++) kfree(txq->cmd[i]); @@ -410,6 +434,26 @@ out_free_arrays: } EXPORT_SYMBOL(iwl_tx_queue_init); +void iwl_tx_queue_reset(struct iwl_priv *priv, struct iwl_tx_queue *txq, + int slots_num, u32 txq_id) +{ + int actual_slots = slots_num; + + if (txq_id == IWL_CMD_QUEUE_NUM) + actual_slots++; + + memset(txq->meta, 0, sizeof(struct iwl_cmd_meta) * actual_slots); + + txq->need_update = 0; + + /* Initialize queue's high/low-water marks, and head/tail indexes */ + iwl_queue_init(priv, &txq->q, TFD_QUEUE_SIZE_MAX, slots_num, txq_id); + + /* Tell device where to find queue */ + priv->cfg->ops->lib->txq_init(priv, txq); +} +EXPORT_SYMBOL(iwl_tx_queue_reset); + /** * iwl_hw_txq_ctx_free - Free TXQ Context * @@ -421,8 +465,7 @@ void iwl_hw_txq_ctx_free(struct iwl_priv *priv) /* Tx queues */ if (priv->txq) { - for (txq_id = 0; txq_id < priv->hw_params.max_txq_num; - txq_id++) + for (txq_id = 0; txq_id < priv->hw_params.max_txq_num; txq_id++) if (txq_id == IWL_CMD_QUEUE_NUM) iwl_cmd_queue_free(priv); else @@ -438,15 +481,15 @@ void iwl_hw_txq_ctx_free(struct iwl_priv *priv) EXPORT_SYMBOL(iwl_hw_txq_ctx_free); /** - * iwl_txq_ctx_reset - Reset TX queue context - * Destroys all DMA structures and initialize them again + * iwl_txq_ctx_alloc - allocate TX queue context + * Allocate all Tx DMA structures and initialize them * * @param priv * @return error code */ -int iwl_txq_ctx_reset(struct iwl_priv *priv) +int iwl_txq_ctx_alloc(struct iwl_priv *priv) { - int ret = 0; + int ret; int txq_id, slots_num; unsigned long flags; @@ -504,8 +547,31 @@ int iwl_txq_ctx_reset(struct iwl_priv *priv) return ret; } +void iwl_txq_ctx_reset(struct iwl_priv *priv) +{ + int txq_id, slots_num; + unsigned long flags; + + spin_lock_irqsave(&priv->lock, flags); + + /* Turn off all Tx DMA fifos */ + priv->cfg->ops->lib->txq_set_sched(priv, 0); + + /* Tell NIC where to find the "keep warm" buffer */ + iwl_write_direct32(priv, FH_KW_MEM_ADDR_REG, priv->kw.dma >> 4); + + spin_unlock_irqrestore(&priv->lock, flags); + + /* Alloc and init all Tx queues, including the command queue (#4) */ + for (txq_id = 0; txq_id < priv->hw_params.max_txq_num; txq_id++) { + slots_num = txq_id == IWL_CMD_QUEUE_NUM ? + TFD_CMD_SLOTS : TFD_TX_CMD_SLOTS; + iwl_tx_queue_reset(priv, &priv->txq[txq_id], slots_num, txq_id); + } +} + /** - * iwl_txq_ctx_stop - Stop all Tx DMA channels, free Tx queue memory + * iwl_txq_ctx_stop - Stop all Tx DMA channels */ void iwl_txq_ctx_stop(struct iwl_priv *priv) { @@ -525,9 +591,6 @@ void iwl_txq_ctx_stop(struct iwl_priv *priv) 1000); } spin_unlock_irqrestore(&priv->lock, flags); - - /* Deallocate memory for all Tx queues */ - iwl_hw_txq_ctx_free(priv); } EXPORT_SYMBOL(iwl_txq_ctx_stop); @@ -1050,6 +1113,14 @@ int iwl_enqueue_hcmd(struct iwl_priv *priv, struct iwl_host_cmd *cmd) spin_lock_irqsave(&priv->hcmd_lock, flags); + /* If this is a huge cmd, mark the huge flag also on the meta.flags + * of the _original_ cmd. This is used for DMA mapping clean up. + */ + if (cmd->flags & CMD_SIZE_HUGE) { + idx = get_cmd_index(q, q->write_ptr, 0); + txq->meta[idx].flags = CMD_SIZE_HUGE; + } + idx = get_cmd_index(q, q->write_ptr, cmd->flags & CMD_SIZE_HUGE); out_cmd = txq->cmd[idx]; out_meta = &txq->meta[idx]; @@ -1227,6 +1298,7 @@ void iwl_tx_cmd_complete(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb) bool huge = !!(pkt->hdr.sequence & SEQ_HUGE_FRAME); struct iwl_device_cmd *cmd; struct iwl_cmd_meta *meta; + struct iwl_tx_queue *txq = &priv->txq[IWL_CMD_QUEUE_NUM]; /* If a Tx command is being handled and it isn't in the actual * command queue then there a command routing bug has been introduced @@ -1240,9 +1312,17 @@ void iwl_tx_cmd_complete(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb) return; } - cmd_index = get_cmd_index(&priv->txq[IWL_CMD_QUEUE_NUM].q, index, huge); - cmd = priv->txq[IWL_CMD_QUEUE_NUM].cmd[cmd_index]; - meta = &priv->txq[IWL_CMD_QUEUE_NUM].meta[cmd_index]; + /* If this is a huge cmd, clear the huge flag on the meta.flags + * of the _original_ cmd. So that iwl_cmd_queue_free won't unmap + * the DMA buffer for the scan (huge) command. + */ + if (huge) { + cmd_index = get_cmd_index(&txq->q, index, 0); + txq->meta[cmd_index].flags = 0; + } + cmd_index = get_cmd_index(&txq->q, index, huge); + cmd = txq->cmd[cmd_index]; + meta = &txq->meta[cmd_index]; pci_unmap_single(priv->pci_dev, pci_unmap_addr(meta, mapping), @@ -1264,6 +1344,7 @@ void iwl_tx_cmd_complete(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb) get_cmd_string(cmd->hdr.cmd)); wake_up_interruptible(&priv->wait_command_queue); } + meta->flags = 0; } EXPORT_SYMBOL(iwl_tx_cmd_complete); diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index 166b67ea622f..7581dbe456da 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -186,14 +186,14 @@ int op_cpu_buffer_write_commit(struct op_entry *entry) struct op_sample *op_cpu_buffer_read_entry(struct op_entry *entry, int cpu) { struct ring_buffer_event *e; - e = ring_buffer_consume(op_ring_buffer_read, cpu, NULL); + e = ring_buffer_consume(op_ring_buffer_read, cpu, NULL, NULL); if (e) goto event; if (ring_buffer_swap_cpu(op_ring_buffer_read, op_ring_buffer_write, cpu)) return NULL; - e = ring_buffer_consume(op_ring_buffer_read, cpu, NULL); + e = ring_buffer_consume(op_ring_buffer_read, cpu, NULL, NULL); if (e) goto event; return NULL; diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 5be11c99e18f..e69d238c5af0 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -236,6 +236,10 @@ static int vq_memory_access_ok(void __user *log_base, struct vhost_memory *mem, int log_all) { int i; + + if (!mem) + return 0; + for (i = 0; i < mem->nregions; ++i) { struct vhost_memory_region *m = mem->regions + i; unsigned long a = m->userspace_addr; diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index 0e8468ffd100..0bf5020d0d32 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -194,10 +194,10 @@ config EP93XX_WATCHDOG config OMAP_WATCHDOG tristate "OMAP Watchdog" - depends on ARCH_OMAP16XX || ARCH_OMAP2 || ARCH_OMAP3 + depends on ARCH_OMAP16XX || ARCH_OMAP2PLUS help - Support for TI OMAP1610/OMAP1710/OMAP2420/OMAP3430 watchdog. Say 'Y' - here to enable the OMAP1610/OMAP1710/OMAP2420/OMAP3430 watchdog timer. + Support for TI OMAP1610/OMAP1710/OMAP2420/OMAP3430/OMAP4430 watchdog. Say 'Y' + here to enable the OMAP1610/OMAP1710/OMAP2420/OMAP3430/OMAP4430 watchdog timer. config PNX4008_WATCHDOG tristate "PNX4008 Watchdog" @@ -302,7 +302,7 @@ config TS72XX_WATCHDOG config MAX63XX_WATCHDOG tristate "Max63xx watchdog" - depends on ARM + depends on ARM && HAS_IOMEM help Support for memory mapped max63{69,70,71,72,73,74} watchdog timer. diff --git a/drivers/watchdog/booke_wdt.c b/drivers/watchdog/booke_wdt.c index 8b724aad6825..500d38342e1e 100644 --- a/drivers/watchdog/booke_wdt.c +++ b/drivers/watchdog/booke_wdt.c @@ -44,7 +44,7 @@ u32 booke_wdt_period = WDT_PERIOD_DEFAULT; #ifdef CONFIG_FSL_BOOKE #define WDTP(x) ((((x)&0x3)<<30)|(((x)&0x3c)<<15)) -#define WDTP_MASK (WDTP(0)) +#define WDTP_MASK (WDTP(0x3f)) #else #define WDTP(x) (TCR_WP(x)) #define WDTP_MASK (TCR_WP_MASK) diff --git a/drivers/watchdog/max63xx_wdt.c b/drivers/watchdog/max63xx_wdt.c index 75f3a83c0361..3053ff05ca41 100644 --- a/drivers/watchdog/max63xx_wdt.c +++ b/drivers/watchdog/max63xx_wdt.c @@ -154,9 +154,14 @@ static void max63xx_wdt_enable(struct max63xx_timeout *entry) static void max63xx_wdt_disable(void) { + u8 val; + spin_lock(&io_lock); - __raw_writeb(3, wdt_base); + val = __raw_readb(wdt_base); + val &= ~MAX6369_WDSET; + val |= 3; + __raw_writeb(val, wdt_base); spin_unlock(&io_lock); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index aa3cd7cc3e40..412593703d1e 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -337,16 +337,15 @@ out: /* * Get ref for the oldest snapc for an inode with dirty data... that is, the * only snap context we are allowed to write back. - * - * Caller holds i_lock. */ -static struct ceph_snap_context *__get_oldest_context(struct inode *inode, - u64 *snap_size) +static struct ceph_snap_context *get_oldest_context(struct inode *inode, + u64 *snap_size) { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_snap_context *snapc = NULL; struct ceph_cap_snap *capsnap = NULL; + spin_lock(&inode->i_lock); list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap, capsnap->context, capsnap->dirty_pages); @@ -357,21 +356,11 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode, break; } } - if (!snapc && ci->i_snap_realm) { - snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context); + if (!snapc && ci->i_head_snapc) { + snapc = ceph_get_snap_context(ci->i_head_snapc); dout(" head snapc %p has %d dirty pages\n", snapc, ci->i_wrbuffer_ref_head); } - return snapc; -} - -static struct ceph_snap_context *get_oldest_context(struct inode *inode, - u64 *snap_size) -{ - struct ceph_snap_context *snapc = NULL; - - spin_lock(&inode->i_lock); - snapc = __get_oldest_context(inode, snap_size); spin_unlock(&inode->i_lock); return snapc; } @@ -392,7 +381,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) int len = PAGE_CACHE_SIZE; loff_t i_size; int err = 0; - struct ceph_snap_context *snapc; + struct ceph_snap_context *snapc, *oldest; u64 snap_size = 0; long writeback_stat; @@ -413,13 +402,16 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) dout("writepage %p page %p not dirty?\n", inode, page); goto out; } - if (snapc != get_oldest_context(inode, &snap_size)) { + oldest = get_oldest_context(inode, &snap_size); + if (snapc->seq > oldest->seq) { dout("writepage %p page %p snapc %p not writeable - noop\n", inode, page, (void *)page->private); /* we should only noop if called by kswapd */ WARN_ON((current->flags & PF_MEMALLOC) == 0); + ceph_put_snap_context(oldest); goto out; } + ceph_put_snap_context(oldest); /* is this a partial page at end of file? */ if (snap_size) @@ -458,7 +450,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) ClearPagePrivate(page); end_page_writeback(page); ceph_put_wrbuffer_cap_refs(ci, 1, snapc); - ceph_put_snap_context(snapc); + ceph_put_snap_context(snapc); /* page's reference */ out: return err; } @@ -558,9 +550,9 @@ static void writepages_finish(struct ceph_osd_request *req, dout("inode %p skipping page %p\n", inode, page); wbc->pages_skipped++; } + ceph_put_snap_context((void *)page->private); page->private = 0; ClearPagePrivate(page); - ceph_put_snap_context(snapc); dout("unlocking %d %p\n", i, page); end_page_writeback(page); @@ -618,7 +610,7 @@ static int ceph_writepages_start(struct address_space *mapping, int range_whole = 0; int should_loop = 1; pgoff_t max_pages = 0, max_pages_ever = 0; - struct ceph_snap_context *snapc = NULL, *last_snapc = NULL; + struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc; struct pagevec pvec; int done = 0; int rc = 0; @@ -770,9 +762,10 @@ get_more_pages: } /* only if matching snap context */ - if (snapc != (void *)page->private) { - dout("page snapc %p != oldest %p\n", - (void *)page->private, snapc); + pgsnapc = (void *)page->private; + if (pgsnapc->seq > snapc->seq) { + dout("page snapc %p %lld > oldest %p %lld\n", + pgsnapc, pgsnapc->seq, snapc, snapc->seq); unlock_page(page); if (!locked_pages) continue; /* keep looking for snap */ @@ -914,7 +907,10 @@ static int context_is_writeable_or_written(struct inode *inode, struct ceph_snap_context *snapc) { struct ceph_snap_context *oldest = get_oldest_context(inode, NULL); - return !oldest || snapc->seq <= oldest->seq; + int ret = !oldest || snapc->seq <= oldest->seq; + + ceph_put_snap_context(oldest); + return ret; } /* @@ -936,8 +932,8 @@ static int ceph_update_writeable_page(struct file *file, int pos_in_page = pos & ~PAGE_CACHE_MASK; int end_in_page = pos_in_page + len; loff_t i_size; - struct ceph_snap_context *snapc; int r; + struct ceph_snap_context *snapc, *oldest; retry_locked: /* writepages currently holds page lock, but if we change that later, */ @@ -947,23 +943,24 @@ retry_locked: BUG_ON(!ci->i_snap_realm); down_read(&mdsc->snap_rwsem); BUG_ON(!ci->i_snap_realm->cached_context); - if (page->private && - (void *)page->private != ci->i_snap_realm->cached_context) { + snapc = (void *)page->private; + if (snapc && snapc != ci->i_head_snapc) { /* * this page is already dirty in another (older) snap * context! is it writeable now? */ - snapc = get_oldest_context(inode, NULL); + oldest = get_oldest_context(inode, NULL); up_read(&mdsc->snap_rwsem); - if (snapc != (void *)page->private) { + if (snapc->seq > oldest->seq) { + ceph_put_snap_context(oldest); dout(" page %p snapc %p not current or oldest\n", - page, (void *)page->private); + page, snapc); /* * queue for writeback, and wait for snapc to * be writeable or written */ - snapc = ceph_get_snap_context((void *)page->private); + snapc = ceph_get_snap_context(snapc); unlock_page(page); ceph_queue_writeback(inode); r = wait_event_interruptible(ci->i_cap_wq, @@ -973,6 +970,7 @@ retry_locked: return r; return -EAGAIN; } + ceph_put_snap_context(oldest); /* yay, writeable, do it now (without dropping page lock) */ dout(" page %p snapc %p not current, but oldest\n", diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 3710e077a857..aa2239fa9a3b 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1205,6 +1205,12 @@ retry: if (capsnap->dirty_pages || capsnap->writing) continue; + /* + * if cap writeback already occurred, we should have dropped + * the capsnap in ceph_put_wrbuffer_cap_refs. + */ + BUG_ON(capsnap->dirty == 0); + /* pick mds, take s_mutex */ mds = __ceph_get_cap_mds(ci, &mseq); if (session && session->s_mds != mds) { @@ -2118,8 +2124,8 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) } spin_unlock(&inode->i_lock); - dout("put_cap_refs %p had %s %s\n", inode, ceph_cap_string(had), - last ? "last" : ""); + dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had), + last ? " last" : "", put ? " put" : ""); if (last && !flushsnaps) ceph_check_caps(ci, 0, NULL); @@ -2143,7 +2149,8 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, { struct inode *inode = &ci->vfs_inode; int last = 0; - int last_snap = 0; + int complete_capsnap = 0; + int drop_capsnap = 0; int found = 0; struct ceph_cap_snap *capsnap = NULL; @@ -2166,19 +2173,32 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { if (capsnap->context == snapc) { found = 1; - capsnap->dirty_pages -= nr; - last_snap = !capsnap->dirty_pages; break; } } BUG_ON(!found); + capsnap->dirty_pages -= nr; + if (capsnap->dirty_pages == 0) { + complete_capsnap = 1; + if (capsnap->dirty == 0) + /* cap writeback completed before we created + * the cap_snap; no FLUSHSNAP is needed */ + drop_capsnap = 1; + } dout("put_wrbuffer_cap_refs on %p cap_snap %p " - " snap %lld %d/%d -> %d/%d %s%s\n", + " snap %lld %d/%d -> %d/%d %s%s%s\n", inode, capsnap, capsnap->context->seq, ci->i_wrbuffer_ref+nr, capsnap->dirty_pages + nr, ci->i_wrbuffer_ref, capsnap->dirty_pages, last ? " (wrbuffer last)" : "", - last_snap ? " (capsnap last)" : ""); + complete_capsnap ? " (complete capsnap)" : "", + drop_capsnap ? " (drop capsnap)" : ""); + if (drop_capsnap) { + ceph_put_snap_context(capsnap->context); + list_del(&capsnap->ci_item); + list_del(&capsnap->flushing_item); + ceph_put_cap_snap(capsnap); + } } spin_unlock(&inode->i_lock); @@ -2186,10 +2206,12 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, if (last) { ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); iput(inode); - } else if (last_snap) { + } else if (complete_capsnap) { ceph_flush_snaps(ci); wake_up(&ci->i_cap_wq); } + if (drop_capsnap) + iput(inode); } /* @@ -2465,8 +2487,8 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, break; } WARN_ON(capsnap->dirty_pages || capsnap->writing); - dout(" removing cap_snap %p follows %lld\n", - capsnap, follows); + dout(" removing %p cap_snap %p follows %lld\n", + inode, capsnap, follows); ceph_put_snap_context(capsnap->context); list_del(&capsnap->ci_item); list_del(&capsnap->flushing_item); diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 7261dc6c2ead..ea8ee2e526aa 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -171,11 +171,11 @@ more: spin_lock(&inode->i_lock); spin_lock(&dcache_lock); + last = dentry; + if (err < 0) goto out_unlock; - last = dentry; - p = p->prev; filp->f_pos++; @@ -312,7 +312,7 @@ more: req->r_readdir_offset = fi->next_offset; req->r_args.readdir.frag = cpu_to_le32(frag); req->r_args.readdir.max_entries = cpu_to_le32(max_entries); - req->r_num_caps = max_entries; + req->r_num_caps = max_entries + 1; err = ceph_mdsc_do_request(mdsc, NULL, req); if (err < 0) { ceph_mdsc_put_request(req); @@ -489,6 +489,7 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req, struct inode *inode = ceph_get_snapdir(parent); dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n", dentry, dentry->d_name.len, dentry->d_name.name, inode); + BUG_ON(!d_unhashed(dentry)); d_add(dentry, inode); err = 0; } diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index aca82d55cc53..26f883c275e8 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -886,6 +886,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, struct inode *in = NULL; struct ceph_mds_reply_inode *ininfo; struct ceph_vino vino; + struct ceph_client *client = ceph_sb_to_client(sb); int i = 0; int err = 0; @@ -949,7 +950,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, return err; } - if (rinfo->head->is_dentry && !req->r_aborted) { + /* + * ignore null lease/binding on snapdir ENOENT, or else we + * will have trouble splicing in the virtual snapdir later + */ + if (rinfo->head->is_dentry && !req->r_aborted && + (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name, + client->mount_args->snapdir_name, + req->r_dentry->d_name.len))) { /* * lookup link rename : null -> possibly existing inode * mknod symlink mkdir : null -> new inode diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c index 8f1715ffbe4b..cdaaa131add3 100644 --- a/fs/ceph/messenger.c +++ b/fs/ceph/messenger.c @@ -30,6 +30,10 @@ static char tag_msg = CEPH_MSGR_TAG_MSG; static char tag_ack = CEPH_MSGR_TAG_ACK; static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE; +#ifdef CONFIG_LOCKDEP +static struct lock_class_key socket_class; +#endif + static void queue_con(struct ceph_connection *con); static void con_work(struct work_struct *); @@ -228,6 +232,10 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con) con->sock = sock; sock->sk->sk_allocation = GFP_NOFS; +#ifdef CONFIG_LOCKDEP + lockdep_set_class(&sock->sk->sk_lock, &socket_class); +#endif + set_sock_callbacks(sock, con); dout("connect %s\n", pr_addr(&con->peer_addr.in_addr)); @@ -333,6 +341,7 @@ static void reset_connection(struct ceph_connection *con) con->out_msg = NULL; } con->in_seq = 0; + con->in_seq_acked = 0; } /* diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c index 21c6623c4b07..2e2c15eed82a 100644 --- a/fs/ceph/osdmap.c +++ b/fs/ceph/osdmap.c @@ -314,71 +314,6 @@ bad: return ERR_PTR(err); } - -/* - * osd map - */ -void ceph_osdmap_destroy(struct ceph_osdmap *map) -{ - dout("osdmap_destroy %p\n", map); - if (map->crush) - crush_destroy(map->crush); - while (!RB_EMPTY_ROOT(&map->pg_temp)) { - struct ceph_pg_mapping *pg = - rb_entry(rb_first(&map->pg_temp), - struct ceph_pg_mapping, node); - rb_erase(&pg->node, &map->pg_temp); - kfree(pg); - } - while (!RB_EMPTY_ROOT(&map->pg_pools)) { - struct ceph_pg_pool_info *pi = - rb_entry(rb_first(&map->pg_pools), - struct ceph_pg_pool_info, node); - rb_erase(&pi->node, &map->pg_pools); - kfree(pi); - } - kfree(map->osd_state); - kfree(map->osd_weight); - kfree(map->osd_addr); - kfree(map); -} - -/* - * adjust max osd value. reallocate arrays. - */ -static int osdmap_set_max_osd(struct ceph_osdmap *map, int max) -{ - u8 *state; - struct ceph_entity_addr *addr; - u32 *weight; - - state = kcalloc(max, sizeof(*state), GFP_NOFS); - addr = kcalloc(max, sizeof(*addr), GFP_NOFS); - weight = kcalloc(max, sizeof(*weight), GFP_NOFS); - if (state == NULL || addr == NULL || weight == NULL) { - kfree(state); - kfree(addr); - kfree(weight); - return -ENOMEM; - } - - /* copy old? */ - if (map->osd_state) { - memcpy(state, map->osd_state, map->max_osd*sizeof(*state)); - memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr)); - memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight)); - kfree(map->osd_state); - kfree(map->osd_addr); - kfree(map->osd_weight); - } - - map->osd_state = state; - map->osd_weight = weight; - map->osd_addr = addr; - map->max_osd = max; - return 0; -} - /* * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid * to a set of osds) @@ -482,6 +417,13 @@ static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id) return NULL; } +static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi) +{ + rb_erase(&pi->node, root); + kfree(pi->name); + kfree(pi); +} + void __decode_pool(void **p, struct ceph_pg_pool_info *pi) { ceph_decode_copy(p, &pi->v, sizeof(pi->v)); @@ -490,6 +432,98 @@ void __decode_pool(void **p, struct ceph_pg_pool_info *pi) *p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2; } +static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map) +{ + struct ceph_pg_pool_info *pi; + u32 num, len, pool; + + ceph_decode_32_safe(p, end, num, bad); + dout(" %d pool names\n", num); + while (num--) { + ceph_decode_32_safe(p, end, pool, bad); + ceph_decode_32_safe(p, end, len, bad); + dout(" pool %d len %d\n", pool, len); + pi = __lookup_pg_pool(&map->pg_pools, pool); + if (pi) { + kfree(pi->name); + pi->name = kmalloc(len + 1, GFP_NOFS); + if (pi->name) { + memcpy(pi->name, *p, len); + pi->name[len] = '\0'; + dout(" name is %s\n", pi->name); + } + } + *p += len; + } + return 0; + +bad: + return -EINVAL; +} + +/* + * osd map + */ +void ceph_osdmap_destroy(struct ceph_osdmap *map) +{ + dout("osdmap_destroy %p\n", map); + if (map->crush) + crush_destroy(map->crush); + while (!RB_EMPTY_ROOT(&map->pg_temp)) { + struct ceph_pg_mapping *pg = + rb_entry(rb_first(&map->pg_temp), + struct ceph_pg_mapping, node); + rb_erase(&pg->node, &map->pg_temp); + kfree(pg); + } + while (!RB_EMPTY_ROOT(&map->pg_pools)) { + struct ceph_pg_pool_info *pi = + rb_entry(rb_first(&map->pg_pools), + struct ceph_pg_pool_info, node); + __remove_pg_pool(&map->pg_pools, pi); + } + kfree(map->osd_state); + kfree(map->osd_weight); + kfree(map->osd_addr); + kfree(map); +} + +/* + * adjust max osd value. reallocate arrays. + */ +static int osdmap_set_max_osd(struct ceph_osdmap *map, int max) +{ + u8 *state; + struct ceph_entity_addr *addr; + u32 *weight; + + state = kcalloc(max, sizeof(*state), GFP_NOFS); + addr = kcalloc(max, sizeof(*addr), GFP_NOFS); + weight = kcalloc(max, sizeof(*weight), GFP_NOFS); + if (state == NULL || addr == NULL || weight == NULL) { + kfree(state); + kfree(addr); + kfree(weight); + return -ENOMEM; + } + + /* copy old? */ + if (map->osd_state) { + memcpy(state, map->osd_state, map->max_osd*sizeof(*state)); + memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr)); + memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight)); + kfree(map->osd_state); + kfree(map->osd_addr); + kfree(map->osd_weight); + } + + map->osd_state = state; + map->osd_weight = weight; + map->osd_addr = addr; + map->max_osd = max; + return 0; +} + /* * decode a full map. */ @@ -526,7 +560,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end) ceph_decode_32_safe(p, end, max, bad); while (max--) { ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad); - pi = kmalloc(sizeof(*pi), GFP_NOFS); + pi = kzalloc(sizeof(*pi), GFP_NOFS); if (!pi) goto bad; pi->id = ceph_decode_32(p); @@ -539,6 +573,10 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end) __decode_pool(p, pi); __insert_pg_pool(&map->pg_pools, pi); } + + if (version >= 5 && __decode_pool_names(p, end, map) < 0) + goto bad; + ceph_decode_32_safe(p, end, map->pool_max, bad); ceph_decode_32_safe(p, end, map->flags, bad); @@ -712,7 +750,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, } pi = __lookup_pg_pool(&map->pg_pools, pool); if (!pi) { - pi = kmalloc(sizeof(*pi), GFP_NOFS); + pi = kzalloc(sizeof(*pi), GFP_NOFS); if (!pi) { err = -ENOMEM; goto bad; @@ -722,6 +760,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, } __decode_pool(p, pi); } + if (version >= 5 && __decode_pool_names(p, end, map) < 0) + goto bad; /* old_pool */ ceph_decode_32_safe(p, end, len, bad); @@ -730,10 +770,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, ceph_decode_32_safe(p, end, pool, bad); pi = __lookup_pg_pool(&map->pg_pools, pool); - if (pi) { - rb_erase(&pi->node, &map->pg_pools); - kfree(pi); - } + if (pi) + __remove_pg_pool(&map->pg_pools, pi); } /* new_up */ diff --git a/fs/ceph/osdmap.h b/fs/ceph/osdmap.h index 1fb55afb2642..8bc9f1e4f562 100644 --- a/fs/ceph/osdmap.h +++ b/fs/ceph/osdmap.h @@ -23,6 +23,7 @@ struct ceph_pg_pool_info { int id; struct ceph_pg_pool v; int pg_num_mask, pgp_num_mask, lpg_num_mask, lpgp_num_mask; + char *name; }; struct ceph_pg_mapping { diff --git a/fs/ceph/rados.h b/fs/ceph/rados.h index 26ac8b89a676..a1fc1d017b58 100644 --- a/fs/ceph/rados.h +++ b/fs/ceph/rados.h @@ -11,8 +11,10 @@ /* * osdmap encoding versions */ -#define CEPH_OSDMAP_INC_VERSION 4 -#define CEPH_OSDMAP_VERSION 4 +#define CEPH_OSDMAP_INC_VERSION 5 +#define CEPH_OSDMAP_INC_VERSION_EXT 5 +#define CEPH_OSDMAP_VERSION 5 +#define CEPH_OSDMAP_VERSION_EXT 5 /* * fs id diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index e6f9bc57d472..2b881262ef67 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -431,8 +431,7 @@ static int dup_array(u64 **dst, __le64 *src, int num) * Caller must hold snap_rwsem for read (i.e., the realm topology won't * change). */ -void ceph_queue_cap_snap(struct ceph_inode_info *ci, - struct ceph_snap_context *snapc) +void ceph_queue_cap_snap(struct ceph_inode_info *ci) { struct inode *inode = &ci->vfs_inode; struct ceph_cap_snap *capsnap; @@ -451,10 +450,11 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci, as no new writes are allowed to start when pending, so any writes in progress now were started before the previous cap_snap. lucky us. */ - dout("queue_cap_snap %p snapc %p seq %llu used %d" - " already pending\n", inode, snapc, snapc->seq, used); + dout("queue_cap_snap %p already pending\n", inode); kfree(capsnap); } else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR)) { + struct ceph_snap_context *snapc = ci->i_head_snapc; + igrab(inode); atomic_set(&capsnap->nref, 1); @@ -463,7 +463,6 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci, INIT_LIST_HEAD(&capsnap->flushing_item); capsnap->follows = snapc->seq - 1; - capsnap->context = ceph_get_snap_context(snapc); capsnap->issued = __ceph_caps_issued(ci, NULL); capsnap->dirty = __ceph_caps_dirty(ci); @@ -480,7 +479,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci, snapshot. */ capsnap->dirty_pages = ci->i_wrbuffer_ref_head; ci->i_wrbuffer_ref_head = 0; - ceph_put_snap_context(ci->i_head_snapc); + capsnap->context = snapc; ci->i_head_snapc = NULL; list_add_tail(&capsnap->ci_item, &ci->i_cap_snaps); @@ -522,15 +521,17 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, capsnap->ctime = inode->i_ctime; capsnap->time_warp_seq = ci->i_time_warp_seq; if (capsnap->dirty_pages) { - dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu " + dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu " "still has %d dirty pages\n", inode, capsnap, capsnap->context, capsnap->context->seq, - capsnap->size, capsnap->dirty_pages); + ceph_cap_string(capsnap->dirty), capsnap->size, + capsnap->dirty_pages); return 0; } - dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu clean\n", + dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu\n", inode, capsnap, capsnap->context, - capsnap->context->seq, capsnap->size); + capsnap->context->seq, ceph_cap_string(capsnap->dirty), + capsnap->size); spin_lock(&mdsc->snap_flush_lock); list_add_tail(&ci->i_snap_flush_item, &mdsc->snap_flush_list); @@ -602,7 +603,7 @@ more: if (lastinode) iput(lastinode); lastinode = inode; - ceph_queue_cap_snap(ci, realm->cached_context); + ceph_queue_cap_snap(ci); spin_lock(&realm->inodes_with_caps_lock); } spin_unlock(&realm->inodes_with_caps_lock); @@ -824,8 +825,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, spin_unlock(&realm->inodes_with_caps_lock); spin_unlock(&inode->i_lock); - ceph_queue_cap_snap(ci, - ci->i_snap_realm->cached_context); + ceph_queue_cap_snap(ci); iput(inode); continue; diff --git a/fs/ceph/super.h b/fs/ceph/super.h index ca702c67bc66..e30dfbb056c3 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -715,8 +715,7 @@ extern int ceph_update_snap_trace(struct ceph_mds_client *m, extern void ceph_handle_snap(struct ceph_mds_client *mdsc, struct ceph_mds_session *session, struct ceph_msg *msg); -extern void ceph_queue_cap_snap(struct ceph_inode_info *ci, - struct ceph_snap_context *snapc); +extern void ceph_queue_cap_snap(struct ceph_inode_info *ci); extern int __ceph_finish_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap *capsnap); extern void ceph_cleanup_empty_realms(struct ceph_mds_client *mdsc); diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 2a3d352c0bff..a8766c4ef2e0 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1294,7 +1294,8 @@ static int nfs4_init_server(struct nfs_server *server, /* Initialise the client representation from the mount data */ server->flags = data->flags; - server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR; + server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR| + NFS_CAP_POSIX_LOCK; server->options = data->options; /* Get a client record */ diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index c6f2750648f4..be46f26c9a56 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1025,12 +1025,12 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry res = NULL; goto out; /* This turned out not to be a regular file */ + case -EISDIR: case -ENOTDIR: goto no_open; case -ELOOP: if (!(nd->intent.open.flags & O_NOFOLLOW)) goto no_open; - /* case -EISDIR: */ /* case -EINVAL: */ default: goto out; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 737128f777f3..50a56edca0b5 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -623,10 +623,10 @@ struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_c list_for_each_entry(pos, &nfsi->open_files, list) { if (cred != NULL && pos->cred != cred) continue; - if ((pos->mode & mode) == mode) { - ctx = get_nfs_open_context(pos); - break; - } + if ((pos->mode & (FMODE_READ|FMODE_WRITE)) != mode) + continue; + ctx = get_nfs_open_context(pos); + break; } spin_unlock(&inode->i_lock); return ctx; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index fe0cd9eb1d4d..638067007c65 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1523,6 +1523,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) nfs_post_op_update_inode(dir, o_res->dir_attr); } else nfs_refresh_inode(dir, o_res->dir_attr); + if ((o_res->rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) == 0) + server->caps &= ~NFS_CAP_POSIX_LOCK; if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) { status = _nfs4_proc_open_confirm(data); if (status != 0) @@ -1664,7 +1666,7 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in status = PTR_ERR(state); if (IS_ERR(state)) goto err_opendata_put; - if ((opendata->o_res.rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) != 0) + if (server->caps & NFS_CAP_POSIX_LOCK) set_bit(NFS_STATE_POSIX_LOCKS, &state->flags); nfs4_opendata_put(opendata); nfs4_put_state_owner(sp); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 53ff70e23993..de38d63aa920 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -201,6 +201,7 @@ static int nfs_set_page_writeback(struct page *page) struct inode *inode = page->mapping->host; struct nfs_server *nfss = NFS_SERVER(inode); + page_cache_get(page); if (atomic_long_inc_return(&nfss->writeback) > NFS_CONGESTION_ON_THRESH) { set_bdi_congested(&nfss->backing_dev_info, @@ -216,6 +217,7 @@ static void nfs_end_page_writeback(struct page *page) struct nfs_server *nfss = NFS_SERVER(inode); end_page_writeback(page); + page_cache_release(page); if (atomic_long_dec_return(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) clear_bdi_congested(&nfss->backing_dev_info, BLK_RW_ASYNC); } @@ -421,6 +423,7 @@ static void nfs_mark_request_dirty(struct nfs_page *req) { __set_page_dirty_nobuffers(req->wb_page); + __mark_inode_dirty(req->wb_page->mapping->host, I_DIRTY_DATASYNC); } #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) @@ -660,9 +663,11 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, req = nfs_setup_write_request(ctx, page, offset, count); if (IS_ERR(req)) return PTR_ERR(req); + nfs_mark_request_dirty(req); /* Update file length */ nfs_grow_file(page, offset, count); nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes); + nfs_mark_request_dirty(req); nfs_clear_page_tag_locked(req); return 0; } @@ -739,8 +744,6 @@ int nfs_updatepage(struct file *file, struct page *page, status = nfs_writepage_setup(ctx, page, offset, count); if (status < 0) nfs_set_pageerror(page); - else - __set_page_dirty_nobuffers(page); dprintk("NFS: nfs_updatepage returns %d (isize %lld)\n", status, (long long)i_size_read(inode)); @@ -749,13 +752,12 @@ int nfs_updatepage(struct file *file, struct page *page, static void nfs_writepage_release(struct nfs_page *req) { + struct page *page = req->wb_page; - if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req)) { - nfs_end_page_writeback(req->wb_page); + if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req)) nfs_inode_remove_request(req); - } else - nfs_end_page_writeback(req->wb_page); nfs_clear_page_tag_locked(req); + nfs_end_page_writeback(page); } static int flush_task_priority(int how) @@ -779,7 +781,6 @@ static int nfs_write_rpcsetup(struct nfs_page *req, int how) { struct inode *inode = req->wb_context->path.dentry->d_inode; - int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; int priority = flush_task_priority(how); struct rpc_task *task; struct rpc_message msg = { @@ -794,9 +795,10 @@ static int nfs_write_rpcsetup(struct nfs_page *req, .callback_ops = call_ops, .callback_data = data, .workqueue = nfsiod_workqueue, - .flags = flags, + .flags = RPC_TASK_ASYNC, .priority = priority, }; + int ret = 0; /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ @@ -835,10 +837,18 @@ static int nfs_write_rpcsetup(struct nfs_page *req, (unsigned long long)data->args.offset); task = rpc_run_task(&task_setup_data); - if (IS_ERR(task)) - return PTR_ERR(task); + if (IS_ERR(task)) { + ret = PTR_ERR(task); + goto out; + } + if (how & FLUSH_SYNC) { + ret = rpc_wait_for_completion_task(task); + if (ret == 0) + ret = task->tk_status; + } rpc_put_task(task); - return 0; +out: + return ret; } /* If a nfs_flush_* function fails, it should remove reqs from @head and @@ -847,9 +857,11 @@ static int nfs_write_rpcsetup(struct nfs_page *req, */ static void nfs_redirty_request(struct nfs_page *req) { + struct page *page = req->wb_page; + nfs_mark_request_dirty(req); - nfs_end_page_writeback(req->wb_page); nfs_clear_page_tag_locked(req); + nfs_end_page_writeback(page); } /* @@ -1084,16 +1096,15 @@ static void nfs_writeback_release_full(void *calldata) if (nfs_write_need_commit(data)) { memcpy(&req->wb_verf, &data->verf, sizeof(req->wb_verf)); nfs_mark_request_commit(req); - nfs_end_page_writeback(page); dprintk(" marked for commit\n"); goto next; } dprintk(" OK\n"); remove_request: - nfs_end_page_writeback(page); nfs_inode_remove_request(req); next: nfs_clear_page_tag_locked(req); + nfs_end_page_writeback(page); } nfs_writedata_release(calldata); } @@ -1207,7 +1218,6 @@ static int nfs_commit_rpcsetup(struct list_head *head, { struct nfs_page *first = nfs_list_entry(head->next); struct inode *inode = first->wb_context->path.dentry->d_inode; - int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; int priority = flush_task_priority(how); struct rpc_task *task; struct rpc_message msg = { @@ -1222,7 +1232,7 @@ static int nfs_commit_rpcsetup(struct list_head *head, .callback_ops = &nfs_commit_ops, .callback_data = data, .workqueue = nfsiod_workqueue, - .flags = flags, + .flags = RPC_TASK_ASYNC, .priority = priority, }; @@ -1252,6 +1262,8 @@ static int nfs_commit_rpcsetup(struct list_head *head, task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); + if (how & FLUSH_SYNC) + rpc_wait_for_completion_task(task); rpc_put_task(task); return 0; } diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 05cd85317f6f..fd9698215759 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -820,10 +820,10 @@ xfs_reclaim_inode( * call into reclaim to find it in a clean state instead of waiting for * it now. We also don't return errors here - if the error is transient * then the next reclaim pass will flush the inode, and if the error - * is permanent then the next sync reclaim will relcaim the inode and + * is permanent then the next sync reclaim will reclaim the inode and * pass on the error. */ - if (error && !XFS_FORCED_SHUTDOWN(ip->i_mount)) { + if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) { xfs_fs_cmn_err(CE_WARN, ip->i_mount, "inode 0x%llx background reclaim flush failed with %d", (long long)ip->i_ino, error); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index e8fba92d7cd9..2be019136287 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -745,9 +745,16 @@ xfs_log_move_tail(xfs_mount_t *mp, /* * Determine if we have a transaction that has gone to disk - * that needs to be covered. Log activity needs to be idle (no AIL and - * nothing in the iclogs). And, we need to be in the right state indicating - * something has gone out. + * that needs to be covered. To begin the transition to the idle state + * firstly the log needs to be idle (no AIL and nothing in the iclogs). + * If we are then in a state where covering is needed, the caller is informed + * that dummy transactions are required to move the log into the idle state. + * + * Because this is called as part of the sync process, we should also indicate + * that dummy transactions should be issued in anything but the covered or + * idle states. This ensures that the log tail is accurately reflected in + * the log at the end of the sync, hence if a crash occurrs avoids replay + * of transactions where the metadata is already on disk. */ int xfs_log_need_covered(xfs_mount_t *mp) @@ -759,17 +766,24 @@ xfs_log_need_covered(xfs_mount_t *mp) return 0; spin_lock(&log->l_icloglock); - if (((log->l_covered_state == XLOG_STATE_COVER_NEED) || - (log->l_covered_state == XLOG_STATE_COVER_NEED2)) - && !xfs_trans_ail_tail(log->l_ailp) - && xlog_iclogs_empty(log)) { - if (log->l_covered_state == XLOG_STATE_COVER_NEED) - log->l_covered_state = XLOG_STATE_COVER_DONE; - else { - ASSERT(log->l_covered_state == XLOG_STATE_COVER_NEED2); - log->l_covered_state = XLOG_STATE_COVER_DONE2; + switch (log->l_covered_state) { + case XLOG_STATE_COVER_DONE: + case XLOG_STATE_COVER_DONE2: + case XLOG_STATE_COVER_IDLE: + break; + case XLOG_STATE_COVER_NEED: + case XLOG_STATE_COVER_NEED2: + if (!xfs_trans_ail_tail(log->l_ailp) && + xlog_iclogs_empty(log)) { + if (log->l_covered_state == XLOG_STATE_COVER_NEED) + log->l_covered_state = XLOG_STATE_COVER_DONE; + else + log->l_covered_state = XLOG_STATE_COVER_DONE2; } + /* FALLTHRU */ + default: needed = 1; + break; } spin_unlock(&log->l_icloglock); return needed; diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 4de02b10007f..9f15150ce8d6 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -278,6 +278,27 @@ struct freq_attr { ssize_t (*store)(struct cpufreq_policy *, const char *, size_t count); }; +#define cpufreq_freq_attr_ro(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0444, show_##_name, NULL) + +#define cpufreq_freq_attr_ro_perm(_name, _perm) \ +static struct freq_attr _name = \ +__ATTR(_name, _perm, show_##_name, NULL) + +#define cpufreq_freq_attr_ro_old(_name) \ +static struct freq_attr _name##_old = \ +__ATTR(_name, 0444, show_##_name##_old, NULL) + +#define cpufreq_freq_attr_rw(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0644, show_##_name, store_##_name) + +#define cpufreq_freq_attr_rw_old(_name) \ +static struct freq_attr _name##_old = \ +__ATTR(_name, 0644, show_##_name##_old, store_##_name##_old) + + struct global_attr { struct attribute attr; ssize_t (*show)(struct kobject *kobj, @@ -286,6 +307,15 @@ struct global_attr { const char *c, size_t count); }; +#define define_one_global_ro(_name) \ +static struct global_attr _name = \ +__ATTR(_name, 0444, show_##_name, NULL) + +#define define_one_global_rw(_name) \ +static struct global_attr _name = \ +__ATTR(_name, 0644, show_##_name, store_##_name) + + /********************************************************************* * CPUFREQ 2.6. INTERFACE * *********************************************************************/ diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index a5740fc4d04b..a73454aec333 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -21,8 +21,7 @@ extern int number_of_cpusets; /* How many cpusets are defined in system? */ extern int cpuset_init(void); extern void cpuset_init_smp(void); extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask); -extern void cpuset_cpus_allowed_locked(struct task_struct *p, - struct cpumask *mask); +extern int cpuset_cpus_allowed_fallback(struct task_struct *p); extern nodemask_t cpuset_mems_allowed(struct task_struct *p); #define cpuset_current_mems_allowed (current->mems_allowed) void cpuset_init_current_mems_allowed(void); @@ -69,9 +68,6 @@ struct seq_file; extern void cpuset_task_status_allowed(struct seq_file *m, struct task_struct *task); -extern void cpuset_lock(void); -extern void cpuset_unlock(void); - extern int cpuset_mem_spread_node(void); static inline int cpuset_do_page_mem_spread(void) @@ -105,10 +101,11 @@ static inline void cpuset_cpus_allowed(struct task_struct *p, { cpumask_copy(mask, cpu_possible_mask); } -static inline void cpuset_cpus_allowed_locked(struct task_struct *p, - struct cpumask *mask) + +static inline int cpuset_cpus_allowed_fallback(struct task_struct *p) { - cpumask_copy(mask, cpu_possible_mask); + cpumask_copy(&p->cpus_allowed, cpu_possible_mask); + return cpumask_any(cpu_active_mask); } static inline nodemask_t cpuset_mems_allowed(struct task_struct *p) @@ -157,9 +154,6 @@ static inline void cpuset_task_status_allowed(struct seq_file *m, { } -static inline void cpuset_lock(void) {} -static inline void cpuset_unlock(void) {} - static inline int cpuset_mem_spread_node(void) { return 0; diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 40b11013408e..81f3b14d5d76 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -1,21 +1,26 @@ /* * Char device interface. * - * Copyright (C) 2005-2006 Kristian Hoegsberg <krh@bitplanet.net> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software Foundation, - * Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * Copyright (C) 2005-2007 Kristian Hoegsberg <krh@bitplanet.net> + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. */ #ifndef _LINUX_FIREWIRE_CDEV_H @@ -438,7 +443,7 @@ struct fw_cdev_remove_descriptor { * @type: %FW_CDEV_ISO_CONTEXT_TRANSMIT or %FW_CDEV_ISO_CONTEXT_RECEIVE * @header_size: Header size to strip for receive contexts * @channel: Channel to bind to - * @speed: Speed to transmit at + * @speed: Speed for transmit contexts * @closure: To be returned in &fw_cdev_event_iso_interrupt * @handle: Handle to context, written back by kernel * @@ -451,6 +456,9 @@ struct fw_cdev_remove_descriptor { * If a context was successfully created, the kernel writes back a handle to the * context, which must be passed in for subsequent operations on that context. * + * For receive contexts, @header_size must be at least 4 and must be a multiple + * of 4. + * * Note that the effect of a @header_size > 4 depends on * &fw_cdev_get_info.version, as documented at &fw_cdev_event_iso_interrupt. */ @@ -481,10 +489,34 @@ struct fw_cdev_create_iso_context { * * &struct fw_cdev_iso_packet is used to describe isochronous packet queues. * - * Use the FW_CDEV_ISO_ macros to fill in @control. The sy and tag fields are - * specified by IEEE 1394a and IEC 61883. - * - * FIXME - finish this documentation + * Use the FW_CDEV_ISO_ macros to fill in @control. + * + * For transmit packets, the header length must be a multiple of 4 and specifies + * the numbers of bytes in @header that will be prepended to the packet's + * payload; these bytes are copied into the kernel and will not be accessed + * after the ioctl has returned. The sy and tag fields are copied to the iso + * packet header (these fields are specified by IEEE 1394a and IEC 61883-1). + * The skip flag specifies that no packet is to be sent in a frame; when using + * this, all other fields except the interrupt flag must be zero. + * + * For receive packets, the header length must be a multiple of the context's + * header size; if the header length is larger than the context's header size, + * multiple packets are queued for this entry. The sy and tag fields are + * ignored. If the sync flag is set, the context drops all packets until + * a packet with a matching sy field is received (the sync value to wait for is + * specified in the &fw_cdev_start_iso structure). The payload length defines + * how many payload bytes can be received for one packet (in addition to payload + * quadlets that have been defined as headers and are stripped and returned in + * the &fw_cdev_event_iso_interrupt structure). If more bytes are received, the + * additional bytes are dropped. If less bytes are received, the remaining + * bytes in this part of the payload buffer will not be written to, not even by + * the next packet, i.e., packets received in consecutive frames will not + * necessarily be consecutive in memory. If an entry has queued multiple + * packets, the payload length is divided equally among them. + * + * When a packet with the interrupt flag set has been completed, the + * &fw_cdev_event_iso_interrupt event will be sent. An entry that has queued + * multiple receive packets is completed when its last packet is completed. */ struct fw_cdev_iso_packet { __u32 control; @@ -501,7 +533,7 @@ struct fw_cdev_iso_packet { * Queue a number of isochronous packets for reception or transmission. * This ioctl takes a pointer to an array of &fw_cdev_iso_packet structs, * which describe how to transmit from or receive into a contiguous region - * of a mmap()'ed payload buffer. As part of the packet descriptors, + * of a mmap()'ed payload buffer. As part of transmit packet descriptors, * a series of headers can be supplied, which will be prepended to the * payload during DMA. * @@ -620,8 +652,8 @@ struct fw_cdev_get_cycle_timer2 { * instead of allocated. * An %FW_CDEV_EVENT_ISO_RESOURCE_DEALLOCATED event concludes this operation. * - * To summarize, %FW_CDEV_IOC_DEALLOCATE_ISO_RESOURCE allocates iso resources - * for the lifetime of the fd or handle. + * To summarize, %FW_CDEV_IOC_ALLOCATE_ISO_RESOURCE allocates iso resources + * for the lifetime of the fd or @handle. * In contrast, %FW_CDEV_IOC_ALLOCATE_ISO_RESOURCE_ONCE allocates iso resources * for the duration of a bus generation. * diff --git a/include/linux/firewire-constants.h b/include/linux/firewire-constants.h index b316770a43fd..9c63f06e67f2 100644 --- a/include/linux/firewire-constants.h +++ b/include/linux/firewire-constants.h @@ -1,3 +1,28 @@ +/* + * IEEE 1394 constants. + * + * Copyright (C) 2005-2007 Kristian Hoegsberg <krh@bitplanet.net> + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + #ifndef _LINUX_FIREWIRE_CONSTANTS_H #define _LINUX_FIREWIRE_CONSTANTS_H @@ -21,7 +46,7 @@ #define EXTCODE_WRAP_ADD 0x6 #define EXTCODE_VENDOR_DEPENDENT 0x7 -/* Juju specific tcodes */ +/* Linux firewire-core (Juju) specific tcodes */ #define TCODE_LOCK_MASK_SWAP (0x10 | EXTCODE_MASK_SWAP) #define TCODE_LOCK_COMPARE_SWAP (0x10 | EXTCODE_COMPARE_SWAP) #define TCODE_LOCK_FETCH_ADD (0x10 | EXTCODE_FETCH_ADD) @@ -36,7 +61,7 @@ #define RCODE_TYPE_ERROR 0x6 #define RCODE_ADDRESS_ERROR 0x7 -/* Juju specific rcodes */ +/* Linux firewire-core (Juju) specific rcodes */ #define RCODE_SEND_ERROR 0x10 #define RCODE_CANCELLED 0x11 #define RCODE_BUSY 0x12 diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 01e6adea07ec..cc12b3c556b3 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -504,18 +504,6 @@ extern int ftrace_dump_on_oops; #define INIT_TRACE_RECURSION #endif -#ifdef CONFIG_HW_BRANCH_TRACER - -void trace_hw_branch(u64 from, u64 to); -void trace_hw_branch_oops(void); - -#else /* CONFIG_HW_BRANCH_TRACER */ - -static inline void trace_hw_branch(u64 from, u64 to) {} -static inline void trace_hw_branch_oops(void) {} - -#endif /* CONFIG_HW_BRANCH_TRACER */ - #ifdef CONFIG_FTRACE_SYSCALLS unsigned long arch_syscall_addr(int nr); diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index c0f4b364c711..39e71b0a3bfd 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -58,6 +58,7 @@ struct trace_iterator { /* The below is zeroed out in pipe_read */ struct trace_seq seq; struct trace_entry *ent; + unsigned long lost_events; int leftover; int cpu; u64 ts; diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 5d86fb2309d2..fd0c1b857d3d 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -422,6 +422,8 @@ extern void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, extern int schedule_hrtimeout_range(ktime_t *expires, unsigned long delta, const enum hrtimer_mode mode); +extern int schedule_hrtimeout_range_clock(ktime_t *expires, + unsigned long delta, const enum hrtimer_mode mode, int clock); extern int schedule_hrtimeout(ktime_t *expires, const enum hrtimer_mode mode); /* Soft interrupt function to run the hrtimer queues: */ diff --git a/include/linux/input/matrix_keypad.h b/include/linux/input/matrix_keypad.h index 3bd018baae20..c964cd7f436a 100644 --- a/include/linux/input/matrix_keypad.h +++ b/include/linux/input/matrix_keypad.h @@ -44,6 +44,7 @@ struct matrix_keymap_data { * @active_low: gpio polarity * @wakeup: controls whether the device should be set up as wakeup * source + * @no_autorepeat: disable key autorepeat * * This structure represents platform-specific data that use used by * matrix_keypad driver to perform proper initialization. @@ -64,6 +65,7 @@ struct matrix_keypad_platform_data { bool active_low; bool wakeup; + bool no_autorepeat; }; /** diff --git a/include/linux/mm.h b/include/linux/mm.h index 462acaf36f3a..fb19bb92b809 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -19,7 +19,6 @@ struct anon_vma; struct file_ra_state; struct user_struct; struct writeback_control; -struct rlimit; #ifndef CONFIG_DISCONTIGMEM /* Don't use mapnrs, do it properly */ extern unsigned long max_mapnr; @@ -1449,9 +1448,6 @@ int vmemmap_populate_basepages(struct page *start_page, int vmemmap_populate(struct page *start_page, unsigned long pages, int node); void vmemmap_populate_print_last(void); -extern int account_locked_memory(struct mm_struct *mm, struct rlimit *rlim, - size_t size); -extern void refund_locked_memory(struct mm_struct *mm, size_t size); enum mf_flags { MF_COUNT_INCREASED = 1 << 0, diff --git a/include/linux/module.h b/include/linux/module.h index 515d53ae6a79..6914fcad4673 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -465,8 +465,7 @@ static inline void __module_get(struct module *module) if (module) { preempt_disable(); __this_cpu_inc(module->refptr->incs); - trace_module_get(module, _THIS_IP_, - __this_cpu_read(module->refptr->incs)); + trace_module_get(module, _THIS_IP_); preempt_enable(); } } @@ -480,8 +479,7 @@ static inline int try_module_get(struct module *module) if (likely(module_is_live(module))) { __this_cpu_inc(module->refptr->incs); - trace_module_get(module, _THIS_IP_, - __this_cpu_read(module->refptr->incs)); + trace_module_get(module, _THIS_IP_); } else ret = 0; diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 717a5e54eb1d..e82957acea56 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -176,6 +176,7 @@ struct nfs_server { #define NFS_CAP_ATIME (1U << 11) #define NFS_CAP_CTIME (1U << 12) #define NFS_CAP_MTIME (1U << 13) +#define NFS_CAP_POSIX_LOCK (1U << 14) /* maximum number of slots to use */ diff --git a/include/linux/nmi.h b/include/linux/nmi.h index b752e807adde..22cc7960b649 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -20,10 +20,14 @@ extern void touch_nmi_watchdog(void); extern void acpi_nmi_disable(void); extern void acpi_nmi_enable(void); #else +#ifndef CONFIG_NMI_WATCHDOG static inline void touch_nmi_watchdog(void) { touch_softlockup_watchdog(); } +#else +extern void touch_nmi_watchdog(void); +#endif static inline void acpi_nmi_disable(void) { } static inline void acpi_nmi_enable(void) { } #endif @@ -47,4 +51,13 @@ static inline bool trigger_all_cpu_backtrace(void) } #endif +#ifdef CONFIG_NMI_WATCHDOG +int hw_nmi_is_cpu_stuck(struct pt_regs *); +u64 hw_nmi_get_sample_period(void); +extern int nmi_watchdog_enabled; +struct ctl_table; +extern int proc_nmi_enabled(struct ctl_table *, int , + void __user *, size_t *, loff_t *); +#endif + #endif diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index c8e375440403..bf896d0b2e9c 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -203,8 +203,9 @@ struct perf_event_attr { enable_on_exec : 1, /* next exec enables */ task : 1, /* trace fork/exit */ watermark : 1, /* wakeup_watermark */ + precise : 1, /* OoO invariant counter */ - __reserved_1 : 49; + __reserved_1 : 48; union { __u32 wakeup_events; /* wakeup every n events */ @@ -293,6 +294,12 @@ struct perf_event_mmap_page { #define PERF_RECORD_MISC_USER (2 << 0) #define PERF_RECORD_MISC_HYPERVISOR (3 << 0) +#define PERF_RECORD_MISC_EXACT (1 << 14) +/* + * Reserve the last bit to indicate some extended misc field + */ +#define PERF_RECORD_MISC_EXT_RESERVED (1 << 15) + struct perf_event_header { __u32 type; __u16 misc; @@ -468,6 +475,17 @@ struct perf_raw_record { void *data; }; +struct perf_branch_entry { + __u64 from; + __u64 to; + __u64 flags; +}; + +struct perf_branch_stack { + __u64 nr; + struct perf_branch_entry entries[0]; +}; + struct task_struct; /** @@ -571,6 +589,14 @@ enum perf_group_flag { PERF_GROUP_SOFTWARE = 0x1, }; +#define SWEVENT_HLIST_BITS 8 +#define SWEVENT_HLIST_SIZE (1 << SWEVENT_HLIST_BITS) + +struct swevent_hlist { + struct hlist_head heads[SWEVENT_HLIST_SIZE]; + struct rcu_head rcu_head; +}; + /** * struct perf_event - performance event kernel representation: */ @@ -579,6 +605,7 @@ struct perf_event { struct list_head group_entry; struct list_head event_entry; struct list_head sibling_list; + struct hlist_node hlist_entry; int nr_siblings; int group_flags; struct perf_event *group_leader; @@ -726,6 +753,9 @@ struct perf_cpu_context { int active_oncpu; int max_pertask; int exclusive; + struct swevent_hlist *swevent_hlist; + struct mutex hlist_mutex; + int hlist_refcount; /* * Recursion avoidance: diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index e1fb60729979..4272521e29e9 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -345,18 +345,6 @@ static inline void user_single_step_siginfo(struct task_struct *tsk, #define arch_ptrace_stop(code, info) do { } while (0) #endif -#ifndef arch_ptrace_untrace -/* - * Do machine-specific work before untracing child. - * - * This is called for a normal detach as well as from ptrace_exit() - * when the tracing task dies. - * - * Called with write_lock(&tasklist_lock) held. - */ -#define arch_ptrace_untrace(task) do { } while (0) -#endif - extern int task_current_syscall(struct task_struct *target, long *callno, unsigned long args[6], unsigned int maxargs, unsigned long *sp, unsigned long *pc); diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h index 5210a5c60877..fe1872e5b37e 100644 --- a/include/linux/rbtree.h +++ b/include/linux/rbtree.h @@ -110,6 +110,7 @@ struct rb_node struct rb_root { struct rb_node *rb_node; + void (*augment_cb)(struct rb_node *node); }; @@ -129,7 +130,9 @@ static inline void rb_set_color(struct rb_node *rb, int color) rb->rb_parent_color = (rb->rb_parent_color & ~1) | color; } -#define RB_ROOT (struct rb_root) { NULL, } +#define RB_ROOT (struct rb_root) { NULL, NULL, } +#define RB_AUGMENT_ROOT(x) (struct rb_root) { NULL, x} + #define rb_entry(ptr, type, member) container_of(ptr, type, member) #define RB_EMPTY_ROOT(root) ((root)->rb_node == NULL) diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 872a98e13d6a..07db2feb8572 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -101,10 +101,7 @@ extern struct lockdep_map rcu_sched_lock_map; # define rcu_read_release_sched() \ lock_release(&rcu_sched_lock_map, 1, _THIS_IP_) -static inline int debug_lockdep_rcu_enabled(void) -{ - return likely(rcu_scheduler_active && debug_locks); -} +extern int debug_lockdep_rcu_enabled(void); /** * rcu_read_lock_held - might we be in RCU read-side critical section? @@ -195,12 +192,30 @@ static inline int rcu_read_lock_sched_held(void) /** * rcu_dereference_check - rcu_dereference with debug checking + * @p: The pointer to read, prior to dereferencing + * @c: The conditions under which the dereference will take place + * + * Do an rcu_dereference(), but check that the conditions under which the + * dereference will take place are correct. Typically the conditions indicate + * the various locking conditions that should be held at that point. The check + * should return true if the conditions are satisfied. + * + * For example: + * + * bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() || + * lockdep_is_held(&foo->lock)); * - * Do an rcu_dereference(), but check that the context is correct. - * For example, rcu_dereference_check(gp, rcu_read_lock_held()) to - * ensure that the rcu_dereference_check() executes within an RCU - * read-side critical section. It is also possible to check for - * locks being held, for example, by using lockdep_is_held(). + * could be used to indicate to lockdep that foo->bar may only be dereferenced + * if either the RCU read lock is held, or that the lock required to replace + * the bar struct at foo->bar is held. + * + * Note that the list of conditions may also include indications of when a lock + * need not be held, for example during initialisation or destruction of the + * target struct: + * + * bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() || + * lockdep_is_held(&foo->lock) || + * atomic_read(&foo->usage) == 0); */ #define rcu_dereference_check(p, c) \ ({ \ @@ -209,13 +224,45 @@ static inline int rcu_read_lock_sched_held(void) rcu_dereference_raw(p); \ }) +/** + * rcu_dereference_protected - fetch RCU pointer when updates prevented + * + * Return the value of the specified RCU-protected pointer, but omit + * both the smp_read_barrier_depends() and the ACCESS_ONCE(). This + * is useful in cases where update-side locks prevent the value of the + * pointer from changing. Please note that this primitive does -not- + * prevent the compiler from repeating this reference or combining it + * with other references, so it should not be used without protection + * of appropriate locks. + */ +#define rcu_dereference_protected(p, c) \ + ({ \ + if (debug_lockdep_rcu_enabled() && !(c)) \ + lockdep_rcu_dereference(__FILE__, __LINE__); \ + (p); \ + }) + #else /* #ifdef CONFIG_PROVE_RCU */ #define rcu_dereference_check(p, c) rcu_dereference_raw(p) +#define rcu_dereference_protected(p, c) (p) #endif /* #else #ifdef CONFIG_PROVE_RCU */ /** + * rcu_access_pointer - fetch RCU pointer with no dereferencing + * + * Return the value of the specified RCU-protected pointer, but omit the + * smp_read_barrier_depends() and keep the ACCESS_ONCE(). This is useful + * when the value of this pointer is accessed, but the pointer is not + * dereferenced, for example, when testing an RCU-protected pointer against + * NULL. This may also be used in cases where update-side locks prevent + * the value of the pointer from changing, but rcu_dereference_protected() + * is a lighter-weight primitive for this use case. + */ +#define rcu_access_pointer(p) ACCESS_ONCE(p) + +/** * rcu_read_lock - mark the beginning of an RCU read-side critical section. * * When synchronize_rcu() is invoked on one CPU while other CPUs diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 5fcc31ed5771..c8297761e414 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -120,9 +120,11 @@ int ring_buffer_write(struct ring_buffer *buffer, unsigned long length, void *data); struct ring_buffer_event * -ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts); +ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts, + unsigned long *lost_events); struct ring_buffer_event * -ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts); +ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, + unsigned long *lost_events); struct ring_buffer_iter * ring_buffer_read_start(struct ring_buffer *buffer, int cpu); diff --git a/include/linux/sched.h b/include/linux/sched.h index dad7f668ebf7..c46b6e5ad685 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -99,7 +99,6 @@ struct futex_pi_state; struct robust_list_head; struct bio_list; struct fs_struct; -struct bts_context; struct perf_event_context; /* @@ -275,11 +274,17 @@ extern cpumask_var_t nohz_cpu_mask; #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ) extern int select_nohz_load_balancer(int cpu); extern int get_nohz_load_balancer(void); +extern int nohz_ratelimit(int cpu); #else static inline int select_nohz_load_balancer(int cpu) { return 0; } + +static inline int nohz_ratelimit(int cpu) +{ + return 0; +} #endif /* @@ -1026,12 +1031,17 @@ struct sched_domain; #define WF_SYNC 0x01 /* waker goes to sleep after wakup */ #define WF_FORK 0x02 /* child wakeup after fork */ +#define ENQUEUE_WAKEUP 1 +#define ENQUEUE_WAKING 2 +#define ENQUEUE_HEAD 4 + +#define DEQUEUE_SLEEP 1 + struct sched_class { const struct sched_class *next; - void (*enqueue_task) (struct rq *rq, struct task_struct *p, int wakeup, - bool head); - void (*dequeue_task) (struct rq *rq, struct task_struct *p, int sleep); + void (*enqueue_task) (struct rq *rq, struct task_struct *p, int flags); + void (*dequeue_task) (struct rq *rq, struct task_struct *p, int flags); void (*yield_task) (struct rq *rq); void (*check_preempt_curr) (struct rq *rq, struct task_struct *p, int flags); @@ -1040,7 +1050,8 @@ struct sched_class { void (*put_prev_task) (struct rq *rq, struct task_struct *p); #ifdef CONFIG_SMP - int (*select_task_rq)(struct task_struct *p, int sd_flag, int flags); + int (*select_task_rq)(struct rq *rq, struct task_struct *p, + int sd_flag, int flags); void (*pre_schedule) (struct rq *this_rq, struct task_struct *task); void (*post_schedule) (struct rq *this_rq); @@ -1077,36 +1088,8 @@ struct load_weight { unsigned long weight, inv_weight; }; -/* - * CFS stats for a schedulable entity (task, task-group etc) - * - * Current field usage histogram: - * - * 4 se->block_start - * 4 se->run_node - * 4 se->sleep_start - * 6 se->load.weight - */ -struct sched_entity { - struct load_weight load; /* for load-balancing */ - struct rb_node run_node; - struct list_head group_node; - unsigned int on_rq; - - u64 exec_start; - u64 sum_exec_runtime; - u64 vruntime; - u64 prev_sum_exec_runtime; - - u64 last_wakeup; - u64 avg_overlap; - - u64 nr_migrations; - - u64 start_runtime; - u64 avg_wakeup; - #ifdef CONFIG_SCHEDSTATS +struct sched_statistics { u64 wait_start; u64 wait_max; u64 wait_count; @@ -1138,6 +1121,24 @@ struct sched_entity { u64 nr_wakeups_affine_attempts; u64 nr_wakeups_passive; u64 nr_wakeups_idle; +}; +#endif + +struct sched_entity { + struct load_weight load; /* for load-balancing */ + struct rb_node run_node; + struct list_head group_node; + unsigned int on_rq; + + u64 exec_start; + u64 sum_exec_runtime; + u64 vruntime; + u64 prev_sum_exec_runtime; + + u64 nr_migrations; + +#ifdef CONFIG_SCHEDSTATS + struct sched_statistics statistics; #endif #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1272,12 +1273,6 @@ struct task_struct { struct list_head ptraced; struct list_head ptrace_entry; - /* - * This is the tracer handle for the ptrace BTS extension. - * This field actually belongs to the ptracer task. - */ - struct bts_context *bts; - /* PID/PID hash table linkage. */ struct pid_link pids[PIDTYPE_MAX]; struct list_head thread_group; @@ -1847,6 +1842,7 @@ extern void sched_clock_idle_sleep_event(void); extern void sched_clock_idle_wakeup_event(u64 delta_ns); #ifdef CONFIG_HOTPLUG_CPU +extern void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p); extern void idle_task_exit(void); #else static inline void idle_task_exit(void) {} @@ -2123,10 +2119,8 @@ extern void set_task_comm(struct task_struct *tsk, char *from); extern char *get_task_comm(char *to, struct task_struct *tsk); #ifdef CONFIG_SMP -extern void wait_task_context_switch(struct task_struct *p); extern unsigned long wait_task_inactive(struct task_struct *, long match_state); #else -static inline void wait_task_context_switch(struct task_struct *p) {} static inline unsigned long wait_task_inactive(struct task_struct *p, long match_state) { diff --git a/include/linux/time.h b/include/linux/time.h index 6e026e45a179..ea3559f0b3f2 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -150,7 +150,6 @@ extern struct timespec timespec_trunc(struct timespec t, unsigned gran); extern int timekeeping_valid_for_hres(void); extern u64 timekeeping_max_deferment(void); extern void update_wall_time(void); -extern void update_xtime_cache(u64 nsec); extern void timekeeping_leap_insert(int leapsecond); struct tms; diff --git a/include/linux/timer.h b/include/linux/timer.h index a2d1eb6cb3f0..ea965b857a50 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -10,13 +10,19 @@ struct tvec_base; struct timer_list { + /* + * All fields that change during normal runtime grouped to the + * same cacheline + */ struct list_head entry; unsigned long expires; + struct tvec_base *base; void (*function)(unsigned long); unsigned long data; - struct tvec_base *base; + int slack; + #ifdef CONFIG_TIMER_STATS void *start_site; char start_comm[16]; @@ -165,6 +171,8 @@ extern int mod_timer(struct timer_list *timer, unsigned long expires); extern int mod_timer_pending(struct timer_list *timer, unsigned long expires); extern int mod_timer_pinned(struct timer_list *timer, unsigned long expires); +extern void set_timer_slack(struct timer_list *time, int slack_hz); + #define TIMER_NOT_PINNED 0 #define TIMER_PINNED 1 /* diff --git a/include/linux/timex.h b/include/linux/timex.h index 7a082b32d8e1..32d852f8cbe4 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -232,13 +232,11 @@ struct timex { */ extern unsigned long tick_usec; /* USER_HZ period (usec) */ extern unsigned long tick_nsec; /* ACTHZ period (nsec) */ -extern int tickadj; /* amount of adjustment per tick */ /* * phase-lock loop variables */ extern int time_status; /* clock synchronization status bits */ -extern long time_adjust; /* The amount of adjtime left */ extern void ntp_init(void); extern void ntp_clear(void); @@ -271,9 +269,6 @@ extern void second_overflow(void); extern void update_ntp_one_tick(void); extern int do_adjtimex(struct timex *); -/* Don't use! Compatibility define for existing users. */ -#define tickadj (500/HZ ? : 1) - int read_current_timer(unsigned long *timer_val); /* The clock frequency of the i8253/i8254 PIT */ diff --git a/include/net/x25.h b/include/net/x25.h index 15ef9624ab75..468551ea4f1d 100644 --- a/include/net/x25.h +++ b/include/net/x25.h @@ -183,6 +183,10 @@ extern int sysctl_x25_clear_request_timeout; extern int sysctl_x25_ack_holdback_timeout; extern int sysctl_x25_forward; +extern int x25_parse_address_block(struct sk_buff *skb, + struct x25_address *called_addr, + struct x25_address *calling_addr); + extern int x25_addr_ntoa(unsigned char *, struct x25_address *, struct x25_address *); extern int x25_addr_aton(unsigned char *, struct x25_address *, diff --git a/include/trace/events/module.h b/include/trace/events/module.h index 4b0f48ba16a6..c7bb2f0482fe 100644 --- a/include/trace/events/module.h +++ b/include/trace/events/module.h @@ -51,11 +51,14 @@ TRACE_EVENT(module_free, TP_printk("%s", __get_str(name)) ); +#ifdef CONFIG_MODULE_UNLOAD +/* trace_module_get/put are only used if CONFIG_MODULE_UNLOAD is defined */ + DECLARE_EVENT_CLASS(module_refcnt, - TP_PROTO(struct module *mod, unsigned long ip, int refcnt), + TP_PROTO(struct module *mod, unsigned long ip), - TP_ARGS(mod, ip, refcnt), + TP_ARGS(mod, ip), TP_STRUCT__entry( __field( unsigned long, ip ) @@ -65,7 +68,7 @@ DECLARE_EVENT_CLASS(module_refcnt, TP_fast_assign( __entry->ip = ip; - __entry->refcnt = refcnt; + __entry->refcnt = __this_cpu_read(mod->refptr->incs) + __this_cpu_read(mod->refptr->decs); __assign_str(name, mod->name); ), @@ -75,17 +78,18 @@ DECLARE_EVENT_CLASS(module_refcnt, DEFINE_EVENT(module_refcnt, module_get, - TP_PROTO(struct module *mod, unsigned long ip, int refcnt), + TP_PROTO(struct module *mod, unsigned long ip), - TP_ARGS(mod, ip, refcnt) + TP_ARGS(mod, ip) ); DEFINE_EVENT(module_refcnt, module_put, - TP_PROTO(struct module *mod, unsigned long ip, int refcnt), + TP_PROTO(struct module *mod, unsigned long ip), - TP_ARGS(mod, ip, refcnt) + TP_ARGS(mod, ip) ); +#endif /* CONFIG_MODULE_UNLOAD */ TRACE_EVENT(module_request, diff --git a/include/trace/events/signal.h b/include/trace/events/signal.h index a510b75ac304..814566c99d29 100644 --- a/include/trace/events/signal.h +++ b/include/trace/events/signal.h @@ -100,18 +100,7 @@ TRACE_EVENT(signal_deliver, __entry->sa_handler, __entry->sa_flags) ); -/** - * signal_overflow_fail - called when signal queue is overflow - * @sig: signal number - * @group: signal to process group or not (bool) - * @info: pointer to struct siginfo - * - * Kernel fails to generate 'sig' signal with 'info' siginfo, because - * siginfo queue is overflow, and the signal is dropped. - * 'group' is not 0 if the signal will be sent to a process group. - * 'sig' is always one of RT signals. - */ -TRACE_EVENT(signal_overflow_fail, +DECLARE_EVENT_CLASS(signal_queue_overflow, TP_PROTO(int sig, int group, struct siginfo *info), @@ -135,6 +124,24 @@ TRACE_EVENT(signal_overflow_fail, ); /** + * signal_overflow_fail - called when signal queue is overflow + * @sig: signal number + * @group: signal to process group or not (bool) + * @info: pointer to struct siginfo + * + * Kernel fails to generate 'sig' signal with 'info' siginfo, because + * siginfo queue is overflow, and the signal is dropped. + * 'group' is not 0 if the signal will be sent to a process group. + * 'sig' is always one of RT signals. + */ +DEFINE_EVENT(signal_queue_overflow, signal_overflow_fail, + + TP_PROTO(int sig, int group, struct siginfo *info), + + TP_ARGS(sig, group, info) +); + +/** * signal_lose_info - called when siginfo is lost * @sig: signal number * @group: signal to process group or not (bool) @@ -145,28 +152,13 @@ TRACE_EVENT(signal_overflow_fail, * 'group' is not 0 if the signal will be sent to a process group. * 'sig' is always one of non-RT signals. */ -TRACE_EVENT(signal_lose_info, +DEFINE_EVENT(signal_queue_overflow, signal_lose_info, TP_PROTO(int sig, int group, struct siginfo *info), - TP_ARGS(sig, group, info), - - TP_STRUCT__entry( - __field( int, sig ) - __field( int, group ) - __field( int, errno ) - __field( int, code ) - ), - - TP_fast_assign( - __entry->sig = sig; - __entry->group = group; - TP_STORE_SIGINFO(__entry, info); - ), - - TP_printk("sig=%d group=%d errno=%d code=%d", - __entry->sig, __entry->group, __entry->errno, __entry->code) + TP_ARGS(sig, group, info) ); + #endif /* _TRACE_SIGNAL_H */ /* This part must be outside protection */ diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index ea6f9d4a20e9..16253db38d73 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -154,9 +154,11 @@ * * field = (typeof(field))entry; * - * p = get_cpu_var(ftrace_event_seq); + * p = &get_cpu_var(ftrace_event_seq); * trace_seq_init(p); - * ret = trace_seq_printf(s, <TP_printk> "\n"); + * ret = trace_seq_printf(s, "%s: ", <call>); + * if (ret) + * ret = trace_seq_printf(s, <TP_printk> "\n"); * put_cpu(); * if (!ret) * return TRACE_TYPE_PARTIAL_LINE; @@ -450,38 +452,38 @@ perf_trace_disable_##name(struct ftrace_event_call *unused) \ * * static void ftrace_raw_event_<call>(proto) * { + * struct ftrace_data_offsets_<call> __maybe_unused __data_offsets; * struct ring_buffer_event *event; * struct ftrace_raw_<call> *entry; <-- defined in stage 1 * struct ring_buffer *buffer; * unsigned long irq_flags; + * int __data_size; * int pc; * * local_save_flags(irq_flags); * pc = preempt_count(); * + * __data_size = ftrace_get_offsets_<call>(&__data_offsets, args); + * * event = trace_current_buffer_lock_reserve(&buffer, * event_<call>.id, - * sizeof(struct ftrace_raw_<call>), + * sizeof(*entry) + __data_size, * irq_flags, pc); * if (!event) * return; * entry = ring_buffer_event_data(event); * - * <assign>; <-- Here we assign the entries by the __field and - * __array macros. + * { <assign>; } <-- Here we assign the entries by the __field and + * __array macros. * - * trace_current_buffer_unlock_commit(buffer, event, irq_flags, pc); + * if (!filter_current_check_discard(buffer, event_call, entry, event)) + * trace_current_buffer_unlock_commit(buffer, + * event, irq_flags, pc); * } * * static int ftrace_raw_reg_event_<call>(struct ftrace_event_call *unused) * { - * int ret; - * - * ret = register_trace_<call>(ftrace_raw_event_<call>); - * if (!ret) - * pr_info("event trace: Could not activate trace point " - * "probe to <call>"); - * return ret; + * return register_trace_<call>(ftrace_raw_event_<call>); * } * * static void ftrace_unreg_event_<call>(struct ftrace_event_call *unused) @@ -493,6 +495,8 @@ perf_trace_disable_##name(struct ftrace_event_call *unused) \ * .trace = ftrace_raw_output_<call>, <-- stage 2 * }; * + * static const char print_fmt_<call>[] = <TP_printk>; + * * static struct ftrace_event_call __used * __attribute__((__aligned__(4))) * __attribute__((section("_ftrace_events"))) event_<call> = { @@ -501,6 +505,8 @@ perf_trace_disable_##name(struct ftrace_event_call *unused) \ * .raw_init = trace_event_raw_init, * .regfunc = ftrace_reg_event_<call>, * .unregfunc = ftrace_unreg_event_<call>, + * .print_fmt = print_fmt_<call>, + * .define_fields = ftrace_define_fields_<call>, * } * */ @@ -569,7 +575,6 @@ ftrace_raw_event_id_##call(struct ftrace_event_call *event_call, \ return; \ entry = ring_buffer_event_data(event); \ \ - \ tstruct \ \ { assign; } \ @@ -758,13 +763,12 @@ __attribute__((section("_ftrace_events"))) event_##call = { \ #define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print) \ static notrace void \ perf_trace_templ_##call(struct ftrace_event_call *event_call, \ - proto) \ + struct pt_regs *__regs, proto) \ { \ struct ftrace_data_offsets_##call __maybe_unused __data_offsets;\ struct ftrace_raw_##call *entry; \ u64 __addr = 0, __count = 1; \ unsigned long irq_flags; \ - struct pt_regs *__regs; \ int __entry_size; \ int __data_size; \ int rctx; \ @@ -785,20 +789,22 @@ perf_trace_templ_##call(struct ftrace_event_call *event_call, \ \ { assign; } \ \ - __regs = &__get_cpu_var(perf_trace_regs); \ - perf_fetch_caller_regs(__regs, 2); \ - \ perf_trace_buf_submit(entry, __entry_size, rctx, __addr, \ __count, irq_flags, __regs); \ } #undef DEFINE_EVENT -#define DEFINE_EVENT(template, call, proto, args) \ -static notrace void perf_trace_##call(proto) \ -{ \ - struct ftrace_event_call *event_call = &event_##call; \ - \ - perf_trace_templ_##template(event_call, args); \ +#define DEFINE_EVENT(template, call, proto, args) \ +static notrace void perf_trace_##call(proto) \ +{ \ + struct ftrace_event_call *event_call = &event_##call; \ + struct pt_regs *__regs = &get_cpu_var(perf_trace_regs); \ + \ + perf_fetch_caller_regs(__regs, 1); \ + \ + perf_trace_templ_##template(event_call, __regs, args); \ + \ + put_cpu_var(perf_trace_regs); \ } #undef DEFINE_EVENT_PRINT diff --git a/init/Kconfig b/init/Kconfig index eb77e8ccde1c..f1730180fe2e 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -604,8 +604,7 @@ config RT_GROUP_SCHED default n help This feature lets you explicitly allocate real CPU bandwidth - to users or control groups (depending on the "Basis for grouping tasks" - setting below. If enabled, it will also make it impossible to + to task groups. If enabled, it will also make it impossible to schedule realtime tasks for non-root users until you allocate realtime bandwidth for them. See Documentation/scheduler/sched-rt-group.txt for more information. @@ -942,6 +941,11 @@ config PERF_USE_VMALLOC help See tools/perf/design.txt for details +config PERF_EVENTS_NMI + bool + help + Arch has support for nmi_watchdog + menu "Kernel Performance Events And Counters" config PERF_EVENTS diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 722b0130aa94..d6c09c46ad06 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -429,7 +429,7 @@ static void wq_add(struct mqueue_inode_info *info, int sr, * sr: SEND or RECV */ static int wq_sleep(struct mqueue_inode_info *info, int sr, - long timeout, struct ext_wait_queue *ewp) + ktime_t *timeout, struct ext_wait_queue *ewp) { int retval; signed long time; @@ -440,7 +440,8 @@ static int wq_sleep(struct mqueue_inode_info *info, int sr, set_current_state(TASK_INTERRUPTIBLE); spin_unlock(&info->lock); - time = schedule_timeout(timeout); + time = schedule_hrtimeout_range_clock(timeout, + HRTIMER_MODE_ABS, 0, CLOCK_REALTIME); while (ewp->state == STATE_PENDING) cpu_relax(); @@ -552,31 +553,16 @@ static void __do_notify(struct mqueue_inode_info *info) wake_up(&info->wait_q); } -static long prepare_timeout(struct timespec *p) +static int prepare_timeout(const struct timespec __user *u_abs_timeout, + ktime_t *expires, struct timespec *ts) { - struct timespec nowts; - long timeout; - - if (p) { - if (unlikely(p->tv_nsec < 0 || p->tv_sec < 0 - || p->tv_nsec >= NSEC_PER_SEC)) - return -EINVAL; - nowts = CURRENT_TIME; - /* first subtract as jiffies can't be too big */ - p->tv_sec -= nowts.tv_sec; - if (p->tv_nsec < nowts.tv_nsec) { - p->tv_nsec += NSEC_PER_SEC; - p->tv_sec--; - } - p->tv_nsec -= nowts.tv_nsec; - if (p->tv_sec < 0) - return 0; - - timeout = timespec_to_jiffies(p) + 1; - } else - return MAX_SCHEDULE_TIMEOUT; + if (copy_from_user(ts, u_abs_timeout, sizeof(struct timespec))) + return -EFAULT; + if (!timespec_valid(ts)) + return -EINVAL; - return timeout; + *expires = timespec_to_ktime(*ts); + return 0; } static void remove_notification(struct mqueue_inode_info *info) @@ -862,22 +848,21 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, struct ext_wait_queue *receiver; struct msg_msg *msg_ptr; struct mqueue_inode_info *info; - struct timespec ts, *p = NULL; - long timeout; + ktime_t expires, *timeout = NULL; + struct timespec ts; int ret; if (u_abs_timeout) { - if (copy_from_user(&ts, u_abs_timeout, - sizeof(struct timespec))) - return -EFAULT; - p = &ts; + int res = prepare_timeout(u_abs_timeout, &expires, &ts); + if (res) + return res; + timeout = &expires; } if (unlikely(msg_prio >= (unsigned long) MQ_PRIO_MAX)) return -EINVAL; - audit_mq_sendrecv(mqdes, msg_len, msg_prio, p); - timeout = prepare_timeout(p); + audit_mq_sendrecv(mqdes, msg_len, msg_prio, timeout ? &ts : NULL); filp = fget(mqdes); if (unlikely(!filp)) { @@ -919,9 +904,6 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, if (filp->f_flags & O_NONBLOCK) { spin_unlock(&info->lock); ret = -EAGAIN; - } else if (unlikely(timeout < 0)) { - spin_unlock(&info->lock); - ret = timeout; } else { wait.task = current; wait.msg = (void *) msg_ptr; @@ -954,24 +936,23 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, size_t, msg_len, unsigned int __user *, u_msg_prio, const struct timespec __user *, u_abs_timeout) { - long timeout; ssize_t ret; struct msg_msg *msg_ptr; struct file *filp; struct inode *inode; struct mqueue_inode_info *info; struct ext_wait_queue wait; - struct timespec ts, *p = NULL; + ktime_t expires, *timeout = NULL; + struct timespec ts; if (u_abs_timeout) { - if (copy_from_user(&ts, u_abs_timeout, - sizeof(struct timespec))) - return -EFAULT; - p = &ts; + int res = prepare_timeout(u_abs_timeout, &expires, &ts); + if (res) + return res; + timeout = &expires; } - audit_mq_sendrecv(mqdes, msg_len, 0, p); - timeout = prepare_timeout(p); + audit_mq_sendrecv(mqdes, msg_len, 0, timeout ? &ts : NULL); filp = fget(mqdes); if (unlikely(!filp)) { @@ -1003,11 +984,6 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, if (filp->f_flags & O_NONBLOCK) { spin_unlock(&info->lock); ret = -EAGAIN; - msg_ptr = NULL; - } else if (unlikely(timeout < 0)) { - spin_unlock(&info->lock); - ret = timeout; - msg_ptr = NULL; } else { wait.task = current; wait.state = STATE_NONE; diff --git a/kernel/Makefile b/kernel/Makefile index a987aa1676b5..d5c30060ac14 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -77,6 +77,7 @@ obj-$(CONFIG_AUDIT_TREE) += audit_tree.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o +obj-$(CONFIG_NMI_WATCHDOG) += nmi_watchdog.o obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o diff --git a/kernel/capability.c b/kernel/capability.c index 9e4697e9b276..2f05303715a5 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -15,7 +15,6 @@ #include <linux/syscalls.h> #include <linux/pid_namespace.h> #include <asm/uaccess.h> -#include "cred-internals.h" /* * Leveraged for setting/resetting capabilities diff --git a/kernel/cpu.c b/kernel/cpu.c index 25bba73b1be3..914aedcde849 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -164,6 +164,7 @@ static inline void check_for_tasks(int cpu) } struct take_cpu_down_param { + struct task_struct *caller; unsigned long mod; void *hcpu; }; @@ -172,6 +173,7 @@ struct take_cpu_down_param { static int __ref take_cpu_down(void *_param) { struct take_cpu_down_param *param = _param; + unsigned int cpu = (unsigned long)param->hcpu; int err; /* Ensure this CPU doesn't handle any more interrupts. */ @@ -182,6 +184,8 @@ static int __ref take_cpu_down(void *_param) raw_notifier_call_chain(&cpu_chain, CPU_DYING | param->mod, param->hcpu); + if (task_cpu(param->caller) == cpu) + move_task_off_dead_cpu(cpu, param->caller); /* Force idle task to run as soon as we yield: it should immediately notice cpu is offline and die quickly. */ sched_idle_next(); @@ -192,10 +196,10 @@ static int __ref take_cpu_down(void *_param) static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) { int err, nr_calls = 0; - cpumask_var_t old_allowed; void *hcpu = (void *)(long)cpu; unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0; struct take_cpu_down_param tcd_param = { + .caller = current, .mod = mod, .hcpu = hcpu, }; @@ -206,9 +210,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) if (!cpu_online(cpu)) return -EINVAL; - if (!alloc_cpumask_var(&old_allowed, GFP_KERNEL)) - return -ENOMEM; - cpu_hotplug_begin(); set_cpu_active(cpu, false); err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod, @@ -225,10 +226,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) goto out_release; } - /* Ensure that we are not runnable on dying cpu */ - cpumask_copy(old_allowed, ¤t->cpus_allowed); - set_cpus_allowed_ptr(current, cpu_active_mask); - err = __stop_machine(take_cpu_down, &tcd_param, cpumask_of(cpu)); if (err) { set_cpu_active(cpu, true); @@ -237,7 +234,7 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) hcpu) == NOTIFY_BAD) BUG(); - goto out_allowed; + goto out_release; } BUG_ON(cpu_online(cpu)); @@ -255,8 +252,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) check_for_tasks(cpu); -out_allowed: - set_cpus_allowed_ptr(current, old_allowed); out_release: cpu_hotplug_done(); if (!err) { @@ -264,7 +259,6 @@ out_release: hcpu) == NOTIFY_BAD) BUG(); } - free_cpumask_var(old_allowed); return err; } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index d10946748ec2..9a50c5f6e727 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2182,19 +2182,52 @@ void __init cpuset_init_smp(void) void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask) { mutex_lock(&callback_mutex); - cpuset_cpus_allowed_locked(tsk, pmask); + task_lock(tsk); + guarantee_online_cpus(task_cs(tsk), pmask); + task_unlock(tsk); mutex_unlock(&callback_mutex); } -/** - * cpuset_cpus_allowed_locked - return cpus_allowed mask from a tasks cpuset. - * Must be called with callback_mutex held. - **/ -void cpuset_cpus_allowed_locked(struct task_struct *tsk, struct cpumask *pmask) +int cpuset_cpus_allowed_fallback(struct task_struct *tsk) { - task_lock(tsk); - guarantee_online_cpus(task_cs(tsk), pmask); - task_unlock(tsk); + const struct cpuset *cs; + int cpu; + + rcu_read_lock(); + cs = task_cs(tsk); + if (cs) + cpumask_copy(&tsk->cpus_allowed, cs->cpus_allowed); + rcu_read_unlock(); + + /* + * We own tsk->cpus_allowed, nobody can change it under us. + * + * But we used cs && cs->cpus_allowed lockless and thus can + * race with cgroup_attach_task() or update_cpumask() and get + * the wrong tsk->cpus_allowed. However, both cases imply the + * subsequent cpuset_change_cpumask()->set_cpus_allowed_ptr() + * which takes task_rq_lock(). + * + * If we are called after it dropped the lock we must see all + * changes in tsk_cs()->cpus_allowed. Otherwise we can temporary + * set any mask even if it is not right from task_cs() pov, + * the pending set_cpus_allowed_ptr() will fix things. + */ + + cpu = cpumask_any_and(&tsk->cpus_allowed, cpu_active_mask); + if (cpu >= nr_cpu_ids) { + /* + * Either tsk->cpus_allowed is wrong (see above) or it + * is actually empty. The latter case is only possible + * if we are racing with remove_tasks_in_empty_cpuset(). + * Like above we can temporary set any mask and rely on + * set_cpus_allowed_ptr() as synchronization point. + */ + cpumask_copy(&tsk->cpus_allowed, cpu_possible_mask); + cpu = cpumask_any(cpu_active_mask); + } + + return cpu; } void cpuset_init_current_mems_allowed(void) @@ -2383,22 +2416,6 @@ int __cpuset_node_allowed_hardwall(int node, gfp_t gfp_mask) } /** - * cpuset_lock - lock out any changes to cpuset structures - * - * The out of memory (oom) code needs to mutex_lock cpusets - * from being changed while it scans the tasklist looking for a - * task in an overlapping cpuset. Expose callback_mutex via this - * cpuset_lock() routine, so the oom code can lock it, before - * locking the task list. The tasklist_lock is a spinlock, so - * must be taken inside callback_mutex. - */ - -void cpuset_lock(void) -{ - mutex_lock(&callback_mutex); -} - -/** * cpuset_unlock - release lock on cpuset changes * * Undo the lock taken in a previous cpuset_lock() call. diff --git a/kernel/cred-internals.h b/kernel/cred-internals.h deleted file mode 100644 index 2dc4fc2d0bf1..000000000000 --- a/kernel/cred-internals.h +++ /dev/null @@ -1,21 +0,0 @@ -/* Internal credentials stuff - * - * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ - -/* - * user.c - */ -static inline void sched_switch_user(struct task_struct *p) -{ -#ifdef CONFIG_USER_SCHED - sched_move_task(p); -#endif /* CONFIG_USER_SCHED */ -} - diff --git a/kernel/cred.c b/kernel/cred.c index e1dbe9eef800..4f483be5944c 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -17,7 +17,6 @@ #include <linux/init_task.h> #include <linux/security.h> #include <linux/cn_proc.h> -#include "cred-internals.h" #if 0 #define kdebug(FMT, ...) \ @@ -558,8 +557,6 @@ int commit_creds(struct cred *new) atomic_dec(&old->user->processes); alter_cred_subscribers(old, -2); - sched_switch_user(task); - /* send notifications */ if (new->uid != old->uid || new->euid != old->euid || diff --git a/kernel/exit.c b/kernel/exit.c index 7f2683a10ac4..eabca5a73a85 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -55,7 +55,6 @@ #include <asm/unistd.h> #include <asm/pgtable.h> #include <asm/mmu_context.h> -#include "cred-internals.h" static void exit_mm(struct task_struct * tsk); diff --git a/kernel/fork.c b/kernel/fork.c index 44b0791b0a2e..5d3592deaf71 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1111,9 +1111,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->memcg_batch.do_batch = 0; p->memcg_batch.memcg = NULL; #endif - - p->bts = NULL; - p->stack_start = stack_start; /* Perform scheduler related setup. Assign this task to a CPU. */ diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 0086628b6e97..b9b134b35088 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -1749,35 +1749,15 @@ void __init hrtimers_init(void) } /** - * schedule_hrtimeout_range - sleep until timeout + * schedule_hrtimeout_range_clock - sleep until timeout * @expires: timeout value (ktime_t) * @delta: slack in expires timeout (ktime_t) * @mode: timer mode, HRTIMER_MODE_ABS or HRTIMER_MODE_REL - * - * Make the current task sleep until the given expiry time has - * elapsed. The routine will return immediately unless - * the current task state has been set (see set_current_state()). - * - * The @delta argument gives the kernel the freedom to schedule the - * actual wakeup to a time that is both power and performance friendly. - * The kernel give the normal best effort behavior for "@expires+@delta", - * but may decide to fire the timer earlier, but no earlier than @expires. - * - * You can set the task state as follows - - * - * %TASK_UNINTERRUPTIBLE - at least @timeout time is guaranteed to - * pass before the routine returns. - * - * %TASK_INTERRUPTIBLE - the routine may return early if a signal is - * delivered to the current task. - * - * The current task state is guaranteed to be TASK_RUNNING when this - * routine returns. - * - * Returns 0 when the timer has expired otherwise -EINTR + * @clock: timer clock, CLOCK_MONOTONIC or CLOCK_REALTIME */ -int __sched schedule_hrtimeout_range(ktime_t *expires, unsigned long delta, - const enum hrtimer_mode mode) +int __sched +schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta, + const enum hrtimer_mode mode, int clock) { struct hrtimer_sleeper t; @@ -1799,7 +1779,7 @@ int __sched schedule_hrtimeout_range(ktime_t *expires, unsigned long delta, return -EINTR; } - hrtimer_init_on_stack(&t.timer, CLOCK_MONOTONIC, mode); + hrtimer_init_on_stack(&t.timer, clock, mode); hrtimer_set_expires_range_ns(&t.timer, *expires, delta); hrtimer_init_sleeper(&t, current); @@ -1818,6 +1798,41 @@ int __sched schedule_hrtimeout_range(ktime_t *expires, unsigned long delta, return !t.task ? 0 : -EINTR; } + +/** + * schedule_hrtimeout_range - sleep until timeout + * @expires: timeout value (ktime_t) + * @delta: slack in expires timeout (ktime_t) + * @mode: timer mode, HRTIMER_MODE_ABS or HRTIMER_MODE_REL + * + * Make the current task sleep until the given expiry time has + * elapsed. The routine will return immediately unless + * the current task state has been set (see set_current_state()). + * + * The @delta argument gives the kernel the freedom to schedule the + * actual wakeup to a time that is both power and performance friendly. + * The kernel give the normal best effort behavior for "@expires+@delta", + * but may decide to fire the timer earlier, but no earlier than @expires. + * + * You can set the task state as follows - + * + * %TASK_UNINTERRUPTIBLE - at least @timeout time is guaranteed to + * pass before the routine returns. + * + * %TASK_INTERRUPTIBLE - the routine may return early if a signal is + * delivered to the current task. + * + * The current task state is guaranteed to be TASK_RUNNING when this + * routine returns. + * + * Returns 0 when the timer has expired otherwise -EINTR + */ +int __sched schedule_hrtimeout_range(ktime_t *expires, unsigned long delta, + const enum hrtimer_mode mode) +{ + return schedule_hrtimeout_range_clock(expires, delta, mode, + CLOCK_MONOTONIC); +} EXPORT_SYMBOL_GPL(schedule_hrtimeout_range); /** diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 2594e1ce41cb..78325f8f1139 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -431,20 +431,7 @@ static struct stack_trace lockdep_init_trace = { /* * Various lockdep statistics: */ -atomic_t chain_lookup_hits; -atomic_t chain_lookup_misses; -atomic_t hardirqs_on_events; -atomic_t hardirqs_off_events; -atomic_t redundant_hardirqs_on; -atomic_t redundant_hardirqs_off; -atomic_t softirqs_on_events; -atomic_t softirqs_off_events; -atomic_t redundant_softirqs_on; -atomic_t redundant_softirqs_off; -atomic_t nr_unused_locks; -atomic_t nr_cyclic_checks; -atomic_t nr_find_usage_forwards_checks; -atomic_t nr_find_usage_backwards_checks; +DEFINE_PER_CPU(struct lockdep_stats, lockdep_stats); #endif /* @@ -748,7 +735,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) return NULL; } class = lock_classes + nr_lock_classes++; - debug_atomic_inc(&nr_unused_locks); + debug_atomic_inc(nr_unused_locks); class->key = key; class->name = lock->name; class->subclass = subclass; @@ -1205,7 +1192,7 @@ check_noncircular(struct lock_list *root, struct lock_class *target, { int result; - debug_atomic_inc(&nr_cyclic_checks); + debug_atomic_inc(nr_cyclic_checks); result = __bfs_forwards(root, target, class_equal, target_entry); @@ -1242,7 +1229,7 @@ find_usage_forwards(struct lock_list *root, enum lock_usage_bit bit, { int result; - debug_atomic_inc(&nr_find_usage_forwards_checks); + debug_atomic_inc(nr_find_usage_forwards_checks); result = __bfs_forwards(root, (void *)bit, usage_match, target_entry); @@ -1265,7 +1252,7 @@ find_usage_backwards(struct lock_list *root, enum lock_usage_bit bit, { int result; - debug_atomic_inc(&nr_find_usage_backwards_checks); + debug_atomic_inc(nr_find_usage_backwards_checks); result = __bfs_backwards(root, (void *)bit, usage_match, target_entry); @@ -1825,7 +1812,7 @@ static inline int lookup_chain_cache(struct task_struct *curr, list_for_each_entry(chain, hash_head, entry) { if (chain->chain_key == chain_key) { cache_hit: - debug_atomic_inc(&chain_lookup_hits); + debug_atomic_inc(chain_lookup_hits); if (very_verbose(class)) printk("\nhash chain already cached, key: " "%016Lx tail class: [%p] %s\n", @@ -1890,7 +1877,7 @@ cache_hit: chain_hlocks[chain->base + j] = class - lock_classes; } list_add_tail_rcu(&chain->entry, hash_head); - debug_atomic_inc(&chain_lookup_misses); + debug_atomic_inc(chain_lookup_misses); inc_chains(); return 1; @@ -2311,7 +2298,7 @@ void trace_hardirqs_on_caller(unsigned long ip) return; if (unlikely(curr->hardirqs_enabled)) { - debug_atomic_inc(&redundant_hardirqs_on); + debug_atomic_inc(redundant_hardirqs_on); return; } /* we'll do an OFF -> ON transition: */ @@ -2338,7 +2325,7 @@ void trace_hardirqs_on_caller(unsigned long ip) curr->hardirq_enable_ip = ip; curr->hardirq_enable_event = ++curr->irq_events; - debug_atomic_inc(&hardirqs_on_events); + debug_atomic_inc(hardirqs_on_events); } EXPORT_SYMBOL(trace_hardirqs_on_caller); @@ -2370,9 +2357,9 @@ void trace_hardirqs_off_caller(unsigned long ip) curr->hardirqs_enabled = 0; curr->hardirq_disable_ip = ip; curr->hardirq_disable_event = ++curr->irq_events; - debug_atomic_inc(&hardirqs_off_events); + debug_atomic_inc(hardirqs_off_events); } else - debug_atomic_inc(&redundant_hardirqs_off); + debug_atomic_inc(redundant_hardirqs_off); } EXPORT_SYMBOL(trace_hardirqs_off_caller); @@ -2396,7 +2383,7 @@ void trace_softirqs_on(unsigned long ip) return; if (curr->softirqs_enabled) { - debug_atomic_inc(&redundant_softirqs_on); + debug_atomic_inc(redundant_softirqs_on); return; } @@ -2406,7 +2393,7 @@ void trace_softirqs_on(unsigned long ip) curr->softirqs_enabled = 1; curr->softirq_enable_ip = ip; curr->softirq_enable_event = ++curr->irq_events; - debug_atomic_inc(&softirqs_on_events); + debug_atomic_inc(softirqs_on_events); /* * We are going to turn softirqs on, so set the * usage bit for all held locks, if hardirqs are @@ -2436,10 +2423,10 @@ void trace_softirqs_off(unsigned long ip) curr->softirqs_enabled = 0; curr->softirq_disable_ip = ip; curr->softirq_disable_event = ++curr->irq_events; - debug_atomic_inc(&softirqs_off_events); + debug_atomic_inc(softirqs_off_events); DEBUG_LOCKS_WARN_ON(!softirq_count()); } else - debug_atomic_inc(&redundant_softirqs_off); + debug_atomic_inc(redundant_softirqs_off); } static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) @@ -2644,7 +2631,7 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, return 0; break; case LOCK_USED: - debug_atomic_dec(&nr_unused_locks); + debug_atomic_dec(nr_unused_locks); break; default: if (!debug_locks_off_graph_unlock()) @@ -2750,7 +2737,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (!class) return 0; } - debug_atomic_inc((atomic_t *)&class->ops); + atomic_inc((atomic_t *)&class->ops); if (very_verbose(class)) { printk("\nacquire class [%p] %s", class->key, class->name); if (class->name_version > 1) diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index a2ee95ad1313..8d7d4b6c741a 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -110,29 +110,61 @@ lockdep_count_backward_deps(struct lock_class *class) #endif #ifdef CONFIG_DEBUG_LOCKDEP + +#include <asm/local.h> /* - * Various lockdep statistics: + * Various lockdep statistics. + * We want them per cpu as they are often accessed in fast path + * and we want to avoid too much cache bouncing. */ -extern atomic_t chain_lookup_hits; -extern atomic_t chain_lookup_misses; -extern atomic_t hardirqs_on_events; -extern atomic_t hardirqs_off_events; -extern atomic_t redundant_hardirqs_on; -extern atomic_t redundant_hardirqs_off; -extern atomic_t softirqs_on_events; -extern atomic_t softirqs_off_events; -extern atomic_t redundant_softirqs_on; -extern atomic_t redundant_softirqs_off; -extern atomic_t nr_unused_locks; -extern atomic_t nr_cyclic_checks; -extern atomic_t nr_cyclic_check_recursions; -extern atomic_t nr_find_usage_forwards_checks; -extern atomic_t nr_find_usage_forwards_recursions; -extern atomic_t nr_find_usage_backwards_checks; -extern atomic_t nr_find_usage_backwards_recursions; -# define debug_atomic_inc(ptr) atomic_inc(ptr) -# define debug_atomic_dec(ptr) atomic_dec(ptr) -# define debug_atomic_read(ptr) atomic_read(ptr) +struct lockdep_stats { + int chain_lookup_hits; + int chain_lookup_misses; + int hardirqs_on_events; + int hardirqs_off_events; + int redundant_hardirqs_on; + int redundant_hardirqs_off; + int softirqs_on_events; + int softirqs_off_events; + int redundant_softirqs_on; + int redundant_softirqs_off; + int nr_unused_locks; + int nr_cyclic_checks; + int nr_cyclic_check_recursions; + int nr_find_usage_forwards_checks; + int nr_find_usage_forwards_recursions; + int nr_find_usage_backwards_checks; + int nr_find_usage_backwards_recursions; +}; + +DECLARE_PER_CPU(struct lockdep_stats, lockdep_stats); + +#define debug_atomic_inc(ptr) { \ + struct lockdep_stats *__cpu_lockdep_stats; \ + \ + WARN_ON_ONCE(!irqs_disabled()); \ + __cpu_lockdep_stats = &__get_cpu_var(lockdep_stats); \ + __cpu_lockdep_stats->ptr++; \ +} + +#define debug_atomic_dec(ptr) { \ + struct lockdep_stats *__cpu_lockdep_stats; \ + \ + WARN_ON_ONCE(!irqs_disabled()); \ + __cpu_lockdep_stats = &__get_cpu_var(lockdep_stats); \ + __cpu_lockdep_stats->ptr--; \ +} + +#define debug_atomic_read(ptr) ({ \ + struct lockdep_stats *__cpu_lockdep_stats; \ + unsigned long long __total = 0; \ + int __cpu; \ + for_each_possible_cpu(__cpu) { \ + __cpu_lockdep_stats = &per_cpu(lockdep_stats, __cpu); \ + __total += __cpu_lockdep_stats->ptr; \ + } \ + __total; \ +}) #else # define debug_atomic_inc(ptr) do { } while (0) # define debug_atomic_dec(ptr) do { } while (0) diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index d4aba4f3584c..59b76c8ce9d7 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -184,34 +184,34 @@ static const struct file_operations proc_lockdep_chains_operations = { static void lockdep_stats_debug_show(struct seq_file *m) { #ifdef CONFIG_DEBUG_LOCKDEP - unsigned int hi1 = debug_atomic_read(&hardirqs_on_events), - hi2 = debug_atomic_read(&hardirqs_off_events), - hr1 = debug_atomic_read(&redundant_hardirqs_on), - hr2 = debug_atomic_read(&redundant_hardirqs_off), - si1 = debug_atomic_read(&softirqs_on_events), - si2 = debug_atomic_read(&softirqs_off_events), - sr1 = debug_atomic_read(&redundant_softirqs_on), - sr2 = debug_atomic_read(&redundant_softirqs_off); - - seq_printf(m, " chain lookup misses: %11u\n", - debug_atomic_read(&chain_lookup_misses)); - seq_printf(m, " chain lookup hits: %11u\n", - debug_atomic_read(&chain_lookup_hits)); - seq_printf(m, " cyclic checks: %11u\n", - debug_atomic_read(&nr_cyclic_checks)); - seq_printf(m, " find-mask forwards checks: %11u\n", - debug_atomic_read(&nr_find_usage_forwards_checks)); - seq_printf(m, " find-mask backwards checks: %11u\n", - debug_atomic_read(&nr_find_usage_backwards_checks)); - - seq_printf(m, " hardirq on events: %11u\n", hi1); - seq_printf(m, " hardirq off events: %11u\n", hi2); - seq_printf(m, " redundant hardirq ons: %11u\n", hr1); - seq_printf(m, " redundant hardirq offs: %11u\n", hr2); - seq_printf(m, " softirq on events: %11u\n", si1); - seq_printf(m, " softirq off events: %11u\n", si2); - seq_printf(m, " redundant softirq ons: %11u\n", sr1); - seq_printf(m, " redundant softirq offs: %11u\n", sr2); + unsigned long long hi1 = debug_atomic_read(hardirqs_on_events), + hi2 = debug_atomic_read(hardirqs_off_events), + hr1 = debug_atomic_read(redundant_hardirqs_on), + hr2 = debug_atomic_read(redundant_hardirqs_off), + si1 = debug_atomic_read(softirqs_on_events), + si2 = debug_atomic_read(softirqs_off_events), + sr1 = debug_atomic_read(redundant_softirqs_on), + sr2 = debug_atomic_read(redundant_softirqs_off); + + seq_printf(m, " chain lookup misses: %11llu\n", + debug_atomic_read(chain_lookup_misses)); + seq_printf(m, " chain lookup hits: %11llu\n", + debug_atomic_read(chain_lookup_hits)); + seq_printf(m, " cyclic checks: %11llu\n", + debug_atomic_read(nr_cyclic_checks)); + seq_printf(m, " find-mask forwards checks: %11llu\n", + debug_atomic_read(nr_find_usage_forwards_checks)); + seq_printf(m, " find-mask backwards checks: %11llu\n", + debug_atomic_read(nr_find_usage_backwards_checks)); + + seq_printf(m, " hardirq on events: %11llu\n", hi1); + seq_printf(m, " hardirq off events: %11llu\n", hi2); + seq_printf(m, " redundant hardirq ons: %11llu\n", hr1); + seq_printf(m, " redundant hardirq offs: %11llu\n", hr2); + seq_printf(m, " softirq on events: %11llu\n", si1); + seq_printf(m, " softirq off events: %11llu\n", si2); + seq_printf(m, " redundant softirq ons: %11llu\n", sr1); + seq_printf(m, " redundant softirq offs: %11llu\n", sr2); #endif } @@ -263,7 +263,7 @@ static int lockdep_stats_show(struct seq_file *m, void *v) #endif } #ifdef CONFIG_DEBUG_LOCKDEP - DEBUG_LOCKS_WARN_ON(debug_atomic_read(&nr_unused_locks) != nr_unused); + DEBUG_LOCKS_WARN_ON(debug_atomic_read(nr_unused_locks) != nr_unused); #endif seq_printf(m, " lock-classes: %11lu [max: %lu]\n", nr_lock_classes, MAX_LOCKDEP_KEYS); diff --git a/kernel/module.c b/kernel/module.c index 1016b75b026a..b8a1e313448c 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -59,8 +59,6 @@ #define CREATE_TRACE_POINTS #include <trace/events/module.h> -EXPORT_TRACEPOINT_SYMBOL(module_get); - #if 0 #define DEBUGP printk #else @@ -515,6 +513,9 @@ MODINFO_ATTR(srcversion); static char last_unloaded_module[MODULE_NAME_LEN+1]; #ifdef CONFIG_MODULE_UNLOAD + +EXPORT_TRACEPOINT_SYMBOL(module_get); + /* Init the unload section of the module. */ static void module_unload_init(struct module *mod) { @@ -867,8 +868,7 @@ void module_put(struct module *module) smp_wmb(); /* see comment in module_refcount */ __this_cpu_inc(module->refptr->decs); - trace_module_put(module, _RET_IP_, - __this_cpu_read(module->refptr->decs)); + trace_module_put(module, _RET_IP_); /* Maybe they're waiting for us to drop reference? */ if (unlikely(!module_is_live(module))) wake_up_process(module->waiter); diff --git a/kernel/nmi_watchdog.c b/kernel/nmi_watchdog.c new file mode 100644 index 000000000000..a79d211c30df --- /dev/null +++ b/kernel/nmi_watchdog.c @@ -0,0 +1,259 @@ +/* + * Detect Hard Lockups using the NMI + * + * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc. + * + * this code detects hard lockups: incidents in where on a CPU + * the kernel does not respond to anything except NMI. + * + * Note: Most of this code is borrowed heavily from softlockup.c, + * so thanks to Ingo for the initial implementation. + * Some chunks also taken from arch/x86/kernel/apic/nmi.c, thanks + * to those contributors as well. + */ + +#include <linux/mm.h> +#include <linux/cpu.h> +#include <linux/nmi.h> +#include <linux/init.h> +#include <linux/delay.h> +#include <linux/freezer.h> +#include <linux/lockdep.h> +#include <linux/notifier.h> +#include <linux/module.h> +#include <linux/sysctl.h> + +#include <asm/irq_regs.h> +#include <linux/perf_event.h> + +static DEFINE_PER_CPU(struct perf_event *, nmi_watchdog_ev); +static DEFINE_PER_CPU(int, nmi_watchdog_touch); +static DEFINE_PER_CPU(long, alert_counter); + +static int panic_on_timeout; + +void touch_nmi_watchdog(void) +{ + __raw_get_cpu_var(nmi_watchdog_touch) = 1; + touch_softlockup_watchdog(); +} +EXPORT_SYMBOL(touch_nmi_watchdog); + +void touch_all_nmi_watchdog(void) +{ + int cpu; + + for_each_online_cpu(cpu) + per_cpu(nmi_watchdog_touch, cpu) = 1; + touch_softlockup_watchdog(); +} + +static int __init setup_nmi_watchdog(char *str) +{ + if (!strncmp(str, "panic", 5)) { + panic_on_timeout = 1; + str = strchr(str, ','); + if (!str) + return 1; + ++str; + } + return 1; +} +__setup("nmi_watchdog=", setup_nmi_watchdog); + +struct perf_event_attr wd_hw_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .size = sizeof(struct perf_event_attr), + .pinned = 1, + .disabled = 1, +}; + +struct perf_event_attr wd_sw_attr = { + .type = PERF_TYPE_SOFTWARE, + .config = PERF_COUNT_SW_CPU_CLOCK, + .size = sizeof(struct perf_event_attr), + .pinned = 1, + .disabled = 1, +}; + +void wd_overflow(struct perf_event *event, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + int cpu = smp_processor_id(); + int touched = 0; + + if (__get_cpu_var(nmi_watchdog_touch)) { + per_cpu(nmi_watchdog_touch, cpu) = 0; + touched = 1; + } + + /* check to see if the cpu is doing anything */ + if (!touched && hw_nmi_is_cpu_stuck(regs)) { + /* + * Ayiee, looks like this CPU is stuck ... + * wait a few IRQs (5 seconds) before doing the oops ... + */ + per_cpu(alert_counter, cpu) += 1; + if (per_cpu(alert_counter, cpu) == 5) { + if (panic_on_timeout) + panic("NMI Watchdog detected LOCKUP on cpu %d", cpu); + else + WARN(1, "NMI Watchdog detected LOCKUP on cpu %d", cpu); + } + } else { + per_cpu(alert_counter, cpu) = 0; + } + + return; +} + +static int enable_nmi_watchdog(int cpu) +{ + struct perf_event *event; + struct perf_event_attr *wd_attr; + + event = per_cpu(nmi_watchdog_ev, cpu); + if (event && event->state > PERF_EVENT_STATE_OFF) + return 0; + + if (event == NULL) { + /* Try to register using hardware perf events first */ + wd_attr = &wd_hw_attr; + wd_attr->sample_period = hw_nmi_get_sample_period(); + event = perf_event_create_kernel_counter(wd_attr, cpu, -1, wd_overflow); + if (IS_ERR(event)) { + /* hardware doesn't exist or not supported, fallback to software events */ + printk(KERN_INFO "nmi_watchdog: hardware not available, trying software events\n"); + wd_attr = &wd_sw_attr; + wd_attr->sample_period = NSEC_PER_SEC; + event = perf_event_create_kernel_counter(wd_attr, cpu, -1, wd_overflow); + if (IS_ERR(event)) { + printk(KERN_ERR "nmi watchdog failed to create perf event on %i: %p\n", cpu, event); + return -1; + } + } + per_cpu(nmi_watchdog_ev, cpu) = event; + } + perf_event_enable(per_cpu(nmi_watchdog_ev, cpu)); + return 0; +} + +static void disable_nmi_watchdog(int cpu) +{ + struct perf_event *event; + + event = per_cpu(nmi_watchdog_ev, cpu); + if (event) { + perf_event_disable(per_cpu(nmi_watchdog_ev, cpu)); + per_cpu(nmi_watchdog_ev, cpu) = NULL; + perf_event_release_kernel(event); + } +} + +#ifdef CONFIG_SYSCTL +/* + * proc handler for /proc/sys/kernel/nmi_watchdog + */ +int nmi_watchdog_enabled; + +int proc_nmi_enabled(struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) +{ + int cpu; + + if (!write) { + struct perf_event *event; + for_each_online_cpu(cpu) { + event = per_cpu(nmi_watchdog_ev, cpu); + if (event && event->state > PERF_EVENT_STATE_OFF) { + nmi_watchdog_enabled = 1; + break; + } + } + proc_dointvec(table, write, buffer, length, ppos); + return 0; + } + + touch_all_nmi_watchdog(); + proc_dointvec(table, write, buffer, length, ppos); + if (nmi_watchdog_enabled) { + for_each_online_cpu(cpu) + if (enable_nmi_watchdog(cpu)) { + printk(KERN_ERR "NMI watchdog failed configuration, " + " can not be enabled\n"); + } + } else { + for_each_online_cpu(cpu) + disable_nmi_watchdog(cpu); + } + return 0; +} + +#endif /* CONFIG_SYSCTL */ + +/* + * Create/destroy watchdog threads as CPUs come and go: + */ +static int __cpuinit +cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) +{ + int hotcpu = (unsigned long)hcpu; + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + per_cpu(nmi_watchdog_touch, hotcpu) = 0; + break; + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + if (enable_nmi_watchdog(hotcpu)) + return NOTIFY_BAD; + break; +#ifdef CONFIG_HOTPLUG_CPU + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + disable_nmi_watchdog(hotcpu); + case CPU_DEAD: + case CPU_DEAD_FROZEN: + break; +#endif /* CONFIG_HOTPLUG_CPU */ + } + return NOTIFY_OK; +} + +static struct notifier_block __cpuinitdata cpu_nfb = { + .notifier_call = cpu_callback +}; + +static int __initdata nonmi_watchdog; + +static int __init nonmi_watchdog_setup(char *str) +{ + nonmi_watchdog = 1; + return 1; +} +__setup("nonmi_watchdog", nonmi_watchdog_setup); + +static int __init spawn_nmi_watchdog_task(void) +{ + void *cpu = (void *)(long)smp_processor_id(); + int err; + + if (nonmi_watchdog) + return 0; + + printk(KERN_INFO "NMI watchdog enabled, takes one hw-pmu counter.\n"); + + err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); + if (err == NOTIFY_BAD) { + BUG(); + return 1; + } + cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); + register_cpu_notifier(&cpu_nfb); + + return 0; +} +early_initcall(spawn_nmi_watchdog_task); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 2f3fbf84215a..07b7a435bf03 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -16,6 +16,7 @@ #include <linux/file.h> #include <linux/poll.h> #include <linux/slab.h> +#include <linux/hash.h> #include <linux/sysfs.h> #include <linux/dcache.h> #include <linux/percpu.h> @@ -1367,6 +1368,8 @@ void perf_event_task_sched_in(struct task_struct *task) if (cpuctx->task_ctx == ctx) return; + perf_disable(); + /* * We want to keep the following priority order: * cpu pinned (that don't need to move), task pinned, @@ -1379,6 +1382,8 @@ void perf_event_task_sched_in(struct task_struct *task) ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE); cpuctx->task_ctx = ctx; + + perf_enable(); } #define MAX_INTERRUPTS (~0ULL) @@ -2642,6 +2647,7 @@ static int perf_fasync(int fd, struct file *filp, int on) } static const struct file_operations perf_fops = { + .llseek = no_llseek, .release = perf_release, .read = perf_read, .poll = perf_poll, @@ -3961,36 +3967,6 @@ static void perf_swevent_add(struct perf_event *event, u64 nr, perf_swevent_overflow(event, 0, nmi, data, regs); } -static int perf_swevent_is_counting(struct perf_event *event) -{ - /* - * The event is active, we're good! - */ - if (event->state == PERF_EVENT_STATE_ACTIVE) - return 1; - - /* - * The event is off/error, not counting. - */ - if (event->state != PERF_EVENT_STATE_INACTIVE) - return 0; - - /* - * The event is inactive, if the context is active - * we're part of a group that didn't make it on the 'pmu', - * not counting. - */ - if (event->ctx->is_active) - return 0; - - /* - * We're inactive and the context is too, this means the - * task is scheduled out, we're counting events that happen - * to us, like migration events. - */ - return 1; -} - static int perf_tp_event_match(struct perf_event *event, struct perf_sample_data *data); @@ -4014,12 +3990,6 @@ static int perf_swevent_match(struct perf_event *event, struct perf_sample_data *data, struct pt_regs *regs) { - if (event->cpu != -1 && event->cpu != smp_processor_id()) - return 0; - - if (!perf_swevent_is_counting(event)) - return 0; - if (event->attr.type != type) return 0; @@ -4036,18 +4006,53 @@ static int perf_swevent_match(struct perf_event *event, return 1; } -static void perf_swevent_ctx_event(struct perf_event_context *ctx, - enum perf_type_id type, - u32 event_id, u64 nr, int nmi, - struct perf_sample_data *data, - struct pt_regs *regs) +static inline u64 swevent_hash(u64 type, u32 event_id) +{ + u64 val = event_id | (type << 32); + + return hash_64(val, SWEVENT_HLIST_BITS); +} + +static struct hlist_head * +find_swevent_head(struct perf_cpu_context *ctx, u64 type, u32 event_id) { + u64 hash; + struct swevent_hlist *hlist; + + hash = swevent_hash(type, event_id); + + hlist = rcu_dereference(ctx->swevent_hlist); + if (!hlist) + return NULL; + + return &hlist->heads[hash]; +} + +static void do_perf_sw_event(enum perf_type_id type, u32 event_id, + u64 nr, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct perf_cpu_context *cpuctx; struct perf_event *event; + struct hlist_node *node; + struct hlist_head *head; - list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + cpuctx = &__get_cpu_var(perf_cpu_context); + + rcu_read_lock(); + + head = find_swevent_head(cpuctx, type, event_id); + + if (!head) + goto end; + + hlist_for_each_entry_rcu(event, node, head, hlist_entry) { if (perf_swevent_match(event, type, event_id, data, regs)) perf_swevent_add(event, nr, nmi, data, regs); } +end: + rcu_read_unlock(); } int perf_swevent_get_recursion_context(void) @@ -4085,27 +4090,6 @@ void perf_swevent_put_recursion_context(int rctx) } EXPORT_SYMBOL_GPL(perf_swevent_put_recursion_context); -static void do_perf_sw_event(enum perf_type_id type, u32 event_id, - u64 nr, int nmi, - struct perf_sample_data *data, - struct pt_regs *regs) -{ - struct perf_cpu_context *cpuctx; - struct perf_event_context *ctx; - - cpuctx = &__get_cpu_var(perf_cpu_context); - rcu_read_lock(); - perf_swevent_ctx_event(&cpuctx->ctx, type, event_id, - nr, nmi, data, regs); - /* - * doesn't really matter which of the child contexts the - * events ends up in. - */ - ctx = rcu_dereference(current->perf_event_ctxp); - if (ctx) - perf_swevent_ctx_event(ctx, type, event_id, nr, nmi, data, regs); - rcu_read_unlock(); -} void __perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr) @@ -4131,16 +4115,28 @@ static void perf_swevent_read(struct perf_event *event) static int perf_swevent_enable(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; + struct perf_cpu_context *cpuctx; + struct hlist_head *head; + + cpuctx = &__get_cpu_var(perf_cpu_context); if (hwc->sample_period) { hwc->last_period = hwc->sample_period; perf_swevent_set_period(event); } + + head = find_swevent_head(cpuctx, event->attr.type, event->attr.config); + if (WARN_ON_ONCE(!head)) + return -EINVAL; + + hlist_add_head_rcu(&event->hlist_entry, head); + return 0; } static void perf_swevent_disable(struct perf_event *event) { + hlist_del_rcu(&event->hlist_entry); } static const struct pmu perf_ops_generic = { @@ -4168,15 +4164,8 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) perf_sample_data_init(&data, 0); data.period = event->hw.last_period; regs = get_irq_regs(); - /* - * In case we exclude kernel IPs or are somehow not in interrupt - * context, provide the next best thing, the user IP. - */ - if ((event->attr.exclude_kernel || !regs) && - !event->attr.exclude_user) - regs = task_pt_regs(current); - if (regs) { + if (regs && !perf_exclude_event(event, regs)) { if (!(event->attr.exclude_idle && current->pid == 0)) if (perf_event_overflow(event, 0, &data, regs)) ret = HRTIMER_NORESTART; @@ -4324,6 +4313,105 @@ static const struct pmu perf_ops_task_clock = { .read = task_clock_perf_event_read, }; +static void swevent_hlist_release_rcu(struct rcu_head *rcu_head) +{ + struct swevent_hlist *hlist; + + hlist = container_of(rcu_head, struct swevent_hlist, rcu_head); + kfree(hlist); +} + +static void swevent_hlist_release(struct perf_cpu_context *cpuctx) +{ + struct swevent_hlist *hlist; + + if (!cpuctx->swevent_hlist) + return; + + hlist = cpuctx->swevent_hlist; + rcu_assign_pointer(cpuctx->swevent_hlist, NULL); + call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu); +} + +static void swevent_hlist_put_cpu(struct perf_event *event, int cpu) +{ + struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + + mutex_lock(&cpuctx->hlist_mutex); + + if (!--cpuctx->hlist_refcount) + swevent_hlist_release(cpuctx); + + mutex_unlock(&cpuctx->hlist_mutex); +} + +static void swevent_hlist_put(struct perf_event *event) +{ + int cpu; + + if (event->cpu != -1) { + swevent_hlist_put_cpu(event, event->cpu); + return; + } + + for_each_possible_cpu(cpu) + swevent_hlist_put_cpu(event, cpu); +} + +static int swevent_hlist_get_cpu(struct perf_event *event, int cpu) +{ + struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + int err = 0; + + mutex_lock(&cpuctx->hlist_mutex); + + if (!cpuctx->swevent_hlist && cpu_online(cpu)) { + struct swevent_hlist *hlist; + + hlist = kzalloc(sizeof(*hlist), GFP_KERNEL); + if (!hlist) { + err = -ENOMEM; + goto exit; + } + rcu_assign_pointer(cpuctx->swevent_hlist, hlist); + } + cpuctx->hlist_refcount++; + exit: + mutex_unlock(&cpuctx->hlist_mutex); + + return err; +} + +static int swevent_hlist_get(struct perf_event *event) +{ + int err; + int cpu, failed_cpu; + + if (event->cpu != -1) + return swevent_hlist_get_cpu(event, event->cpu); + + get_online_cpus(); + for_each_possible_cpu(cpu) { + err = swevent_hlist_get_cpu(event, cpu); + if (err) { + failed_cpu = cpu; + goto fail; + } + } + put_online_cpus(); + + return 0; + fail: + for_each_possible_cpu(cpu) { + if (cpu == failed_cpu) + break; + swevent_hlist_put_cpu(event, cpu); + } + + put_online_cpus(); + return err; +} + #ifdef CONFIG_EVENT_TRACING void perf_tp_event(int event_id, u64 addr, u64 count, void *record, @@ -4357,10 +4445,13 @@ static int perf_tp_event_match(struct perf_event *event, static void tp_perf_event_destroy(struct perf_event *event) { perf_trace_disable(event->attr.config); + swevent_hlist_put(event); } static const struct pmu *tp_perf_event_init(struct perf_event *event) { + int err; + /* * Raw tracepoint data is a severe data leak, only allow root to * have these. @@ -4374,6 +4465,11 @@ static const struct pmu *tp_perf_event_init(struct perf_event *event) return NULL; event->destroy = tp_perf_event_destroy; + err = swevent_hlist_get(event); + if (err) { + perf_trace_disable(event->attr.config); + return ERR_PTR(err); + } return &perf_ops_generic; } @@ -4474,6 +4570,7 @@ static void sw_perf_event_destroy(struct perf_event *event) WARN_ON(event->parent); atomic_dec(&perf_swevent_enabled[event_id]); + swevent_hlist_put(event); } static const struct pmu *sw_perf_event_init(struct perf_event *event) @@ -4512,6 +4609,12 @@ static const struct pmu *sw_perf_event_init(struct perf_event *event) case PERF_COUNT_SW_ALIGNMENT_FAULTS: case PERF_COUNT_SW_EMULATION_FAULTS: if (!event->parent) { + int err; + + err = swevent_hlist_get(event); + if (err) + return ERR_PTR(err); + atomic_inc(&perf_swevent_enabled[event_id]); event->destroy = sw_perf_event_destroy; } @@ -5384,6 +5487,7 @@ static void __init perf_event_init_all_cpus(void) for_each_possible_cpu(cpu) { cpuctx = &per_cpu(perf_cpu_context, cpu); + mutex_init(&cpuctx->hlist_mutex); __perf_event_init_context(&cpuctx->ctx, NULL); } } @@ -5397,6 +5501,16 @@ static void __cpuinit perf_event_init_cpu(int cpu) spin_lock(&perf_resource_lock); cpuctx->max_pertask = perf_max_events - perf_reserved_percpu; spin_unlock(&perf_resource_lock); + + mutex_lock(&cpuctx->hlist_mutex); + if (cpuctx->hlist_refcount > 0) { + struct swevent_hlist *hlist; + + hlist = kzalloc(sizeof(*hlist), GFP_KERNEL); + WARN_ON_ONCE(!hlist); + rcu_assign_pointer(cpuctx->swevent_hlist, hlist); + } + mutex_unlock(&cpuctx->hlist_mutex); } #ifdef CONFIG_HOTPLUG_CPU @@ -5416,6 +5530,10 @@ static void perf_event_exit_cpu(int cpu) struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); struct perf_event_context *ctx = &cpuctx->ctx; + mutex_lock(&cpuctx->hlist_mutex); + swevent_hlist_release(cpuctx); + mutex_unlock(&cpuctx->hlist_mutex); + mutex_lock(&ctx->mutex); smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1); mutex_unlock(&ctx->mutex); diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index bc7704b3a443..799f360d1475 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -11,19 +11,18 @@ #include <trace/events/timer.h> /* - * Called after updating RLIMIT_CPU to set timer expiration if necessary. + * Called after updating RLIMIT_CPU to run cpu timer and update + * tsk->signal->cputime_expires expiration cache if necessary. Needs + * siglock protection since other code may update expiration cache as + * well. */ void update_rlimit_cpu(unsigned long rlim_new) { cputime_t cputime = secs_to_cputime(rlim_new); - struct signal_struct *const sig = current->signal; - if (cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) || - cputime_gt(sig->it[CPUCLOCK_PROF].expires, cputime)) { - spin_lock_irq(¤t->sighand->siglock); - set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL); - spin_unlock_irq(¤t->sighand->siglock); - } + spin_lock_irq(¤t->sighand->siglock); + set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL); + spin_unlock_irq(¤t->sighand->siglock); } static int check_clock(const clockid_t which_clock) @@ -548,111 +547,62 @@ static inline int expires_gt(cputime_t expires, cputime_t new_exp) cputime_gt(expires, new_exp); } -static inline int expires_le(cputime_t expires, cputime_t new_exp) -{ - return !cputime_eq(expires, cputime_zero) && - cputime_le(expires, new_exp); -} /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held - * for reading, and interrupts disabled. + * for reading, interrupts disabled and p->sighand->siglock taken. */ -static void arm_timer(struct k_itimer *timer, union cpu_time_count now) +static void arm_timer(struct k_itimer *timer) { struct task_struct *p = timer->it.cpu.task; struct list_head *head, *listpos; + struct task_cputime *cputime_expires; struct cpu_timer_list *const nt = &timer->it.cpu; struct cpu_timer_list *next; - unsigned long i; - head = (CPUCLOCK_PERTHREAD(timer->it_clock) ? - p->cpu_timers : p->signal->cpu_timers); + if (CPUCLOCK_PERTHREAD(timer->it_clock)) { + head = p->cpu_timers; + cputime_expires = &p->cputime_expires; + } else { + head = p->signal->cpu_timers; + cputime_expires = &p->signal->cputime_expires; + } head += CPUCLOCK_WHICH(timer->it_clock); - BUG_ON(!irqs_disabled()); - spin_lock(&p->sighand->siglock); - listpos = head; - if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) { - list_for_each_entry(next, head, entry) { - if (next->expires.sched > nt->expires.sched) - break; - listpos = &next->entry; - } - } else { - list_for_each_entry(next, head, entry) { - if (cputime_gt(next->expires.cpu, nt->expires.cpu)) - break; - listpos = &next->entry; - } + list_for_each_entry(next, head, entry) { + if (cpu_time_before(timer->it_clock, nt->expires, next->expires)) + break; + listpos = &next->entry; } list_add(&nt->entry, listpos); if (listpos == head) { + union cpu_time_count *exp = &nt->expires; + /* - * We are the new earliest-expiring timer. - * If we are a thread timer, there can always - * be a process timer telling us to stop earlier. + * We are the new earliest-expiring POSIX 1.b timer, hence + * need to update expiration cache. Take into account that + * for process timers we share expiration cache with itimers + * and RLIMIT_CPU and for thread timers with RLIMIT_RTTIME. */ - if (CPUCLOCK_PERTHREAD(timer->it_clock)) { - union cpu_time_count *exp = &nt->expires; - - switch (CPUCLOCK_WHICH(timer->it_clock)) { - default: - BUG(); - case CPUCLOCK_PROF: - if (expires_gt(p->cputime_expires.prof_exp, - exp->cpu)) - p->cputime_expires.prof_exp = exp->cpu; - break; - case CPUCLOCK_VIRT: - if (expires_gt(p->cputime_expires.virt_exp, - exp->cpu)) - p->cputime_expires.virt_exp = exp->cpu; - break; - case CPUCLOCK_SCHED: - if (p->cputime_expires.sched_exp == 0 || - p->cputime_expires.sched_exp > exp->sched) - p->cputime_expires.sched_exp = - exp->sched; - break; - } - } else { - struct signal_struct *const sig = p->signal; - union cpu_time_count *exp = &timer->it.cpu.expires; - - /* - * For a process timer, set the cached expiration time. - */ - switch (CPUCLOCK_WHICH(timer->it_clock)) { - default: - BUG(); - case CPUCLOCK_VIRT: - if (expires_le(sig->it[CPUCLOCK_VIRT].expires, - exp->cpu)) - break; - sig->cputime_expires.virt_exp = exp->cpu; - break; - case CPUCLOCK_PROF: - if (expires_le(sig->it[CPUCLOCK_PROF].expires, - exp->cpu)) - break; - i = sig->rlim[RLIMIT_CPU].rlim_cur; - if (i != RLIM_INFINITY && - i <= cputime_to_secs(exp->cpu)) - break; - sig->cputime_expires.prof_exp = exp->cpu; - break; - case CPUCLOCK_SCHED: - sig->cputime_expires.sched_exp = exp->sched; - break; - } + switch (CPUCLOCK_WHICH(timer->it_clock)) { + case CPUCLOCK_PROF: + if (expires_gt(cputime_expires->prof_exp, exp->cpu)) + cputime_expires->prof_exp = exp->cpu; + break; + case CPUCLOCK_VIRT: + if (expires_gt(cputime_expires->virt_exp, exp->cpu)) + cputime_expires->virt_exp = exp->cpu; + break; + case CPUCLOCK_SCHED: + if (cputime_expires->sched_exp == 0 || + cputime_expires->sched_exp > exp->sched) + cputime_expires->sched_exp = exp->sched; + break; } } - - spin_unlock(&p->sighand->siglock); } /* @@ -660,7 +610,12 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) */ static void cpu_timer_fire(struct k_itimer *timer) { - if (unlikely(timer->sigq == NULL)) { + if ((timer->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE) { + /* + * User don't want any signal. + */ + timer->it.cpu.expires.sched = 0; + } else if (unlikely(timer->sigq == NULL)) { /* * This a special case for clock_nanosleep, * not a normal timer from sys_timer_create. @@ -721,7 +676,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, struct itimerspec *new, struct itimerspec *old) { struct task_struct *p = timer->it.cpu.task; - union cpu_time_count old_expires, new_expires, val; + union cpu_time_count old_expires, new_expires, old_incr, val; int ret; if (unlikely(p == NULL)) { @@ -752,6 +707,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, BUG_ON(!irqs_disabled()); ret = 0; + old_incr = timer->it.cpu.incr; spin_lock(&p->sighand->siglock); old_expires = timer->it.cpu.expires; if (unlikely(timer->it.cpu.firing)) { @@ -759,7 +715,6 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, ret = TIMER_RETRY; } else list_del_init(&timer->it.cpu.entry); - spin_unlock(&p->sighand->siglock); /* * We need to sample the current value to convert the new @@ -813,6 +768,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, * disable this firing since we are already reporting * it as an overrun (thanks to bump_cpu_timer above). */ + spin_unlock(&p->sighand->siglock); read_unlock(&tasklist_lock); goto out; } @@ -828,11 +784,11 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, */ timer->it.cpu.expires = new_expires; if (new_expires.sched != 0 && - (timer->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE && cpu_time_before(timer->it_clock, val, new_expires)) { - arm_timer(timer, val); + arm_timer(timer); } + spin_unlock(&p->sighand->siglock); read_unlock(&tasklist_lock); /* @@ -853,7 +809,6 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, timer->it_overrun = -1; if (new_expires.sched != 0 && - (timer->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE && !cpu_time_before(timer->it_clock, val, new_expires)) { /* * The designated time already passed, so we notify @@ -867,7 +822,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, out: if (old) { sample_to_timespec(timer->it_clock, - timer->it.cpu.incr, &old->it_interval); + old_incr, &old->it_interval); } return ret; } @@ -927,25 +882,6 @@ void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) read_unlock(&tasklist_lock); } - if ((timer->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE) { - if (timer->it.cpu.incr.sched == 0 && - cpu_time_before(timer->it_clock, - timer->it.cpu.expires, now)) { - /* - * Do-nothing timer expired and has no reload, - * so it's as if it was never set. - */ - timer->it.cpu.expires.sched = 0; - itp->it_value.tv_sec = itp->it_value.tv_nsec = 0; - return; - } - /* - * Account for any expirations and reloads that should - * have happened. - */ - bump_cpu_timer(timer, now); - } - if (unlikely(clear_dead)) { /* * We've noticed that the thread is dead, but @@ -1270,6 +1206,7 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) goto out; } read_lock(&tasklist_lock); /* arm_timer needs it. */ + spin_lock(&p->sighand->siglock); } else { read_lock(&tasklist_lock); if (unlikely(p->signal == NULL)) { @@ -1290,6 +1227,7 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) clear_dead_task(timer, now); goto out_unlock; } + spin_lock(&p->sighand->siglock); cpu_timer_sample_group(timer->it_clock, p, &now); bump_cpu_timer(timer, now); /* Leave the tasklist_lock locked for the call below. */ @@ -1298,7 +1236,9 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) /* * Now re-arm for the new expiry time. */ - arm_timer(timer, now); + BUG_ON(!irqs_disabled()); + arm_timer(timer); + spin_unlock(&p->sighand->siglock); out_unlock: read_unlock(&tasklist_lock); @@ -1390,7 +1330,7 @@ static inline int fastpath_timer_check(struct task_struct *tsk) return 1; } - return sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY; + return 0; } /* @@ -1456,21 +1396,23 @@ void run_posix_cpu_timers(struct task_struct *tsk) } /* - * Set one of the process-wide special case CPU timers. + * Set one of the process-wide special case CPU timers or RLIMIT_CPU. * The tsk->sighand->siglock must be held by the caller. - * The *newval argument is relative and we update it to be absolute, *oldval - * is absolute and we update it to be relative. */ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, cputime_t *newval, cputime_t *oldval) { union cpu_time_count now; - struct list_head *head; BUG_ON(clock_idx == CPUCLOCK_SCHED); cpu_timer_sample_group(clock_idx, tsk, &now); if (oldval) { + /* + * We are setting itimer. The *oldval is absolute and we update + * it to be relative, *newval argument is relative and we update + * it to be absolute. + */ if (!cputime_eq(*oldval, cputime_zero)) { if (cputime_le(*oldval, now.cpu)) { /* Just about to fire. */ @@ -1483,33 +1425,21 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, if (cputime_eq(*newval, cputime_zero)) return; *newval = cputime_add(*newval, now.cpu); - - /* - * If the RLIMIT_CPU timer will expire before the - * ITIMER_PROF timer, we have nothing else to do. - */ - if (tsk->signal->rlim[RLIMIT_CPU].rlim_cur - < cputime_to_secs(*newval)) - return; } /* - * Check whether there are any process timers already set to fire - * before this one. If so, we don't have anything more to do. + * Update expiration cache if we are the earliest timer, or eventually + * RLIMIT_CPU limit is earlier than prof_exp cpu timer expire. */ - head = &tsk->signal->cpu_timers[clock_idx]; - if (list_empty(head) || - cputime_ge(list_first_entry(head, - struct cpu_timer_list, entry)->expires.cpu, - *newval)) { - switch (clock_idx) { - case CPUCLOCK_PROF: + switch (clock_idx) { + case CPUCLOCK_PROF: + if (expires_gt(tsk->signal->cputime_expires.prof_exp, *newval)) tsk->signal->cputime_expires.prof_exp = *newval; - break; - case CPUCLOCK_VIRT: + break; + case CPUCLOCK_VIRT: + if (expires_gt(tsk->signal->cputime_expires.virt_exp, *newval)) tsk->signal->cputime_expires.virt_exp = *newval; - break; - } + break; } } diff --git a/kernel/power/user.c b/kernel/power/user.c index 4d2289626a84..a8c96212bc1b 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -420,7 +420,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, * User space encodes device types as two-byte values, * so we need to recode them */ - swdev = old_decode_dev(swap_area.dev); + swdev = new_decode_dev(swap_area.dev); if (swdev) { offset = swap_area.offset; data->swap = swap_type_of(swdev, offset, NULL); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 42ad8ae729a0..9fb51237b18c 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -76,7 +76,6 @@ void __ptrace_unlink(struct task_struct *child) child->parent = child->real_parent; list_del_init(&child->ptrace_entry); - arch_ptrace_untrace(child); if (task_is_traced(child)) ptrace_untrace(child); } diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 63fe25433980..03a7ea1579f6 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -69,6 +69,13 @@ EXPORT_SYMBOL_GPL(rcu_scheduler_active); #ifdef CONFIG_DEBUG_LOCK_ALLOC +int debug_lockdep_rcu_enabled(void) +{ + return rcu_scheduler_active && debug_locks && + current->lockdep_recursion == 0; +} +EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled); + /** * rcu_read_lock_bh_held - might we be in RCU-bh read-side critical section? * diff --git a/kernel/sched.c b/kernel/sched.c index 6af210a7de70..95eaecc19914 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -493,8 +493,11 @@ struct rq { #define CPU_LOAD_IDX_MAX 5 unsigned long cpu_load[CPU_LOAD_IDX_MAX]; #ifdef CONFIG_NO_HZ + u64 nohz_stamp; unsigned char in_nohz_recently; #endif + unsigned int skip_clock_update; + /* capture load from *all* tasks on this cpu: */ struct load_weight load; unsigned long nr_load_updates; @@ -592,6 +595,13 @@ static inline void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) { rq->curr->sched_class->check_preempt_curr(rq, p, flags); + + /* + * A queue event has occurred, and we're going to schedule. In + * this case, we can save a useless back to back clock update. + */ + if (test_tsk_need_resched(p)) + rq->skip_clock_update = 1; } static inline int cpu_of(struct rq *rq) @@ -626,7 +636,8 @@ static inline int cpu_of(struct rq *rq) inline void update_rq_clock(struct rq *rq) { - rq->clock = sched_clock_cpu(cpu_of(rq)); + if (!rq->skip_clock_update) + rq->clock = sched_clock_cpu(cpu_of(rq)); } /* @@ -904,16 +915,12 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) #endif /* __ARCH_WANT_UNLOCKED_CTXSW */ /* - * Check whether the task is waking, we use this to synchronize against - * ttwu() so that task_cpu() reports a stable number. - * - * We need to make an exception for PF_STARTING tasks because the fork - * path might require task_rq_lock() to work, eg. it can call - * set_cpus_allowed_ptr() from the cpuset clone_ns code. + * Check whether the task is waking, we use this to synchronize ->cpus_allowed + * against ttwu(). */ static inline int task_is_waking(struct task_struct *p) { - return unlikely((p->state == TASK_WAKING) && !(p->flags & PF_STARTING)); + return unlikely(p->state == TASK_WAKING); } /* @@ -926,11 +933,9 @@ static inline struct rq *__task_rq_lock(struct task_struct *p) struct rq *rq; for (;;) { - while (task_is_waking(p)) - cpu_relax(); rq = task_rq(p); raw_spin_lock(&rq->lock); - if (likely(rq == task_rq(p) && !task_is_waking(p))) + if (likely(rq == task_rq(p))) return rq; raw_spin_unlock(&rq->lock); } @@ -947,12 +952,10 @@ static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) struct rq *rq; for (;;) { - while (task_is_waking(p)) - cpu_relax(); local_irq_save(*flags); rq = task_rq(p); raw_spin_lock(&rq->lock); - if (likely(rq == task_rq(p) && !task_is_waking(p))) + if (likely(rq == task_rq(p))) return rq; raw_spin_unlock_irqrestore(&rq->lock, *flags); } @@ -1229,6 +1232,17 @@ void wake_up_idle_cpu(int cpu) if (!tsk_is_polling(rq->idle)) smp_send_reschedule(cpu); } + +int nohz_ratelimit(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + u64 diff = rq->clock - rq->nohz_stamp; + + rq->nohz_stamp = rq->clock; + + return diff < (NSEC_PER_SEC / HZ) >> 1; +} + #endif /* CONFIG_NO_HZ */ static u64 sched_avg_period(void) @@ -1771,8 +1785,6 @@ static void double_rq_lock(struct rq *rq1, struct rq *rq2) raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); } } - update_rq_clock(rq1); - update_rq_clock(rq2); } /* @@ -1860,62 +1872,43 @@ static void set_load_weight(struct task_struct *p) p->se.load.inv_weight = prio_to_wmult[p->static_prio - MAX_RT_PRIO]; } -static void update_avg(u64 *avg, u64 sample) -{ - s64 diff = sample - *avg; - *avg += diff >> 3; -} - -static void -enqueue_task(struct rq *rq, struct task_struct *p, int wakeup, bool head) +static void enqueue_task(struct rq *rq, struct task_struct *p, int flags) { - if (wakeup) - p->se.start_runtime = p->se.sum_exec_runtime; - + update_rq_clock(rq); sched_info_queued(p); - p->sched_class->enqueue_task(rq, p, wakeup, head); + p->sched_class->enqueue_task(rq, p, flags); p->se.on_rq = 1; } -static void dequeue_task(struct rq *rq, struct task_struct *p, int sleep) +static void dequeue_task(struct rq *rq, struct task_struct *p, int flags) { - if (sleep) { - if (p->se.last_wakeup) { - update_avg(&p->se.avg_overlap, - p->se.sum_exec_runtime - p->se.last_wakeup); - p->se.last_wakeup = 0; - } else { - update_avg(&p->se.avg_wakeup, - sysctl_sched_wakeup_granularity); - } - } - + update_rq_clock(rq); sched_info_dequeued(p); - p->sched_class->dequeue_task(rq, p, sleep); + p->sched_class->dequeue_task(rq, p, flags); p->se.on_rq = 0; } /* * activate_task - move a task to the runqueue. */ -static void activate_task(struct rq *rq, struct task_struct *p, int wakeup) +static void activate_task(struct rq *rq, struct task_struct *p, int flags) { if (task_contributes_to_load(p)) rq->nr_uninterruptible--; - enqueue_task(rq, p, wakeup, false); + enqueue_task(rq, p, flags); inc_nr_running(rq); } /* * deactivate_task - remove a task from the runqueue. */ -static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep) +static void deactivate_task(struct rq *rq, struct task_struct *p, int flags) { if (task_contributes_to_load(p)) rq->nr_uninterruptible++; - dequeue_task(rq, p, sleep); + dequeue_task(rq, p, flags); dec_nr_running(rq); } @@ -2078,49 +2071,6 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req) } /* - * wait_task_context_switch - wait for a thread to complete at least one - * context switch. - * - * @p must not be current. - */ -void wait_task_context_switch(struct task_struct *p) -{ - unsigned long nvcsw, nivcsw, flags; - int running; - struct rq *rq; - - nvcsw = p->nvcsw; - nivcsw = p->nivcsw; - for (;;) { - /* - * The runqueue is assigned before the actual context - * switch. We need to take the runqueue lock. - * - * We could check initially without the lock but it is - * very likely that we need to take the lock in every - * iteration. - */ - rq = task_rq_lock(p, &flags); - running = task_running(rq, p); - task_rq_unlock(rq, &flags); - - if (likely(!running)) - break; - /* - * The switch count is incremented before the actual - * context switch. We thus wait for two switches to be - * sure at least one completed. - */ - if ((p->nvcsw - nvcsw) > 1) - break; - if ((p->nivcsw - nivcsw) > 1) - break; - - cpu_relax(); - } -} - -/* * wait_task_inactive - wait for a thread to unschedule. * * If @match_state is nonzero, it's the @p->state value just checked and @@ -2273,6 +2223,9 @@ void task_oncpu_function_call(struct task_struct *p, } #ifdef CONFIG_SMP +/* + * ->cpus_allowed is protected by either TASK_WAKING or rq->lock held. + */ static int select_fallback_rq(int cpu, struct task_struct *p) { int dest_cpu; @@ -2289,12 +2242,8 @@ static int select_fallback_rq(int cpu, struct task_struct *p) return dest_cpu; /* No more Mr. Nice Guy. */ - if (dest_cpu >= nr_cpu_ids) { - rcu_read_lock(); - cpuset_cpus_allowed_locked(p, &p->cpus_allowed); - rcu_read_unlock(); - dest_cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed); - + if (unlikely(dest_cpu >= nr_cpu_ids)) { + dest_cpu = cpuset_cpus_allowed_fallback(p); /* * Don't tell them about moving exiting tasks or * kernel threads (both mm NULL), since they never @@ -2311,17 +2260,12 @@ static int select_fallback_rq(int cpu, struct task_struct *p) } /* - * Gets called from 3 sites (exec, fork, wakeup), since it is called without - * holding rq->lock we need to ensure ->cpus_allowed is stable, this is done - * by: - * - * exec: is unstable, retry loop - * fork & wake-up: serialize ->cpus_allowed against TASK_WAKING + * The caller (fork, wakeup) owns TASK_WAKING, ->cpus_allowed is stable. */ static inline -int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) +int select_task_rq(struct rq *rq, struct task_struct *p, int sd_flags, int wake_flags) { - int cpu = p->sched_class->select_task_rq(p, sd_flags, wake_flags); + int cpu = p->sched_class->select_task_rq(rq, p, sd_flags, wake_flags); /* * In order not to call set_task_cpu() on a blocking task we need @@ -2339,6 +2283,12 @@ int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) return cpu; } + +static void update_avg(u64 *avg, u64 sample) +{ + s64 diff = sample - *avg; + *avg += diff >> 3; +} #endif /*** @@ -2360,16 +2310,13 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, { int cpu, orig_cpu, this_cpu, success = 0; unsigned long flags; + unsigned long en_flags = ENQUEUE_WAKEUP; struct rq *rq; - if (!sched_feat(SYNC_WAKEUPS)) - wake_flags &= ~WF_SYNC; - this_cpu = get_cpu(); smp_wmb(); rq = task_rq_lock(p, &flags); - update_rq_clock(rq); if (!(p->state & state)) goto out; @@ -2389,28 +2336,26 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, * * First fix up the nr_uninterruptible count: */ - if (task_contributes_to_load(p)) - rq->nr_uninterruptible--; + if (task_contributes_to_load(p)) { + if (likely(cpu_online(orig_cpu))) + rq->nr_uninterruptible--; + else + this_rq()->nr_uninterruptible--; + } p->state = TASK_WAKING; - if (p->sched_class->task_waking) + if (p->sched_class->task_waking) { p->sched_class->task_waking(rq, p); + en_flags |= ENQUEUE_WAKING; + } - __task_rq_unlock(rq); - - cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags); - if (cpu != orig_cpu) { - /* - * Since we migrate the task without holding any rq->lock, - * we need to be careful with task_rq_lock(), since that - * might end up locking an invalid rq. - */ + cpu = select_task_rq(rq, p, SD_BALANCE_WAKE, wake_flags); + if (cpu != orig_cpu) set_task_cpu(p, cpu); - } + __task_rq_unlock(rq); rq = cpu_rq(cpu); raw_spin_lock(&rq->lock); - update_rq_clock(rq); /* * We migrated the task without holding either rq->lock, however @@ -2438,34 +2383,18 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, out_activate: #endif /* CONFIG_SMP */ - schedstat_inc(p, se.nr_wakeups); + schedstat_inc(p, se.statistics.nr_wakeups); if (wake_flags & WF_SYNC) - schedstat_inc(p, se.nr_wakeups_sync); + schedstat_inc(p, se.statistics.nr_wakeups_sync); if (orig_cpu != cpu) - schedstat_inc(p, se.nr_wakeups_migrate); + schedstat_inc(p, se.statistics.nr_wakeups_migrate); if (cpu == this_cpu) - schedstat_inc(p, se.nr_wakeups_local); + schedstat_inc(p, se.statistics.nr_wakeups_local); else - schedstat_inc(p, se.nr_wakeups_remote); - activate_task(rq, p, 1); + schedstat_inc(p, se.statistics.nr_wakeups_remote); + activate_task(rq, p, en_flags); success = 1; - /* - * Only attribute actual wakeups done by this task. - */ - if (!in_interrupt()) { - struct sched_entity *se = ¤t->se; - u64 sample = se->sum_exec_runtime; - - if (se->last_wakeup) - sample -= se->last_wakeup; - else - sample -= se->start_runtime; - update_avg(&se->avg_wakeup, sample); - - se->last_wakeup = se->sum_exec_runtime; - } - out_running: trace_sched_wakeup(rq, p, success); check_preempt_curr(rq, p, wake_flags); @@ -2527,42 +2456,9 @@ static void __sched_fork(struct task_struct *p) p->se.sum_exec_runtime = 0; p->se.prev_sum_exec_runtime = 0; p->se.nr_migrations = 0; - p->se.last_wakeup = 0; - p->se.avg_overlap = 0; - p->se.start_runtime = 0; - p->se.avg_wakeup = sysctl_sched_wakeup_granularity; #ifdef CONFIG_SCHEDSTATS - p->se.wait_start = 0; - p->se.wait_max = 0; - p->se.wait_count = 0; - p->se.wait_sum = 0; - - p->se.sleep_start = 0; - p->se.sleep_max = 0; - p->se.sum_sleep_runtime = 0; - - p->se.block_start = 0; - p->se.block_max = 0; - p->se.exec_max = 0; - p->se.slice_max = 0; - - p->se.nr_migrations_cold = 0; - p->se.nr_failed_migrations_affine = 0; - p->se.nr_failed_migrations_running = 0; - p->se.nr_failed_migrations_hot = 0; - p->se.nr_forced_migrations = 0; - - p->se.nr_wakeups = 0; - p->se.nr_wakeups_sync = 0; - p->se.nr_wakeups_migrate = 0; - p->se.nr_wakeups_local = 0; - p->se.nr_wakeups_remote = 0; - p->se.nr_wakeups_affine = 0; - p->se.nr_wakeups_affine_attempts = 0; - p->se.nr_wakeups_passive = 0; - p->se.nr_wakeups_idle = 0; - + memset(&p->se.statistics, 0, sizeof(p->se.statistics)); #endif INIT_LIST_HEAD(&p->rt.run_list); @@ -2583,11 +2479,11 @@ void sched_fork(struct task_struct *p, int clone_flags) __sched_fork(p); /* - * We mark the process as waking here. This guarantees that + * We mark the process as running here. This guarantees that * nobody will actually run it, and a signal or other external * event cannot wake it up and insert it on the runqueue either. */ - p->state = TASK_WAKING; + p->state = TASK_RUNNING; /* * Revert to default priority/policy on fork if requested. @@ -2654,29 +2550,25 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) int cpu __maybe_unused = get_cpu(); #ifdef CONFIG_SMP + rq = task_rq_lock(p, &flags); + p->state = TASK_WAKING; + /* * Fork balancing, do it here and not earlier because: * - cpus_allowed can change in the fork path * - any previously selected cpu might disappear through hotplug * - * We still have TASK_WAKING but PF_STARTING is gone now, meaning - * ->cpus_allowed is stable, we have preemption disabled, meaning - * cpu_online_mask is stable. + * We set TASK_WAKING so that select_task_rq() can drop rq->lock + * without people poking at ->cpus_allowed. */ - cpu = select_task_rq(p, SD_BALANCE_FORK, 0); + cpu = select_task_rq(rq, p, SD_BALANCE_FORK, 0); set_task_cpu(p, cpu); -#endif - - /* - * Since the task is not on the rq and we still have TASK_WAKING set - * nobody else will migrate this task. - */ - rq = cpu_rq(cpu); - raw_spin_lock_irqsave(&rq->lock, flags); - BUG_ON(p->state != TASK_WAKING); p->state = TASK_RUNNING; - update_rq_clock(rq); + task_rq_unlock(rq, &flags); +#endif + + rq = task_rq_lock(p, &flags); activate_task(rq, p, 0); trace_sched_wakeup_new(rq, p, 1); check_preempt_curr(rq, p, WF_FORK); @@ -3122,32 +3014,21 @@ void sched_exec(void) { struct task_struct *p = current; struct migration_req req; - int dest_cpu, this_cpu; unsigned long flags; struct rq *rq; - -again: - this_cpu = get_cpu(); - dest_cpu = select_task_rq(p, SD_BALANCE_EXEC, 0); - if (dest_cpu == this_cpu) { - put_cpu(); - return; - } + int dest_cpu; rq = task_rq_lock(p, &flags); - put_cpu(); + dest_cpu = p->sched_class->select_task_rq(rq, p, SD_BALANCE_EXEC, 0); + if (dest_cpu == smp_processor_id()) + goto unlock; /* * select_task_rq() can race against ->cpus_allowed */ - if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed) - || unlikely(!cpu_active(dest_cpu))) { - task_rq_unlock(rq, &flags); - goto again; - } - - /* force the process onto the specified CPU */ - if (migrate_task(p, dest_cpu, &req)) { + if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed) && + likely(cpu_active(dest_cpu)) && + migrate_task(p, dest_cpu, &req)) { /* Need to wait for migration thread (might exit: take ref). */ struct task_struct *mt = rq->migration_thread; @@ -3159,6 +3040,7 @@ again: return; } +unlock: task_rq_unlock(rq, &flags); } @@ -3630,23 +3512,9 @@ static inline void schedule_debug(struct task_struct *prev) static void put_prev_task(struct rq *rq, struct task_struct *prev) { - if (prev->state == TASK_RUNNING) { - u64 runtime = prev->se.sum_exec_runtime; - - runtime -= prev->se.prev_sum_exec_runtime; - runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); - - /* - * In order to avoid avg_overlap growing stale when we are - * indeed overlapping and hence not getting put to sleep, grow - * the avg_overlap on preemption. - * - * We use the average preemption runtime because that - * correlates to the amount of cache footprint a task can - * build up. - */ - update_avg(&prev->se.avg_overlap, runtime); - } + if (prev->se.on_rq) + update_rq_clock(rq); + rq->skip_clock_update = 0; prev->sched_class->put_prev_task(rq, prev); } @@ -3709,14 +3577,13 @@ need_resched_nonpreemptible: hrtick_clear(rq); raw_spin_lock_irq(&rq->lock); - update_rq_clock(rq); clear_tsk_need_resched(prev); if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { if (unlikely(signal_pending_state(prev->state, prev))) prev->state = TASK_RUNNING; else - deactivate_task(rq, prev, 1); + deactivate_task(rq, prev, DEQUEUE_SLEEP); switch_count = &prev->nvcsw; } @@ -4266,7 +4133,6 @@ void rt_mutex_setprio(struct task_struct *p, int prio) BUG_ON(prio < 0 || prio > MAX_PRIO); rq = task_rq_lock(p, &flags); - update_rq_clock(rq); oldprio = p->prio; prev_class = p->sched_class; @@ -4287,7 +4153,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio) if (running) p->sched_class->set_curr_task(rq); if (on_rq) { - enqueue_task(rq, p, 0, oldprio < prio); + enqueue_task(rq, p, oldprio < prio ? ENQUEUE_HEAD : 0); check_class_changed(rq, p, prev_class, oldprio, running); } @@ -4309,7 +4175,6 @@ void set_user_nice(struct task_struct *p, long nice) * the task might be in the middle of scheduling on another CPU. */ rq = task_rq_lock(p, &flags); - update_rq_clock(rq); /* * The RT priorities are set via sched_setscheduler(), but we still * allow the 'normal' nice value to be set - but as expected @@ -4331,7 +4196,7 @@ void set_user_nice(struct task_struct *p, long nice) delta = p->prio - old_prio; if (on_rq) { - enqueue_task(rq, p, 0, false); + enqueue_task(rq, p, 0); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -4592,7 +4457,6 @@ recheck: raw_spin_unlock_irqrestore(&p->pi_lock, flags); goto recheck; } - update_rq_clock(rq); on_rq = p->se.on_rq; running = task_current(rq, p); if (on_rq) @@ -5358,7 +5222,18 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) struct rq *rq; int ret = 0; + /* + * Serialize against TASK_WAKING so that ttwu() and wunt() can + * drop the rq->lock and still rely on ->cpus_allowed. + */ +again: + while (task_is_waking(p)) + cpu_relax(); rq = task_rq_lock(p, &flags); + if (task_is_waking(p)) { + task_rq_unlock(rq, &flags); + goto again; + } if (!cpumask_intersects(new_mask, cpu_active_mask)) { ret = -EINVAL; @@ -5516,30 +5391,29 @@ static int migration_thread(void *data) } #ifdef CONFIG_HOTPLUG_CPU - -static int __migrate_task_irq(struct task_struct *p, int src_cpu, int dest_cpu) -{ - int ret; - - local_irq_disable(); - ret = __migrate_task(p, src_cpu, dest_cpu); - local_irq_enable(); - return ret; -} - /* * Figure out where task on dead CPU should go, use force if necessary. */ -static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) +void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) { - int dest_cpu; + struct rq *rq = cpu_rq(dead_cpu); + int needs_cpu, uninitialized_var(dest_cpu); + unsigned long flags; -again: - dest_cpu = select_fallback_rq(dead_cpu, p); + local_irq_save(flags); - /* It can have affinity changed while we were choosing. */ - if (unlikely(!__migrate_task_irq(p, dead_cpu, dest_cpu))) - goto again; + raw_spin_lock(&rq->lock); + needs_cpu = (task_cpu(p) == dead_cpu) && (p->state != TASK_WAKING); + if (needs_cpu) + dest_cpu = select_fallback_rq(dead_cpu, p); + raw_spin_unlock(&rq->lock); + /* + * It can only fail if we race with set_cpus_allowed(), + * in the racer should migrate the task anyway. + */ + if (needs_cpu) + __migrate_task(p, dead_cpu, dest_cpu); + local_irq_restore(flags); } /* @@ -5603,7 +5477,6 @@ void sched_idle_next(void) __setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1); - update_rq_clock(rq); activate_task(rq, p, 0); raw_spin_unlock_irqrestore(&rq->lock, flags); @@ -5658,7 +5531,6 @@ static void migrate_dead_tasks(unsigned int dead_cpu) for ( ; ; ) { if (!rq->nr_running) break; - update_rq_clock(rq); next = pick_next_task(rq); if (!next) break; @@ -5934,7 +5806,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) case CPU_DEAD: case CPU_DEAD_FROZEN: - cpuset_lock(); /* around calls to cpuset_cpus_allowed_lock() */ migrate_live_tasks(cpu); rq = cpu_rq(cpu); kthread_stop(rq->migration_thread); @@ -5942,13 +5813,11 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) rq->migration_thread = NULL; /* Idle task back to normal (off runqueue, low prio) */ raw_spin_lock_irq(&rq->lock); - update_rq_clock(rq); deactivate_task(rq, rq->idle, 0); __setscheduler(rq, rq->idle, SCHED_NORMAL, 0); rq->idle->sched_class = &idle_sched_class; migrate_dead_tasks(cpu); raw_spin_unlock_irq(&rq->lock); - cpuset_unlock(); migrate_nr_uninterruptible(rq); BUG_ON(rq->nr_running != 0); calc_global_load_remove(rq); @@ -7892,7 +7761,6 @@ static void normalize_task(struct rq *rq, struct task_struct *p) { int on_rq; - update_rq_clock(rq); on_rq = p->se.on_rq; if (on_rq) deactivate_task(rq, p, 0); @@ -7919,9 +7787,9 @@ void normalize_rt_tasks(void) p->se.exec_start = 0; #ifdef CONFIG_SCHEDSTATS - p->se.wait_start = 0; - p->se.sleep_start = 0; - p->se.block_start = 0; + p->se.statistics.wait_start = 0; + p->se.statistics.sleep_start = 0; + p->se.statistics.block_start = 0; #endif if (!rt_task(p)) { @@ -8254,8 +8122,6 @@ void sched_move_task(struct task_struct *tsk) rq = task_rq_lock(tsk, &flags); - update_rq_clock(rq); - running = task_current(rq, tsk); on_rq = tsk->se.on_rq; @@ -8274,7 +8140,7 @@ void sched_move_task(struct task_struct *tsk) if (unlikely(running)) tsk->sched_class->set_curr_task(rq); if (on_rq) - enqueue_task(rq, tsk, 0, false); + enqueue_task(rq, tsk, 0); task_rq_unlock(rq, &flags); } diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 9b49db144037..9cf1baf6616a 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -70,16 +70,16 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, PN(se->vruntime); PN(se->sum_exec_runtime); #ifdef CONFIG_SCHEDSTATS - PN(se->wait_start); - PN(se->sleep_start); - PN(se->block_start); - PN(se->sleep_max); - PN(se->block_max); - PN(se->exec_max); - PN(se->slice_max); - PN(se->wait_max); - PN(se->wait_sum); - P(se->wait_count); + PN(se->statistics.wait_start); + PN(se->statistics.sleep_start); + PN(se->statistics.block_start); + PN(se->statistics.sleep_max); + PN(se->statistics.block_max); + PN(se->statistics.exec_max); + PN(se->statistics.slice_max); + PN(se->statistics.wait_max); + PN(se->statistics.wait_sum); + P(se->statistics.wait_count); #endif P(se->load.weight); #undef PN @@ -104,7 +104,7 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) SEQ_printf(m, "%9Ld.%06ld %9Ld.%06ld %9Ld.%06ld", SPLIT_NS(p->se.vruntime), SPLIT_NS(p->se.sum_exec_runtime), - SPLIT_NS(p->se.sum_sleep_runtime)); + SPLIT_NS(p->se.statistics.sum_sleep_runtime)); #else SEQ_printf(m, "%15Ld %15Ld %15Ld.%06ld %15Ld.%06ld %15Ld.%06ld", 0LL, 0LL, 0LL, 0L, 0LL, 0L, 0LL, 0L); @@ -173,11 +173,6 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) task_group_path(tg, path, sizeof(path)); SEQ_printf(m, "\ncfs_rq[%d]:%s\n", cpu, path); -#elif defined(CONFIG_USER_SCHED) && defined(CONFIG_FAIR_GROUP_SCHED) - { - uid_t uid = cfs_rq->tg->uid; - SEQ_printf(m, "\ncfs_rq[%d] for UID: %u\n", cpu, uid); - } #else SEQ_printf(m, "\ncfs_rq[%d]:\n", cpu); #endif @@ -407,40 +402,38 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.exec_start); PN(se.vruntime); PN(se.sum_exec_runtime); - PN(se.avg_overlap); - PN(se.avg_wakeup); nr_switches = p->nvcsw + p->nivcsw; #ifdef CONFIG_SCHEDSTATS - PN(se.wait_start); - PN(se.sleep_start); - PN(se.block_start); - PN(se.sleep_max); - PN(se.block_max); - PN(se.exec_max); - PN(se.slice_max); - PN(se.wait_max); - PN(se.wait_sum); - P(se.wait_count); - PN(se.iowait_sum); - P(se.iowait_count); + PN(se.statistics.wait_start); + PN(se.statistics.sleep_start); + PN(se.statistics.block_start); + PN(se.statistics.sleep_max); + PN(se.statistics.block_max); + PN(se.statistics.exec_max); + PN(se.statistics.slice_max); + PN(se.statistics.wait_max); + PN(se.statistics.wait_sum); + P(se.statistics.wait_count); + PN(se.statistics.iowait_sum); + P(se.statistics.iowait_count); P(sched_info.bkl_count); P(se.nr_migrations); - P(se.nr_migrations_cold); - P(se.nr_failed_migrations_affine); - P(se.nr_failed_migrations_running); - P(se.nr_failed_migrations_hot); - P(se.nr_forced_migrations); - P(se.nr_wakeups); - P(se.nr_wakeups_sync); - P(se.nr_wakeups_migrate); - P(se.nr_wakeups_local); - P(se.nr_wakeups_remote); - P(se.nr_wakeups_affine); - P(se.nr_wakeups_affine_attempts); - P(se.nr_wakeups_passive); - P(se.nr_wakeups_idle); + P(se.statistics.nr_migrations_cold); + P(se.statistics.nr_failed_migrations_affine); + P(se.statistics.nr_failed_migrations_running); + P(se.statistics.nr_failed_migrations_hot); + P(se.statistics.nr_forced_migrations); + P(se.statistics.nr_wakeups); + P(se.statistics.nr_wakeups_sync); + P(se.statistics.nr_wakeups_migrate); + P(se.statistics.nr_wakeups_local); + P(se.statistics.nr_wakeups_remote); + P(se.statistics.nr_wakeups_affine); + P(se.statistics.nr_wakeups_affine_attempts); + P(se.statistics.nr_wakeups_passive); + P(se.statistics.nr_wakeups_idle); { u64 avg_atom, avg_per_cpu; @@ -491,31 +484,6 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) void proc_sched_set_task(struct task_struct *p) { #ifdef CONFIG_SCHEDSTATS - p->se.wait_max = 0; - p->se.wait_sum = 0; - p->se.wait_count = 0; - p->se.iowait_sum = 0; - p->se.iowait_count = 0; - p->se.sleep_max = 0; - p->se.sum_sleep_runtime = 0; - p->se.block_max = 0; - p->se.exec_max = 0; - p->se.slice_max = 0; - p->se.nr_migrations = 0; - p->se.nr_migrations_cold = 0; - p->se.nr_failed_migrations_affine = 0; - p->se.nr_failed_migrations_running = 0; - p->se.nr_failed_migrations_hot = 0; - p->se.nr_forced_migrations = 0; - p->se.nr_wakeups = 0; - p->se.nr_wakeups_sync = 0; - p->se.nr_wakeups_migrate = 0; - p->se.nr_wakeups_local = 0; - p->se.nr_wakeups_remote = 0; - p->se.nr_wakeups_affine = 0; - p->se.nr_wakeups_affine_attempts = 0; - p->se.nr_wakeups_passive = 0; - p->se.nr_wakeups_idle = 0; - p->sched_info.bkl_count = 0; + memset(&p->se.statistics, 0, sizeof(p->se.statistics)); #endif } diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 5a5ea2cd924f..88d3053ac7c2 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -35,8 +35,8 @@ * (to see the precise effective timeslice length of your workload, * run vmstat and monitor the context-switches (cs) field) */ -unsigned int sysctl_sched_latency = 5000000ULL; -unsigned int normalized_sysctl_sched_latency = 5000000ULL; +unsigned int sysctl_sched_latency = 6000000ULL; +unsigned int normalized_sysctl_sched_latency = 6000000ULL; /* * The initial- and re-scaling of tunables is configurable @@ -52,15 +52,15 @@ enum sched_tunable_scaling sysctl_sched_tunable_scaling /* * Minimal preemption granularity for CPU-bound tasks: - * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 1000000ULL; -unsigned int normalized_sysctl_sched_min_granularity = 1000000ULL; +unsigned int sysctl_sched_min_granularity = 2000000ULL; +unsigned int normalized_sysctl_sched_min_granularity = 2000000ULL; /* * is kept at sysctl_sched_latency / sysctl_sched_min_granularity */ -static unsigned int sched_nr_latency = 5; +static unsigned int sched_nr_latency = 3; /* * After fork, child runs first. If set to 0 (default) then @@ -505,7 +505,8 @@ __update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr, { unsigned long delta_exec_weighted; - schedstat_set(curr->exec_max, max((u64)delta_exec, curr->exec_max)); + schedstat_set(curr->statistics.exec_max, + max((u64)delta_exec, curr->statistics.exec_max)); curr->sum_exec_runtime += delta_exec; schedstat_add(cfs_rq, exec_clock, delta_exec); @@ -548,7 +549,7 @@ static void update_curr(struct cfs_rq *cfs_rq) static inline void update_stats_wait_start(struct cfs_rq *cfs_rq, struct sched_entity *se) { - schedstat_set(se->wait_start, rq_of(cfs_rq)->clock); + schedstat_set(se->statistics.wait_start, rq_of(cfs_rq)->clock); } /* @@ -567,18 +568,18 @@ static void update_stats_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) static void update_stats_wait_end(struct cfs_rq *cfs_rq, struct sched_entity *se) { - schedstat_set(se->wait_max, max(se->wait_max, - rq_of(cfs_rq)->clock - se->wait_start)); - schedstat_set(se->wait_count, se->wait_count + 1); - schedstat_set(se->wait_sum, se->wait_sum + - rq_of(cfs_rq)->clock - se->wait_start); + schedstat_set(se->statistics.wait_max, max(se->statistics.wait_max, + rq_of(cfs_rq)->clock - se->statistics.wait_start)); + schedstat_set(se->statistics.wait_count, se->statistics.wait_count + 1); + schedstat_set(se->statistics.wait_sum, se->statistics.wait_sum + + rq_of(cfs_rq)->clock - se->statistics.wait_start); #ifdef CONFIG_SCHEDSTATS if (entity_is_task(se)) { trace_sched_stat_wait(task_of(se), - rq_of(cfs_rq)->clock - se->wait_start); + rq_of(cfs_rq)->clock - se->statistics.wait_start); } #endif - schedstat_set(se->wait_start, 0); + schedstat_set(se->statistics.wait_start, 0); } static inline void @@ -657,39 +658,39 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) if (entity_is_task(se)) tsk = task_of(se); - if (se->sleep_start) { - u64 delta = rq_of(cfs_rq)->clock - se->sleep_start; + if (se->statistics.sleep_start) { + u64 delta = rq_of(cfs_rq)->clock - se->statistics.sleep_start; if ((s64)delta < 0) delta = 0; - if (unlikely(delta > se->sleep_max)) - se->sleep_max = delta; + if (unlikely(delta > se->statistics.sleep_max)) + se->statistics.sleep_max = delta; - se->sleep_start = 0; - se->sum_sleep_runtime += delta; + se->statistics.sleep_start = 0; + se->statistics.sum_sleep_runtime += delta; if (tsk) { account_scheduler_latency(tsk, delta >> 10, 1); trace_sched_stat_sleep(tsk, delta); } } - if (se->block_start) { - u64 delta = rq_of(cfs_rq)->clock - se->block_start; + if (se->statistics.block_start) { + u64 delta = rq_of(cfs_rq)->clock - se->statistics.block_start; if ((s64)delta < 0) delta = 0; - if (unlikely(delta > se->block_max)) - se->block_max = delta; + if (unlikely(delta > se->statistics.block_max)) + se->statistics.block_max = delta; - se->block_start = 0; - se->sum_sleep_runtime += delta; + se->statistics.block_start = 0; + se->statistics.sum_sleep_runtime += delta; if (tsk) { if (tsk->in_iowait) { - se->iowait_sum += delta; - se->iowait_count++; + se->statistics.iowait_sum += delta; + se->statistics.iowait_count++; trace_sched_stat_iowait(tsk, delta); } @@ -737,20 +738,10 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) vruntime += sched_vslice(cfs_rq, se); /* sleeps up to a single latency don't count. */ - if (!initial && sched_feat(FAIR_SLEEPERS)) { + if (!initial) { unsigned long thresh = sysctl_sched_latency; /* - * Convert the sleeper threshold into virtual time. - * SCHED_IDLE is a special sub-class. We care about - * fairness only relative to other SCHED_IDLE tasks, - * all of which have the same weight. - */ - if (sched_feat(NORMALIZED_SLEEPER) && (!entity_is_task(se) || - task_of(se)->policy != SCHED_IDLE)) - thresh = calc_delta_fair(thresh, se); - - /* * Halve their sleep time's effect, to allow * for a gentler effect of sleepers: */ @@ -766,9 +757,6 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) se->vruntime = vruntime; } -#define ENQUEUE_WAKEUP 1 -#define ENQUEUE_MIGRATE 2 - static void enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { @@ -776,7 +764,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * Update the normalized vruntime before updating min_vruntime * through callig update_curr(). */ - if (!(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_MIGRATE)) + if (!(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_WAKING)) se->vruntime += cfs_rq->min_vruntime; /* @@ -812,7 +800,7 @@ static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) } static void -dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int sleep) +dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { /* * Update run-time statistics of the 'current'. @@ -820,15 +808,15 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int sleep) update_curr(cfs_rq); update_stats_dequeue(cfs_rq, se); - if (sleep) { + if (flags & DEQUEUE_SLEEP) { #ifdef CONFIG_SCHEDSTATS if (entity_is_task(se)) { struct task_struct *tsk = task_of(se); if (tsk->state & TASK_INTERRUPTIBLE) - se->sleep_start = rq_of(cfs_rq)->clock; + se->statistics.sleep_start = rq_of(cfs_rq)->clock; if (tsk->state & TASK_UNINTERRUPTIBLE) - se->block_start = rq_of(cfs_rq)->clock; + se->statistics.block_start = rq_of(cfs_rq)->clock; } #endif } @@ -845,7 +833,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int sleep) * update can refer to the ->curr item and we need to reflect this * movement in our normalized position. */ - if (!sleep) + if (!(flags & DEQUEUE_SLEEP)) se->vruntime -= cfs_rq->min_vruntime; } @@ -912,7 +900,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) * when there are only lesser-weight tasks around): */ if (rq_of(cfs_rq)->load.weight >= 2*se->load.weight) { - se->slice_max = max(se->slice_max, + se->statistics.slice_max = max(se->statistics.slice_max, se->sum_exec_runtime - se->prev_sum_exec_runtime); } #endif @@ -1054,16 +1042,10 @@ static inline void hrtick_update(struct rq *rq) * then put the task into the rbtree: */ static void -enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup, bool head) +enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) { struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se; - int flags = 0; - - if (wakeup) - flags |= ENQUEUE_WAKEUP; - if (p->state == TASK_WAKING) - flags |= ENQUEUE_MIGRATE; for_each_sched_entity(se) { if (se->on_rq) @@ -1081,18 +1063,18 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup, bool head) * decreased. We remove the task from the rbtree and * update the fair scheduling stats: */ -static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int sleep) +static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) { struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se; for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - dequeue_entity(cfs_rq, se, sleep); + dequeue_entity(cfs_rq, se, flags); /* Don't dequeue parent if it has other entities besides us */ if (cfs_rq->load.weight) break; - sleep = 1; + flags |= DEQUEUE_SLEEP; } hrtick_update(rq); @@ -1240,7 +1222,6 @@ static inline unsigned long effective_load(struct task_group *tg, int cpu, static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) { - struct task_struct *curr = current; unsigned long this_load, load; int idx, this_cpu, prev_cpu; unsigned long tl_per_task; @@ -1255,18 +1236,6 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) load = source_load(prev_cpu, idx); this_load = target_load(this_cpu, idx); - if (sync) { - if (sched_feat(SYNC_LESS) && - (curr->se.avg_overlap > sysctl_sched_migration_cost || - p->se.avg_overlap > sysctl_sched_migration_cost)) - sync = 0; - } else { - if (sched_feat(SYNC_MORE) && - (curr->se.avg_overlap < sysctl_sched_migration_cost && - p->se.avg_overlap < sysctl_sched_migration_cost)) - sync = 1; - } - /* * If sync wakeup then subtract the (maximum possible) * effect of the currently running task from the load @@ -1306,7 +1275,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) if (sync && balanced) return 1; - schedstat_inc(p, se.nr_wakeups_affine_attempts); + schedstat_inc(p, se.statistics.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); if (balanced || @@ -1318,7 +1287,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) * there is no bad imbalance. */ schedstat_inc(sd, ttwu_move_affine); - schedstat_inc(p, se.nr_wakeups_affine); + schedstat_inc(p, se.statistics.nr_wakeups_affine); return 1; } @@ -1445,19 +1414,19 @@ select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target) * * preempt must be disabled. */ -static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) +static int +select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_flags) { struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL; int cpu = smp_processor_id(); int prev_cpu = task_cpu(p); int new_cpu = cpu; - int want_affine = 0; + int want_affine = 0, cpu_idle = !current->pid; int want_sd = 1; int sync = wake_flags & WF_SYNC; if (sd_flag & SD_BALANCE_WAKE) { - if (sched_feat(AFFINE_WAKEUPS) && - cpumask_test_cpu(cpu, &p->cpus_allowed)) + if (cpumask_test_cpu(cpu, &p->cpus_allowed)) want_affine = 1; new_cpu = prev_cpu; } @@ -1509,13 +1478,15 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag * If there's an idle sibling in this domain, make that * the wake_affine target instead of the current cpu. */ - if (tmp->flags & SD_SHARE_PKG_RESOURCES) + if (!cpu_idle && tmp->flags & SD_SHARE_PKG_RESOURCES) target = select_idle_sibling(p, tmp, target); if (target >= 0) { if (tmp->flags & SD_WAKE_AFFINE) { affine_sd = tmp; want_affine = 0; + if (target != cpu) + cpu_idle = 1; } cpu = target; } @@ -1531,6 +1502,7 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag sd = tmp; } +#ifdef CONFIG_FAIR_GROUP_SCHED if (sched_feat(LB_SHARES_UPDATE)) { /* * Pick the largest domain to update shares over @@ -1541,12 +1513,18 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag cpumask_weight(sched_domain_span(sd)))) tmp = affine_sd; - if (tmp) + if (tmp) { + raw_spin_unlock(&rq->lock); update_shares(tmp); + raw_spin_lock(&rq->lock); + } } +#endif - if (affine_sd && wake_affine(affine_sd, p, sync)) - return cpu; + if (affine_sd) { + if (cpu_idle || cpu == prev_cpu || wake_affine(affine_sd, p, sync)) + return cpu; + } while (sd) { int load_idx = sd->forkexec_idx; @@ -1591,63 +1569,26 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag } #endif /* CONFIG_SMP */ -/* - * Adaptive granularity - * - * se->avg_wakeup gives the average time a task runs until it does a wakeup, - * with the limit of wakeup_gran -- when it never does a wakeup. - * - * So the smaller avg_wakeup is the faster we want this task to preempt, - * but we don't want to treat the preemptee unfairly and therefore allow it - * to run for at least the amount of time we'd like to run. - * - * NOTE: we use 2*avg_wakeup to increase the probability of actually doing one - * - * NOTE: we use *nr_running to scale with load, this nicely matches the - * degrading latency on load. - */ -static unsigned long -adaptive_gran(struct sched_entity *curr, struct sched_entity *se) -{ - u64 this_run = curr->sum_exec_runtime - curr->prev_sum_exec_runtime; - u64 expected_wakeup = 2*se->avg_wakeup * cfs_rq_of(se)->nr_running; - u64 gran = 0; - - if (this_run < expected_wakeup) - gran = expected_wakeup - this_run; - - return min_t(s64, gran, sysctl_sched_wakeup_granularity); -} - static unsigned long wakeup_gran(struct sched_entity *curr, struct sched_entity *se) { unsigned long gran = sysctl_sched_wakeup_granularity; - if (cfs_rq_of(curr)->curr && sched_feat(ADAPTIVE_GRAN)) - gran = adaptive_gran(curr, se); - /* * Since its curr running now, convert the gran from real-time * to virtual-time in his units. + * + * By using 'se' instead of 'curr' we penalize light tasks, so + * they get preempted easier. That is, if 'se' < 'curr' then + * the resulting gran will be larger, therefore penalizing the + * lighter, if otoh 'se' > 'curr' then the resulting gran will + * be smaller, again penalizing the lighter task. + * + * This is especially important for buddies when the leftmost + * task is higher priority than the buddy. */ - if (sched_feat(ASYM_GRAN)) { - /* - * By using 'se' instead of 'curr' we penalize light tasks, so - * they get preempted easier. That is, if 'se' < 'curr' then - * the resulting gran will be larger, therefore penalizing the - * lighter, if otoh 'se' > 'curr' then the resulting gran will - * be smaller, again penalizing the lighter task. - * - * This is especially important for buddies when the leftmost - * task is higher priority than the buddy. - */ - if (unlikely(se->load.weight != NICE_0_LOAD)) - gran = calc_delta_fair(gran, se); - } else { - if (unlikely(curr->load.weight != NICE_0_LOAD)) - gran = calc_delta_fair(gran, curr); - } + if (unlikely(se->load.weight != NICE_0_LOAD)) + gran = calc_delta_fair(gran, se); return gran; } @@ -1705,7 +1646,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ struct task_struct *curr = rq->curr; struct sched_entity *se = &curr->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(curr); - int sync = wake_flags & WF_SYNC; int scale = cfs_rq->nr_running >= sched_nr_latency; if (unlikely(rt_prio(p->prio))) @@ -1738,14 +1678,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (unlikely(curr->policy == SCHED_IDLE)) goto preempt; - if (sched_feat(WAKEUP_SYNC) && sync) - goto preempt; - - if (sched_feat(WAKEUP_OVERLAP) && - se->avg_overlap < sysctl_sched_migration_cost && - pse->avg_overlap < sysctl_sched_migration_cost) - goto preempt; - if (!sched_feat(WAKEUP_PREEMPT)) return; @@ -1844,13 +1776,13 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, * 3) are cache-hot on their current CPU. */ if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) { - schedstat_inc(p, se.nr_failed_migrations_affine); + schedstat_inc(p, se.statistics.nr_failed_migrations_affine); return 0; } *all_pinned = 0; if (task_running(rq, p)) { - schedstat_inc(p, se.nr_failed_migrations_running); + schedstat_inc(p, se.statistics.nr_failed_migrations_running); return 0; } @@ -1866,14 +1798,14 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, #ifdef CONFIG_SCHEDSTATS if (tsk_cache_hot) { schedstat_inc(sd, lb_hot_gained[idle]); - schedstat_inc(p, se.nr_forced_migrations); + schedstat_inc(p, se.statistics.nr_forced_migrations); } #endif return 1; } if (tsk_cache_hot) { - schedstat_inc(p, se.nr_failed_migrations_hot); + schedstat_inc(p, se.statistics.nr_failed_migrations_hot); return 0; } return 1; @@ -3112,8 +3044,6 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) /* move a task from busiest_rq to target_rq */ double_lock_balance(busiest_rq, target_rq); - update_rq_clock(busiest_rq); - update_rq_clock(target_rq); /* Search for an sd spanning us and the target CPU. */ for_each_domain(target_cpu, sd) { diff --git a/kernel/sched_features.h b/kernel/sched_features.h index d5059fd761d9..83c66e8ad3ee 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -1,11 +1,4 @@ /* - * Disregards a certain amount of sleep time (sched_latency_ns) and - * considers the task to be running during that period. This gives it - * a service deficit on wakeup, allowing it to run sooner. - */ -SCHED_FEAT(FAIR_SLEEPERS, 1) - -/* * Only give sleepers 50% of their service deficit. This allows * them to run sooner, but does not allow tons of sleepers to * rip the spread apart. @@ -13,13 +6,6 @@ SCHED_FEAT(FAIR_SLEEPERS, 1) SCHED_FEAT(GENTLE_FAIR_SLEEPERS, 1) /* - * By not normalizing the sleep time, heavy tasks get an effective - * longer period, and lighter task an effective shorter period they - * are considered running. - */ -SCHED_FEAT(NORMALIZED_SLEEPER, 0) - -/* * Place new tasks ahead so that they do not starve already running * tasks */ @@ -31,37 +17,6 @@ SCHED_FEAT(START_DEBIT, 1) SCHED_FEAT(WAKEUP_PREEMPT, 1) /* - * Compute wakeup_gran based on task behaviour, clipped to - * [0, sched_wakeup_gran_ns] - */ -SCHED_FEAT(ADAPTIVE_GRAN, 1) - -/* - * When converting the wakeup granularity to virtual time, do it such - * that heavier tasks preempting a lighter task have an edge. - */ -SCHED_FEAT(ASYM_GRAN, 1) - -/* - * Always wakeup-preempt SYNC wakeups, see SYNC_WAKEUPS. - */ -SCHED_FEAT(WAKEUP_SYNC, 0) - -/* - * Wakeup preempt based on task behaviour. Tasks that do not overlap - * don't get preempted. - */ -SCHED_FEAT(WAKEUP_OVERLAP, 0) - -/* - * Use the SYNC wakeup hint, pipes and the likes use this to indicate - * the remote end is likely to consume the data we just wrote, and - * therefore has cache benefit from being placed on the same cpu, see - * also AFFINE_WAKEUPS. - */ -SCHED_FEAT(SYNC_WAKEUPS, 1) - -/* * Based on load and program behaviour, see if it makes sense to place * a newly woken task on the same cpu as the task that woke it -- * improve cache locality. Typically used with SYNC wakeups as @@ -70,16 +25,6 @@ SCHED_FEAT(SYNC_WAKEUPS, 1) SCHED_FEAT(AFFINE_WAKEUPS, 1) /* - * Weaken SYNC hint based on overlap - */ -SCHED_FEAT(SYNC_LESS, 1) - -/* - * Add SYNC hint based on overlap - */ -SCHED_FEAT(SYNC_MORE, 0) - -/* * Prefer to schedule the task we woke last (assuming it failed * wakeup-preemption), since its likely going to consume data we * touched, increases cache locality. diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index a8a6d8a50947..bea2b8f12024 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -6,7 +6,8 @@ */ #ifdef CONFIG_SMP -static int select_task_rq_idle(struct task_struct *p, int sd_flag, int flags) +static int +select_task_rq_idle(struct rq *rq, struct task_struct *p, int sd_flag, int flags) { return task_cpu(p); /* IDLE tasks as never migrated */ } @@ -32,7 +33,7 @@ static struct task_struct *pick_next_task_idle(struct rq *rq) * message if some code attempts to do it: */ static void -dequeue_task_idle(struct rq *rq, struct task_struct *p, int sleep) +dequeue_task_idle(struct rq *rq, struct task_struct *p, int flags) { raw_spin_unlock_irq(&rq->lock); printk(KERN_ERR "bad: scheduling from the idle thread!\n"); diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index b5b920ae2ea7..8afb953e31c6 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -613,7 +613,7 @@ static void update_curr_rt(struct rq *rq) if (unlikely((s64)delta_exec < 0)) delta_exec = 0; - schedstat_set(curr->se.exec_max, max(curr->se.exec_max, delta_exec)); + schedstat_set(curr->se.statistics.exec_max, max(curr->se.statistics.exec_max, delta_exec)); curr->se.sum_exec_runtime += delta_exec; account_group_exec_runtime(curr, delta_exec); @@ -888,20 +888,20 @@ static void dequeue_rt_entity(struct sched_rt_entity *rt_se) * Adding/removing a task to/from a priority array: */ static void -enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup, bool head) +enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags) { struct sched_rt_entity *rt_se = &p->rt; - if (wakeup) + if (flags & ENQUEUE_WAKEUP) rt_se->timeout = 0; - enqueue_rt_entity(rt_se, head); + enqueue_rt_entity(rt_se, flags & ENQUEUE_HEAD); if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); } -static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) +static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags) { struct sched_rt_entity *rt_se = &p->rt; @@ -948,10 +948,9 @@ static void yield_task_rt(struct rq *rq) #ifdef CONFIG_SMP static int find_lowest_rq(struct task_struct *task); -static int select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) +static int +select_task_rq_rt(struct rq *rq, struct task_struct *p, int sd_flag, int flags) { - struct rq *rq = task_rq(p); - if (sd_flag != SD_BALANCE_WAKE) return smp_processor_id(); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8686b0f5fc12..a38af430f0d8 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -74,6 +74,10 @@ #include <scsi/sg.h> #endif +#ifdef CONFIG_NMI_WATCHDOG +#include <linux/nmi.h> +#endif + #if defined(CONFIG_SYSCTL) @@ -682,7 +686,16 @@ static struct ctl_table kern_table[] = { .mode = 0444, .proc_handler = proc_dointvec, }, -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) +#if defined(CONFIG_NMI_WATCHDOG) + { + .procname = "nmi_watchdog", + .data = &nmi_watchdog_enabled, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = proc_nmi_enabled, + }, +#endif +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) && !defined(CONFIG_NMI_WATCHDOG) { .procname = "unknown_nmi_panic", .data = &unknown_nmi_panic, diff --git a/kernel/time.c b/kernel/time.c index 656dccfe1cbb..50612faa9baf 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -132,12 +132,11 @@ SYSCALL_DEFINE2(gettimeofday, struct timeval __user *, tv, */ static inline void warp_clock(void) { - write_seqlock_irq(&xtime_lock); - wall_to_monotonic.tv_sec -= sys_tz.tz_minuteswest * 60; - xtime.tv_sec += sys_tz.tz_minuteswest * 60; - update_xtime_cache(0); - write_sequnlock_irq(&xtime_lock); - clock_was_set(); + struct timespec delta, adjust; + delta.tv_sec = sys_tz.tz_minuteswest * 60; + delta.tv_nsec = 0; + adjust = timespec_add_safe(current_kernel_time(), delta); + do_settimeofday(&adjust); } /* diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 7c0f180d6e9d..c63116863a80 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -69,7 +69,7 @@ static s64 time_freq; /* time at last adjustment (secs): */ static long time_reftime; -long time_adjust; +static long time_adjust; /* constant (boot-param configurable) NTP tick adjustment (upscaled) */ static s64 ntp_tick_adj; diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index f992762d7f51..f25735a767af 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -262,6 +262,9 @@ void tick_nohz_stop_sched_tick(int inidle) goto end; } + if (nohz_ratelimit(cpu)) + goto end; + ts->idle_calls++; /* Read jiffies and the time when jiffies were updated last */ do { diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 39f6177fafac..caf8d4d4f5c8 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -165,13 +165,6 @@ struct timespec raw_time; /* flag for if timekeeping is suspended */ int __read_mostly timekeeping_suspended; -static struct timespec xtime_cache __attribute__ ((aligned (16))); -void update_xtime_cache(u64 nsec) -{ - xtime_cache = xtime; - timespec_add_ns(&xtime_cache, nsec); -} - /* must hold xtime_lock */ void timekeeping_leap_insert(int leapsecond) { @@ -332,8 +325,6 @@ int do_settimeofday(struct timespec *tv) xtime = *tv; - update_xtime_cache(0); - timekeeper.ntp_error = 0; ntp_clear(); @@ -559,7 +550,6 @@ void __init timekeeping_init(void) } set_normalized_timespec(&wall_to_monotonic, -boot.tv_sec, -boot.tv_nsec); - update_xtime_cache(0); total_sleep_time.tv_sec = 0; total_sleep_time.tv_nsec = 0; write_sequnlock_irqrestore(&xtime_lock, flags); @@ -593,7 +583,6 @@ static int timekeeping_resume(struct sys_device *dev) wall_to_monotonic = timespec_sub(wall_to_monotonic, ts); total_sleep_time = timespec_add_safe(total_sleep_time, ts); } - update_xtime_cache(0); /* re-base the last cycle value */ timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock); timekeeper.ntp_error = 0; @@ -788,7 +777,6 @@ void update_wall_time(void) { struct clocksource *clock; cycle_t offset; - u64 nsecs; int shift = 0, maxshift; /* Make sure we're fully resumed: */ @@ -847,7 +835,9 @@ void update_wall_time(void) timekeeper.ntp_error += neg << timekeeper.ntp_error_shift; } - /* store full nanoseconds into xtime after rounding it up and + + /* + * Store full nanoseconds into xtime after rounding it up and * add the remainder to the error difference. */ xtime.tv_nsec = ((s64) timekeeper.xtime_nsec >> timekeeper.shift) + 1; @@ -855,8 +845,15 @@ void update_wall_time(void) timekeeper.ntp_error += timekeeper.xtime_nsec << timekeeper.ntp_error_shift; - nsecs = clocksource_cyc2ns(offset, timekeeper.mult, timekeeper.shift); - update_xtime_cache(nsecs); + /* + * Finally, make sure that after the rounding + * xtime.tv_nsec isn't larger then NSEC_PER_SEC + */ + if (unlikely(xtime.tv_nsec >= NSEC_PER_SEC)) { + xtime.tv_nsec -= NSEC_PER_SEC; + xtime.tv_sec++; + second_overflow(); + } /* check to see if there is a new clocksource to use */ update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); @@ -896,13 +893,13 @@ EXPORT_SYMBOL_GPL(monotonic_to_bootbased); unsigned long get_seconds(void) { - return xtime_cache.tv_sec; + return xtime.tv_sec; } EXPORT_SYMBOL(get_seconds); struct timespec __current_kernel_time(void) { - return xtime_cache; + return xtime; } struct timespec current_kernel_time(void) @@ -913,7 +910,7 @@ struct timespec current_kernel_time(void) do { seq = read_seqbegin(&xtime_lock); - now = xtime_cache; + now = xtime; } while (read_seqretry(&xtime_lock, seq)); return now; @@ -928,7 +925,7 @@ struct timespec get_monotonic_coarse(void) do { seq = read_seqbegin(&xtime_lock); - now = xtime_cache; + now = xtime; mono = wall_to_monotonic; } while (read_seqretry(&xtime_lock, seq)); diff --git a/kernel/timer.c b/kernel/timer.c index aeb6a54f2771..9199f3c52215 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -319,6 +319,24 @@ unsigned long round_jiffies_up_relative(unsigned long j) } EXPORT_SYMBOL_GPL(round_jiffies_up_relative); +/** + * set_timer_slack - set the allowed slack for a timer + * @slack_hz: the amount of time (in jiffies) allowed for rounding + * + * Set the amount of time, in jiffies, that a certain timer has + * in terms of slack. By setting this value, the timer subsystem + * will schedule the actual timer somewhere between + * the time mod_timer() asks for, and that time plus the slack. + * + * By setting the slack to -1, a percentage of the delay is used + * instead. + */ +void set_timer_slack(struct timer_list *timer, int slack_hz) +{ + timer->slack = slack_hz; +} +EXPORT_SYMBOL_GPL(set_timer_slack); + static inline void set_running_timer(struct tvec_base *base, struct timer_list *timer) @@ -550,6 +568,7 @@ static void __init_timer(struct timer_list *timer, { timer->entry.next = NULL; timer->base = __raw_get_cpu_var(tvec_bases); + timer->slack = -1; #ifdef CONFIG_TIMER_STATS timer->start_site = NULL; timer->start_pid = -1; @@ -715,6 +734,41 @@ int mod_timer_pending(struct timer_list *timer, unsigned long expires) } EXPORT_SYMBOL(mod_timer_pending); +/* + * Decide where to put the timer while taking the slack into account + * + * Algorithm: + * 1) calculate the maximum (absolute) time + * 2) calculate the highest bit where the expires and new max are different + * 3) use this bit to make a mask + * 4) use the bitmask to round down the maximum time, so that all last + * bits are zeros + */ +static inline +unsigned long apply_slack(struct timer_list *timer, unsigned long expires) +{ + unsigned long expires_limit, mask; + int bit; + + expires_limit = expires + timer->slack; + + if (timer->slack < 0) /* auto slack: use 0.4% */ + expires_limit = expires + (expires - jiffies)/256; + + mask = expires ^ expires_limit; + + if (mask == 0) + return expires; + + bit = find_last_bit(&mask, BITS_PER_LONG); + + mask = (1 << bit) - 1; + + expires_limit = expires_limit & ~(mask); + + return expires_limit; +} + /** * mod_timer - modify a timer's timeout * @timer: the timer to be modified @@ -745,6 +799,8 @@ int mod_timer(struct timer_list *timer, unsigned long expires) if (timer_pending(timer) && timer->expires == expires) return 1; + expires = apply_slack(timer, expires); + return __mod_timer(timer, expires, false, TIMER_NOT_PINNED); } EXPORT_SYMBOL(mod_timer); @@ -955,6 +1011,47 @@ static int cascade(struct tvec_base *base, struct tvec *tv, int index) return index; } +static void call_timer_fn(struct timer_list *timer, void (*fn)(unsigned long), + unsigned long data) +{ + int preempt_count = preempt_count(); + +#ifdef CONFIG_LOCKDEP + /* + * It is permissible to free the timer from inside the + * function that is called from it, this we need to take into + * account for lockdep too. To avoid bogus "held lock freed" + * warnings as well as problems when looking into + * timer->lockdep_map, make a copy and use that here. + */ + struct lockdep_map lockdep_map = timer->lockdep_map; +#endif + /* + * Couple the lock chain with the lock chain at + * del_timer_sync() by acquiring the lock_map around the fn() + * call here and in del_timer_sync(). + */ + lock_map_acquire(&lockdep_map); + + trace_timer_expire_entry(timer); + fn(data); + trace_timer_expire_exit(timer); + + lock_map_release(&lockdep_map); + + if (preempt_count != preempt_count()) { + WARN_ONCE(1, "timer: %pF preempt leak: %08x -> %08x\n", + fn, preempt_count, preempt_count()); + /* + * Restore the preempt count. That gives us a decent + * chance to survive and extract information. If the + * callback kept a lock held, bad luck, but not worse + * than the BUG() we had. + */ + preempt_count() = preempt_count; + } +} + #define INDEX(N) ((base->timer_jiffies >> (TVR_BITS + (N) * TVN_BITS)) & TVN_MASK) /** @@ -998,45 +1095,7 @@ static inline void __run_timers(struct tvec_base *base) detach_timer(timer, 1); spin_unlock_irq(&base->lock); - { - int preempt_count = preempt_count(); - -#ifdef CONFIG_LOCKDEP - /* - * It is permissible to free the timer from - * inside the function that is called from - * it, this we need to take into account for - * lockdep too. To avoid bogus "held lock - * freed" warnings as well as problems when - * looking into timer->lockdep_map, make a - * copy and use that here. - */ - struct lockdep_map lockdep_map = - timer->lockdep_map; -#endif - /* - * Couple the lock chain with the lock chain at - * del_timer_sync() by acquiring the lock_map - * around the fn() call here and in - * del_timer_sync(). - */ - lock_map_acquire(&lockdep_map); - - trace_timer_expire_entry(timer); - fn(data); - trace_timer_expire_exit(timer); - - lock_map_release(&lockdep_map); - - if (preempt_count != preempt_count()) { - printk(KERN_ERR "huh, entered %p " - "with preempt_count %08x, exited" - " with %08x?\n", - fn, preempt_count, - preempt_count()); - BUG(); - } - } + call_timer_fn(timer, fn, data); spin_lock_irq(&base->lock); } } diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 13e13d428cd3..8b1797c4545b 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -44,9 +44,6 @@ config HAVE_FTRACE_MCOUNT_RECORD help See Documentation/trace/ftrace-design.txt -config HAVE_HW_BRANCH_TRACER - bool - config HAVE_SYSCALL_TRACEPOINTS bool help @@ -374,14 +371,6 @@ config STACK_TRACER Say N if unsure. -config HW_BRANCH_TRACER - depends on HAVE_HW_BRANCH_TRACER - bool "Trace hw branches" - select GENERIC_TRACER - help - This tracer records all branches on the system in a circular - buffer, giving access to the last N branches for each cpu. - config KMEMTRACE bool "Trace SLAB allocations" select GENERIC_TRACER diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 78edc6490038..ffb1a5b0550e 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -41,7 +41,6 @@ obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o obj-$(CONFIG_BOOT_TRACER) += trace_boot.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += trace_functions_graph.o obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o -obj-$(CONFIG_HW_BRANCH_TRACER) += trace_hw_branches.o obj-$(CONFIG_KMEMTRACE) += kmemtrace.o obj-$(CONFIG_WORKQUEUE_TRACER) += trace_workqueue.o obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 41ca394feb22..5885cdfc41f3 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -319,6 +319,11 @@ EXPORT_SYMBOL_GPL(ring_buffer_event_data); #define TS_MASK ((1ULL << TS_SHIFT) - 1) #define TS_DELTA_TEST (~TS_MASK) +/* Flag when events were overwritten */ +#define RB_MISSED_EVENTS (1 << 31) +/* Missed count stored at end */ +#define RB_MISSED_STORED (1 << 30) + struct buffer_data_page { u64 time_stamp; /* page time stamp */ local_t commit; /* write committed index */ @@ -338,6 +343,7 @@ struct buffer_page { local_t write; /* index for next write */ unsigned read; /* index for next read */ local_t entries; /* entries on this page */ + unsigned long real_end; /* real end of data */ struct buffer_data_page *page; /* Actual data page */ }; @@ -417,6 +423,12 @@ int ring_buffer_print_page_header(struct trace_seq *s) (unsigned int)sizeof(field.commit), (unsigned int)is_signed_type(long)); + ret = trace_seq_printf(s, "\tfield: int overwrite;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), commit), + 1, + (unsigned int)is_signed_type(long)); + ret = trace_seq_printf(s, "\tfield: char data;\t" "offset:%u;\tsize:%u;\tsigned:%u;\n", (unsigned int)offsetof(typeof(field), data), @@ -440,6 +452,8 @@ struct ring_buffer_per_cpu { struct buffer_page *tail_page; /* write to tail */ struct buffer_page *commit_page; /* committed pages */ struct buffer_page *reader_page; + unsigned long lost_events; + unsigned long last_overrun; local_t commit_overrun; local_t overrun; local_t entries; @@ -1762,6 +1776,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, kmemcheck_annotate_bitfield(event, bitfield); /* + * Save the original length to the meta data. + * This will be used by the reader to add lost event + * counter. + */ + tail_page->real_end = tail; + + /* * If this event is bigger than the minimum size, then * we need to be careful that we don't subtract the * write counter enough to allow another writer to slip @@ -2838,6 +2859,7 @@ static struct buffer_page * rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) { struct buffer_page *reader = NULL; + unsigned long overwrite; unsigned long flags; int nr_loops = 0; int ret; @@ -2879,6 +2901,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) local_set(&cpu_buffer->reader_page->write, 0); local_set(&cpu_buffer->reader_page->entries, 0); local_set(&cpu_buffer->reader_page->page->commit, 0); + cpu_buffer->reader_page->real_end = 0; spin: /* @@ -2899,6 +2922,18 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) rb_set_list_to_head(cpu_buffer, &cpu_buffer->reader_page->list); /* + * We want to make sure we read the overruns after we set up our + * pointers to the next object. The writer side does a + * cmpxchg to cross pages which acts as the mb on the writer + * side. Note, the reader will constantly fail the swap + * while the writer is updating the pointers, so this + * guarantees that the overwrite recorded here is the one we + * want to compare with the last_overrun. + */ + smp_mb(); + overwrite = local_read(&(cpu_buffer->overrun)); + + /* * Here's the tricky part. * * We need to move the pointer past the header page. @@ -2929,6 +2964,11 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->reader_page = reader; rb_reset_reader_page(cpu_buffer); + if (overwrite != cpu_buffer->last_overrun) { + cpu_buffer->lost_events = overwrite - cpu_buffer->last_overrun; + cpu_buffer->last_overrun = overwrite; + } + goto again; out: @@ -3005,8 +3045,14 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) rb_advance_iter(iter); } +static int rb_lost_events(struct ring_buffer_per_cpu *cpu_buffer) +{ + return cpu_buffer->lost_events; +} + static struct ring_buffer_event * -rb_buffer_peek(struct ring_buffer_per_cpu *cpu_buffer, u64 *ts) +rb_buffer_peek(struct ring_buffer_per_cpu *cpu_buffer, u64 *ts, + unsigned long *lost_events) { struct ring_buffer_event *event; struct buffer_page *reader; @@ -3058,6 +3104,8 @@ rb_buffer_peek(struct ring_buffer_per_cpu *cpu_buffer, u64 *ts) ring_buffer_normalize_time_stamp(cpu_buffer->buffer, cpu_buffer->cpu, ts); } + if (lost_events) + *lost_events = rb_lost_events(cpu_buffer); return event; default: @@ -3168,12 +3216,14 @@ static inline int rb_ok_to_lock(void) * @buffer: The ring buffer to read * @cpu: The cpu to peak at * @ts: The timestamp counter of this event. + * @lost_events: a variable to store if events were lost (may be NULL) * * This will return the event that will be read next, but does * not consume the data. */ struct ring_buffer_event * -ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) +ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts, + unsigned long *lost_events) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; @@ -3188,7 +3238,7 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) local_irq_save(flags); if (dolock) spin_lock(&cpu_buffer->reader_lock); - event = rb_buffer_peek(cpu_buffer, ts); + event = rb_buffer_peek(cpu_buffer, ts, lost_events); if (event && event->type_len == RINGBUF_TYPE_PADDING) rb_advance_reader(cpu_buffer); if (dolock) @@ -3230,13 +3280,17 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) /** * ring_buffer_consume - return an event and consume it * @buffer: The ring buffer to get the next event from + * @cpu: the cpu to read the buffer from + * @ts: a variable to store the timestamp (may be NULL) + * @lost_events: a variable to store if events were lost (may be NULL) * * Returns the next event in the ring buffer, and that event is consumed. * Meaning, that sequential reads will keep returning a different event, * and eventually empty the ring buffer if the producer is slower. */ struct ring_buffer_event * -ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts) +ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, + unsigned long *lost_events) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event = NULL; @@ -3257,9 +3311,11 @@ ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts) if (dolock) spin_lock(&cpu_buffer->reader_lock); - event = rb_buffer_peek(cpu_buffer, ts); - if (event) + event = rb_buffer_peek(cpu_buffer, ts, lost_events); + if (event) { + cpu_buffer->lost_events = 0; rb_advance_reader(cpu_buffer); + } if (dolock) spin_unlock(&cpu_buffer->reader_lock); @@ -3408,6 +3464,9 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->write_stamp = 0; cpu_buffer->read_stamp = 0; + cpu_buffer->lost_events = 0; + cpu_buffer->last_overrun = 0; + rb_head_page_activate(cpu_buffer); } @@ -3683,6 +3742,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer, struct ring_buffer_event *event; struct buffer_data_page *bpage; struct buffer_page *reader; + unsigned long missed_events; unsigned long flags; unsigned int commit; unsigned int read; @@ -3719,6 +3779,9 @@ int ring_buffer_read_page(struct ring_buffer *buffer, read = reader->read; commit = rb_page_commit(reader); + /* Check if any events were dropped */ + missed_events = cpu_buffer->lost_events; + /* * If this page has been partially read or * if len is not big enough to read the rest of the page or @@ -3779,9 +3842,35 @@ int ring_buffer_read_page(struct ring_buffer *buffer, local_set(&reader->entries, 0); reader->read = 0; *data_page = bpage; + + /* + * Use the real_end for the data size, + * This gives us a chance to store the lost events + * on the page. + */ + if (reader->real_end) + local_set(&bpage->commit, reader->real_end); } ret = read; + cpu_buffer->lost_events = 0; + /* + * Set a flag in the commit field if we lost events + */ + if (missed_events) { + commit = local_read(&bpage->commit); + + /* If there is room at the end of the page to save the + * missed events, then record it there. + */ + if (BUF_PAGE_SIZE - commit >= sizeof(missed_events)) { + memcpy(&bpage->data[commit], &missed_events, + sizeof(missed_events)); + local_add(RB_MISSED_STORED, &bpage->commit); + } + local_add(RB_MISSED_EVENTS, &bpage->commit); + } + out_unlock: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index df74c7982255..dc56556b55a2 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -81,7 +81,7 @@ static enum event_status read_event(int cpu) int *entry; u64 ts; - event = ring_buffer_consume(buffer, cpu, &ts); + event = ring_buffer_consume(buffer, cpu, &ts, NULL); if (!event) return EVENT_DROPPED; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 44f916a04065..bed83cab6da2 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1545,7 +1545,8 @@ static void trace_iterator_increment(struct trace_iterator *iter) } static struct trace_entry * -peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts) +peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts, + unsigned long *lost_events) { struct ring_buffer_event *event; struct ring_buffer_iter *buf_iter = iter->buffer_iter[cpu]; @@ -1556,7 +1557,8 @@ peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts) if (buf_iter) event = ring_buffer_iter_peek(buf_iter, ts); else - event = ring_buffer_peek(iter->tr->buffer, cpu, ts); + event = ring_buffer_peek(iter->tr->buffer, cpu, ts, + lost_events); ftrace_enable_cpu(); @@ -1564,10 +1566,12 @@ peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts) } static struct trace_entry * -__find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) +__find_next_entry(struct trace_iterator *iter, int *ent_cpu, + unsigned long *missing_events, u64 *ent_ts) { struct ring_buffer *buffer = iter->tr->buffer; struct trace_entry *ent, *next = NULL; + unsigned long lost_events = 0, next_lost = 0; int cpu_file = iter->cpu_file; u64 next_ts = 0, ts; int next_cpu = -1; @@ -1580,7 +1584,7 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) if (cpu_file > TRACE_PIPE_ALL_CPU) { if (ring_buffer_empty_cpu(buffer, cpu_file)) return NULL; - ent = peek_next_entry(iter, cpu_file, ent_ts); + ent = peek_next_entry(iter, cpu_file, ent_ts, missing_events); if (ent_cpu) *ent_cpu = cpu_file; @@ -1592,7 +1596,7 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) if (ring_buffer_empty_cpu(buffer, cpu)) continue; - ent = peek_next_entry(iter, cpu, &ts); + ent = peek_next_entry(iter, cpu, &ts, &lost_events); /* * Pick the entry with the smallest timestamp: @@ -1601,6 +1605,7 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) next = ent; next_cpu = cpu; next_ts = ts; + next_lost = lost_events; } } @@ -1610,6 +1615,9 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) if (ent_ts) *ent_ts = next_ts; + if (missing_events) + *missing_events = next_lost; + return next; } @@ -1617,13 +1625,14 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) { - return __find_next_entry(iter, ent_cpu, ent_ts); + return __find_next_entry(iter, ent_cpu, NULL, ent_ts); } /* Find the next real entry, and increment the iterator to the next entry */ static void *find_next_entry_inc(struct trace_iterator *iter) { - iter->ent = __find_next_entry(iter, &iter->cpu, &iter->ts); + iter->ent = __find_next_entry(iter, &iter->cpu, + &iter->lost_events, &iter->ts); if (iter->ent) trace_iterator_increment(iter); @@ -1635,7 +1644,8 @@ static void trace_consume(struct trace_iterator *iter) { /* Don't allow ftrace to trace into the ring buffers */ ftrace_disable_cpu(); - ring_buffer_consume(iter->tr->buffer, iter->cpu, &iter->ts); + ring_buffer_consume(iter->tr->buffer, iter->cpu, &iter->ts, + &iter->lost_events); ftrace_enable_cpu(); } @@ -2030,6 +2040,10 @@ static enum print_line_t print_trace_line(struct trace_iterator *iter) { enum print_line_t ret; + if (iter->lost_events) + trace_seq_printf(&iter->seq, "CPU:%d [LOST %lu EVENTS]\n", + iter->cpu, iter->lost_events); + if (iter->trace && iter->trace->print_line) { ret = iter->trace->print_line(iter); if (ret != TRACE_TYPE_UNHANDLED) diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 2825ef2c0b15..3ebdb6bd2362 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -34,7 +34,6 @@ enum trace_type { TRACE_GRAPH_RET, TRACE_GRAPH_ENT, TRACE_USER_STACK, - TRACE_HW_BRANCHES, TRACE_KMEM_ALLOC, TRACE_KMEM_FREE, TRACE_BLK, @@ -103,29 +102,17 @@ struct syscall_trace_exit { long ret; }; -struct kprobe_trace_entry { +struct kprobe_trace_entry_head { struct trace_entry ent; unsigned long ip; - int nargs; - unsigned long args[]; }; -#define SIZEOF_KPROBE_TRACE_ENTRY(n) \ - (offsetof(struct kprobe_trace_entry, args) + \ - (sizeof(unsigned long) * (n))) - -struct kretprobe_trace_entry { +struct kretprobe_trace_entry_head { struct trace_entry ent; unsigned long func; unsigned long ret_ip; - int nargs; - unsigned long args[]; }; -#define SIZEOF_KRETPROBE_TRACE_ENTRY(n) \ - (offsetof(struct kretprobe_trace_entry, args) + \ - (sizeof(unsigned long) * (n))) - /* * trace_flag_type is an enumeration that holds different * states when a trace occurs. These are: @@ -229,7 +216,6 @@ extern void __ftrace_bad_type(void); TRACE_GRAPH_ENT); \ IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry, \ TRACE_GRAPH_RET); \ - IF_ASSIGN(var, ent, struct hw_branch_entry, TRACE_HW_BRANCHES);\ IF_ASSIGN(var, ent, struct kmemtrace_alloc_entry, \ TRACE_KMEM_ALLOC); \ IF_ASSIGN(var, ent, struct kmemtrace_free_entry, \ @@ -467,8 +453,6 @@ extern int trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr); -extern int trace_selftest_startup_hw_branches(struct tracer *trace, - struct trace_array *tr); extern int trace_selftest_startup_ksym(struct tracer *trace, struct trace_array *tr); #endif /* CONFIG_FTRACE_STARTUP_TEST */ diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index c16a08f399df..dc008c1240da 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -318,18 +318,6 @@ FTRACE_ENTRY(branch, trace_branch, __entry->func, __entry->file, __entry->correct) ); -FTRACE_ENTRY(hw_branch, hw_branch_entry, - - TRACE_HW_BRANCHES, - - F_STRUCT( - __field( u64, from ) - __field( u64, to ) - ), - - F_printk("from: %llx to: %llx", __entry->from, __entry->to) -); - FTRACE_ENTRY(kmem_alloc, kmemtrace_alloc_entry, TRACE_KMEM_ALLOC, diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 9aed1a5cf553..669b9c31861d 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -490,9 +490,10 @@ get_return_for_leaf(struct trace_iterator *iter, * We need to consume the current entry to see * the next one. */ - ring_buffer_consume(iter->tr->buffer, iter->cpu, NULL); + ring_buffer_consume(iter->tr->buffer, iter->cpu, + NULL, NULL); event = ring_buffer_peek(iter->tr->buffer, iter->cpu, - NULL); + NULL, NULL); } if (!event) diff --git a/kernel/trace/trace_hw_branches.c b/kernel/trace/trace_hw_branches.c deleted file mode 100644 index 7b97000745f5..000000000000 --- a/kernel/trace/trace_hw_branches.c +++ /dev/null @@ -1,312 +0,0 @@ -/* - * h/w branch tracer for x86 based on BTS - * - * Copyright (C) 2008-2009 Intel Corporation. - * Markus Metzger <markus.t.metzger@gmail.com>, 2008-2009 - */ -#include <linux/kallsyms.h> -#include <linux/debugfs.h> -#include <linux/ftrace.h> -#include <linux/module.h> -#include <linux/cpu.h> -#include <linux/smp.h> -#include <linux/fs.h> - -#include <asm/ds.h> - -#include "trace_output.h" -#include "trace.h" - - -#define BTS_BUFFER_SIZE (1 << 13) - -static DEFINE_PER_CPU(struct bts_tracer *, hwb_tracer); -static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], hwb_buffer); - -#define this_tracer per_cpu(hwb_tracer, smp_processor_id()) - -static int trace_hw_branches_enabled __read_mostly; -static int trace_hw_branches_suspended __read_mostly; -static struct trace_array *hw_branch_trace __read_mostly; - - -static void bts_trace_init_cpu(int cpu) -{ - per_cpu(hwb_tracer, cpu) = - ds_request_bts_cpu(cpu, per_cpu(hwb_buffer, cpu), - BTS_BUFFER_SIZE, NULL, (size_t)-1, - BTS_KERNEL); - - if (IS_ERR(per_cpu(hwb_tracer, cpu))) - per_cpu(hwb_tracer, cpu) = NULL; -} - -static int bts_trace_init(struct trace_array *tr) -{ - int cpu; - - hw_branch_trace = tr; - trace_hw_branches_enabled = 0; - - get_online_cpus(); - for_each_online_cpu(cpu) { - bts_trace_init_cpu(cpu); - - if (likely(per_cpu(hwb_tracer, cpu))) - trace_hw_branches_enabled = 1; - } - trace_hw_branches_suspended = 0; - put_online_cpus(); - - /* If we could not enable tracing on a single cpu, we fail. */ - return trace_hw_branches_enabled ? 0 : -EOPNOTSUPP; -} - -static void bts_trace_reset(struct trace_array *tr) -{ - int cpu; - - get_online_cpus(); - for_each_online_cpu(cpu) { - if (likely(per_cpu(hwb_tracer, cpu))) { - ds_release_bts(per_cpu(hwb_tracer, cpu)); - per_cpu(hwb_tracer, cpu) = NULL; - } - } - trace_hw_branches_enabled = 0; - trace_hw_branches_suspended = 0; - put_online_cpus(); -} - -static void bts_trace_start(struct trace_array *tr) -{ - int cpu; - - get_online_cpus(); - for_each_online_cpu(cpu) - if (likely(per_cpu(hwb_tracer, cpu))) - ds_resume_bts(per_cpu(hwb_tracer, cpu)); - trace_hw_branches_suspended = 0; - put_online_cpus(); -} - -static void bts_trace_stop(struct trace_array *tr) -{ - int cpu; - - get_online_cpus(); - for_each_online_cpu(cpu) - if (likely(per_cpu(hwb_tracer, cpu))) - ds_suspend_bts(per_cpu(hwb_tracer, cpu)); - trace_hw_branches_suspended = 1; - put_online_cpus(); -} - -static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb, - unsigned long action, void *hcpu) -{ - int cpu = (long)hcpu; - - switch (action) { - case CPU_ONLINE: - case CPU_DOWN_FAILED: - /* The notification is sent with interrupts enabled. */ - if (trace_hw_branches_enabled) { - bts_trace_init_cpu(cpu); - - if (trace_hw_branches_suspended && - likely(per_cpu(hwb_tracer, cpu))) - ds_suspend_bts(per_cpu(hwb_tracer, cpu)); - } - break; - - case CPU_DOWN_PREPARE: - /* The notification is sent with interrupts enabled. */ - if (likely(per_cpu(hwb_tracer, cpu))) { - ds_release_bts(per_cpu(hwb_tracer, cpu)); - per_cpu(hwb_tracer, cpu) = NULL; - } - } - - return NOTIFY_DONE; -} - -static struct notifier_block bts_hotcpu_notifier __cpuinitdata = { - .notifier_call = bts_hotcpu_handler -}; - -static void bts_trace_print_header(struct seq_file *m) -{ - seq_puts(m, "# CPU# TO <- FROM\n"); -} - -static enum print_line_t bts_trace_print_line(struct trace_iterator *iter) -{ - unsigned long symflags = TRACE_ITER_SYM_OFFSET; - struct trace_entry *entry = iter->ent; - struct trace_seq *seq = &iter->seq; - struct hw_branch_entry *it; - - trace_assign_type(it, entry); - - if (entry->type == TRACE_HW_BRANCHES) { - if (trace_seq_printf(seq, "%4d ", iter->cpu) && - seq_print_ip_sym(seq, it->to, symflags) && - trace_seq_printf(seq, "\t <- ") && - seq_print_ip_sym(seq, it->from, symflags) && - trace_seq_printf(seq, "\n")) - return TRACE_TYPE_HANDLED; - return TRACE_TYPE_PARTIAL_LINE; - } - return TRACE_TYPE_UNHANDLED; -} - -void trace_hw_branch(u64 from, u64 to) -{ - struct ftrace_event_call *call = &event_hw_branch; - struct trace_array *tr = hw_branch_trace; - struct ring_buffer_event *event; - struct ring_buffer *buf; - struct hw_branch_entry *entry; - unsigned long irq1; - int cpu; - - if (unlikely(!tr)) - return; - - if (unlikely(!trace_hw_branches_enabled)) - return; - - local_irq_save(irq1); - cpu = raw_smp_processor_id(); - if (atomic_inc_return(&tr->data[cpu]->disabled) != 1) - goto out; - - buf = tr->buffer; - event = trace_buffer_lock_reserve(buf, TRACE_HW_BRANCHES, - sizeof(*entry), 0, 0); - if (!event) - goto out; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, from); - entry->ent.type = TRACE_HW_BRANCHES; - entry->from = from; - entry->to = to; - if (!filter_check_discard(call, entry, buf, event)) - trace_buffer_unlock_commit(buf, event, 0, 0); - - out: - atomic_dec(&tr->data[cpu]->disabled); - local_irq_restore(irq1); -} - -static void trace_bts_at(const struct bts_trace *trace, void *at) -{ - struct bts_struct bts; - int err = 0; - - WARN_ON_ONCE(!trace->read); - if (!trace->read) - return; - - err = trace->read(this_tracer, at, &bts); - if (err < 0) - return; - - switch (bts.qualifier) { - case BTS_BRANCH: - trace_hw_branch(bts.variant.lbr.from, bts.variant.lbr.to); - break; - } -} - -/* - * Collect the trace on the current cpu and write it into the ftrace buffer. - * - * pre: tracing must be suspended on the current cpu - */ -static void trace_bts_cpu(void *arg) -{ - struct trace_array *tr = (struct trace_array *)arg; - const struct bts_trace *trace; - unsigned char *at; - - if (unlikely(!tr)) - return; - - if (unlikely(atomic_read(&tr->data[raw_smp_processor_id()]->disabled))) - return; - - if (unlikely(!this_tracer)) - return; - - trace = ds_read_bts(this_tracer); - if (!trace) - return; - - for (at = trace->ds.top; (void *)at < trace->ds.end; - at += trace->ds.size) - trace_bts_at(trace, at); - - for (at = trace->ds.begin; (void *)at < trace->ds.top; - at += trace->ds.size) - trace_bts_at(trace, at); -} - -static void trace_bts_prepare(struct trace_iterator *iter) -{ - int cpu; - - get_online_cpus(); - for_each_online_cpu(cpu) - if (likely(per_cpu(hwb_tracer, cpu))) - ds_suspend_bts(per_cpu(hwb_tracer, cpu)); - /* - * We need to collect the trace on the respective cpu since ftrace - * implicitly adds the record for the current cpu. - * Once that is more flexible, we could collect the data from any cpu. - */ - on_each_cpu(trace_bts_cpu, iter->tr, 1); - - for_each_online_cpu(cpu) - if (likely(per_cpu(hwb_tracer, cpu))) - ds_resume_bts(per_cpu(hwb_tracer, cpu)); - put_online_cpus(); -} - -static void trace_bts_close(struct trace_iterator *iter) -{ - tracing_reset_online_cpus(iter->tr); -} - -void trace_hw_branch_oops(void) -{ - if (this_tracer) { - ds_suspend_bts_noirq(this_tracer); - trace_bts_cpu(hw_branch_trace); - ds_resume_bts_noirq(this_tracer); - } -} - -struct tracer bts_tracer __read_mostly = -{ - .name = "hw-branch-tracer", - .init = bts_trace_init, - .reset = bts_trace_reset, - .print_header = bts_trace_print_header, - .print_line = bts_trace_print_line, - .start = bts_trace_start, - .stop = bts_trace_stop, - .open = trace_bts_prepare, - .close = trace_bts_close, -#ifdef CONFIG_FTRACE_SELFTEST - .selftest = trace_selftest_startup_hw_branches, -#endif /* CONFIG_FTRACE_SELFTEST */ -}; - -__init static int init_bts_trace(void) -{ - register_hotcpu_notifier(&bts_hotcpu_notifier); - return register_tracer(&bts_tracer); -} -device_initcall(init_bts_trace); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 1251e367bae9..a7514326052b 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -29,6 +29,8 @@ #include <linux/ctype.h> #include <linux/ptrace.h> #include <linux/perf_event.h> +#include <linux/stringify.h> +#include <asm/bitsperlong.h> #include "trace.h" #include "trace_output.h" @@ -40,7 +42,6 @@ /* Reserved field names */ #define FIELD_STRING_IP "__probe_ip" -#define FIELD_STRING_NARGS "__probe_nargs" #define FIELD_STRING_RETIP "__probe_ret_ip" #define FIELD_STRING_FUNC "__probe_func" @@ -52,56 +53,102 @@ const char *reserved_field_names[] = { "common_tgid", "common_lock_depth", FIELD_STRING_IP, - FIELD_STRING_NARGS, FIELD_STRING_RETIP, FIELD_STRING_FUNC, }; -struct fetch_func { - unsigned long (*func)(struct pt_regs *, void *); +/* Printing function type */ +typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *); +#define PRINT_TYPE_FUNC_NAME(type) print_type_##type +#define PRINT_TYPE_FMT_NAME(type) print_type_format_##type + +/* Printing in basic type function template */ +#define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt, cast) \ +static __kprobes int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s, \ + const char *name, void *data)\ +{ \ + return trace_seq_printf(s, " %s=" fmt, name, (cast)*(type *)data);\ +} \ +static const char PRINT_TYPE_FMT_NAME(type)[] = fmt; + +DEFINE_BASIC_PRINT_TYPE_FUNC(u8, "%x", unsigned int) +DEFINE_BASIC_PRINT_TYPE_FUNC(u16, "%x", unsigned int) +DEFINE_BASIC_PRINT_TYPE_FUNC(u32, "%lx", unsigned long) +DEFINE_BASIC_PRINT_TYPE_FUNC(u64, "%llx", unsigned long long) +DEFINE_BASIC_PRINT_TYPE_FUNC(s8, "%d", int) +DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d", int) +DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%ld", long) +DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%lld", long long) + +/* Data fetch function type */ +typedef void (*fetch_func_t)(struct pt_regs *, void *, void *); + +struct fetch_param { + fetch_func_t fn; void *data; }; -static __kprobes unsigned long call_fetch(struct fetch_func *f, - struct pt_regs *regs) +static __kprobes void call_fetch(struct fetch_param *fprm, + struct pt_regs *regs, void *dest) { - return f->func(regs, f->data); + return fprm->fn(regs, fprm->data, dest); } -/* fetch handlers */ -static __kprobes unsigned long fetch_register(struct pt_regs *regs, - void *offset) -{ - return regs_get_register(regs, (unsigned int)((unsigned long)offset)); +#define FETCH_FUNC_NAME(kind, type) fetch_##kind##_##type +/* + * Define macro for basic types - we don't need to define s* types, because + * we have to care only about bitwidth at recording time. + */ +#define DEFINE_BASIC_FETCH_FUNCS(kind) \ +DEFINE_FETCH_##kind(u8) \ +DEFINE_FETCH_##kind(u16) \ +DEFINE_FETCH_##kind(u32) \ +DEFINE_FETCH_##kind(u64) + +#define CHECK_BASIC_FETCH_FUNCS(kind, fn) \ + ((FETCH_FUNC_NAME(kind, u8) == fn) || \ + (FETCH_FUNC_NAME(kind, u16) == fn) || \ + (FETCH_FUNC_NAME(kind, u32) == fn) || \ + (FETCH_FUNC_NAME(kind, u64) == fn)) + +/* Data fetch function templates */ +#define DEFINE_FETCH_reg(type) \ +static __kprobes void FETCH_FUNC_NAME(reg, type)(struct pt_regs *regs, \ + void *offset, void *dest) \ +{ \ + *(type *)dest = (type)regs_get_register(regs, \ + (unsigned int)((unsigned long)offset)); \ } - -static __kprobes unsigned long fetch_stack(struct pt_regs *regs, - void *num) -{ - return regs_get_kernel_stack_nth(regs, - (unsigned int)((unsigned long)num)); +DEFINE_BASIC_FETCH_FUNCS(reg) + +#define DEFINE_FETCH_stack(type) \ +static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\ + void *offset, void *dest) \ +{ \ + *(type *)dest = (type)regs_get_kernel_stack_nth(regs, \ + (unsigned int)((unsigned long)offset)); \ } +DEFINE_BASIC_FETCH_FUNCS(stack) -static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr) -{ - unsigned long retval; - - if (probe_kernel_address(addr, retval)) - return 0; - return retval; +#define DEFINE_FETCH_retval(type) \ +static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\ + void *dummy, void *dest) \ +{ \ + *(type *)dest = (type)regs_return_value(regs); \ } - -static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs, - void *dummy) -{ - return regs_return_value(regs); -} - -static __kprobes unsigned long fetch_stack_address(struct pt_regs *regs, - void *dummy) -{ - return kernel_stack_pointer(regs); +DEFINE_BASIC_FETCH_FUNCS(retval) + +#define DEFINE_FETCH_memory(type) \ +static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\ + void *addr, void *dest) \ +{ \ + type retval; \ + if (probe_kernel_address(addr, retval)) \ + *(type *)dest = 0; \ + else \ + *(type *)dest = retval; \ } +DEFINE_BASIC_FETCH_FUNCS(memory) /* Memory fetching by symbol */ struct symbol_cache { @@ -145,51 +192,126 @@ static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset) return sc; } -static __kprobes unsigned long fetch_symbol(struct pt_regs *regs, void *data) -{ - struct symbol_cache *sc = data; - - if (sc->addr) - return fetch_memory(regs, (void *)sc->addr); - else - return 0; +#define DEFINE_FETCH_symbol(type) \ +static __kprobes void FETCH_FUNC_NAME(symbol, type)(struct pt_regs *regs,\ + void *data, void *dest) \ +{ \ + struct symbol_cache *sc = data; \ + if (sc->addr) \ + fetch_memory_##type(regs, (void *)sc->addr, dest); \ + else \ + *(type *)dest = 0; \ } +DEFINE_BASIC_FETCH_FUNCS(symbol) -/* Special indirect memory access interface */ -struct indirect_fetch_data { - struct fetch_func orig; +/* Dereference memory access function */ +struct deref_fetch_param { + struct fetch_param orig; long offset; }; -static __kprobes unsigned long fetch_indirect(struct pt_regs *regs, void *data) -{ - struct indirect_fetch_data *ind = data; - unsigned long addr; - - addr = call_fetch(&ind->orig, regs); - if (addr) { - addr += ind->offset; - return fetch_memory(regs, (void *)addr); - } else - return 0; +#define DEFINE_FETCH_deref(type) \ +static __kprobes void FETCH_FUNC_NAME(deref, type)(struct pt_regs *regs,\ + void *data, void *dest) \ +{ \ + struct deref_fetch_param *dprm = data; \ + unsigned long addr; \ + call_fetch(&dprm->orig, regs, &addr); \ + if (addr) { \ + addr += dprm->offset; \ + fetch_memory_##type(regs, (void *)addr, dest); \ + } else \ + *(type *)dest = 0; \ } +DEFINE_BASIC_FETCH_FUNCS(deref) -static __kprobes void free_indirect_fetch_data(struct indirect_fetch_data *data) +static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data) { - if (data->orig.func == fetch_indirect) - free_indirect_fetch_data(data->orig.data); - else if (data->orig.func == fetch_symbol) + if (CHECK_BASIC_FETCH_FUNCS(deref, data->orig.fn)) + free_deref_fetch_param(data->orig.data); + else if (CHECK_BASIC_FETCH_FUNCS(symbol, data->orig.fn)) free_symbol_cache(data->orig.data); kfree(data); } +/* Default (unsigned long) fetch type */ +#define __DEFAULT_FETCH_TYPE(t) u##t +#define _DEFAULT_FETCH_TYPE(t) __DEFAULT_FETCH_TYPE(t) +#define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG) +#define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE) + +#define ASSIGN_FETCH_FUNC(kind, type) \ + .kind = FETCH_FUNC_NAME(kind, type) + +#define ASSIGN_FETCH_TYPE(ptype, ftype, sign) \ + {.name = #ptype, \ + .size = sizeof(ftype), \ + .is_signed = sign, \ + .print = PRINT_TYPE_FUNC_NAME(ptype), \ + .fmt = PRINT_TYPE_FMT_NAME(ptype), \ +ASSIGN_FETCH_FUNC(reg, ftype), \ +ASSIGN_FETCH_FUNC(stack, ftype), \ +ASSIGN_FETCH_FUNC(retval, ftype), \ +ASSIGN_FETCH_FUNC(memory, ftype), \ +ASSIGN_FETCH_FUNC(symbol, ftype), \ +ASSIGN_FETCH_FUNC(deref, ftype), \ + } + +/* Fetch type information table */ +static const struct fetch_type { + const char *name; /* Name of type */ + size_t size; /* Byte size of type */ + int is_signed; /* Signed flag */ + print_type_func_t print; /* Print functions */ + const char *fmt; /* Fromat string */ + /* Fetch functions */ + fetch_func_t reg; + fetch_func_t stack; + fetch_func_t retval; + fetch_func_t memory; + fetch_func_t symbol; + fetch_func_t deref; +} fetch_type_table[] = { + ASSIGN_FETCH_TYPE(u8, u8, 0), + ASSIGN_FETCH_TYPE(u16, u16, 0), + ASSIGN_FETCH_TYPE(u32, u32, 0), + ASSIGN_FETCH_TYPE(u64, u64, 0), + ASSIGN_FETCH_TYPE(s8, u8, 1), + ASSIGN_FETCH_TYPE(s16, u16, 1), + ASSIGN_FETCH_TYPE(s32, u32, 1), + ASSIGN_FETCH_TYPE(s64, u64, 1), +}; + +static const struct fetch_type *find_fetch_type(const char *type) +{ + int i; + + if (!type) + type = DEFAULT_FETCH_TYPE_STR; + + for (i = 0; i < ARRAY_SIZE(fetch_type_table); i++) + if (strcmp(type, fetch_type_table[i].name) == 0) + return &fetch_type_table[i]; + return NULL; +} + +/* Special function : only accept unsigned long */ +static __kprobes void fetch_stack_address(struct pt_regs *regs, + void *dummy, void *dest) +{ + *(unsigned long *)dest = kernel_stack_pointer(regs); +} + /** * Kprobe event core functions */ struct probe_arg { - struct fetch_func fetch; - const char *name; + struct fetch_param fetch; + unsigned int offset; /* Offset from argument entry */ + const char *name; /* Name of this argument */ + const char *comm; /* Command of this argument */ + const struct fetch_type *type; /* Type of this argument */ }; /* Flags for trace_probe */ @@ -204,6 +326,7 @@ struct trace_probe { const char *symbol; /* symbol name */ struct ftrace_event_call call; struct trace_event event; + ssize_t size; /* trace entry size */ unsigned int nr_args; struct probe_arg args[]; }; @@ -212,6 +335,7 @@ struct trace_probe { (offsetof(struct trace_probe, args) + \ (sizeof(struct probe_arg) * (n))) + static __kprobes int probe_is_return(struct trace_probe *tp) { return tp->rp.handler != NULL; @@ -222,49 +346,6 @@ static __kprobes const char *probe_symbol(struct trace_probe *tp) return tp->symbol ? tp->symbol : "unknown"; } -static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff) -{ - int ret = -EINVAL; - - if (ff->func == fetch_register) { - const char *name; - name = regs_query_register_name((unsigned int)((long)ff->data)); - ret = snprintf(buf, n, "%%%s", name); - } else if (ff->func == fetch_stack) - ret = snprintf(buf, n, "$stack%lu", (unsigned long)ff->data); - else if (ff->func == fetch_memory) - ret = snprintf(buf, n, "@0x%p", ff->data); - else if (ff->func == fetch_symbol) { - struct symbol_cache *sc = ff->data; - if (sc->offset) - ret = snprintf(buf, n, "@%s%+ld", sc->symbol, - sc->offset); - else - ret = snprintf(buf, n, "@%s", sc->symbol); - } else if (ff->func == fetch_retvalue) - ret = snprintf(buf, n, "$retval"); - else if (ff->func == fetch_stack_address) - ret = snprintf(buf, n, "$stack"); - else if (ff->func == fetch_indirect) { - struct indirect_fetch_data *id = ff->data; - size_t l = 0; - ret = snprintf(buf, n, "%+ld(", id->offset); - if (ret >= n) - goto end; - l += ret; - ret = probe_arg_string(buf + l, n - l, &id->orig); - if (ret < 0) - goto end; - l += ret; - ret = snprintf(buf + l, n - l, ")"); - ret += l; - } -end: - if (ret >= n) - return -ENOSPC; - return ret; -} - static int register_probe_event(struct trace_probe *tp); static void unregister_probe_event(struct trace_probe *tp); @@ -347,11 +428,12 @@ error: static void free_probe_arg(struct probe_arg *arg) { - if (arg->fetch.func == fetch_symbol) + if (CHECK_BASIC_FETCH_FUNCS(deref, arg->fetch.fn)) + free_deref_fetch_param(arg->fetch.data); + else if (CHECK_BASIC_FETCH_FUNCS(symbol, arg->fetch.fn)) free_symbol_cache(arg->fetch.data); - else if (arg->fetch.func == fetch_indirect) - free_indirect_fetch_data(arg->fetch.data); kfree(arg->name); + kfree(arg->comm); } static void free_trace_probe(struct trace_probe *tp) @@ -457,28 +539,30 @@ static int split_symbol_offset(char *symbol, unsigned long *offset) #define PARAM_MAX_ARGS 16 #define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long)) -static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return) +static int parse_probe_vars(char *arg, const struct fetch_type *t, + struct fetch_param *f, int is_return) { int ret = 0; unsigned long param; if (strcmp(arg, "retval") == 0) { - if (is_return) { - ff->func = fetch_retvalue; - ff->data = NULL; - } else + if (is_return) + f->fn = t->retval; + else ret = -EINVAL; } else if (strncmp(arg, "stack", 5) == 0) { if (arg[5] == '\0') { - ff->func = fetch_stack_address; - ff->data = NULL; + if (strcmp(t->name, DEFAULT_FETCH_TYPE_STR) == 0) + f->fn = fetch_stack_address; + else + ret = -EINVAL; } else if (isdigit(arg[5])) { ret = strict_strtoul(arg + 5, 10, ¶m); if (ret || param > PARAM_MAX_STACK) ret = -EINVAL; else { - ff->func = fetch_stack; - ff->data = (void *)param; + f->fn = t->stack; + f->data = (void *)param; } } else ret = -EINVAL; @@ -488,7 +572,8 @@ static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return) } /* Recursive argument parser */ -static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) +static int __parse_probe_arg(char *arg, const struct fetch_type *t, + struct fetch_param *f, int is_return) { int ret = 0; unsigned long param; @@ -497,13 +582,13 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) switch (arg[0]) { case '$': - ret = parse_probe_vars(arg + 1, ff, is_return); + ret = parse_probe_vars(arg + 1, t, f, is_return); break; case '%': /* named register */ ret = regs_query_register_offset(arg + 1); if (ret >= 0) { - ff->func = fetch_register; - ff->data = (void *)(unsigned long)ret; + f->fn = t->reg; + f->data = (void *)(unsigned long)ret; ret = 0; } break; @@ -512,26 +597,22 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) ret = strict_strtoul(arg + 1, 0, ¶m); if (ret) break; - ff->func = fetch_memory; - ff->data = (void *)param; + f->fn = t->memory; + f->data = (void *)param; } else { ret = split_symbol_offset(arg + 1, &offset); if (ret) break; - ff->data = alloc_symbol_cache(arg + 1, offset); - if (ff->data) - ff->func = fetch_symbol; - else - ret = -EINVAL; + f->data = alloc_symbol_cache(arg + 1, offset); + if (f->data) + f->fn = t->symbol; } break; - case '+': /* indirect memory */ + case '+': /* deref memory */ case '-': tmp = strchr(arg, '('); - if (!tmp) { - ret = -EINVAL; + if (!tmp) break; - } *tmp = '\0'; ret = strict_strtol(arg + 1, 0, &offset); if (ret) @@ -541,38 +622,58 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) arg = tmp + 1; tmp = strrchr(arg, ')'); if (tmp) { - struct indirect_fetch_data *id; + struct deref_fetch_param *dprm; + const struct fetch_type *t2 = find_fetch_type(NULL); *tmp = '\0'; - id = kzalloc(sizeof(struct indirect_fetch_data), - GFP_KERNEL); - if (!id) + dprm = kzalloc(sizeof(struct deref_fetch_param), + GFP_KERNEL); + if (!dprm) return -ENOMEM; - id->offset = offset; - ret = __parse_probe_arg(arg, &id->orig, is_return); + dprm->offset = offset; + ret = __parse_probe_arg(arg, t2, &dprm->orig, + is_return); if (ret) - kfree(id); + kfree(dprm); else { - ff->func = fetch_indirect; - ff->data = (void *)id; + f->fn = t->deref; + f->data = (void *)dprm; } - } else - ret = -EINVAL; + } break; - default: - /* TODO: support custom handler */ - ret = -EINVAL; } + if (!ret && !f->fn) + ret = -EINVAL; return ret; } /* String length checking wrapper */ -static int parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) +static int parse_probe_arg(char *arg, struct trace_probe *tp, + struct probe_arg *parg, int is_return) { + const char *t; + if (strlen(arg) > MAX_ARGSTR_LEN) { pr_info("Argument is too long.: %s\n", arg); return -ENOSPC; } - return __parse_probe_arg(arg, ff, is_return); + parg->comm = kstrdup(arg, GFP_KERNEL); + if (!parg->comm) { + pr_info("Failed to allocate memory for command '%s'.\n", arg); + return -ENOMEM; + } + t = strchr(parg->comm, ':'); + if (t) { + arg[t - parg->comm] = '\0'; + t++; + } + parg->type = find_fetch_type(t); + if (!parg->type) { + pr_info("Unsupported type: %s\n", t); + return -EINVAL; + } + parg->offset = tp->size; + tp->size += parg->type->size; + return __parse_probe_arg(arg, parg->type, &parg->fetch, is_return); } /* Return 1 if name is reserved or already used by another argument */ @@ -602,15 +703,18 @@ static int create_trace_probe(int argc, char **argv) * @ADDR : fetch memory at ADDR (ADDR should be in kernel) * @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol) * %REG : fetch register REG - * Indirect memory fetch: + * Dereferencing memory fetch: * +|-offs(ARG) : fetch memory at ARG +|- offs address. * Alias name of args: * NAME=FETCHARG : set NAME as alias of FETCHARG. + * Type of args: + * FETCHARG:TYPE : use TYPE instead of unsigned long. */ struct trace_probe *tp; int i, ret = 0; int is_return = 0, is_delete = 0; - char *symbol = NULL, *event = NULL, *arg = NULL, *group = NULL; + char *symbol = NULL, *event = NULL, *group = NULL; + char *arg, *tmp; unsigned long offset = 0; void *addr = NULL; char buf[MAX_EVENT_NAME_LEN]; @@ -723,13 +827,6 @@ static int create_trace_probe(int argc, char **argv) else arg = argv[i]; - if (conflict_field_name(argv[i], tp->args, i)) { - pr_info("Argument%d name '%s' conflicts with " - "another field.\n", i, argv[i]); - ret = -EINVAL; - goto error; - } - tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); if (!tp->args[i].name) { pr_info("Failed to allocate argument%d name '%s'.\n", @@ -737,9 +834,19 @@ static int create_trace_probe(int argc, char **argv) ret = -ENOMEM; goto error; } + tmp = strchr(tp->args[i].name, ':'); + if (tmp) + *tmp = '_'; /* convert : to _ */ + + if (conflict_field_name(tp->args[i].name, tp->args, i)) { + pr_info("Argument%d name '%s' conflicts with " + "another field.\n", i, argv[i]); + ret = -EINVAL; + goto error; + } /* Parse fetch argument */ - ret = parse_probe_arg(arg, &tp->args[i].fetch, is_return); + ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { pr_info("Parse error at argument%d. (%d)\n", i, ret); kfree(tp->args[i].name); @@ -794,8 +901,7 @@ static void probes_seq_stop(struct seq_file *m, void *v) static int probes_seq_show(struct seq_file *m, void *v) { struct trace_probe *tp = v; - int i, ret; - char buf[MAX_ARGSTR_LEN + 1]; + int i; seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p'); seq_printf(m, ":%s/%s", tp->call.system, tp->call.name); @@ -807,15 +913,10 @@ static int probes_seq_show(struct seq_file *m, void *v) else seq_printf(m, " %s", probe_symbol(tp)); - for (i = 0; i < tp->nr_args; i++) { - ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i].fetch); - if (ret < 0) { - pr_warning("Argument%d decoding error(%d).\n", i, ret); - return ret; - } - seq_printf(m, " %s=%s", tp->args[i].name, buf); - } + for (i = 0; i < tp->nr_args; i++) + seq_printf(m, " %s=%s", tp->args[i].name, tp->args[i].comm); seq_printf(m, "\n"); + return 0; } @@ -945,9 +1046,10 @@ static const struct file_operations kprobe_profile_ops = { static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) { struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); - struct kprobe_trace_entry *entry; + struct kprobe_trace_entry_head *entry; struct ring_buffer_event *event; struct ring_buffer *buffer; + u8 *data; int size, i, pc; unsigned long irq_flags; struct ftrace_event_call *call = &tp->call; @@ -957,7 +1059,7 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) local_save_flags(irq_flags); pc = preempt_count(); - size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args); + size = sizeof(*entry) + tp->size; event = trace_current_buffer_lock_reserve(&buffer, call->id, size, irq_flags, pc); @@ -965,10 +1067,10 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) return; entry = ring_buffer_event_data(event); - entry->nargs = tp->nr_args; entry->ip = (unsigned long)kp->addr; + data = (u8 *)&entry[1]; for (i = 0; i < tp->nr_args; i++) - entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); if (!filter_current_check_discard(buffer, call, entry, event)) trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); @@ -979,9 +1081,10 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, struct pt_regs *regs) { struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); - struct kretprobe_trace_entry *entry; + struct kretprobe_trace_entry_head *entry; struct ring_buffer_event *event; struct ring_buffer *buffer; + u8 *data; int size, i, pc; unsigned long irq_flags; struct ftrace_event_call *call = &tp->call; @@ -989,7 +1092,7 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, local_save_flags(irq_flags); pc = preempt_count(); - size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args); + size = sizeof(*entry) + tp->size; event = trace_current_buffer_lock_reserve(&buffer, call->id, size, irq_flags, pc); @@ -997,11 +1100,11 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, return; entry = ring_buffer_event_data(event); - entry->nargs = tp->nr_args; entry->func = (unsigned long)tp->rp.kp.addr; entry->ret_ip = (unsigned long)ri->ret_addr; + data = (u8 *)&entry[1]; for (i = 0; i < tp->nr_args; i++) - entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); if (!filter_current_check_discard(buffer, call, entry, event)) trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); @@ -1011,13 +1114,14 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, enum print_line_t print_kprobe_event(struct trace_iterator *iter, int flags) { - struct kprobe_trace_entry *field; + struct kprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; struct trace_event *event; struct trace_probe *tp; + u8 *data; int i; - field = (struct kprobe_trace_entry *)iter->ent; + field = (struct kprobe_trace_entry_head *)iter->ent; event = ftrace_find_event(field->ent.type); tp = container_of(event, struct trace_probe, event); @@ -1030,9 +1134,10 @@ print_kprobe_event(struct trace_iterator *iter, int flags) if (!trace_seq_puts(s, ")")) goto partial; - for (i = 0; i < field->nargs; i++) - if (!trace_seq_printf(s, " %s=%lx", - tp->args[i].name, field->args[i])) + data = (u8 *)&field[1]; + for (i = 0; i < tp->nr_args; i++) + if (!tp->args[i].type->print(s, tp->args[i].name, + data + tp->args[i].offset)) goto partial; if (!trace_seq_puts(s, "\n")) @@ -1046,13 +1151,14 @@ partial: enum print_line_t print_kretprobe_event(struct trace_iterator *iter, int flags) { - struct kretprobe_trace_entry *field; + struct kretprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; struct trace_event *event; struct trace_probe *tp; + u8 *data; int i; - field = (struct kretprobe_trace_entry *)iter->ent; + field = (struct kretprobe_trace_entry_head *)iter->ent; event = ftrace_find_event(field->ent.type); tp = container_of(event, struct trace_probe, event); @@ -1071,9 +1177,10 @@ print_kretprobe_event(struct trace_iterator *iter, int flags) if (!trace_seq_puts(s, ")")) goto partial; - for (i = 0; i < field->nargs; i++) - if (!trace_seq_printf(s, " %s=%lx", - tp->args[i].name, field->args[i])) + data = (u8 *)&field[1]; + for (i = 0; i < tp->nr_args; i++) + if (!tp->args[i].type->print(s, tp->args[i].name, + data + tp->args[i].offset)) goto partial; if (!trace_seq_puts(s, "\n")) @@ -1129,29 +1236,43 @@ static int probe_event_raw_init(struct ftrace_event_call *event_call) static int kprobe_event_define_fields(struct ftrace_event_call *event_call) { int ret, i; - struct kprobe_trace_entry field; + struct kprobe_trace_entry_head field; struct trace_probe *tp = (struct trace_probe *)event_call->data; DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); - DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); /* Set argument names as fields */ - for (i = 0; i < tp->nr_args; i++) - DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0); + for (i = 0; i < tp->nr_args; i++) { + ret = trace_define_field(event_call, tp->args[i].type->name, + tp->args[i].name, + sizeof(field) + tp->args[i].offset, + tp->args[i].type->size, + tp->args[i].type->is_signed, + FILTER_OTHER); + if (ret) + return ret; + } return 0; } static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) { int ret, i; - struct kretprobe_trace_entry field; + struct kretprobe_trace_entry_head field; struct trace_probe *tp = (struct trace_probe *)event_call->data; DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); - DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); /* Set argument names as fields */ - for (i = 0; i < tp->nr_args; i++) - DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0); + for (i = 0; i < tp->nr_args; i++) { + ret = trace_define_field(event_call, tp->args[i].type->name, + tp->args[i].name, + sizeof(field) + tp->args[i].offset, + tp->args[i].type->size, + tp->args[i].type->is_signed, + FILTER_OTHER); + if (ret) + return ret; + } return 0; } @@ -1176,8 +1297,8 @@ static int __set_print_fmt(struct trace_probe *tp, char *buf, int len) pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt); for (i = 0; i < tp->nr_args; i++) { - pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%%lx", - tp->args[i].name); + pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%s", + tp->args[i].name, tp->args[i].type->fmt); } pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg); @@ -1219,12 +1340,13 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp, { struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); struct ftrace_event_call *call = &tp->call; - struct kprobe_trace_entry *entry; + struct kprobe_trace_entry_head *entry; + u8 *data; int size, __size, i; unsigned long irq_flags; int rctx; - __size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args); + __size = sizeof(*entry) + tp->size; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, @@ -1235,10 +1357,10 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp, if (!entry) return; - entry->nargs = tp->nr_args; entry->ip = (unsigned long)kp->addr; + data = (u8 *)&entry[1]; for (i = 0; i < tp->nr_args; i++) - entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, irq_flags, regs); } @@ -1249,12 +1371,13 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri, { struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); struct ftrace_event_call *call = &tp->call; - struct kretprobe_trace_entry *entry; + struct kretprobe_trace_entry_head *entry; + u8 *data; int size, __size, i; unsigned long irq_flags; int rctx; - __size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args); + __size = sizeof(*entry) + tp->size; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, @@ -1265,11 +1388,11 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri, if (!entry) return; - entry->nargs = tp->nr_args; entry->func = (unsigned long)tp->rp.kp.addr; entry->ret_ip = (unsigned long)ri->ret_addr; + data = (u8 *)&entry[1]; for (i = 0; i < tp->nr_args; i++) - entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, irq_flags, regs); diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 81003b4d617f..71fa771ee4d7 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -17,7 +17,6 @@ static inline int trace_valid_entry(struct trace_entry *entry) case TRACE_BRANCH: case TRACE_GRAPH_ENT: case TRACE_GRAPH_RET: - case TRACE_HW_BRANCHES: case TRACE_KSYM: return 1; } @@ -30,7 +29,7 @@ static int trace_test_buffer_cpu(struct trace_array *tr, int cpu) struct trace_entry *entry; unsigned int loops = 0; - while ((event = ring_buffer_consume(tr->buffer, cpu, NULL))) { + while ((event = ring_buffer_consume(tr->buffer, cpu, NULL, NULL))) { entry = ring_buffer_event_data(event); /* @@ -755,62 +754,6 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) } #endif /* CONFIG_BRANCH_TRACER */ -#ifdef CONFIG_HW_BRANCH_TRACER -int -trace_selftest_startup_hw_branches(struct tracer *trace, - struct trace_array *tr) -{ - struct trace_iterator *iter; - struct tracer tracer; - unsigned long count; - int ret; - - if (!trace->open) { - printk(KERN_CONT "missing open function..."); - return -1; - } - - ret = tracer_init(trace, tr); - if (ret) { - warn_failed_init_tracer(trace, ret); - return ret; - } - - /* - * The hw-branch tracer needs to collect the trace from the various - * cpu trace buffers - before tracing is stopped. - */ - iter = kzalloc(sizeof(*iter), GFP_KERNEL); - if (!iter) - return -ENOMEM; - - memcpy(&tracer, trace, sizeof(tracer)); - - iter->trace = &tracer; - iter->tr = tr; - iter->pos = -1; - mutex_init(&iter->mutex); - - trace->open(iter); - - mutex_destroy(&iter->mutex); - kfree(iter); - - tracing_stop(); - - ret = trace_test_buffer(tr, &count); - trace->reset(tr); - tracing_start(); - - if (!ret && !count) { - printk(KERN_CONT "no entries found.."); - ret = -1; - } - - return ret; -} -#endif /* CONFIG_HW_BRANCH_TRACER */ - #ifdef CONFIG_KSYM_TRACER static int ksym_selftest_dummy; diff --git a/kernel/user.c b/kernel/user.c index 766467b3bcb7..8e1c8c0a496c 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -16,7 +16,6 @@ #include <linux/interrupt.h> #include <linux/module.h> #include <linux/user_namespace.h> -#include "cred-internals.h" struct user_namespace init_user_ns = { .kref = { @@ -137,9 +136,7 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid) struct hlist_head *hashent = uidhashentry(ns, uid); struct user_struct *up, *new; - /* Make uid_hash_find() + uids_user_create() + uid_hash_insert() - * atomic. - */ + /* Make uid_hash_find() + uid_hash_insert() atomic. */ spin_lock_irq(&uidhash_lock); up = uid_hash_find(uid, hashent); spin_unlock_irq(&uidhash_lock); @@ -161,11 +158,6 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid) spin_lock_irq(&uidhash_lock); up = uid_hash_find(uid, hashent); if (up) { - /* This case is not possible when CONFIG_USER_SCHED - * is defined, since we serialize alloc_uid() using - * uids_mutex. Hence no need to call - * sched_destroy_user() or remove_user_sysfs_dir(). - */ key_put(new->uid_keyring); key_put(new->session_keyring); kmem_cache_free(uid_cachep, new); @@ -178,8 +170,6 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid) return up; - put_user_ns(new->user_ns); - kmem_cache_free(uid_cachep, new); out_unlock: return NULL; } diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index ff017108700d..220ae6063b6f 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -171,6 +171,18 @@ config DETECT_SOFTLOCKUP can be detected via the NMI-watchdog, on platforms that support it.) +config NMI_WATCHDOG + bool "Detect Hard Lockups with an NMI Watchdog" + depends on DEBUG_KERNEL && PERF_EVENTS && PERF_EVENTS_NMI + help + Say Y here to enable the kernel to use the NMI as a watchdog + to detect hard lockups. This is useful when a cpu hangs for no + reason but can still respond to NMIs. A backtrace is displayed + for reviewing and reporting. + + The overhead should be minimal, just an extra NMI every few + seconds. + config BOOTPARAM_SOFTLOCKUP_PANIC bool "Panic (Reboot) On Soft Lockups" depends on DETECT_SOFTLOCKUP @@ -356,7 +368,7 @@ config SLUB_STATS config DEBUG_KMEMLEAK bool "Kernel memory leak detector" depends on DEBUG_KERNEL && EXPERIMENTAL && !MEMORY_HOTPLUG && \ - (X86 || ARM || PPC || S390 || SUPERH || MICROBLAZE) + (X86 || ARM || PPC || S390 || SPARC64 || SUPERH || MICROBLAZE) select DEBUG_FS if SYSFS select STACKTRACE if STACKTRACE_SUPPORT diff --git a/lib/debugobjects.c b/lib/debugobjects.c index b862b30369ff..bf007a43c053 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -774,7 +774,7 @@ static int __init fixup_free(void *addr, enum debug_obj_state state) } } -static int +static int __init check_results(void *addr, enum debug_obj_state state, int fixups, int warnings) { struct debug_bucket *db; @@ -917,7 +917,7 @@ void __init debug_objects_early_init(void) /* * Convert the statically allocated objects to dynamic ones: */ -static int debug_objects_replace_static_objects(void) +static int __init debug_objects_replace_static_objects(void) { struct debug_bucket *db = obj_hash; struct hlist_node *node, *tmp; diff --git a/lib/dma-debug.c b/lib/dma-debug.c index ba8b67039d13..01e64270e246 100644 --- a/lib/dma-debug.c +++ b/lib/dma-debug.c @@ -570,7 +570,7 @@ static ssize_t filter_write(struct file *file, const char __user *userbuf, * Now parse out the first token and use it as the name for the * driver to filter for. */ - for (i = 0; i < NAME_MAX_LEN; ++i) { + for (i = 0; i < NAME_MAX_LEN - 1; ++i) { current_driver_name[i] = buf[i]; if (isspace(buf[i]) || buf[i] == ' ' || buf[i] == 0) break; diff --git a/lib/rbtree.c b/lib/rbtree.c index e2aa3be29858..15e10b1afdd2 100644 --- a/lib/rbtree.c +++ b/lib/rbtree.c @@ -44,6 +44,11 @@ static void __rb_rotate_left(struct rb_node *node, struct rb_root *root) else root->rb_node = right; rb_set_parent(node, right); + + if (root->augment_cb) { + root->augment_cb(node); + root->augment_cb(right); + } } static void __rb_rotate_right(struct rb_node *node, struct rb_root *root) @@ -67,12 +72,20 @@ static void __rb_rotate_right(struct rb_node *node, struct rb_root *root) else root->rb_node = left; rb_set_parent(node, left); + + if (root->augment_cb) { + root->augment_cb(node); + root->augment_cb(left); + } } void rb_insert_color(struct rb_node *node, struct rb_root *root) { struct rb_node *parent, *gparent; + if (root->augment_cb) + root->augment_cb(node); + while ((parent = rb_parent(node)) && rb_is_red(parent)) { gparent = rb_parent(parent); @@ -227,12 +240,15 @@ void rb_erase(struct rb_node *node, struct rb_root *root) else { struct rb_node *old = node, *left; + int old_parent_cb = 0; + int successor_parent_cb = 0; node = node->rb_right; while ((left = node->rb_left) != NULL) node = left; if (rb_parent(old)) { + old_parent_cb = 1; if (rb_parent(old)->rb_left == old) rb_parent(old)->rb_left = node; else @@ -247,8 +263,10 @@ void rb_erase(struct rb_node *node, struct rb_root *root) if (parent == old) { parent = node; } else { + successor_parent_cb = 1; if (child) rb_set_parent(child, parent); + parent->rb_left = child; node->rb_right = old->rb_right; @@ -259,6 +277,24 @@ void rb_erase(struct rb_node *node, struct rb_root *root) node->rb_left = old->rb_left; rb_set_parent(old->rb_left, node); + if (root->augment_cb) { + /* + * Here, three different nodes can have new children. + * The parent of the successor node that was selected + * to replace the node to be erased. + * The node that is getting erased and is now replaced + * by its successor. + * The parent of the node getting erased-replaced. + */ + if (successor_parent_cb) + root->augment_cb(parent); + + root->augment_cb(node); + + if (old_parent_cb) + root->augment_cb(rb_parent(old)); + } + goto color; } @@ -267,15 +303,19 @@ void rb_erase(struct rb_node *node, struct rb_root *root) if (child) rb_set_parent(child, parent); - if (parent) - { + + if (parent) { if (parent->rb_left == node) parent->rb_left = child; else parent->rb_right = child; - } - else + + if (root->augment_cb) + root->augment_cb(parent); + + } else { root->rb_node = child; + } color: if (color == RB_BLACK) diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 24112e5a5780..7376b7c55ffe 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -408,12 +408,12 @@ enum format_type { }; struct printf_spec { - u16 type; - s16 field_width; /* width of output field */ + u8 type; /* format_type enum */ u8 flags; /* flags to number() */ - u8 base; - s8 precision; /* # of digits/chars */ - u8 qualifier; + u8 base; /* number base, 8, 10 or 16 only */ + u8 qualifier; /* number qualifier, one of 'hHlLtzZ' */ + s16 field_width; /* width of output field */ + s16 precision; /* # of digits/chars */ }; static char *number(char *buf, char *end, unsigned long long num, diff --git a/mm/mlock.c b/mm/mlock.c index 8f4e2dfceec1..3f82720e0515 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -607,44 +607,3 @@ void user_shm_unlock(size_t size, struct user_struct *user) spin_unlock(&shmlock_user_lock); free_uid(user); } - -int account_locked_memory(struct mm_struct *mm, struct rlimit *rlim, - size_t size) -{ - unsigned long lim, vm, pgsz; - int error = -ENOMEM; - - pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; - - down_write(&mm->mmap_sem); - - lim = ACCESS_ONCE(rlim[RLIMIT_AS].rlim_cur) >> PAGE_SHIFT; - vm = mm->total_vm + pgsz; - if (lim < vm) - goto out; - - lim = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur) >> PAGE_SHIFT; - vm = mm->locked_vm + pgsz; - if (lim < vm) - goto out; - - mm->total_vm += pgsz; - mm->locked_vm += pgsz; - - error = 0; - out: - up_write(&mm->mmap_sem); - return error; -} - -void refund_locked_memory(struct mm_struct *mm, size_t size) -{ - unsigned long pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; - - down_write(&mm->mmap_sem); - - mm->total_vm -= pgsz; - mm->locked_vm -= pgsz; - - up_write(&mm->mmap_sem); -} diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 6980625537ca..f29ada827a6a 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -723,7 +723,7 @@ static int br_multicast_igmp3_report(struct net_bridge *br, if (!pskb_may_pull(skb, len)) return -EINVAL; - grec = (void *)(skb->data + len); + grec = (void *)(skb->data + len - sizeof(*grec)); group = grec->grec_mca; type = grec->grec_type; diff --git a/net/can/raw.c b/net/can/raw.c index 3a7dffb6519c..da99cf153b33 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -445,7 +445,7 @@ static int raw_setsockopt(struct socket *sock, int level, int optname, return -EFAULT; } } else if (count == 1) { - if (copy_from_user(&sfilter, optval, optlen)) + if (copy_from_user(&sfilter, optval, sizeof(sfilter))) return -EFAULT; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 954bbfb39dff..8fef859db35d 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -472,8 +472,8 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, if (hslot->count < hslot2->count) goto begin; - result = udp4_lib_lookup2(net, INADDR_ANY, sport, - daddr, hnum, dif, + result = udp4_lib_lookup2(net, saddr, sport, + INADDR_ANY, hnum, dif, hslot2, slot2); } rcu_read_unlock(); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index c177aea88c0b..90824852f598 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -259,8 +259,8 @@ static struct sock *__udp6_lib_lookup(struct net *net, if (hslot->count < hslot2->count) goto begin; - result = udp6_lib_lookup2(net, &in6addr_any, sport, - daddr, hnum, dif, + result = udp6_lib_lookup2(net, saddr, sport, + &in6addr_any, hnum, dif, hslot2, slot2); } rcu_read_unlock(); diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 06c33b68d8e5..b887e484ae04 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -225,11 +225,11 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata, switch (sdata->vif.type) { case NL80211_IFTYPE_AP: sdata->vif.bss_conf.enable_beacon = - !!rcu_dereference(sdata->u.ap.beacon); + !!sdata->u.ap.beacon; break; case NL80211_IFTYPE_ADHOC: sdata->vif.bss_conf.enable_beacon = - !!rcu_dereference(sdata->u.ibss.presp); + !!sdata->u.ibss.presp; break; case NL80211_IFTYPE_MESH_POINT: sdata->vif.bss_conf.enable_beacon = true; diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 58e3e3a61d99..859ee5f3d941 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -750,9 +750,6 @@ ieee80211_mesh_rx_mgmt(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb) switch (fc & IEEE80211_FCTL_STYPE) { case IEEE80211_STYPE_ACTION: - if (skb->len < IEEE80211_MIN_ACTION_SIZE) - return RX_DROP_MONITOR; - /* fall through */ case IEEE80211_STYPE_PROBE_RESP: case IEEE80211_STYPE_BEACON: skb_queue_tail(&ifmsh->skb_queue, skb); diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index f0accf622cd7..04ea07f0e78a 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1974,6 +1974,11 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) goto handled; } break; + case MESH_PLINK_CATEGORY: + case MESH_PATH_SEL_CATEGORY: + if (ieee80211_vif_is_mesh(&sdata->vif)) + return ieee80211_mesh_rx_mgmt(sdata, rx->skb); + break; } /* diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 56422d894351..fb12cec4d333 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -93,12 +93,18 @@ struct sta_info *sta_info_get(struct ieee80211_sub_if_data *sdata, struct ieee80211_local *local = sdata->local; struct sta_info *sta; - sta = rcu_dereference(local->sta_hash[STA_HASH(addr)]); + sta = rcu_dereference_check(local->sta_hash[STA_HASH(addr)], + rcu_read_lock_held() || + lockdep_is_held(&local->sta_lock) || + lockdep_is_held(&local->sta_mtx)); while (sta) { if (sta->sdata == sdata && memcmp(sta->sta.addr, addr, ETH_ALEN) == 0) break; - sta = rcu_dereference(sta->hnext); + sta = rcu_dereference_check(sta->hnext, + rcu_read_lock_held() || + lockdep_is_held(&local->sta_lock) || + lockdep_is_held(&local->sta_mtx)); } return sta; } @@ -113,13 +119,19 @@ struct sta_info *sta_info_get_bss(struct ieee80211_sub_if_data *sdata, struct ieee80211_local *local = sdata->local; struct sta_info *sta; - sta = rcu_dereference(local->sta_hash[STA_HASH(addr)]); + sta = rcu_dereference_check(local->sta_hash[STA_HASH(addr)], + rcu_read_lock_held() || + lockdep_is_held(&local->sta_lock) || + lockdep_is_held(&local->sta_mtx)); while (sta) { if ((sta->sdata == sdata || sta->sdata->bss == sdata->bss) && memcmp(sta->sta.addr, addr, ETH_ALEN) == 0) break; - sta = rcu_dereference(sta->hnext); + sta = rcu_dereference_check(sta->hnext, + rcu_read_lock_held() || + lockdep_is_held(&local->sta_lock) || + lockdep_is_held(&local->sta_mtx)); } return sta; } diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index e56f711baccc..cbddd0cb83f1 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -83,6 +83,41 @@ struct compat_x25_subscrip_struct { }; #endif + +int x25_parse_address_block(struct sk_buff *skb, + struct x25_address *called_addr, + struct x25_address *calling_addr) +{ + unsigned char len; + int needed; + int rc; + + if (skb->len < 1) { + /* packet has no address block */ + rc = 0; + goto empty; + } + + len = *skb->data; + needed = 1 + (len >> 4) + (len & 0x0f); + + if (skb->len < needed) { + /* packet is too short to hold the addresses it claims + to hold */ + rc = -1; + goto empty; + } + + return x25_addr_ntoa(skb->data, called_addr, calling_addr); + +empty: + *called_addr->x25_addr = 0; + *calling_addr->x25_addr = 0; + + return rc; +} + + int x25_addr_ntoa(unsigned char *p, struct x25_address *called_addr, struct x25_address *calling_addr) { @@ -554,7 +589,8 @@ static int x25_create(struct net *net, struct socket *sock, int protocol, x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE; x25->facilities.pacsize_in = X25_DEFAULT_PACKET_SIZE; x25->facilities.pacsize_out = X25_DEFAULT_PACKET_SIZE; - x25->facilities.throughput = X25_DEFAULT_THROUGHPUT; + x25->facilities.throughput = 0; /* by default don't negotiate + throughput */ x25->facilities.reverse = X25_DEFAULT_REVERSE; x25->dte_facilities.calling_len = 0; x25->dte_facilities.called_len = 0; @@ -922,16 +958,26 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, /* * Extract the X.25 addresses and convert them to ASCII strings, * and remove them. + * + * Address block is mandatory in call request packets */ - addr_len = x25_addr_ntoa(skb->data, &source_addr, &dest_addr); + addr_len = x25_parse_address_block(skb, &source_addr, &dest_addr); + if (addr_len <= 0) + goto out_clear_request; skb_pull(skb, addr_len); /* * Get the length of the facilities, skip past them for the moment * get the call user data because this is needed to determine * the correct listener + * + * Facilities length is mandatory in call request packets */ + if (skb->len < 1) + goto out_clear_request; len = skb->data[0] + 1; + if (skb->len < len) + goto out_clear_request; skb_pull(skb,len); /* @@ -1415,9 +1461,20 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) if (facilities.winsize_in < 1 || facilities.winsize_in > 127) break; - if (facilities.throughput < 0x03 || - facilities.throughput > 0xDD) - break; + if (facilities.throughput) { + int out = facilities.throughput & 0xf0; + int in = facilities.throughput & 0x0f; + if (!out) + facilities.throughput |= + X25_DEFAULT_THROUGHPUT << 4; + else if (out < 0x30 || out > 0xD0) + break; + if (!in) + facilities.throughput |= + X25_DEFAULT_THROUGHPUT; + else if (in < 0x03 || in > 0x0D) + break; + } if (facilities.reverse && (facilities.reverse & 0x81) != 0x81) break; diff --git a/net/x25/x25_facilities.c b/net/x25/x25_facilities.c index a21f6646eb3a..771bab00754b 100644 --- a/net/x25/x25_facilities.c +++ b/net/x25/x25_facilities.c @@ -35,7 +35,7 @@ int x25_parse_facilities(struct sk_buff *skb, struct x25_facilities *facilities, struct x25_dte_facilities *dte_facs, unsigned long *vc_fac_mask) { unsigned char *p = skb->data; - unsigned int len = *p++; + unsigned int len; *vc_fac_mask = 0; @@ -50,6 +50,14 @@ int x25_parse_facilities(struct sk_buff *skb, struct x25_facilities *facilities, memset(dte_facs->called_ae, '\0', sizeof(dte_facs->called_ae)); memset(dte_facs->calling_ae, '\0', sizeof(dte_facs->calling_ae)); + if (skb->len < 1) + return 0; + + len = *p++; + + if (len >= skb->len) + return -1; + while (len > 0) { switch (*p & X25_FAC_CLASS_MASK) { case X25_FAC_CLASS_A: @@ -247,6 +255,8 @@ int x25_negotiate_facilities(struct sk_buff *skb, struct sock *sk, memcpy(new, ours, sizeof(*new)); len = x25_parse_facilities(skb, &theirs, dte, &x25->vc_facil_mask); + if (len < 0) + return len; /* * They want reverse charging, we won't accept it. @@ -259,9 +269,18 @@ int x25_negotiate_facilities(struct sk_buff *skb, struct sock *sk, new->reverse = theirs.reverse; if (theirs.throughput) { - if (theirs.throughput < ours->throughput) { - SOCK_DEBUG(sk, "X.25: throughput negotiated down\n"); - new->throughput = theirs.throughput; + int theirs_in = theirs.throughput & 0x0f; + int theirs_out = theirs.throughput & 0xf0; + int ours_in = ours->throughput & 0x0f; + int ours_out = ours->throughput & 0xf0; + if (!ours_in || theirs_in < ours_in) { + SOCK_DEBUG(sk, "X.25: inbound throughput negotiated\n"); + new->throughput = (new->throughput & 0xf0) | theirs_in; + } + if (!ours_out || theirs_out < ours_out) { + SOCK_DEBUG(sk, + "X.25: outbound throughput negotiated\n"); + new->throughput = (new->throughput & 0x0f) | theirs_out; } } diff --git a/net/x25/x25_in.c b/net/x25/x25_in.c index a31b3b9e5966..372ac226e648 100644 --- a/net/x25/x25_in.c +++ b/net/x25/x25_in.c @@ -90,6 +90,7 @@ static int x25_queue_rx_frame(struct sock *sk, struct sk_buff *skb, int more) static int x25_state1_machine(struct sock *sk, struct sk_buff *skb, int frametype) { struct x25_address source_addr, dest_addr; + int len; switch (frametype) { case X25_CALL_ACCEPTED: { @@ -107,11 +108,17 @@ static int x25_state1_machine(struct sock *sk, struct sk_buff *skb, int frametyp * Parse the data in the frame. */ skb_pull(skb, X25_STD_MIN_LEN); - skb_pull(skb, x25_addr_ntoa(skb->data, &source_addr, &dest_addr)); - skb_pull(skb, - x25_parse_facilities(skb, &x25->facilities, + + len = x25_parse_address_block(skb, &source_addr, + &dest_addr); + if (len > 0) + skb_pull(skb, len); + + len = x25_parse_facilities(skb, &x25->facilities, &x25->dte_facilities, - &x25->vc_facil_mask)); + &x25->vc_facil_mask); + if (len > 0) + skb_pull(skb, len); /* * Copy any Call User Data. */ diff --git a/security/selinux/ss/avtab.h b/security/selinux/ss/avtab.h index 8da6a8428086..cd4f734e2749 100644 --- a/security/selinux/ss/avtab.h +++ b/security/selinux/ss/avtab.h @@ -82,7 +82,7 @@ struct avtab_node *avtab_search_node_next(struct avtab_node *node, int specified void avtab_cache_init(void); void avtab_cache_destroy(void); -#define MAX_AVTAB_HASH_BITS 13 +#define MAX_AVTAB_HASH_BITS 11 #define MAX_AVTAB_HASH_BUCKETS (1 << MAX_AVTAB_HASH_BITS) #define MAX_AVTAB_HASH_MASK (MAX_AVTAB_HASH_BUCKETS-1) #define MAX_AVTAB_SIZE MAX_AVTAB_HASH_BUCKETS diff --git a/sound/arm/aaci.c b/sound/arm/aaci.c index 656e474dca47..91acc9a243ec 100644 --- a/sound/arm/aaci.c +++ b/sound/arm/aaci.c @@ -863,7 +863,6 @@ static int __devinit aaci_probe_ac97(struct aaci *aaci) struct snd_ac97 *ac97; int ret; - writel(0, aaci->base + AC97_POWERDOWN); /* * Assert AACIRESET for 2us */ @@ -1047,7 +1046,11 @@ static int __devinit aaci_probe(struct amba_device *dev, struct amba_id *id) writel(0x1fff, aaci->base + AACI_INTCLR); writel(aaci->maincr, aaci->base + AACI_MAINCR); - + /* + * Fix: ac97 read back fail errors by reading + * from any arbitrary aaci register. + */ + readl(aaci->base + AACI_CSCH1); ret = aaci_probe_ac97(aaci); if (ret) goto out; diff --git a/sound/drivers/pcsp/pcsp.h b/sound/drivers/pcsp/pcsp.h index 1e123077923d..4ff6c8cc5077 100644 --- a/sound/drivers/pcsp/pcsp.h +++ b/sound/drivers/pcsp/pcsp.h @@ -16,7 +16,7 @@ #include <asm/i8253.h> #else #include <asm/8253pit.h> -static DEFINE_SPINLOCK(i8253_lock); +static DEFINE_RAW_SPINLOCK(i8253_lock); #endif #define PCSP_SOUND_VERSION 0x400 /* read 4.00 */ diff --git a/sound/drivers/pcsp/pcsp_input.c b/sound/drivers/pcsp/pcsp_input.c index 0444cdeb4bec..b5e2b54c2604 100644 --- a/sound/drivers/pcsp/pcsp_input.c +++ b/sound/drivers/pcsp/pcsp_input.c @@ -21,7 +21,7 @@ static void pcspkr_do_sound(unsigned int count) { unsigned long flags; - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); if (count) { /* set command for counter 2, 2 byte write */ @@ -36,7 +36,7 @@ static void pcspkr_do_sound(unsigned int count) outb(inb_p(0x61) & 0xFC, 0x61); } - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); } void pcspkr_stop_sound(void) diff --git a/sound/drivers/pcsp/pcsp_lib.c b/sound/drivers/pcsp/pcsp_lib.c index d77ffa9a9387..ce9e7d170c0d 100644 --- a/sound/drivers/pcsp/pcsp_lib.c +++ b/sound/drivers/pcsp/pcsp_lib.c @@ -66,7 +66,7 @@ static u64 pcsp_timer_update(struct snd_pcsp *chip) timer_cnt = val * CUR_DIV() / 256; if (timer_cnt && chip->enable) { - spin_lock_irqsave(&i8253_lock, flags); + raw_spin_lock_irqsave(&i8253_lock, flags); if (!nforce_wa) { outb_p(chip->val61, 0x61); outb_p(timer_cnt, 0x42); @@ -75,7 +75,7 @@ static u64 pcsp_timer_update(struct snd_pcsp *chip) outb(chip->val61 ^ 2, 0x61); chip->thalf = 1; } - spin_unlock_irqrestore(&i8253_lock, flags); + raw_spin_unlock_irqrestore(&i8253_lock, flags); } chip->ns_rem = PCSP_PERIOD_NS(); @@ -159,10 +159,10 @@ static int pcsp_start_playing(struct snd_pcsp *chip) return -EIO; } - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); chip->val61 = inb(0x61) | 0x03; outb_p(0x92, 0x43); /* binary, mode 1, LSB only, ch 2 */ - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); atomic_set(&chip->timer_active, 1); chip->thalf = 0; @@ -179,11 +179,11 @@ static void pcsp_stop_playing(struct snd_pcsp *chip) return; atomic_set(&chip->timer_active, 0); - spin_lock(&i8253_lock); + raw_spin_lock(&i8253_lock); /* restore the timer */ outb_p(0xb6, 0x43); /* binary, mode 3, LSB/MSB, ch 2 */ outb(chip->val61 & 0xFC, 0x61); - spin_unlock(&i8253_lock); + raw_spin_unlock(&i8253_lock); } /* diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index f8fd586ae024..f669442b7c82 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2272,6 +2272,7 @@ static struct snd_pci_quirk position_fix_list[] __devinitdata = { SND_PCI_QUIRK(0x1458, 0xa022, "ga-ma770-ud3", POS_FIX_LPIB), SND_PCI_QUIRK(0x1462, 0x1002, "MSI Wind U115", POS_FIX_LPIB), SND_PCI_QUIRK(0x1565, 0x820f, "Biostar Microtech", POS_FIX_LPIB), + SND_PCI_QUIRK(0x1565, 0x8218, "Biostar Microtech", POS_FIX_LPIB), SND_PCI_QUIRK(0x8086, 0xd601, "eMachines T5212", POS_FIX_LPIB), {} }; diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index c7730dbb9ddb..aad1627f56f1 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -230,6 +230,7 @@ enum { ALC888_ACER_ASPIRE_7730G, ALC883_MEDION, ALC883_MEDION_MD2, + ALC883_MEDION_WIM2160, ALC883_LAPTOP_EAPD, ALC883_LENOVO_101E_2ch, ALC883_LENOVO_NB0763, @@ -1389,22 +1390,31 @@ struct alc_fixup { static void alc_pick_fixup(struct hda_codec *codec, const struct snd_pci_quirk *quirk, - const struct alc_fixup *fix) + const struct alc_fixup *fix, + int pre_init) { const struct alc_pincfg *cfg; quirk = snd_pci_quirk_lookup(codec->bus->pci, quirk); if (!quirk) return; - fix += quirk->value; cfg = fix->pins; - if (cfg) { + if (pre_init && cfg) { +#ifdef CONFIG_SND_DEBUG_VERBOSE + snd_printdd(KERN_INFO "hda_codec: %s: Apply pincfg for %s\n", + codec->chip_name, quirk->name); +#endif for (; cfg->nid; cfg++) snd_hda_codec_set_pincfg(codec, cfg->nid, cfg->val); } - if (fix->verbs) + if (!pre_init && fix->verbs) { +#ifdef CONFIG_SND_DEBUG_VERBOSE + snd_printdd(KERN_INFO "hda_codec: %s: Apply fix-verbs for %s\n", + codec->chip_name, quirk->name); +#endif add_verb(codec->spec, fix->verbs); + } } static int alc_read_coef_idx(struct hda_codec *codec, @@ -4808,6 +4818,25 @@ static void alc880_auto_init_analog_input(struct hda_codec *codec) } } +static void alc880_auto_init_input_src(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int c; + + for (c = 0; c < spec->num_adc_nids; c++) { + unsigned int mux_idx; + const struct hda_input_mux *imux; + mux_idx = c >= spec->num_mux_defs ? 0 : c; + imux = &spec->input_mux[mux_idx]; + if (!imux->num_items && mux_idx > 0) + imux = &spec->input_mux[0]; + if (imux) + snd_hda_codec_write(codec, spec->adc_nids[c], 0, + AC_VERB_SET_CONNECT_SEL, + imux->items[0].index); + } +} + /* parse the BIOS configuration and set up the alc_spec */ /* return 1 if successful, 0 if the proper config is not found, * or a negative error code @@ -4886,6 +4915,7 @@ static void alc880_auto_init(struct hda_codec *codec) alc880_auto_init_multi_out(codec); alc880_auto_init_extra_out(codec); alc880_auto_init_analog_input(codec); + alc880_auto_init_input_src(codec); if (spec->unsol_event) alc_inithook(codec); } @@ -6397,6 +6427,8 @@ static void alc260_auto_init_analog_input(struct hda_codec *codec) } } +#define alc260_auto_init_input_src alc880_auto_init_input_src + /* * generic initialization of ADC, input mixers and output mixers */ @@ -6483,6 +6515,7 @@ static void alc260_auto_init(struct hda_codec *codec) struct alc_spec *spec = codec->spec; alc260_auto_init_multi_out(codec); alc260_auto_init_analog_input(codec); + alc260_auto_init_input_src(codec); if (spec->unsol_event) alc_inithook(codec); } @@ -8455,6 +8488,42 @@ static struct snd_kcontrol_new alc883_medion_md2_mixer[] = { { } /* end */ }; +static struct snd_kcontrol_new alc883_medion_wim2160_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_MUTE("Speaker Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x1a, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x08, 0x0, HDA_INPUT), + { } /* end */ +}; + +static struct hda_verb alc883_medion_wim2160_verbs[] = { + /* Unmute front mixer */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* Set speaker pin to front mixer */ + {0x15, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Init headphone pin */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x1a, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x1a, AC_VERB_SET_UNSOLICITED_ENABLE, ALC880_HP_EVENT | AC_USRSP_EN}, + + { } /* end */ +}; + +/* toggle speaker-output according to the hp-jack state */ +static void alc883_medion_wim2160_setup(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + + spec->autocfg.hp_pins[0] = 0x1a; + spec->autocfg.speaker_pins[0] = 0x15; +} + static struct snd_kcontrol_new alc883_acer_aspire_mixer[] = { HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), @@ -9164,6 +9233,7 @@ static const char *alc882_models[ALC882_MODEL_LAST] = { [ALC888_ACER_ASPIRE_7730G] = "acer-aspire-7730g", [ALC883_MEDION] = "medion", [ALC883_MEDION_MD2] = "medion-md2", + [ALC883_MEDION_WIM2160] = "medion-wim2160", [ALC883_LAPTOP_EAPD] = "laptop-eapd", [ALC883_LENOVO_101E_2ch] = "lenovo-101e", [ALC883_LENOVO_NB0763] = "lenovo-nb0763", @@ -9280,6 +9350,7 @@ static struct snd_pci_quirk alc882_cfg_tbl[] = { SND_PCI_QUIRK(0x1462, 0xaa08, "MSI", ALC883_TARGA_2ch_DIG), SND_PCI_QUIRK(0x147b, 0x1083, "Abit IP35-PRO", ALC883_6ST_DIG), + SND_PCI_QUIRK(0x1558, 0x0571, "Clevo laptop M570U", ALC883_3ST_6ch_DIG), SND_PCI_QUIRK(0x1558, 0x0721, "Clevo laptop M720R", ALC883_CLEVO_M720), SND_PCI_QUIRK(0x1558, 0x0722, "Clevo laptop M720SR", ALC883_CLEVO_M720), SND_PCI_QUIRK(0x1558, 0x5409, "Clevo laptop M540R", ALC883_CLEVO_M540R), @@ -9818,6 +9889,21 @@ static struct alc_config_preset alc882_presets[] = { .setup = alc883_medion_md2_setup, .init_hook = alc_automute_amp, }, + [ALC883_MEDION_WIM2160] = { + .mixers = { alc883_medion_wim2160_mixer }, + .init_verbs = { alc883_init_verbs, alc883_medion_wim2160_verbs }, + .num_dacs = ARRAY_SIZE(alc883_dac_nids), + .dac_nids = alc883_dac_nids, + .dig_out_nid = ALC883_DIGOUT_NID, + .num_adc_nids = ARRAY_SIZE(alc883_adc_nids), + .adc_nids = alc883_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc883_3ST_2ch_modes), + .channel_mode = alc883_3ST_2ch_modes, + .input_mux = &alc883_capture_source, + .unsol_event = alc_automute_amp_unsol_event, + .setup = alc883_medion_wim2160_setup, + .init_hook = alc_automute_amp, + }, [ALC883_LAPTOP_EAPD] = { .mixers = { alc883_base_mixer }, .init_verbs = { alc883_init_verbs, alc882_eapd_verbs }, @@ -10363,7 +10449,8 @@ static int patch_alc882(struct hda_codec *codec) board_config = ALC882_AUTO; } - alc_pick_fixup(codec, alc882_fixup_tbl, alc882_fixups); + if (board_config == ALC882_AUTO) + alc_pick_fixup(codec, alc882_fixup_tbl, alc882_fixups, 1); if (board_config == ALC882_AUTO) { /* automatic parse from the BIOS config */ @@ -10436,6 +10523,9 @@ static int patch_alc882(struct hda_codec *codec) set_capture_mixer(codec); set_beep_amp(spec, 0x0b, 0x05, HDA_INPUT); + if (board_config == ALC882_AUTO) + alc_pick_fixup(codec, alc882_fixup_tbl, alc882_fixups, 0); + spec->vmaster_nid = 0x0c; codec->patch_ops = alc_patch_ops; @@ -12816,6 +12906,7 @@ static int alc268_new_analog_output(struct alc_spec *spec, hda_nid_t nid, dac = 0x02; break; case 0x15: + case 0x21: /* ALC269vb has this pin, too */ dac = 0x03; break; default: @@ -13735,19 +13826,19 @@ static void alc269_laptop_unsol_event(struct hda_codec *codec, } } -static void alc269_laptop_dmic_setup(struct hda_codec *codec) +static void alc269_laptop_amic_setup(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; spec->autocfg.hp_pins[0] = 0x15; spec->autocfg.speaker_pins[0] = 0x14; spec->ext_mic.pin = 0x18; spec->ext_mic.mux_idx = 0; - spec->int_mic.pin = 0x12; - spec->int_mic.mux_idx = 5; + spec->int_mic.pin = 0x19; + spec->int_mic.mux_idx = 1; spec->auto_mic = 1; } -static void alc269vb_laptop_dmic_setup(struct hda_codec *codec) +static void alc269_laptop_dmic_setup(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; spec->autocfg.hp_pins[0] = 0x15; @@ -13755,14 +13846,14 @@ static void alc269vb_laptop_dmic_setup(struct hda_codec *codec) spec->ext_mic.pin = 0x18; spec->ext_mic.mux_idx = 0; spec->int_mic.pin = 0x12; - spec->int_mic.mux_idx = 6; + spec->int_mic.mux_idx = 5; spec->auto_mic = 1; } -static void alc269_laptop_amic_setup(struct hda_codec *codec) +static void alc269vb_laptop_amic_setup(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - spec->autocfg.hp_pins[0] = 0x15; + spec->autocfg.hp_pins[0] = 0x21; spec->autocfg.speaker_pins[0] = 0x14; spec->ext_mic.pin = 0x18; spec->ext_mic.mux_idx = 0; @@ -13771,6 +13862,18 @@ static void alc269_laptop_amic_setup(struct hda_codec *codec) spec->auto_mic = 1; } +static void alc269vb_laptop_dmic_setup(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + spec->autocfg.hp_pins[0] = 0x21; + spec->autocfg.speaker_pins[0] = 0x14; + spec->ext_mic.pin = 0x18; + spec->ext_mic.mux_idx = 0; + spec->int_mic.pin = 0x12; + spec->int_mic.mux_idx = 6; + spec->auto_mic = 1; +} + static void alc269_laptop_inithook(struct hda_codec *codec) { alc269_speaker_automute(codec); @@ -13975,6 +14078,27 @@ static void alc269_auto_init(struct hda_codec *codec) alc_inithook(codec); } +enum { + ALC269_FIXUP_SONY_VAIO, +}; + +const static struct hda_verb alc269_sony_vaio_fixup_verbs[] = { + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREFGRD}, + {} +}; + +static const struct alc_fixup alc269_fixups[] = { + [ALC269_FIXUP_SONY_VAIO] = { + .verbs = alc269_sony_vaio_fixup_verbs + }, +}; + +static struct snd_pci_quirk alc269_fixup_tbl[] = { + SND_PCI_QUIRK(0x104d, 0x9071, "Sony VAIO", ALC269_FIXUP_SONY_VAIO), + {} +}; + + /* * configuration and preset */ @@ -14034,7 +14158,7 @@ static struct snd_pci_quirk alc269_cfg_tbl[] = { ALC269_DMIC), SND_PCI_QUIRK(0x1043, 0x8398, "ASUS P1005HA", ALC269_DMIC), SND_PCI_QUIRK(0x1043, 0x83ce, "ASUS P1005HA", ALC269_DMIC), - SND_PCI_QUIRK(0x104d, 0x9071, "SONY XTB", ALC269_DMIC), + SND_PCI_QUIRK(0x104d, 0x9071, "Sony VAIO", ALC269_AUTO), SND_PCI_QUIRK(0x10cf, 0x1475, "Lifebook ICH9M-based", ALC269_LIFEBOOK), SND_PCI_QUIRK(0x152d, 0x1778, "Quanta ON1", ALC269_DMIC), SND_PCI_QUIRK(0x1734, 0x115d, "FSC Amilo", ALC269_FUJITSU), @@ -14108,7 +14232,7 @@ static struct alc_config_preset alc269_presets[] = { .num_channel_mode = ARRAY_SIZE(alc269_modes), .channel_mode = alc269_modes, .unsol_event = alc269_laptop_unsol_event, - .setup = alc269_laptop_amic_setup, + .setup = alc269vb_laptop_amic_setup, .init_hook = alc269_laptop_inithook, }, [ALC269VB_DMIC] = { @@ -14188,6 +14312,9 @@ static int patch_alc269(struct hda_codec *codec) board_config = ALC269_AUTO; } + if (board_config == ALC269_AUTO) + alc_pick_fixup(codec, alc269_fixup_tbl, alc269_fixups, 1); + if (board_config == ALC269_AUTO) { /* automatic parse from the BIOS config */ err = alc269_parse_auto_config(codec); @@ -14240,6 +14367,9 @@ static int patch_alc269(struct hda_codec *codec) set_capture_mixer(codec); set_beep_amp(spec, 0x0b, 0x04, HDA_INPUT); + if (board_config == ALC269_AUTO) + alc_pick_fixup(codec, alc269_fixup_tbl, alc269_fixups, 0); + spec->vmaster_nid = 0x02; codec->patch_ops = alc_patch_ops; @@ -15328,7 +15458,8 @@ static int patch_alc861(struct hda_codec *codec) board_config = ALC861_AUTO; } - alc_pick_fixup(codec, alc861_fixup_tbl, alc861_fixups); + if (board_config == ALC861_AUTO) + alc_pick_fixup(codec, alc861_fixup_tbl, alc861_fixups, 1); if (board_config == ALC861_AUTO) { /* automatic parse from the BIOS config */ @@ -15365,6 +15496,9 @@ static int patch_alc861(struct hda_codec *codec) spec->vmaster_nid = 0x03; + if (board_config == ALC861_AUTO) + alc_pick_fixup(codec, alc861_fixup_tbl, alc861_fixups, 0); + codec->patch_ops = alc_patch_ops; if (board_config == ALC861_AUTO) { spec->init_hook = alc861_auto_init; @@ -16299,7 +16433,8 @@ static int patch_alc861vd(struct hda_codec *codec) board_config = ALC861VD_AUTO; } - alc_pick_fixup(codec, alc861vd_fixup_tbl, alc861vd_fixups); + if (board_config == ALC861VD_AUTO) + alc_pick_fixup(codec, alc861vd_fixup_tbl, alc861vd_fixups, 1); if (board_config == ALC861VD_AUTO) { /* automatic parse from the BIOS config */ @@ -16347,6 +16482,9 @@ static int patch_alc861vd(struct hda_codec *codec) spec->vmaster_nid = 0x02; + if (board_config == ALC861VD_AUTO) + alc_pick_fixup(codec, alc861vd_fixup_tbl, alc861vd_fixups, 0); + codec->patch_ops = alc_patch_ops; if (board_config == ALC861VD_AUTO) diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index 9ddc37300f6b..73453814e098 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -476,7 +476,7 @@ static struct snd_kcontrol_new *via_clone_control(struct via_spec *spec, knew->name = kstrdup(tmpl->name, GFP_KERNEL); if (!knew->name) return NULL; - return 0; + return knew; } static void via_free_kctls(struct hda_codec *codec) @@ -1215,14 +1215,13 @@ static struct snd_kcontrol_new via_hp_mixer[2] = { }, }; -static int via_hp_build(struct via_spec *spec) +static int via_hp_build(struct hda_codec *codec) { + struct via_spec *spec = codec->spec; struct snd_kcontrol_new *knew; hda_nid_t nid; - - knew = via_clone_control(spec, &via_hp_mixer[0]); - if (knew == NULL) - return -ENOMEM; + int nums; + hda_nid_t conn[HDA_MAX_CONNECTIONS]; switch (spec->codec_type) { case VT1718S: @@ -1239,6 +1238,14 @@ static int via_hp_build(struct via_spec *spec) break; } + nums = snd_hda_get_connections(codec, nid, conn, HDA_MAX_CONNECTIONS); + if (nums <= 1) + return 0; + + knew = via_clone_control(spec, &via_hp_mixer[0]); + if (knew == NULL) + return -ENOMEM; + knew->subdevice = HDA_SUBDEV_NID_FLAG | nid; knew->private_value = nid; @@ -2561,7 +2568,7 @@ static int vt1708_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); return 1; @@ -3087,7 +3094,7 @@ static int vt1709_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); return 1; @@ -3654,7 +3661,7 @@ static int vt1708B_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); return 1; @@ -4140,7 +4147,7 @@ static int vt1708S_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); return 1; @@ -4510,7 +4517,7 @@ static int vt1702_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); return 1; } @@ -4930,7 +4937,7 @@ static int vt1718S_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); @@ -5425,7 +5432,7 @@ static int vt1716S_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); via_smart51_build(spec); @@ -5781,7 +5788,7 @@ static int vt2002P_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); return 1; } @@ -6000,12 +6007,12 @@ static int vt1812_auto_create_multi_out_ctls(struct via_spec *spec, /* Line-Out: PortE */ err = via_add_control(spec, VIA_CTL_WIDGET_VOL, - "Master Front Playback Volume", + "Front Playback Volume", HDA_COMPOSE_AMP_VAL(0x8, 3, 0, HDA_OUTPUT)); if (err < 0) return err; err = via_add_control(spec, VIA_CTL_WIDGET_BIND_PIN_MUTE, - "Master Front Playback Switch", + "Front Playback Switch", HDA_COMPOSE_AMP_VAL(0x28, 3, 0, HDA_OUTPUT)); if (err < 0) return err; @@ -6130,7 +6137,7 @@ static int vt1812_parse_auto_config(struct hda_codec *codec) spec->input_mux = &spec->private_imux[0]; if (spec->hp_mux) - via_hp_build(spec); + via_hp_build(codec); return 1; } diff --git a/sound/soc/codecs/wm2000.c b/sound/soc/codecs/wm2000.c index a34cbcf7904f..002e289d1255 100644 --- a/sound/soc/codecs/wm2000.c +++ b/sound/soc/codecs/wm2000.c @@ -23,7 +23,6 @@ #include <linux/module.h> #include <linux/moduleparam.h> -#include <linux/version.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/firmware.h> diff --git a/sound/soc/imx/imx-pcm-dma-mx2.c b/sound/soc/imx/imx-pcm-dma-mx2.c index 2e79d7136298..2b31ac673ea4 100644 --- a/sound/soc/imx/imx-pcm-dma-mx2.c +++ b/sound/soc/imx/imx-pcm-dma-mx2.c @@ -71,7 +71,12 @@ static void imx_ssi_dma_callback(int channel, void *data) static void snd_imx_dma_err_callback(int channel, void *data, int err) { - pr_err("DMA error callback called\n"); + struct snd_pcm_substream *substream = data; + struct snd_soc_pcm_runtime *rtd = substream->private_data; + struct imx_pcm_dma_params *dma_params = rtd->dai->cpu_dai->dma_data; + struct snd_pcm_runtime *runtime = substream->runtime; + struct imx_pcm_runtime_data *iprtd = runtime->private_data; + int ret; pr_err("DMA timeout on channel %d -%s%s%s%s\n", channel, @@ -79,6 +84,14 @@ static void snd_imx_dma_err_callback(int channel, void *data, int err) err & IMX_DMA_ERR_REQUEST ? " request" : "", err & IMX_DMA_ERR_TRANSFER ? " transfer" : "", err & IMX_DMA_ERR_BUFFER ? " buffer" : ""); + + imx_dma_disable(iprtd->dma); + ret = imx_dma_setup_sg(iprtd->dma, iprtd->sg_list, iprtd->sg_count, + IMX_DMA_LENGTH_LOOP, dma_params->dma_addr, + substream->stream == SNDRV_PCM_STREAM_PLAYBACK ? + DMA_MODE_WRITE : DMA_MODE_READ); + if (!ret) + imx_dma_enable(iprtd->dma); } static int imx_ssi_dma_alloc(struct snd_pcm_substream *substream) diff --git a/sound/soc/imx/imx-pcm-fiq.c b/sound/soc/imx/imx-pcm-fiq.c index f96a373699cf..6b518e07eea9 100644 --- a/sound/soc/imx/imx-pcm-fiq.c +++ b/sound/soc/imx/imx-pcm-fiq.c @@ -39,23 +39,24 @@ struct imx_pcm_runtime_data { unsigned long offset; unsigned long last_offset; unsigned long size; - struct timer_list timer; - int poll_time; + struct hrtimer hrt; + int poll_time_ns; + struct snd_pcm_substream *substream; + atomic_t running; }; -static inline void imx_ssi_set_next_poll(struct imx_pcm_runtime_data *iprtd) +static enum hrtimer_restart snd_hrtimer_callback(struct hrtimer *hrt) { - iprtd->timer.expires = jiffies + iprtd->poll_time; -} - -static void imx_ssi_timer_callback(unsigned long data) -{ - struct snd_pcm_substream *substream = (void *)data; + struct imx_pcm_runtime_data *iprtd = + container_of(hrt, struct imx_pcm_runtime_data, hrt); + struct snd_pcm_substream *substream = iprtd->substream; struct snd_pcm_runtime *runtime = substream->runtime; - struct imx_pcm_runtime_data *iprtd = runtime->private_data; struct pt_regs regs; unsigned long delta; + if (!atomic_read(&iprtd->running)) + return HRTIMER_NORESTART; + get_fiq_regs(®s); if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) @@ -72,16 +73,14 @@ static void imx_ssi_timer_callback(unsigned long data) /* If we've transferred at least a period then report it and * reset our poll time */ - if (delta >= runtime->period_size) { + if (delta >= iprtd->period) { snd_pcm_period_elapsed(substream); iprtd->last_offset = iprtd->offset; - - imx_ssi_set_next_poll(iprtd); } - /* Restart the timer; if we didn't report we'll run on the next tick */ - add_timer(&iprtd->timer); + hrtimer_forward_now(hrt, ns_to_ktime(iprtd->poll_time_ns)); + return HRTIMER_RESTART; } static struct fiq_handler fh = { @@ -99,8 +98,8 @@ static int snd_imx_pcm_hw_params(struct snd_pcm_substream *substream, iprtd->period = params_period_bytes(params) ; iprtd->offset = 0; iprtd->last_offset = 0; - iprtd->poll_time = HZ / (params_rate(params) / params_period_size(params)); - + iprtd->poll_time_ns = 1000000000 / params_rate(params) * + params_period_size(params); snd_pcm_set_runtime_buffer(substream, &substream->dma_buffer); return 0; @@ -135,8 +134,9 @@ static int snd_imx_pcm_trigger(struct snd_pcm_substream *substream, int cmd) case SNDRV_PCM_TRIGGER_START: case SNDRV_PCM_TRIGGER_RESUME: case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: - imx_ssi_set_next_poll(iprtd); - add_timer(&iprtd->timer); + atomic_set(&iprtd->running, 1); + hrtimer_start(&iprtd->hrt, ns_to_ktime(iprtd->poll_time_ns), + HRTIMER_MODE_REL); if (++fiq_enable == 1) enable_fiq(imx_pcm_fiq); @@ -145,11 +145,11 @@ static int snd_imx_pcm_trigger(struct snd_pcm_substream *substream, int cmd) case SNDRV_PCM_TRIGGER_STOP: case SNDRV_PCM_TRIGGER_SUSPEND: case SNDRV_PCM_TRIGGER_PAUSE_PUSH: - del_timer(&iprtd->timer); + atomic_set(&iprtd->running, 0); + if (--fiq_enable == 0) disable_fiq(imx_pcm_fiq); - break; default: return -EINVAL; @@ -180,7 +180,7 @@ static struct snd_pcm_hardware snd_imx_hardware = { .buffer_bytes_max = IMX_SSI_DMABUF_SIZE, .period_bytes_min = 128, .period_bytes_max = 16 * 1024, - .periods_min = 2, + .periods_min = 4, .periods_max = 255, .fifo_size = 0, }; @@ -194,9 +194,11 @@ static int snd_imx_open(struct snd_pcm_substream *substream) iprtd = kzalloc(sizeof(*iprtd), GFP_KERNEL); runtime->private_data = iprtd; - init_timer(&iprtd->timer); - iprtd->timer.data = (unsigned long)substream; - iprtd->timer.function = imx_ssi_timer_callback; + iprtd->substream = substream; + + atomic_set(&iprtd->running, 0); + hrtimer_init(&iprtd->hrt, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + iprtd->hrt.function = snd_hrtimer_callback; ret = snd_pcm_hw_constraint_integer(substream->runtime, SNDRV_PCM_HW_PARAM_PERIODS); @@ -212,7 +214,8 @@ static int snd_imx_close(struct snd_pcm_substream *substream) struct snd_pcm_runtime *runtime = substream->runtime; struct imx_pcm_runtime_data *iprtd = runtime->private_data; - del_timer_sync(&iprtd->timer); + hrtimer_cancel(&iprtd->hrt); + kfree(iprtd); return 0; diff --git a/sound/soc/imx/imx-ssi.c b/sound/soc/imx/imx-ssi.c index 0bcc6d7d9471..80b4fee2442b 100644 --- a/sound/soc/imx/imx-ssi.c +++ b/sound/soc/imx/imx-ssi.c @@ -656,7 +656,8 @@ static int imx_ssi_probe(struct platform_device *pdev) dai->private_data = ssi; if ((cpu_is_mx27() || cpu_is_mx21()) && - !(ssi->flags & IMX_SSI_USE_AC97)) { + !(ssi->flags & IMX_SSI_USE_AC97) && + (ssi->flags & IMX_SSI_DMA)) { ssi->flags |= IMX_SSI_DMA; platform = imx_ssi_dma_mx2_init(pdev, ssi); } else diff --git a/sound/usb/usbmidi.c b/sound/usb/usbmidi.c index 2c59afd99611..9e28b20cb2ce 100644 --- a/sound/usb/usbmidi.c +++ b/sound/usb/usbmidi.c @@ -986,6 +986,8 @@ static void snd_usbmidi_output_drain(struct snd_rawmidi_substream *substream) DEFINE_WAIT(wait); long timeout = msecs_to_jiffies(50); + if (ep->umidi->disconnected) + return; /* * The substream buffer is empty, but some data might still be in the * currently active URBs, so we have to wait for those to complete. @@ -1123,14 +1125,21 @@ static int snd_usbmidi_in_endpoint_create(struct snd_usb_midi* umidi, * Frees an output endpoint. * May be called when ep hasn't been initialized completely. */ -static void snd_usbmidi_out_endpoint_delete(struct snd_usb_midi_out_endpoint* ep) +static void snd_usbmidi_out_endpoint_clear(struct snd_usb_midi_out_endpoint *ep) { unsigned int i; for (i = 0; i < OUTPUT_URBS; ++i) - if (ep->urbs[i].urb) + if (ep->urbs[i].urb) { free_urb_and_buffer(ep->umidi, ep->urbs[i].urb, ep->max_transfer); + ep->urbs[i].urb = NULL; + } +} + +static void snd_usbmidi_out_endpoint_delete(struct snd_usb_midi_out_endpoint *ep) +{ + snd_usbmidi_out_endpoint_clear(ep); kfree(ep); } @@ -1262,15 +1271,18 @@ void snd_usbmidi_disconnect(struct list_head* p) usb_kill_urb(ep->out->urbs[j].urb); if (umidi->usb_protocol_ops->finish_out_endpoint) umidi->usb_protocol_ops->finish_out_endpoint(ep->out); + ep->out->active_urbs = 0; + if (ep->out->drain_urbs) { + ep->out->drain_urbs = 0; + wake_up(&ep->out->drain_wait); + } } if (ep->in) for (j = 0; j < INPUT_URBS; ++j) usb_kill_urb(ep->in->urbs[j]); /* free endpoints here; later call can result in Oops */ - if (ep->out) { - snd_usbmidi_out_endpoint_delete(ep->out); - ep->out = NULL; - } + if (ep->out) + snd_usbmidi_out_endpoint_clear(ep->out); if (ep->in) { snd_usbmidi_in_endpoint_delete(ep->in); ep->in = NULL; diff --git a/tools/perf/Documentation/perf-bench.txt b/tools/perf/Documentation/perf-bench.txt index ae525ac5a2ce..0181dddf6b61 100644 --- a/tools/perf/Documentation/perf-bench.txt +++ b/tools/perf/Documentation/perf-bench.txt @@ -19,12 +19,12 @@ COMMON OPTIONS -f:: --format=:: Specify format style. -Current available format styles are, +Current available format styles are: 'default':: Default style. This is mainly for human reading. --------------------- -% perf bench sched pipe # with no style specify +% perf bench sched pipe # with no style specified (executing 1000000 pipe operations between two tasks) Total time:5.855 sec 5.855061 usecs/op @@ -79,7 +79,7 @@ options (20 sender and receiver processes per group) Total time:0.308 sec -% perf bench sched messaging -t -g 20 # be multi-thread,with 20 groups +% perf bench sched messaging -t -g 20 # be multi-thread, with 20 groups (20 sender and receiver threads per group) (20 groups == 800 threads run) diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt index 34202b1be0bb..63c25d304880 100644 --- a/tools/perf/Documentation/perf-probe.txt +++ b/tools/perf/Documentation/perf-probe.txt @@ -57,6 +57,11 @@ OPTIONS --force:: Forcibly add events with existing name. +-n:: +--dry-run:: + Dry run. With this option, --add and --del doesn't execute actual + adding and removal operations. + PROBE SYNTAX ------------ Probe points are defined by following syntax. @@ -74,13 +79,22 @@ Probe points are defined by following syntax. 'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. Currently, event group name is set as 'probe'. 'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, ':RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition. In addition, '@SRC' specifies a source file which has that function. It is also possible to specify a probe point by the source line number or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern. -'ARG' specifies the arguments of this probe point. You can use the name of local variable, or kprobe-tracer argument format (e.g. $retval, %ax, etc). +'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT). + +PROBE ARGUMENT +-------------- +Each probe argument follows below syntax. + + [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE] + +'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.) +'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. LINE SYNTAX ----------- Line range is descripted by following syntax. - "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]" + "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]" FUNC specifies the function name of showing lines. 'RLN' is the start line number from function entry line, and 'RLN2' is the end line number. As same as diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt index fc46c0b40f6e..020d871c7934 100644 --- a/tools/perf/Documentation/perf-record.txt +++ b/tools/perf/Documentation/perf-record.txt @@ -58,7 +58,7 @@ OPTIONS -f:: --force:: - Overwrite existing data file. + Overwrite existing data file. (deprecated) -c:: --count=:: @@ -101,7 +101,7 @@ OPTIONS -R:: --raw-samples:: -Collect raw sample records from all opened counters (typically for tracepoint counters). +Collect raw sample records from all opened counters (default for tracepoint counters). SEE ALSO -------- diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt index 1ce79198997b..8417644a6166 100644 --- a/tools/perf/Documentation/perf-sched.txt +++ b/tools/perf/Documentation/perf-sched.txt @@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -There's four variants of perf sched: +There are four variants of perf sched: 'perf sched record <command>' to record the scheduling events of an arbitrary workload. @@ -27,7 +27,7 @@ There's four variants of perf sched: via perf sched record. (this is done by starting up mockup threads that mimic the workload based on the events in the trace. These threads can then replay the timings (CPU runtime and sleep patterns) - of the workload as it occured when it was recorded - and can repeat + of the workload as it occurred when it was recorded - and can repeat it a number of times, measuring its performance.) OPTIONS diff --git a/tools/perf/Makefile b/tools/perf/Makefile index bc0f670a8338..57b3569716dd 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -1,3 +1,7 @@ +ifeq ("$(origin O)", "command line") + OUTPUT := $(O)/ +endif + # The default target of this Makefile is... all:: @@ -150,10 +154,17 @@ all:: # Define LDFLAGS=-static to build a static binary. # # Define EXTRA_CFLAGS=-m64 or EXTRA_CFLAGS=-m32 as appropriate for cross-builds. +# +# Define NO_DWARF if you do not want debug-info analysis feature at all. -PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE - @$(SHELL_PATH) util/PERF-VERSION-GEN --include PERF-VERSION-FILE +$(shell sh -c 'mkdir -p $(OUTPUT)scripts/python/Perf-Trace-Util/' 2> /dev/null) +$(shell sh -c 'mkdir -p $(OUTPUT)scripts/perl/Perf-Trace-Util/' 2> /dev/null) +$(shell sh -c 'mkdir -p $(OUTPUT)util/scripting-engines/' 2> /dev/null) +$(shell sh -c 'mkdir $(OUTPUT)bench' 2> /dev/null) + +$(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE + @$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT) +-include $(OUTPUT)PERF-VERSION-FILE uname_S := $(shell sh -c 'uname -s 2>/dev/null || echo not') uname_M := $(shell sh -c 'uname -m 2>/dev/null || echo not') @@ -308,7 +319,7 @@ PROGRAMS += $(EXTRA_PROGRAMS) # # Single 'perf' binary right now: # -PROGRAMS += perf +PROGRAMS += $(OUTPUT)perf # List built-in command $C whose implementation cmd_$C() is not in # builtin-$C.o but is linked in as part of some other command. @@ -318,7 +329,7 @@ PROGRAMS += perf ALL_PROGRAMS = $(PROGRAMS) $(SCRIPTS) # what 'all' will build but not install in perfexecdir -OTHER_PROGRAMS = perf$X +OTHER_PROGRAMS = $(OUTPUT)perf$X # Set paths to tools early so that they can be used for version tests. ifndef SHELL_PATH @@ -330,7 +341,7 @@ endif export PERL_PATH -LIB_FILE=libperf.a +LIB_FILE=$(OUTPUT)libperf.a LIB_H += ../../include/linux/perf_event.h LIB_H += ../../include/linux/rbtree.h @@ -375,7 +386,6 @@ LIB_H += util/header.h LIB_H += util/help.h LIB_H += util/session.h LIB_H += util/strbuf.h -LIB_H += util/string.h LIB_H += util/strlist.h LIB_H += util/svghelper.h LIB_H += util/run-command.h @@ -391,77 +401,77 @@ LIB_H += util/probe-finder.h LIB_H += util/probe-event.h LIB_H += util/cpumap.h -LIB_OBJS += util/abspath.o -LIB_OBJS += util/alias.o -LIB_OBJS += util/build-id.o -LIB_OBJS += util/config.o -LIB_OBJS += util/ctype.o -LIB_OBJS += util/debugfs.o -LIB_OBJS += util/environment.o -LIB_OBJS += util/event.o -LIB_OBJS += util/exec_cmd.o -LIB_OBJS += util/help.o -LIB_OBJS += util/levenshtein.o -LIB_OBJS += util/parse-options.o -LIB_OBJS += util/parse-events.o -LIB_OBJS += util/path.o -LIB_OBJS += util/rbtree.o -LIB_OBJS += util/bitmap.o -LIB_OBJS += util/hweight.o -LIB_OBJS += util/find_next_bit.o -LIB_OBJS += util/run-command.o -LIB_OBJS += util/quote.o -LIB_OBJS += util/strbuf.o -LIB_OBJS += util/string.o -LIB_OBJS += util/strlist.o -LIB_OBJS += util/usage.o -LIB_OBJS += util/wrapper.o -LIB_OBJS += util/sigchain.o -LIB_OBJS += util/symbol.o -LIB_OBJS += util/color.o -LIB_OBJS += util/pager.o -LIB_OBJS += util/header.o -LIB_OBJS += util/callchain.o -LIB_OBJS += util/values.o -LIB_OBJS += util/debug.o -LIB_OBJS += util/map.o -LIB_OBJS += util/session.o -LIB_OBJS += util/thread.o -LIB_OBJS += util/trace-event-parse.o -LIB_OBJS += util/trace-event-read.o -LIB_OBJS += util/trace-event-info.o -LIB_OBJS += util/trace-event-scripting.o -LIB_OBJS += util/svghelper.o -LIB_OBJS += util/sort.o -LIB_OBJS += util/hist.o -LIB_OBJS += util/probe-event.o -LIB_OBJS += util/util.o -LIB_OBJS += util/cpumap.o - -BUILTIN_OBJS += builtin-annotate.o - -BUILTIN_OBJS += builtin-bench.o +LIB_OBJS += $(OUTPUT)util/abspath.o +LIB_OBJS += $(OUTPUT)util/alias.o +LIB_OBJS += $(OUTPUT)util/build-id.o +LIB_OBJS += $(OUTPUT)util/config.o +LIB_OBJS += $(OUTPUT)util/ctype.o +LIB_OBJS += $(OUTPUT)util/debugfs.o +LIB_OBJS += $(OUTPUT)util/environment.o +LIB_OBJS += $(OUTPUT)util/event.o +LIB_OBJS += $(OUTPUT)util/exec_cmd.o +LIB_OBJS += $(OUTPUT)util/help.o +LIB_OBJS += $(OUTPUT)util/levenshtein.o +LIB_OBJS += $(OUTPUT)util/parse-options.o +LIB_OBJS += $(OUTPUT)util/parse-events.o +LIB_OBJS += $(OUTPUT)util/path.o +LIB_OBJS += $(OUTPUT)util/rbtree.o +LIB_OBJS += $(OUTPUT)util/bitmap.o +LIB_OBJS += $(OUTPUT)util/hweight.o +LIB_OBJS += $(OUTPUT)util/find_next_bit.o +LIB_OBJS += $(OUTPUT)util/run-command.o +LIB_OBJS += $(OUTPUT)util/quote.o +LIB_OBJS += $(OUTPUT)util/strbuf.o +LIB_OBJS += $(OUTPUT)util/string.o +LIB_OBJS += $(OUTPUT)util/strlist.o +LIB_OBJS += $(OUTPUT)util/usage.o +LIB_OBJS += $(OUTPUT)util/wrapper.o +LIB_OBJS += $(OUTPUT)util/sigchain.o +LIB_OBJS += $(OUTPUT)util/symbol.o +LIB_OBJS += $(OUTPUT)util/color.o +LIB_OBJS += $(OUTPUT)util/pager.o +LIB_OBJS += $(OUTPUT)util/header.o +LIB_OBJS += $(OUTPUT)util/callchain.o +LIB_OBJS += $(OUTPUT)util/values.o +LIB_OBJS += $(OUTPUT)util/debug.o +LIB_OBJS += $(OUTPUT)util/map.o +LIB_OBJS += $(OUTPUT)util/session.o +LIB_OBJS += $(OUTPUT)util/thread.o +LIB_OBJS += $(OUTPUT)util/trace-event-parse.o +LIB_OBJS += $(OUTPUT)util/trace-event-read.o +LIB_OBJS += $(OUTPUT)util/trace-event-info.o +LIB_OBJS += $(OUTPUT)util/trace-event-scripting.o +LIB_OBJS += $(OUTPUT)util/svghelper.o +LIB_OBJS += $(OUTPUT)util/sort.o +LIB_OBJS += $(OUTPUT)util/hist.o +LIB_OBJS += $(OUTPUT)util/probe-event.o +LIB_OBJS += $(OUTPUT)util/util.o +LIB_OBJS += $(OUTPUT)util/cpumap.o + +BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o + +BUILTIN_OBJS += $(OUTPUT)builtin-bench.o # Benchmark modules -BUILTIN_OBJS += bench/sched-messaging.o -BUILTIN_OBJS += bench/sched-pipe.o -BUILTIN_OBJS += bench/mem-memcpy.o - -BUILTIN_OBJS += builtin-diff.o -BUILTIN_OBJS += builtin-help.o -BUILTIN_OBJS += builtin-sched.o -BUILTIN_OBJS += builtin-buildid-list.o -BUILTIN_OBJS += builtin-buildid-cache.o -BUILTIN_OBJS += builtin-list.o -BUILTIN_OBJS += builtin-record.o -BUILTIN_OBJS += builtin-report.o -BUILTIN_OBJS += builtin-stat.o -BUILTIN_OBJS += builtin-timechart.o -BUILTIN_OBJS += builtin-top.o -BUILTIN_OBJS += builtin-trace.o -BUILTIN_OBJS += builtin-probe.o -BUILTIN_OBJS += builtin-kmem.o -BUILTIN_OBJS += builtin-lock.o +BUILTIN_OBJS += $(OUTPUT)bench/sched-messaging.o +BUILTIN_OBJS += $(OUTPUT)bench/sched-pipe.o +BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy.o + +BUILTIN_OBJS += $(OUTPUT)builtin-diff.o +BUILTIN_OBJS += $(OUTPUT)builtin-help.o +BUILTIN_OBJS += $(OUTPUT)builtin-sched.o +BUILTIN_OBJS += $(OUTPUT)builtin-buildid-list.o +BUILTIN_OBJS += $(OUTPUT)builtin-buildid-cache.o +BUILTIN_OBJS += $(OUTPUT)builtin-list.o +BUILTIN_OBJS += $(OUTPUT)builtin-record.o +BUILTIN_OBJS += $(OUTPUT)builtin-report.o +BUILTIN_OBJS += $(OUTPUT)builtin-stat.o +BUILTIN_OBJS += $(OUTPUT)builtin-timechart.o +BUILTIN_OBJS += $(OUTPUT)builtin-top.o +BUILTIN_OBJS += $(OUTPUT)builtin-trace.o +BUILTIN_OBJS += $(OUTPUT)builtin-probe.o +BUILTIN_OBJS += $(OUTPUT)builtin-kmem.o +BUILTIN_OBJS += $(OUTPUT)builtin-lock.o PERFLIBS = $(LIB_FILE) @@ -492,6 +502,10 @@ ifeq ($(uname_S),Darwin) PTHREAD_LIBS = endif +ifneq ($(OUTPUT),) + BASIC_CFLAGS += -I$(OUTPUT) +endif + ifeq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y) ifneq ($(shell sh -c "(echo '\#include <gnu/libc-version.h>'; echo 'int main(void) { const char * version = gnu_get_libc_version(); return (long)version; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y) msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]/glibc-static); @@ -506,11 +520,20 @@ endif ifneq ($(shell sh -c "(echo '\#include <dwarf.h>'; echo '\#include <libdw.h>'; echo 'int main(void) { Dwarf *dbg; dbg = dwarf_begin(0, DWARF_C_READ); return (long)dbg; }') | $(CC) -x c - $(ALL_CFLAGS) -I/usr/include/elfutils -ldw -lelf -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y) msg := $(warning No libdw.h found or old libdw.h found, disables dwarf support. Please install elfutils-devel/elfutils-dev); - BASIC_CFLAGS += -DNO_DWARF_SUPPORT else - BASIC_CFLAGS += -I/usr/include/elfutils +ifndef NO_DWARF + BASIC_CFLAGS += -I/usr/include/elfutils -DDWARF_SUPPORT EXTLIBS += -lelf -ldw - LIB_OBJS += util/probe-finder.o + LIB_OBJS += $(OUTPUT)util/probe-finder.o +endif +endif + +ifneq ($(shell sh -c "(echo '\#include <newt.h>'; echo 'int main(void) { newtInit(); newtCls(); return newtFinished(); }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -lnewt -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y) + msg := $(warning newt not found, disables TUI support. Please install newt-devel or libnewt-dev); + BASIC_CFLAGS += -DNO_NEWT_SUPPORT +else + EXTLIBS += -lnewt + LIB_OBJS += $(OUTPUT)util/newt.o endif ifndef NO_LIBPERL @@ -522,8 +545,8 @@ ifneq ($(shell sh -c "(echo '\#include <EXTERN.h>'; echo '\#include <perl.h>'; e BASIC_CFLAGS += -DNO_LIBPERL else ALL_LDFLAGS += $(PERL_EMBED_LDOPTS) - LIB_OBJS += util/scripting-engines/trace-event-perl.o - LIB_OBJS += scripts/perl/Perf-Trace-Util/Context.o + LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o + LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o endif ifndef NO_LIBPYTHON @@ -531,12 +554,12 @@ PYTHON_EMBED_LDOPTS = `python-config --ldflags 2>/dev/null` PYTHON_EMBED_CCOPTS = `python-config --cflags 2>/dev/null` endif -ifneq ($(shell sh -c "(echo '\#include <Python.h>'; echo 'int main(void) { Py_Initialize(); return 0; }') | $(CC) -x c - $(PYTHON_EMBED_CCOPTS) -o /dev/null $(PYTHON_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y) +ifneq ($(shell sh -c "(echo '\#include <Python.h>'; echo 'int main(void) { Py_Initialize(); return 0; }') | $(CC) -x c - $(PYTHON_EMBED_CCOPTS) -o $(BITBUCKET) $(PYTHON_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y) BASIC_CFLAGS += -DNO_LIBPYTHON else ALL_LDFLAGS += $(PYTHON_EMBED_LDOPTS) - LIB_OBJS += util/scripting-engines/trace-event-python.o - LIB_OBJS += scripts/python/Perf-Trace-Util/Context.o + LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o + LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o endif ifdef NO_DEMANGLE @@ -607,53 +630,53 @@ ifdef NO_C99_FORMAT endif ifdef SNPRINTF_RETURNS_BOGUS COMPAT_CFLAGS += -DSNPRINTF_RETURNS_BOGUS - COMPAT_OBJS += compat/snprintf.o + COMPAT_OBJS += $(OUTPUT)compat/snprintf.o endif ifdef FREAD_READS_DIRECTORIES COMPAT_CFLAGS += -DFREAD_READS_DIRECTORIES - COMPAT_OBJS += compat/fopen.o + COMPAT_OBJS += $(OUTPUT)compat/fopen.o endif ifdef NO_SYMLINK_HEAD BASIC_CFLAGS += -DNO_SYMLINK_HEAD endif ifdef NO_STRCASESTR COMPAT_CFLAGS += -DNO_STRCASESTR - COMPAT_OBJS += compat/strcasestr.o + COMPAT_OBJS += $(OUTPUT)compat/strcasestr.o endif ifdef NO_STRTOUMAX COMPAT_CFLAGS += -DNO_STRTOUMAX - COMPAT_OBJS += compat/strtoumax.o + COMPAT_OBJS += $(OUTPUT)compat/strtoumax.o endif ifdef NO_STRTOULL COMPAT_CFLAGS += -DNO_STRTOULL endif ifdef NO_SETENV COMPAT_CFLAGS += -DNO_SETENV - COMPAT_OBJS += compat/setenv.o + COMPAT_OBJS += $(OUTPUT)compat/setenv.o endif ifdef NO_MKDTEMP COMPAT_CFLAGS += -DNO_MKDTEMP - COMPAT_OBJS += compat/mkdtemp.o + COMPAT_OBJS += $(OUTPUT)compat/mkdtemp.o endif ifdef NO_UNSETENV COMPAT_CFLAGS += -DNO_UNSETENV - COMPAT_OBJS += compat/unsetenv.o + COMPAT_OBJS += $(OUTPUT)compat/unsetenv.o endif ifdef NO_SYS_SELECT_H BASIC_CFLAGS += -DNO_SYS_SELECT_H endif ifdef NO_MMAP COMPAT_CFLAGS += -DNO_MMAP - COMPAT_OBJS += compat/mmap.o + COMPAT_OBJS += $(OUTPUT)compat/mmap.o else ifdef USE_WIN32_MMAP COMPAT_CFLAGS += -DUSE_WIN32_MMAP - COMPAT_OBJS += compat/win32mmap.o + COMPAT_OBJS += $(OUTPUT)compat/win32mmap.o endif endif ifdef NO_PREAD COMPAT_CFLAGS += -DNO_PREAD - COMPAT_OBJS += compat/pread.o + COMPAT_OBJS += $(OUTPUT)compat/pread.o endif ifdef NO_FAST_WORKING_DIRECTORY BASIC_CFLAGS += -DNO_FAST_WORKING_DIRECTORY @@ -675,10 +698,10 @@ else endif endif ifdef NO_INET_NTOP - LIB_OBJS += compat/inet_ntop.o + LIB_OBJS += $(OUTPUT)compat/inet_ntop.o endif ifdef NO_INET_PTON - LIB_OBJS += compat/inet_pton.o + LIB_OBJS += $(OUTPUT)compat/inet_pton.o endif ifdef NO_ICONV @@ -695,15 +718,15 @@ endif ifdef PPC_SHA1 SHA1_HEADER = "ppc/sha1.h" - LIB_OBJS += ppc/sha1.o ppc/sha1ppc.o + LIB_OBJS += $(OUTPUT)ppc/sha1.o ppc/sha1ppc.o else ifdef ARM_SHA1 SHA1_HEADER = "arm/sha1.h" - LIB_OBJS += arm/sha1.o arm/sha1_arm.o + LIB_OBJS += $(OUTPUT)arm/sha1.o $(OUTPUT)arm/sha1_arm.o else ifdef MOZILLA_SHA1 SHA1_HEADER = "mozilla-sha1/sha1.h" - LIB_OBJS += mozilla-sha1/sha1.o + LIB_OBJS += $(OUTPUT)mozilla-sha1/sha1.o else SHA1_HEADER = <openssl/sha.h> EXTLIBS += $(LIB_4_CRYPTO) @@ -715,15 +738,15 @@ ifdef NO_PERL_MAKEMAKER endif ifdef NO_HSTRERROR COMPAT_CFLAGS += -DNO_HSTRERROR - COMPAT_OBJS += compat/hstrerror.o + COMPAT_OBJS += $(OUTPUT)compat/hstrerror.o endif ifdef NO_MEMMEM COMPAT_CFLAGS += -DNO_MEMMEM - COMPAT_OBJS += compat/memmem.o + COMPAT_OBJS += $(OUTPUT)compat/memmem.o endif ifdef INTERNAL_QSORT COMPAT_CFLAGS += -DINTERNAL_QSORT - COMPAT_OBJS += compat/qsort.o + COMPAT_OBJS += $(OUTPUT)compat/qsort.o endif ifdef RUNTIME_PREFIX COMPAT_CFLAGS += -DRUNTIME_PREFIX @@ -803,7 +826,7 @@ export TAR INSTALL DESTDIR SHELL_PATH SHELL = $(SHELL_PATH) -all:: .perf.dev.null shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS +all:: .perf.dev.null shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) $(OUTPUT)PERF-BUILD-OPTIONS ifneq (,$X) $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), test '$p' -ef '$p$X' || $(RM) '$p';) endif @@ -815,39 +838,39 @@ please_set_SHELL_PATH_to_a_more_modern_shell: shell_compatibility_test: please_set_SHELL_PATH_to_a_more_modern_shell -strip: $(PROGRAMS) perf$X - $(STRIP) $(STRIP_OPTS) $(PROGRAMS) perf$X +strip: $(PROGRAMS) $(OUTPUT)perf$X + $(STRIP) $(STRIP_OPTS) $(PROGRAMS) $(OUTPUT)perf$X -perf.o: perf.c common-cmds.h PERF-CFLAGS +$(OUTPUT)perf.o: perf.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS $(QUIET_CC)$(CC) -DPERF_VERSION='"$(PERF_VERSION)"' \ '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ - $(ALL_CFLAGS) -c $(filter %.c,$^) + $(ALL_CFLAGS) -c $(filter %.c,$^) -o $@ -perf$X: perf.o $(BUILTIN_OBJS) $(PERFLIBS) - $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ perf.o \ +$(OUTPUT)perf$X: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS) + $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(OUTPUT)perf.o \ $(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS) -builtin-help.o: builtin-help.c common-cmds.h PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \ +$(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ '-DPERF_MAN_PATH="$(mandir_SQ)"' \ '-DPERF_INFO_PATH="$(infodir_SQ)"' $< -builtin-timechart.o: builtin-timechart.c common-cmds.h PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \ +$(OUTPUT)builtin-timechart.o: builtin-timechart.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ '-DPERF_MAN_PATH="$(mandir_SQ)"' \ '-DPERF_INFO_PATH="$(infodir_SQ)"' $< -$(BUILT_INS): perf$X +$(BUILT_INS): $(OUTPUT)perf$X $(QUIET_BUILT_IN)$(RM) $@ && \ ln perf$X $@ 2>/dev/null || \ ln -s perf$X $@ 2>/dev/null || \ cp perf$X $@ -common-cmds.h: util/generate-cmdlist.sh command-list.txt +$(OUTPUT)common-cmds.h: util/generate-cmdlist.sh command-list.txt -common-cmds.h: $(wildcard Documentation/perf-*.txt) +$(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt) $(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@ $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh @@ -859,7 +882,7 @@ $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh -e 's/@@NO_CURL@@/$(NO_CURL)/g' \ $@.sh >$@+ && \ chmod +x $@+ && \ - mv $@+ $@ + mv $@+ $(OUTPUT)$@ configure: configure.ac $(QUIET_GEN)$(RM) $@ $<+ && \ @@ -869,60 +892,60 @@ configure: configure.ac $(RM) $<+ # These can record PERF_VERSION -perf.o perf.spec \ +$(OUTPUT)perf.o perf.spec \ $(patsubst %.sh,%,$(SCRIPT_SH)) \ $(patsubst %.perl,%,$(SCRIPT_PERL)) \ - : PERF-VERSION-FILE + : $(OUTPUT)PERF-VERSION-FILE -%.o: %.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $< -%.s: %.c PERF-CFLAGS +$(OUTPUT)%.o: %.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $< +$(OUTPUT)%.s: %.c $(OUTPUT)PERF-CFLAGS $(QUIET_CC)$(CC) -S $(ALL_CFLAGS) $< -%.o: %.S - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $< +$(OUTPUT)%.o: %.S + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $< -util/exec_cmd.o: util/exec_cmd.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \ +$(OUTPUT)util/exec_cmd.o: util/exec_cmd.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ '-DPERF_EXEC_PATH="$(perfexecdir_SQ)"' \ '-DBINDIR="$(bindir_relative_SQ)"' \ '-DPREFIX="$(prefix_SQ)"' \ $< -builtin-init-db.o: builtin-init-db.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $< +$(OUTPUT)builtin-init-db.o: builtin-init-db.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $< -util/config.o: util/config.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< +$(OUTPUT)util/config.o: util/config.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< -util/rbtree.o: ../../lib/rbtree.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/rbtree.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< +$(OUTPUT)util/rbtree.o: ../../lib/rbtree.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< # some perf warning policies can't fit to lib/bitmap.c, eg: it warns about variable shadowing # from <string.h> that comes from kernel headers wrapping. KBITMAP_FLAGS=`echo $(ALL_CFLAGS) | sed s/-Wshadow// | sed s/-Wswitch-default// | sed s/-Wextra//` -util/bitmap.o: ../../lib/bitmap.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/bitmap.o -c $(KBITMAP_FLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< +$(OUTPUT)util/bitmap.o: ../../lib/bitmap.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(KBITMAP_FLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< -util/hweight.o: ../../lib/hweight.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/hweight.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< +$(OUTPUT)util/hweight.o: ../../lib/hweight.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< -util/find_next_bit.o: ../../lib/find_next_bit.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/find_next_bit.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< +$(OUTPUT)util/find_next_bit.o: ../../lib/find_next_bit.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< -util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/scripting-engines/trace-event-perl.o -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< +$(OUTPUT)util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< -scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o scripts/perl/Perf-Trace-Util/Context.o -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< +$(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< -util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o util/scripting-engines/trace-event-python.o -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< +$(OUTPUT)util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< -scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c PERF-CFLAGS - $(QUIET_CC)$(CC) -o scripts/python/Perf-Trace-Util/Context.o -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< +$(OUTPUT)scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< -perf-%$X: %.o $(PERFLIBS) +$(OUTPUT)perf-%$X: %.o $(PERFLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) $(LIBS) $(LIB_OBJS) $(BUILTIN_OBJS): $(LIB_H) @@ -963,17 +986,17 @@ cscope: TRACK_CFLAGS = $(subst ','\'',$(ALL_CFLAGS)):\ $(bindir_SQ):$(perfexecdir_SQ):$(template_dir_SQ):$(prefix_SQ) -PERF-CFLAGS: .FORCE-PERF-CFLAGS +$(OUTPUT)PERF-CFLAGS: .FORCE-PERF-CFLAGS @FLAGS='$(TRACK_CFLAGS)'; \ - if test x"$$FLAGS" != x"`cat PERF-CFLAGS 2>/dev/null`" ; then \ + if test x"$$FLAGS" != x"`cat $(OUTPUT)PERF-CFLAGS 2>/dev/null`" ; then \ echo 1>&2 " * new build flags or prefix"; \ - echo "$$FLAGS" >PERF-CFLAGS; \ + echo "$$FLAGS" >$(OUTPUT)PERF-CFLAGS; \ fi # We need to apply sq twice, once to protect from the shell -# that runs PERF-BUILD-OPTIONS, and then again to protect it +# that runs $(OUTPUT)PERF-BUILD-OPTIONS, and then again to protect it # and the first level quoting from the shell that runs "echo". -PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS +$(OUTPUT)PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS @echo SHELL_PATH=\''$(subst ','\'',$(SHELL_PATH_SQ))'\' >$@ @echo TAR=\''$(subst ','\'',$(subst ','\'',$(TAR)))'\' >>$@ @echo NO_CURL=\''$(subst ','\'',$(subst ','\'',$(NO_CURL)))'\' >>$@ @@ -994,7 +1017,7 @@ all:: $(TEST_PROGRAMS) export NO_SVN_TESTS -check: common-cmds.h +check: $(OUTPUT)common-cmds.h if sparse; \ then \ for i in *.c */*.c; \ @@ -1028,10 +1051,10 @@ export perfexec_instdir install: all $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)' - $(INSTALL) perf$X '$(DESTDIR_SQ)$(bindir_SQ)' + $(INSTALL) $(OUTPUT)perf$X '$(DESTDIR_SQ)$(bindir_SQ)' $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace' $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin' - $(INSTALL) perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' + $(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' $(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace' $(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl' $(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin' @@ -1045,7 +1068,7 @@ ifdef BUILT_INS $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' $(INSTALL) $(BUILT_INS) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' ifneq (,$X) - $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';) + $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) $(OUTPUT)perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';) endif endif @@ -1129,14 +1152,14 @@ clean: $(RM) *.o */*.o */*/*.o */*/*/*.o $(LIB_FILE) $(RM) $(ALL_PROGRAMS) $(BUILT_INS) perf$X $(RM) $(TEST_PROGRAMS) - $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo common-cmds.h TAGS tags cscope* + $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope* $(RM) -r autom4te.cache $(RM) config.log config.mak.autogen config.mak.append config.status config.cache $(RM) -r $(PERF_TARNAME) .doc-tmp-dir $(RM) $(PERF_TARNAME).tar.gz perf-core_$(PERF_VERSION)-*.tar.gz $(RM) $(htmldocs).tar.gz $(manpages).tar.gz $(MAKE) -C Documentation/ clean - $(RM) PERF-VERSION-FILE PERF-CFLAGS PERF-BUILD-OPTIONS + $(RM) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)PERF-CFLAGS $(OUTPUT)PERF-BUILD-OPTIONS .PHONY: all install clean strip .PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell diff --git a/tools/perf/bench/mem-memcpy.c b/tools/perf/bench/mem-memcpy.c index 89773178e894..38dae7465142 100644 --- a/tools/perf/bench/mem-memcpy.c +++ b/tools/perf/bench/mem-memcpy.c @@ -10,7 +10,6 @@ #include "../perf.h" #include "../util/util.h" #include "../util/parse-options.h" -#include "../util/string.h" #include "../util/header.h" #include "bench.h" @@ -24,7 +23,7 @@ static const char *length_str = "1MB"; static const char *routine = "default"; -static int use_clock = 0; +static bool use_clock = false; static int clock_fd; static const struct option options[] = { diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c index 81cee78181fa..da1b2e9f01ff 100644 --- a/tools/perf/bench/sched-messaging.c +++ b/tools/perf/bench/sched-messaging.c @@ -31,9 +31,9 @@ #define DATASIZE 100 -static int use_pipes = 0; +static bool use_pipes = false; static unsigned int loops = 100; -static unsigned int thread_mode = 0; +static bool thread_mode = false; static unsigned int num_groups = 10; struct sender_context { diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c index 4f77c7c27640..d9ab3ce446ac 100644 --- a/tools/perf/bench/sched-pipe.c +++ b/tools/perf/bench/sched-pipe.c @@ -93,7 +93,7 @@ int bench_sched_pipe(int argc, const char **argv, switch (bench_format) { case BENCH_FORMAT_DEFAULT: - printf("# Extecuted %d pipe operations between two tasks\n\n", + printf("# Executed %d pipe operations between two tasks\n\n", loops); result_usec = diff.tv_sec * 1000000; diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 6ad7148451c5..06eaebe10d04 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -14,7 +14,6 @@ #include "util/cache.h" #include <linux/rbtree.h> #include "util/symbol.h" -#include "util/string.h" #include "perf.h" #include "util/debug.h" @@ -29,11 +28,11 @@ static char const *input_name = "perf.data"; -static int force; +static bool force; -static int full_paths; +static bool full_paths; -static int print_line; +static bool print_line; struct sym_hist { u64 sum; @@ -69,13 +68,13 @@ static int sym__alloc_hist(struct symbol *self) static int annotate__hist_hit(struct hist_entry *he, u64 ip) { unsigned int sym_size, offset; - struct symbol *sym = he->sym; + struct symbol *sym = he->ms.sym; struct sym_priv *priv; struct sym_hist *h; he->count++; - if (!sym || !he->map) + if (!sym || !he->ms.map) return 0; priv = symbol__priv(sym); @@ -85,7 +84,7 @@ static int annotate__hist_hit(struct hist_entry *he, u64 ip) sym_size = sym->end - sym->start; offset = ip - sym->start; - pr_debug3("%s: ip=%#Lx\n", __func__, he->map->unmap_ip(he->map, ip)); + pr_debug3("%s: ip=%#Lx\n", __func__, he->ms.map->unmap_ip(he->ms.map, ip)); if (offset >= sym_size) return 0; @@ -94,8 +93,8 @@ static int annotate__hist_hit(struct hist_entry *he, u64 ip) h->sum++; h->ip[offset]++; - pr_debug3("%#Lx %s: count++ [ip: %#Lx, %#Lx] => %Ld\n", he->sym->start, - he->sym->name, ip, ip - he->sym->start, h->ip[offset]); + pr_debug3("%#Lx %s: count++ [ip: %#Lx, %#Lx] => %Ld\n", he->ms.sym->start, + he->ms.sym->name, ip, ip - he->ms.sym->start, h->ip[offset]); return 0; } @@ -187,7 +186,7 @@ static struct objdump_line *objdump__get_next_ip_line(struct list_head *head, static int parse_line(FILE *file, struct hist_entry *he, struct list_head *head) { - struct symbol *sym = he->sym; + struct symbol *sym = he->ms.sym; struct objdump_line *objdump_line; char *line = NULL, *tmp, *tmp2; size_t line_len; @@ -226,7 +225,7 @@ static int parse_line(FILE *file, struct hist_entry *he, } if (line_ip != -1) { - u64 start = map__rip_2objdump(he->map, sym->start); + u64 start = map__rip_2objdump(he->ms.map, sym->start); offset = line_ip - start; } @@ -244,7 +243,7 @@ static int objdump_line__print(struct objdump_line *self, struct list_head *head, struct hist_entry *he, u64 len) { - struct symbol *sym = he->sym; + struct symbol *sym = he->ms.sym; static const char *prev_line; static const char *prev_color; @@ -327,7 +326,7 @@ static void insert_source_line(struct sym_ext *sym_ext) static void free_source_line(struct hist_entry *he, int len) { - struct sym_priv *priv = symbol__priv(he->sym); + struct sym_priv *priv = symbol__priv(he->ms.sym); struct sym_ext *sym_ext = priv->ext; int i; @@ -346,7 +345,7 @@ static void free_source_line(struct hist_entry *he, int len) static void get_source_line(struct hist_entry *he, int len, const char *filename) { - struct symbol *sym = he->sym; + struct symbol *sym = he->ms.sym; u64 start; int i; char cmd[PATH_MAX * 2]; @@ -361,7 +360,7 @@ get_source_line(struct hist_entry *he, int len, const char *filename) if (!priv->ext) return; - start = he->map->unmap_ip(he->map, sym->start); + start = he->ms.map->unmap_ip(he->ms.map, sym->start); for (i = 0; i < len; i++) { char *path = NULL; @@ -425,7 +424,7 @@ static void print_summary(const char *filename) static void hist_entry__print_hits(struct hist_entry *self) { - struct symbol *sym = self->sym; + struct symbol *sym = self->ms.sym; struct sym_priv *priv = symbol__priv(sym); struct sym_hist *h = priv->hist; u64 len = sym->end - sym->start, offset; @@ -439,9 +438,9 @@ static void hist_entry__print_hits(struct hist_entry *self) static void annotate_sym(struct hist_entry *he) { - struct map *map = he->map; + struct map *map = he->ms.map; struct dso *dso = map->dso; - struct symbol *sym = he->sym; + struct symbol *sym = he->ms.sym; const char *filename = dso->long_name, *d_filename; u64 len; char command[PATH_MAX*2]; @@ -452,6 +451,16 @@ static void annotate_sym(struct hist_entry *he) if (!filename) return; + if (dso->origin == DSO__ORIG_KERNEL) { + if (dso->annotate_warned) + return; + dso->annotate_warned = 1; + pr_err("Can't annotate %s: No vmlinux file was found in the " + "path:\n", sym->name); + vmlinux_path__fprintf(stderr); + return; + } + pr_debug("%s: filename=%s, sym=%s, start=%#Lx, end=%#Lx\n", __func__, filename, sym->name, map->unmap_ip(map, sym->start), map->unmap_ip(map, sym->end)); @@ -516,17 +525,17 @@ static void perf_session__find_annotations(struct perf_session *self) struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node); struct sym_priv *priv; - if (he->sym == NULL) + if (he->ms.sym == NULL) continue; - priv = symbol__priv(he->sym); + priv = symbol__priv(he->ms.sym); if (priv->hist == NULL) continue; annotate_sym(he); /* * Since we have a hist_entry per IP for the same symbol, free - * he->sym->hist to signal we already processed this symbol. + * he->ms.sym->hist to signal we already processed this symbol. */ free(priv->hist); priv->hist = NULL; @@ -581,10 +590,12 @@ static const char * const annotate_usage[] = { static const struct option options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", + "only consider symbols in these dsos"), OPT_STRING('s', "symbol", &sym_hist_filter, "symbol", "symbol to annotate"), OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c index 30a05f552c96..f8e3d1852029 100644 --- a/tools/perf/builtin-buildid-cache.c +++ b/tools/perf/builtin-buildid-cache.c @@ -27,7 +27,7 @@ static const struct option buildid_cache_options[] = { "file list", "file(s) to add"), OPT_STRING('r', "remove", &remove_name_list_str, "file list", "file(s) to remove"), - OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose"), + OPT_INCR('v', "verbose", &verbose, "be more verbose"), OPT_END() }; diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c index d0675c02f81e..af2ad8b92f76 100644 --- a/tools/perf/builtin-buildid-list.c +++ b/tools/perf/builtin-buildid-list.c @@ -16,7 +16,7 @@ #include "util/symbol.h" static char const *input_name = "perf.data"; -static int force; +static bool force; static bool with_hits; static const char * const buildid_list_usage[] = { @@ -29,7 +29,7 @@ static const struct option options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose"), OPT_END() }; diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c index 1ea15d8aeed1..3a1d94d75dce 100644 --- a/tools/perf/builtin-diff.c +++ b/tools/perf/builtin-diff.c @@ -19,7 +19,7 @@ static char const *input_old = "perf.data.old", *input_new = "perf.data"; static char diff__default_sort_order[] = "dso,symbol"; -static int force; +static bool force; static bool show_displacement; static int perf_session__add_hist_entry(struct perf_session *self, @@ -188,7 +188,7 @@ static const char * const diff_usage[] = { }; static const struct option options[] = { - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('m', "displacement", &show_displacement, "Show position displacement relative to baseline"), diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c index 215b584007b1..81e3ecc40fc7 100644 --- a/tools/perf/builtin-help.c +++ b/tools/perf/builtin-help.c @@ -29,7 +29,7 @@ enum help_format { HELP_FORMAT_WEB, }; -static int show_all = 0; +static bool show_all = false; static enum help_format help_format = HELP_FORMAT_MAN; static struct option builtin_help_options[] = { OPT_BOOLEAN('a', "all", &show_all, "print all available commands"), diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index 924a9518931a..513aa8a55db6 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -363,19 +363,21 @@ static void __print_result(struct rb_root *root, struct perf_session *session, struct alloc_stat *data = rb_entry(next, struct alloc_stat, node); struct symbol *sym = NULL; + struct map *map; char buf[BUFSIZ]; u64 addr; if (is_caller) { addr = data->call_site; if (!raw_ip) - sym = map_groups__find_function(&session->kmaps, addr, NULL); + sym = map_groups__find_function(&session->kmaps, + addr, &map, NULL); } else addr = data->ptr; if (sym != NULL) snprintf(buf, sizeof(buf), "%s+%Lx", sym->name, - addr - sym->start); + addr - map->unmap_ip(map, sym->start)); else snprintf(buf, sizeof(buf), "%#Lx", addr); printf(" %-34s |", buf); @@ -488,6 +490,9 @@ static int __cmd_kmem(void) if (session == NULL) return -ENOMEM; + if (perf_session__create_kernel_maps(session) < 0) + goto out_delete; + if (!perf_session__has_traces(session, "kmem record")) goto out_delete; diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c index e12c844df1e2..6c38e4febf9f 100644 --- a/tools/perf/builtin-lock.c +++ b/tools/perf/builtin-lock.c @@ -744,7 +744,7 @@ static const char * const lock_usage[] = { static const struct option lock_options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), - OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), OPT_END() }; diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c index 152d6c9b1fa4..c1e54035e8cf 100644 --- a/tools/perf/builtin-probe.c +++ b/tools/perf/builtin-probe.c @@ -36,13 +36,10 @@ #include "builtin.h" #include "util/util.h" #include "util/strlist.h" -#include "util/event.h" +#include "util/symbol.h" #include "util/debug.h" #include "util/debugfs.h" -#include "util/symbol.h" -#include "util/thread.h" #include "util/parse-options.h" -#include "util/parse-events.h" /* For debugfs_path */ #include "util/probe-finder.h" #include "util/probe-event.h" @@ -50,103 +47,83 @@ /* Session management structure */ static struct { - bool need_dwarf; bool list_events; bool force_add; bool show_lines; - int nr_probe; - struct probe_point probes[MAX_PROBES]; + int nevents; + struct perf_probe_event events[MAX_PROBES]; struct strlist *dellist; - struct map_groups kmap_groups; - struct map *kmaps[MAP__NR_TYPES]; struct line_range line_range; -} session; +} params; /* Parse an event definition. Note that any error must die. */ -static void parse_probe_event(const char *str) +static int parse_probe_event(const char *str) { - struct probe_point *pp = &session.probes[session.nr_probe]; + struct perf_probe_event *pev = ¶ms.events[params.nevents]; + int ret; - pr_debug("probe-definition(%d): %s\n", session.nr_probe, str); - if (++session.nr_probe == MAX_PROBES) + pr_debug("probe-definition(%d): %s\n", params.nevents, str); + if (++params.nevents == MAX_PROBES) die("Too many probes (> %d) are specified.", MAX_PROBES); - /* Parse perf-probe event into probe_point */ - parse_perf_probe_event(str, pp, &session.need_dwarf); + /* Parse a perf-probe command into event */ + ret = parse_perf_probe_command(str, pev); + pr_debug("%d arguments\n", pev->nargs); - pr_debug("%d arguments\n", pp->nr_args); + return ret; } -static void parse_probe_event_argv(int argc, const char **argv) +static int parse_probe_event_argv(int argc, const char **argv) { - int i, len; + int i, len, ret; char *buf; /* Bind up rest arguments */ len = 0; for (i = 0; i < argc; i++) len += strlen(argv[i]) + 1; - buf = zalloc(len + 1); - if (!buf) - die("Failed to allocate memory for binding arguments."); + buf = xzalloc(len + 1); len = 0; for (i = 0; i < argc; i++) len += sprintf(&buf[len], "%s ", argv[i]); - parse_probe_event(buf); + ret = parse_probe_event(buf); free(buf); + return ret; } static int opt_add_probe_event(const struct option *opt __used, const char *str, int unset __used) { if (str) - parse_probe_event(str); - return 0; + return parse_probe_event(str); + else + return 0; } static int opt_del_probe_event(const struct option *opt __used, const char *str, int unset __used) { if (str) { - if (!session.dellist) - session.dellist = strlist__new(true, NULL); - strlist__add(session.dellist, str); + if (!params.dellist) + params.dellist = strlist__new(true, NULL); + strlist__add(params.dellist, str); } return 0; } -/* Currently just checking function name from symbol map */ -static void evaluate_probe_point(struct probe_point *pp) -{ - struct symbol *sym; - sym = map__find_symbol_by_name(session.kmaps[MAP__FUNCTION], - pp->function, NULL); - if (!sym) - die("Kernel symbol \'%s\' not found - probe not added.", - pp->function); -} - -#ifndef NO_DWARF_SUPPORT -static int open_vmlinux(void) -{ - if (map__load(session.kmaps[MAP__FUNCTION], NULL) < 0) { - pr_debug("Failed to load kernel map.\n"); - return -EINVAL; - } - pr_debug("Try to open %s\n", - session.kmaps[MAP__FUNCTION]->dso->long_name); - return open(session.kmaps[MAP__FUNCTION]->dso->long_name, O_RDONLY); -} - +#ifdef DWARF_SUPPORT static int opt_show_lines(const struct option *opt __used, const char *str, int unset __used) { + int ret = 0; + if (str) - parse_line_range_desc(str, &session.line_range); - INIT_LIST_HEAD(&session.line_range.line_list); - session.show_lines = true; - return 0; + ret = parse_line_range_desc(str, ¶ms.line_range); + INIT_LIST_HEAD(¶ms.line_range.line_list); + params.show_lines = true; + + return ret; } #endif @@ -155,29 +132,25 @@ static const char * const probe_usage[] = { "perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]", "perf probe [<options>] --del '[GROUP:]EVENT' ...", "perf probe --list", -#ifndef NO_DWARF_SUPPORT +#ifdef DWARF_SUPPORT "perf probe --line 'LINEDESC'", #endif NULL }; static const struct option options[] = { - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show parsed arguments, etc)"), -#ifndef NO_DWARF_SUPPORT - OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, - "file", "vmlinux pathname"), -#endif - OPT_BOOLEAN('l', "list", &session.list_events, + OPT_BOOLEAN('l', "list", ¶ms.list_events, "list up current probe events"), OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.", opt_del_probe_event), OPT_CALLBACK('a', "add", NULL, -#ifdef NO_DWARF_SUPPORT - "[EVENT=]FUNC[+OFF|%return] [ARG ...]", -#else +#ifdef DWARF_SUPPORT "[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT" - " [ARG ...]", + " [[NAME=]ARG ...]", +#else + "[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]", #endif "probe point definition, where\n" "\t\tGROUP:\tGroup name (optional)\n" @@ -185,51 +158,33 @@ static const struct option options[] = { "\t\tFUNC:\tFunction name\n" "\t\tOFF:\tOffset from function entry (in byte)\n" "\t\t%return:\tPut the probe at function return\n" -#ifdef NO_DWARF_SUPPORT - "\t\tARG:\tProbe argument (only \n" -#else +#ifdef DWARF_SUPPORT "\t\tSRC:\tSource code path\n" "\t\tRL:\tRelative line number from function entry.\n" "\t\tAL:\tAbsolute line number in file.\n" "\t\tPT:\tLazy expression of line code.\n" "\t\tARG:\tProbe argument (local variable name or\n" -#endif "\t\t\tkprobe-tracer argument format.)\n", +#else + "\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n", +#endif opt_add_probe_event), - OPT_BOOLEAN('f', "force", &session.force_add, "forcibly add events" + OPT_BOOLEAN('f', "force", ¶ms.force_add, "forcibly add events" " with existing name"), -#ifndef NO_DWARF_SUPPORT +#ifdef DWARF_SUPPORT OPT_CALLBACK('L', "line", NULL, - "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]", + "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]", "Show source code lines.", opt_show_lines), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), #endif + OPT__DRY_RUN(&probe_event_dry_run), OPT_END() }; -/* Initialize symbol maps for vmlinux */ -static void init_vmlinux(void) -{ - symbol_conf.sort_by_name = true; - if (symbol_conf.vmlinux_name == NULL) - symbol_conf.try_vmlinux_path = true; - else - pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name); - if (symbol__init() < 0) - die("Failed to init symbol map."); - - map_groups__init(&session.kmap_groups); - if (map_groups__create_kernel_maps(&session.kmap_groups, - session.kmaps) < 0) - die("Failed to create kernel maps."); -} - int cmd_probe(int argc, const char **argv, const char *prefix __used) { - int i, ret; -#ifndef NO_DWARF_SUPPORT - int fd; -#endif - struct probe_point *pp; + int ret; argc = parse_options(argc, argv, options, probe_usage, PARSE_OPT_STOP_AT_NON_OPTION); @@ -238,123 +193,65 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) pr_warning(" Error: '-' is not supported.\n"); usage_with_options(probe_usage, options); } - parse_probe_event_argv(argc, argv); + ret = parse_probe_event_argv(argc, argv); + if (ret < 0) { + pr_err(" Error: Parse Error. (%d)\n", ret); + return ret; + } } - if ((!session.nr_probe && !session.dellist && !session.list_events && - !session.show_lines)) + if ((!params.nevents && !params.dellist && !params.list_events && + !params.show_lines)) usage_with_options(probe_usage, options); - if (debugfs_valid_mountpoint(debugfs_path) < 0) - die("Failed to find debugfs path."); - - if (session.list_events) { - if (session.nr_probe != 0 || session.dellist) { - pr_warning(" Error: Don't use --list with" - " --add/--del.\n"); + if (params.list_events) { + if (params.nevents != 0 || params.dellist) { + pr_err(" Error: Don't use --list with --add/--del.\n"); usage_with_options(probe_usage, options); } - if (session.show_lines) { - pr_warning(" Error: Don't use --list with --line.\n"); + if (params.show_lines) { + pr_err(" Error: Don't use --list with --line.\n"); usage_with_options(probe_usage, options); } - show_perf_probe_events(); - return 0; + ret = show_perf_probe_events(); + if (ret < 0) + pr_err(" Error: Failed to show event list. (%d)\n", + ret); + return ret; } -#ifndef NO_DWARF_SUPPORT - if (session.show_lines) { - if (session.nr_probe != 0 || session.dellist) { +#ifdef DWARF_SUPPORT + if (params.show_lines) { + if (params.nevents != 0 || params.dellist) { pr_warning(" Error: Don't use --line with" " --add/--del.\n"); usage_with_options(probe_usage, options); } - init_vmlinux(); - fd = open_vmlinux(); - if (fd < 0) - die("Could not open debuginfo file."); - ret = find_line_range(fd, &session.line_range); - if (ret <= 0) - die("Source line is not found.\n"); - close(fd); - show_line_range(&session.line_range); - return 0; - } -#endif - if (session.dellist) { - del_trace_kprobe_events(session.dellist); - strlist__delete(session.dellist); - if (session.nr_probe == 0) - return 0; + ret = show_line_range(¶ms.line_range); + if (ret < 0) + pr_err(" Error: Failed to show lines. (%d)\n", ret); + return ret; } +#endif - /* Add probes */ - init_vmlinux(); - - if (session.need_dwarf) -#ifdef NO_DWARF_SUPPORT - die("Debuginfo-analysis is not supported"); -#else /* !NO_DWARF_SUPPORT */ - pr_debug("Some probes require debuginfo.\n"); - - fd = open_vmlinux(); - if (fd < 0) { - if (session.need_dwarf) - die("Could not open debuginfo file."); - - pr_debug("Could not open vmlinux/module file." - " Try to use symbols.\n"); - goto end_dwarf; - } - - /* Searching probe points */ - for (i = 0; i < session.nr_probe; i++) { - pp = &session.probes[i]; - if (pp->found) - continue; - - lseek(fd, SEEK_SET, 0); - ret = find_probe_point(fd, pp); - if (ret > 0) - continue; - if (ret == 0) { /* No error but failed to find probe point. */ - synthesize_perf_probe_point(pp); - die("Probe point '%s' not found. - probe not added.", - pp->probes[0]); - } - /* Error path */ - if (session.need_dwarf) { - if (ret == -ENOENT) - pr_warning("No dwarf info found in the vmlinux - please rebuild with CONFIG_DEBUG_INFO=y.\n"); - die("Could not analyze debuginfo."); + if (params.dellist) { + ret = del_perf_probe_events(params.dellist); + strlist__delete(params.dellist); + if (ret < 0) { + pr_err(" Error: Failed to delete events. (%d)\n", ret); + return ret; } - pr_debug("An error occurred in debuginfo analysis." - " Try to use symbols.\n"); - break; } - close(fd); - -end_dwarf: -#endif /* !NO_DWARF_SUPPORT */ - /* Synthesize probes without dwarf */ - for (i = 0; i < session.nr_probe; i++) { - pp = &session.probes[i]; - if (pp->found) /* This probe is already found. */ - continue; - - evaluate_probe_point(pp); - ret = synthesize_trace_kprobe_event(pp); - if (ret == -E2BIG) - die("probe point definition becomes too long."); - else if (ret < 0) - die("Failed to synthesize a probe point."); + if (params.nevents) { + ret = add_perf_probe_events(params.events, params.nevents, + params.force_add); + if (ret < 0) { + pr_err(" Error: Failed to add events. (%d)\n", ret); + return ret; + } } - - /* Settng up probe points */ - add_trace_kprobe_events(session.probes, session.nr_probe, - session.force_add); return 0; } diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 3b8b6387c47c..a1b99eeac3c0 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -15,7 +15,6 @@ #include "util/util.h" #include "util/parse-options.h" #include "util/parse-events.h" -#include "util/string.h" #include "util/header.h" #include "util/event.h" @@ -27,31 +26,41 @@ #include <unistd.h> #include <sched.h> -static int fd[MAX_NR_CPUS][MAX_COUNTERS]; +enum write_mode_t { + WRITE_FORCE, + WRITE_APPEND +}; + +static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; +static unsigned int user_interval = UINT_MAX; static long default_interval = 0; static int nr_cpus = 0; static unsigned int page_size; static unsigned int mmap_pages = 128; +static unsigned int user_freq = UINT_MAX; static int freq = 1000; static int output; +static int pipe_output = 0; static const char *output_name = "perf.data"; static int group = 0; static unsigned int realtime_prio = 0; -static int raw_samples = 0; -static int system_wide = 0; +static bool raw_samples = false; +static bool system_wide = false; static int profile_cpu = -1; static pid_t target_pid = -1; +static pid_t target_tid = -1; +static pid_t *all_tids = NULL; +static int thread_num = 0; static pid_t child_pid = -1; -static int inherit = 1; -static int force = 0; -static int append_file = 0; -static int call_graph = 0; -static int inherit_stat = 0; -static int no_samples = 0; -static int sample_address = 0; -static int multiplex = 0; +static bool inherit = true; +static enum write_mode_t write_mode = WRITE_FORCE; +static bool call_graph = false; +static bool inherit_stat = false; +static bool no_samples = false; +static bool sample_address = false; +static bool multiplex = false; static int multiplex_fd = -1; static long samples = 0; @@ -60,7 +69,7 @@ static struct timeval this_read; static u64 bytes_written = 0; -static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS]; +static struct pollfd *event_array; static int nr_poll = 0; static int nr_cpu = 0; @@ -77,7 +86,7 @@ struct mmap_data { unsigned int prev; }; -static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS]; +static struct mmap_data *mmap_array[MAX_NR_CPUS][MAX_COUNTERS]; static unsigned long mmap_read_head(struct mmap_data *md) { @@ -101,6 +110,11 @@ static void mmap_write_tail(struct mmap_data *md, unsigned long tail) pc->data_tail = tail; } +static void advance_output(size_t size) +{ + bytes_written += size; +} + static void write_output(void *buf, size_t size) { while (size) { @@ -225,12 +239,13 @@ static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int n return h_attr; } -static void create_counter(int counter, int cpu, pid_t pid) +static void create_counter(int counter, int cpu) { char *filter = filters[counter]; struct perf_event_attr *attr = attrs + counter; struct perf_header_attr *h_attr; int track = !counter; /* only the first counter needs these */ + int thread_index; int ret; struct { u64 count; @@ -248,10 +263,19 @@ static void create_counter(int counter, int cpu, pid_t pid) if (nr_counters > 1) attr->sample_type |= PERF_SAMPLE_ID; - if (freq) { - attr->sample_type |= PERF_SAMPLE_PERIOD; - attr->freq = 1; - attr->sample_freq = freq; + /* + * We default some events to a 1 default interval. But keep + * it a weak assumption overridable by the user. + */ + if (!attr->sample_period || (user_freq != UINT_MAX && + user_interval != UINT_MAX)) { + if (freq) { + attr->sample_type |= PERF_SAMPLE_PERIOD; + attr->freq = 1; + attr->sample_freq = freq; + } else { + attr->sample_period = default_interval; + } } if (no_samples) @@ -275,118 +299,129 @@ static void create_counter(int counter, int cpu, pid_t pid) attr->mmap = track; attr->comm = track; attr->inherit = inherit; - attr->disabled = 1; + if (target_pid == -1 && !system_wide) { + attr->disabled = 1; + attr->enable_on_exec = 1; + } + for (thread_index = 0; thread_index < thread_num; thread_index++) { try_again: - fd[nr_cpu][counter] = sys_perf_event_open(attr, pid, cpu, group_fd, 0); - - if (fd[nr_cpu][counter] < 0) { - int err = errno; - - if (err == EPERM || err == EACCES) - die("Permission error - are you root?\n"); - else if (err == ENODEV && profile_cpu != -1) - die("No such device - did you specify an out-of-range profile CPU?\n"); + fd[nr_cpu][counter][thread_index] = sys_perf_event_open(attr, + all_tids[thread_index], cpu, group_fd, 0); + + if (fd[nr_cpu][counter][thread_index] < 0) { + int err = errno; + + if (err == EPERM || err == EACCES) + die("Permission error - are you root?\n" + "\t Consider tweaking" + " /proc/sys/kernel/perf_event_paranoid.\n"); + else if (err == ENODEV && profile_cpu != -1) { + die("No such device - did you specify" + " an out-of-range profile CPU?\n"); + } - /* - * If it's cycles then fall back to hrtimer - * based cpu-clock-tick sw counter, which - * is always available even if no PMU support: - */ - if (attr->type == PERF_TYPE_HARDWARE - && attr->config == PERF_COUNT_HW_CPU_CYCLES) { - - if (verbose) - warning(" ... trying to fall back to cpu-clock-ticks\n"); - attr->type = PERF_TYPE_SOFTWARE; - attr->config = PERF_COUNT_SW_CPU_CLOCK; - goto try_again; - } - printf("\n"); - error("perfcounter syscall returned with %d (%s)\n", - fd[nr_cpu][counter], strerror(err)); + /* + * If it's cycles then fall back to hrtimer + * based cpu-clock-tick sw counter, which + * is always available even if no PMU support: + */ + if (attr->type == PERF_TYPE_HARDWARE + && attr->config == PERF_COUNT_HW_CPU_CYCLES) { + + if (verbose) + warning(" ... trying to fall back to cpu-clock-ticks\n"); + attr->type = PERF_TYPE_SOFTWARE; + attr->config = PERF_COUNT_SW_CPU_CLOCK; + goto try_again; + } + printf("\n"); + error("perfcounter syscall returned with %d (%s)\n", + fd[nr_cpu][counter][thread_index], strerror(err)); #if defined(__i386__) || defined(__x86_64__) - if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP) - die("No hardware sampling interrupt available. No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it.\n"); + if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP) + die("No hardware sampling interrupt available." + " No APIC? If so then you can boot the kernel" + " with the \"lapic\" boot parameter to" + " force-enable it.\n"); #endif - die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); - exit(-1); - } + die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); + exit(-1); + } - h_attr = get_header_attr(attr, counter); - if (h_attr == NULL) - die("nomem\n"); + h_attr = get_header_attr(attr, counter); + if (h_attr == NULL) + die("nomem\n"); - if (!file_new) { - if (memcmp(&h_attr->attr, attr, sizeof(*attr))) { - fprintf(stderr, "incompatible append\n"); - exit(-1); + if (!file_new) { + if (memcmp(&h_attr->attr, attr, sizeof(*attr))) { + fprintf(stderr, "incompatible append\n"); + exit(-1); + } } - } - if (read(fd[nr_cpu][counter], &read_data, sizeof(read_data)) == -1) { - perror("Unable to read perf file descriptor\n"); - exit(-1); - } + if (read(fd[nr_cpu][counter][thread_index], &read_data, sizeof(read_data)) == -1) { + perror("Unable to read perf file descriptor\n"); + exit(-1); + } - if (perf_header_attr__add_id(h_attr, read_data.id) < 0) { - pr_warning("Not enough memory to add id\n"); - exit(-1); - } + if (perf_header_attr__add_id(h_attr, read_data.id) < 0) { + pr_warning("Not enough memory to add id\n"); + exit(-1); + } - assert(fd[nr_cpu][counter] >= 0); - fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK); + assert(fd[nr_cpu][counter][thread_index] >= 0); + fcntl(fd[nr_cpu][counter][thread_index], F_SETFL, O_NONBLOCK); - /* - * First counter acts as the group leader: - */ - if (group && group_fd == -1) - group_fd = fd[nr_cpu][counter]; - if (multiplex && multiplex_fd == -1) - multiplex_fd = fd[nr_cpu][counter]; + /* + * First counter acts as the group leader: + */ + if (group && group_fd == -1) + group_fd = fd[nr_cpu][counter][thread_index]; + if (multiplex && multiplex_fd == -1) + multiplex_fd = fd[nr_cpu][counter][thread_index]; - if (multiplex && fd[nr_cpu][counter] != multiplex_fd) { + if (multiplex && fd[nr_cpu][counter][thread_index] != multiplex_fd) { - ret = ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd); - assert(ret != -1); - } else { - event_array[nr_poll].fd = fd[nr_cpu][counter]; - event_array[nr_poll].events = POLLIN; - nr_poll++; - - mmap_array[nr_cpu][counter].counter = counter; - mmap_array[nr_cpu][counter].prev = 0; - mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1; - mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size, - PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0); - if (mmap_array[nr_cpu][counter].base == MAP_FAILED) { - error("failed to mmap with %d (%s)\n", errno, strerror(errno)); - exit(-1); + ret = ioctl(fd[nr_cpu][counter][thread_index], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd); + assert(ret != -1); + } else { + event_array[nr_poll].fd = fd[nr_cpu][counter][thread_index]; + event_array[nr_poll].events = POLLIN; + nr_poll++; + + mmap_array[nr_cpu][counter][thread_index].counter = counter; + mmap_array[nr_cpu][counter][thread_index].prev = 0; + mmap_array[nr_cpu][counter][thread_index].mask = mmap_pages*page_size - 1; + mmap_array[nr_cpu][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size, + PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter][thread_index], 0); + if (mmap_array[nr_cpu][counter][thread_index].base == MAP_FAILED) { + error("failed to mmap with %d (%s)\n", errno, strerror(errno)); + exit(-1); + } } - } - if (filter != NULL) { - ret = ioctl(fd[nr_cpu][counter], - PERF_EVENT_IOC_SET_FILTER, filter); - if (ret) { - error("failed to set filter with %d (%s)\n", errno, - strerror(errno)); - exit(-1); + if (filter != NULL) { + ret = ioctl(fd[nr_cpu][counter][thread_index], + PERF_EVENT_IOC_SET_FILTER, filter); + if (ret) { + error("failed to set filter with %d (%s)\n", errno, + strerror(errno)); + exit(-1); + } } } - - ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_ENABLE); } -static void open_counters(int cpu, pid_t pid) +static void open_counters(int cpu) { int counter; group_fd = -1; for (counter = 0; counter < nr_counters; counter++) - create_counter(counter, cpu, pid); + create_counter(counter, cpu); nr_cpu++; } @@ -406,10 +441,19 @@ static int process_buildids(void) static void atexit_header(void) { - session->header.data_size += bytes_written; + if (!pipe_output) { + session->header.data_size += bytes_written; - process_buildids(); - perf_header__write(&session->header, output, true); + process_buildids(); + perf_header__write(&session->header, output, true); + } else { + int err; + + err = event__synthesize_build_ids(process_synthesized_event, + session); + if (err < 0) + pr_err("Couldn't synthesize build ids.\n"); + } } static int __cmd_record(int argc, const char **argv) @@ -421,7 +465,7 @@ static int __cmd_record(int argc, const char **argv) int err; unsigned long waking = 0; int child_ready_pipe[2], go_pipe[2]; - const bool forks = target_pid == -1 && argc > 0; + const bool forks = argc > 0; char buf; page_size = sysconf(_SC_PAGE_SIZE); @@ -435,45 +479,44 @@ static int __cmd_record(int argc, const char **argv) exit(-1); } - if (!stat(output_name, &st) && st.st_size) { - if (!force) { - if (!append_file) { - pr_err("Error, output file %s exists, use -A " - "to append or -f to overwrite.\n", - output_name); - exit(-1); - } - } else { + if (!strcmp(output_name, "-")) + pipe_output = 1; + else if (!stat(output_name, &st) && st.st_size) { + if (write_mode == WRITE_FORCE) { char oldname[PATH_MAX]; snprintf(oldname, sizeof(oldname), "%s.old", output_name); unlink(oldname); rename(output_name, oldname); } - } else { - append_file = 0; + } else if (write_mode == WRITE_APPEND) { + write_mode = WRITE_FORCE; } flags = O_CREAT|O_RDWR; - if (append_file) + if (write_mode == WRITE_APPEND) file_new = 0; else flags |= O_TRUNC; - output = open(output_name, flags, S_IRUSR|S_IWUSR); + if (pipe_output) + output = STDOUT_FILENO; + else + output = open(output_name, flags, S_IRUSR | S_IWUSR); if (output < 0) { perror("failed to create output file"); exit(-1); } - session = perf_session__new(output_name, O_WRONLY, force); + session = perf_session__new(output_name, O_WRONLY, + write_mode == WRITE_FORCE); if (session == NULL) { pr_err("Not enough memory for reading perf file header\n"); return -1; } if (!file_new) { - err = perf_header__read(&session->header, output); + err = perf_header__read(session, output); if (err < 0) return err; } @@ -492,13 +535,15 @@ static int __cmd_record(int argc, const char **argv) atexit(atexit_header); if (forks) { - pid = fork(); + child_pid = fork(); if (pid < 0) { perror("failed to fork"); exit(-1); } - if (!pid) { + if (!child_pid) { + if (pipe_output) + dup2(2, 1); close(child_ready_pipe[0]); close(go_pipe[1]); fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC); @@ -527,10 +572,8 @@ static int __cmd_record(int argc, const char **argv) exit(-1); } - child_pid = pid; - - if (!system_wide) - target_pid = pid; + if (!system_wide && target_tid == -1 && target_pid == -1) + all_tids[0] = child_pid; close(child_ready_pipe[1]); close(go_pipe[0]); @@ -544,16 +587,19 @@ static int __cmd_record(int argc, const char **argv) close(child_ready_pipe[0]); } - if ((!system_wide && !inherit) || profile_cpu != -1) { - open_counters(profile_cpu, target_pid); + open_counters(profile_cpu); } else { nr_cpus = read_cpu_map(); for (i = 0; i < nr_cpus; i++) - open_counters(cpumap[i], target_pid); + open_counters(cpumap[i]); } - if (file_new) { + if (pipe_output) { + err = perf_header__write_pipe(output); + if (err < 0) + return err; + } else if (file_new) { err = perf_header__write(&session->header, output, false); if (err < 0) return err; @@ -561,8 +607,39 @@ static int __cmd_record(int argc, const char **argv) post_processing_offset = lseek(output, 0, SEEK_CUR); + if (pipe_output) { + err = event__synthesize_attrs(&session->header, + process_synthesized_event, + session); + if (err < 0) { + pr_err("Couldn't synthesize attrs.\n"); + return err; + } + + err = event__synthesize_event_types(process_synthesized_event, + session); + if (err < 0) { + pr_err("Couldn't synthesize event_types.\n"); + return err; + } + + err = event__synthesize_tracing_data(output, attrs, + nr_counters, + process_synthesized_event, + session); + if (err <= 0) { + pr_err("Couldn't record tracing data.\n"); + return err; + } + + advance_output(err); + } + err = event__synthesize_kernel_mmap(process_synthesized_event, session, "_text"); + if (err < 0) + err = event__synthesize_kernel_mmap(process_synthesized_event, + session, "_stext"); if (err < 0) { pr_err("Couldn't record kernel reference relocation symbol.\n"); return err; @@ -575,7 +652,7 @@ static int __cmd_record(int argc, const char **argv) } if (!system_wide && profile_cpu == -1) - event__synthesize_thread(target_pid, process_synthesized_event, + event__synthesize_thread(target_tid, process_synthesized_event, session); else event__synthesize_threads(process_synthesized_event, session); @@ -598,11 +675,16 @@ static int __cmd_record(int argc, const char **argv) for (;;) { int hits = samples; + int thread; for (i = 0; i < nr_cpu; i++) { for (counter = 0; counter < nr_counters; counter++) { - if (mmap_array[i][counter].base) - mmap_read(&mmap_array[i][counter]); + for (thread = 0; + thread < thread_num; thread++) { + if (mmap_array[i][counter][thread].base) + mmap_read(&mmap_array[i][counter][thread]); + } + } } @@ -615,8 +697,15 @@ static int __cmd_record(int argc, const char **argv) if (done) { for (i = 0; i < nr_cpu; i++) { - for (counter = 0; counter < nr_counters; counter++) - ioctl(fd[i][counter], PERF_EVENT_IOC_DISABLE); + for (counter = 0; + counter < nr_counters; + counter++) { + for (thread = 0; + thread < thread_num; + thread++) + ioctl(fd[i][counter][thread], + PERF_EVENT_IOC_DISABLE); + } } } } @@ -641,6 +730,8 @@ static const char * const record_usage[] = { NULL }; +static bool force, append_file; + static const struct option options[] = { OPT_CALLBACK('e', "event", NULL, "event", "event selector. use 'perf list' to list available events", @@ -648,7 +739,9 @@ static const struct option options[] = { OPT_CALLBACK(0, "filter", NULL, "filter", "event filter", parse_filter), OPT_INTEGER('p', "pid", &target_pid, - "record events on existing pid"), + "record events on existing process id"), + OPT_INTEGER('t', "tid", &target_tid, + "record events on existing thread id"), OPT_INTEGER('r', "realtime", &realtime_prio, "collect data with this RT SCHED_FIFO priority"), OPT_BOOLEAN('R', "raw-samples", &raw_samples, @@ -660,20 +753,20 @@ static const struct option options[] = { OPT_INTEGER('C', "profile_cpu", &profile_cpu, "CPU to profile on"), OPT_BOOLEAN('f', "force", &force, - "overwrite existing data file"), - OPT_LONG('c', "count", &default_interval, + "overwrite existing data file (deprecated)"), + OPT_LONG('c', "count", &user_interval, "event period to sample"), OPT_STRING('o', "output", &output_name, "file", "output file name"), OPT_BOOLEAN('i', "inherit", &inherit, "child tasks inherit counters"), - OPT_INTEGER('F', "freq", &freq, + OPT_INTEGER('F', "freq", &user_freq, "profile at this frequency"), OPT_INTEGER('m', "mmap-pages", &mmap_pages, "number of mmap data pages"), OPT_BOOLEAN('g', "call-graph", &call_graph, "do call-graph (stack chain/backtrace) recording"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show counter open errors, etc)"), OPT_BOOLEAN('s', "stat", &inherit_stat, "per thread counts"), @@ -688,12 +781,23 @@ static const struct option options[] = { int cmd_record(int argc, const char **argv, const char *prefix __used) { - int counter; + int i,j; argc = parse_options(argc, argv, options, record_usage, PARSE_OPT_STOP_AT_NON_OPTION); - if (!argc && target_pid == -1 && !system_wide && profile_cpu == -1) + if (!argc && target_pid == -1 && target_tid == -1 && + !system_wide && profile_cpu == -1) + usage_with_options(record_usage, options); + + if (force && append_file) { + fprintf(stderr, "Can't overwrite and append at the same time." + " You need to choose between -f and -A"); usage_with_options(record_usage, options); + } else if (append_file) { + write_mode = WRITE_APPEND; + } else { + write_mode = WRITE_FORCE; + } symbol__init(); @@ -703,6 +807,42 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) attrs[0].config = PERF_COUNT_HW_CPU_CYCLES; } + if (target_pid != -1) { + target_tid = target_pid; + thread_num = find_all_tid(target_pid, &all_tids); + if (thread_num <= 0) { + fprintf(stderr, "Can't find all threads of pid %d\n", + target_pid); + usage_with_options(record_usage, options); + } + } else { + all_tids=malloc(sizeof(pid_t)); + if (!all_tids) + return -ENOMEM; + + all_tids[0] = target_tid; + thread_num = 1; + } + + for (i = 0; i < MAX_NR_CPUS; i++) { + for (j = 0; j < MAX_COUNTERS; j++) { + fd[i][j] = malloc(sizeof(int)*thread_num); + mmap_array[i][j] = zalloc( + sizeof(struct mmap_data)*thread_num); + if (!fd[i][j] || !mmap_array[i][j]) + return -ENOMEM; + } + } + event_array = malloc( + sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num); + if (!event_array) + return -ENOMEM; + + if (user_interval != UINT_MAX) + default_interval = user_interval; + if (user_freq != UINT_MAX) + freq = user_freq; + /* * User specified count overrides default frequency. */ @@ -715,12 +855,5 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) exit(EXIT_FAILURE); } - for (counter = 0; counter < nr_counters; counter++) { - if (attrs[counter].sample_period) - continue; - - attrs[counter].sample_period = default_interval; - } - return __cmd_record(argc, argv); } diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index f815de25d0fc..7da5fb365264 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -14,7 +14,6 @@ #include "util/cache.h" #include <linux/rbtree.h> #include "util/symbol.h" -#include "util/string.h" #include "util/callchain.h" #include "util/strlist.h" #include "util/values.h" @@ -33,11 +32,11 @@ static char const *input_name = "perf.data"; -static int force; +static bool force; static bool hide_unresolved; static bool dont_use_callchains; -static int show_threads; +static bool show_threads; static struct perf_read_values show_threads_values; static char default_pretty_printing_style[] = "normal"; @@ -81,15 +80,20 @@ static int perf_session__add_hist_entry(struct perf_session *self, struct addr_location *al, struct sample_data *data) { - struct symbol **syms = NULL, *parent = NULL; + struct map_symbol *syms = NULL; + struct symbol *parent = NULL; bool hit; + int err; struct hist_entry *he; struct event_stat_id *stats; struct perf_event_attr *attr; - if ((sort__has_parent || symbol_conf.use_callchain) && data->callchain) + if ((sort__has_parent || symbol_conf.use_callchain) && data->callchain) { syms = perf_session__resolve_callchain(self, al->thread, data->callchain, &parent); + if (syms == NULL) + return -ENOMEM; + } attr = perf_header__find_attr(data->id, &self->header); if (attr) @@ -108,9 +112,12 @@ static int perf_session__add_hist_entry(struct perf_session *self, if (symbol_conf.use_callchain) { if (!hit) - callchain_init(&he->callchain); - append_chain(&he->callchain, data->callchain, syms); + callchain_init(he->callchain); + err = append_chain(he->callchain, data->callchain, syms); free(syms); + + if (err) + return err; } return 0; @@ -260,13 +267,27 @@ static struct perf_event_ops event_ops = { .fork = event__process_task, .lost = event__process_lost, .read = process_read_event, + .attr = event__process_attr, + .event_type = event__process_event_type, + .tracing_data = event__process_tracing_data, + .build_id = event__process_build_id, }; +extern volatile int session_done; + +static void sig_handler(int sig __attribute__((__unused__))) +{ + session_done = 1; +} + static int __cmd_report(void) { int ret = -EINVAL; struct perf_session *session; struct rb_node *next; + const char *help = "For a higher level overview, try: perf report --sort comm,dso"; + + signal(SIGINT, sig_handler); session = perf_session__new(input_name, O_RDONLY, force); if (session == NULL) @@ -297,34 +318,44 @@ static int __cmd_report(void) next = rb_first(&session->stats_by_id); while (next) { struct event_stat_id *stats; + u64 nr_hists; stats = rb_entry(next, struct event_stat_id, rb_node); perf_session__collapse_resort(&stats->hists); - perf_session__output_resort(&stats->hists, stats->stats.total); - if (rb_first(&session->stats_by_id) == - rb_last(&session->stats_by_id)) - fprintf(stdout, "# Samples: %Ld\n#\n", - stats->stats.total); - else - fprintf(stdout, "# Samples: %Ld %s\n#\n", - stats->stats.total, - __event_name(stats->type, stats->config)); - - perf_session__fprintf_hists(&stats->hists, NULL, false, stdout, + nr_hists = perf_session__output_resort(&stats->hists, + stats->stats.total); + if (use_browser) + perf_session__browse_hists(&stats->hists, nr_hists, + stats->stats.total, help, + input_name); + else { + if (rb_first(&session->stats_by_id) == + rb_last(&session->stats_by_id)) + fprintf(stdout, "# Samples: %Ld\n#\n", + stats->stats.total); + else + fprintf(stdout, "# Samples: %Ld %s\n#\n", + stats->stats.total, + __event_name(stats->type, stats->config)); + + perf_session__fprintf_hists(&stats->hists, NULL, false, stdout, stats->stats.total); - fprintf(stdout, "\n\n"); + fprintf(stdout, "\n\n"); + } + next = rb_next(&stats->rb_node); } - if (sort_order == default_sort_order && - parent_pattern == default_parent_pattern) - fprintf(stdout, "#\n# (For a higher level overview, try: perf report --sort comm,dso)\n#\n"); + if (!use_browser && sort_order == default_sort_order && + parent_pattern == default_parent_pattern) { + fprintf(stdout, "#\n# (%s)\n#\n", help); - if (show_threads) { - bool raw_printing_style = !strcmp(pretty_printing_style, "raw"); - perf_read_values_display(stdout, &show_threads_values, - raw_printing_style); - perf_read_values_destroy(&show_threads_values); + if (show_threads) { + bool style = !strcmp(pretty_printing_style, "raw"); + perf_read_values_display(stdout, &show_threads_values, + style); + perf_read_values_destroy(&show_threads_values); + } } out_delete: perf_session__delete(session); @@ -400,7 +431,7 @@ static const char * const report_usage[] = { static const struct option options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), @@ -447,7 +478,8 @@ int cmd_report(int argc, const char **argv, const char *prefix __used) { argc = parse_options(argc, argv, options, report_usage, 0); - setup_pager(); + if (strcmp(input_name, "-") != 0) + setup_browser(); if (symbol__init() < 0) return -1; @@ -455,7 +487,8 @@ int cmd_report(int argc, const char **argv, const char *prefix __used) setup_sorting(report_usage, options); if (parent_pattern != default_parent_pattern) { - sort_dimension__add("parent"); + if (sort_dimension__add("parent") < 0) + return -1; sort_parent.elide = 1; } else symbol_conf.exclude_other = false; diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 4f5a03e43444..09ddc8e6d8e1 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -68,10 +68,10 @@ enum sched_event_type { struct sched_atom { enum sched_event_type type; + int specific_wait; u64 timestamp; u64 duration; unsigned long nr; - int specific_wait; sem_t *wait_sem; struct task_desc *wakee; }; @@ -1790,7 +1790,7 @@ static const char * const sched_usage[] = { static const struct option sched_options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), @@ -1805,7 +1805,7 @@ static const char * const latency_usage[] = { static const struct option latency_options[] = { OPT_STRING('s', "sort", &sort_order, "key[,key2...]", "sort by key(s): runtime, switch, avg, max"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_INTEGER('C', "CPU", &profile_cpu, "CPU to profile on"), @@ -1822,7 +1822,7 @@ static const char * const replay_usage[] = { static const struct option replay_options[] = { OPT_INTEGER('r', "repeat", &replay_repeat, "repeat the workload replay N times (-1: infinite)"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 95db31cff6fd..e619ac89dff5 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -46,6 +46,7 @@ #include "util/debug.h" #include "util/header.h" #include "util/cpumap.h" +#include "util/thread.h" #include <sys/prctl.h> #include <math.h> @@ -66,18 +67,21 @@ static struct perf_event_attr default_attrs[] = { }; -static int system_wide = 0; +static bool system_wide = false; static unsigned int nr_cpus = 0; static int run_idx = 0; static int run_count = 1; -static int inherit = 1; -static int scale = 1; +static bool inherit = true; +static bool scale = true; static pid_t target_pid = -1; +static pid_t target_tid = -1; +static pid_t *all_tids = NULL; +static int thread_num = 0; static pid_t child_pid = -1; -static int null_run = 0; +static bool null_run = false; -static int fd[MAX_NR_CPUS][MAX_COUNTERS]; +static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; static int event_scaled[MAX_COUNTERS]; @@ -140,9 +144,11 @@ struct stats runtime_branches_stats; #define ERR_PERF_OPEN \ "Error: counter %d, sys_perf_event_open() syscall returned with %d (%s)\n" -static void create_perf_stat_counter(int counter, int pid) +static int create_perf_stat_counter(int counter) { struct perf_event_attr *attr = attrs + counter; + int thread; + int ncreated = 0; if (scale) attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED | @@ -152,21 +158,33 @@ static void create_perf_stat_counter(int counter, int pid) unsigned int cpu; for (cpu = 0; cpu < nr_cpus; cpu++) { - fd[cpu][counter] = sys_perf_event_open(attr, -1, cpumap[cpu], -1, 0); - if (fd[cpu][counter] < 0 && verbose) - fprintf(stderr, ERR_PERF_OPEN, counter, - fd[cpu][counter], strerror(errno)); + fd[cpu][counter][0] = sys_perf_event_open(attr, + -1, cpumap[cpu], -1, 0); + if (fd[cpu][counter][0] < 0) + pr_debug(ERR_PERF_OPEN, counter, + fd[cpu][counter][0], strerror(errno)); + else + ++ncreated; } } else { attr->inherit = inherit; - attr->disabled = 1; - attr->enable_on_exec = 1; - - fd[0][counter] = sys_perf_event_open(attr, pid, -1, -1, 0); - if (fd[0][counter] < 0 && verbose) - fprintf(stderr, ERR_PERF_OPEN, counter, - fd[0][counter], strerror(errno)); + if (target_pid == -1) { + attr->disabled = 1; + attr->enable_on_exec = 1; + } + for (thread = 0; thread < thread_num; thread++) { + fd[0][counter][thread] = sys_perf_event_open(attr, + all_tids[thread], -1, -1, 0); + if (fd[0][counter][thread] < 0) + pr_debug(ERR_PERF_OPEN, counter, + fd[0][counter][thread], + strerror(errno)); + else + ++ncreated; + } } + + return ncreated; } /* @@ -190,25 +208,28 @@ static void read_counter(int counter) unsigned int cpu; size_t res, nv; int scaled; - int i; + int i, thread; count[0] = count[1] = count[2] = 0; nv = scale ? 3 : 1; for (cpu = 0; cpu < nr_cpus; cpu++) { - if (fd[cpu][counter] < 0) - continue; - - res = read(fd[cpu][counter], single_count, nv * sizeof(u64)); - assert(res == nv * sizeof(u64)); - - close(fd[cpu][counter]); - fd[cpu][counter] = -1; - - count[0] += single_count[0]; - if (scale) { - count[1] += single_count[1]; - count[2] += single_count[2]; + for (thread = 0; thread < thread_num; thread++) { + if (fd[cpu][counter][thread] < 0) + continue; + + res = read(fd[cpu][counter][thread], + single_count, nv * sizeof(u64)); + assert(res == nv * sizeof(u64)); + + close(fd[cpu][counter][thread]); + fd[cpu][counter][thread] = -1; + + count[0] += single_count[0]; + if (scale) { + count[1] += single_count[1]; + count[2] += single_count[2]; + } } } @@ -250,10 +271,9 @@ static int run_perf_stat(int argc __used, const char **argv) { unsigned long long t0, t1; int status = 0; - int counter; - int pid = target_pid; + int counter, ncreated = 0; int child_ready_pipe[2], go_pipe[2]; - const bool forks = (target_pid == -1 && argc > 0); + const bool forks = (argc > 0); char buf; if (!system_wide) @@ -265,10 +285,10 @@ static int run_perf_stat(int argc __used, const char **argv) } if (forks) { - if ((pid = fork()) < 0) + if ((child_pid = fork()) < 0) perror("failed to fork"); - if (!pid) { + if (!child_pid) { close(child_ready_pipe[0]); close(go_pipe[1]); fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC); @@ -297,7 +317,8 @@ static int run_perf_stat(int argc __used, const char **argv) exit(-1); } - child_pid = pid; + if (target_tid == -1 && target_pid == -1 && !system_wide) + all_tids[0] = child_pid; /* * Wait for the child to be ready to exec. @@ -310,7 +331,16 @@ static int run_perf_stat(int argc __used, const char **argv) } for (counter = 0; counter < nr_counters; counter++) - create_perf_stat_counter(counter, pid); + ncreated += create_perf_stat_counter(counter); + + if (ncreated == 0) { + pr_err("No permission to collect %sstats.\n" + "Consider tweaking /proc/sys/kernel/perf_event_paranoid.\n", + system_wide ? "system-wide " : ""); + if (child_pid != -1) + kill(child_pid, SIGTERM); + return -1; + } /* * Enable counters and exec the command: @@ -321,7 +351,7 @@ static int run_perf_stat(int argc __used, const char **argv) close(go_pipe[1]); wait(&status); } else { - while(!done); + while(!done) sleep(1); } t1 = rdclock(); @@ -429,12 +459,14 @@ static void print_stat(int argc, const char **argv) fprintf(stderr, "\n"); fprintf(stderr, " Performance counter stats for "); - if(target_pid == -1) { + if(target_pid == -1 && target_tid == -1) { fprintf(stderr, "\'%s", argv[0]); for (i = 1; i < argc; i++) fprintf(stderr, " %s", argv[i]); - }else - fprintf(stderr, "task pid \'%d", target_pid); + } else if (target_pid != -1) + fprintf(stderr, "process id \'%d", target_pid); + else + fprintf(stderr, "thread id \'%d", target_tid); fprintf(stderr, "\'"); if (run_count > 1) @@ -459,7 +491,7 @@ static volatile int signr = -1; static void skip_signal(int signo) { - if(target_pid != -1) + if(child_pid == -1) done = 1; signr = signo; @@ -489,12 +521,14 @@ static const struct option options[] = { OPT_BOOLEAN('i', "inherit", &inherit, "child tasks inherit counters"), OPT_INTEGER('p', "pid", &target_pid, - "stat events on existing pid"), + "stat events on existing process id"), + OPT_INTEGER('t', "tid", &target_tid, + "stat events on existing thread id"), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), OPT_BOOLEAN('c', "scale", &scale, "scale/normalize counters"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show counter open errors, etc)"), OPT_INTEGER('r', "repeat", &run_count, "repeat command and print average + stddev (max: 100)"), @@ -506,10 +540,11 @@ static const struct option options[] = { int cmd_stat(int argc, const char **argv, const char *prefix __used) { int status; + int i,j; argc = parse_options(argc, argv, options, stat_usage, PARSE_OPT_STOP_AT_NON_OPTION); - if (!argc && target_pid == -1) + if (!argc && target_pid == -1 && target_tid == -1) usage_with_options(stat_usage, options); if (run_count <= 0) usage_with_options(stat_usage, options); @@ -525,6 +560,31 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) else nr_cpus = 1; + if (target_pid != -1) { + target_tid = target_pid; + thread_num = find_all_tid(target_pid, &all_tids); + if (thread_num <= 0) { + fprintf(stderr, "Can't find all threads of pid %d\n", + target_pid); + usage_with_options(stat_usage, options); + } + } else { + all_tids=malloc(sizeof(pid_t)); + if (!all_tids) + return -ENOMEM; + + all_tids[0] = target_tid; + thread_num = 1; + } + + for (i = 0; i < MAX_NR_CPUS; i++) { + for (j = 0; j < MAX_COUNTERS; j++) { + fd[i][j] = malloc(sizeof(int)*thread_num); + if (!fd[i][j]) + return -ENOMEM; + } + } + /* * We dont want to block the signals - that would cause * child tasks to inherit that and Ctrl-C would not work. @@ -543,7 +603,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) status = run_perf_stat(argc, argv); } - print_stat(argc, argv); + if (status != -1) + print_stat(argc, argv); return status; } diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c index 0d4d8ff7914b..96f4a092df37 100644 --- a/tools/perf/builtin-timechart.c +++ b/tools/perf/builtin-timechart.c @@ -21,7 +21,6 @@ #include "util/cache.h" #include <linux/rbtree.h> #include "util/symbol.h" -#include "util/string.h" #include "util/callchain.h" #include "util/strlist.h" @@ -43,7 +42,7 @@ static u64 turbo_frequency; static u64 first_time, last_time; -static int power_only; +static bool power_only; struct per_pid; @@ -78,8 +77,6 @@ struct per_pid { struct per_pidcomm *all; struct per_pidcomm *current; - - int painted; }; diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 1f529321607e..40f24dd46ef4 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -55,9 +55,9 @@ #include <linux/unistd.h> #include <linux/types.h> -static int fd[MAX_NR_CPUS][MAX_COUNTERS]; +static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; -static int system_wide = 0; +static bool system_wide = false; static int default_interval = 0; @@ -65,18 +65,21 @@ static int count_filter = 5; static int print_entries; static int target_pid = -1; -static int inherit = 0; +static int target_tid = -1; +static pid_t *all_tids = NULL; +static int thread_num = 0; +static bool inherit = false; static int profile_cpu = -1; static int nr_cpus = 0; static unsigned int realtime_prio = 0; -static int group = 0; +static bool group = false; static unsigned int page_size; static unsigned int mmap_pages = 16; static int freq = 1000; /* 1 KHz */ static int delay_secs = 2; -static int zero = 0; -static int dump_symtab = 0; +static bool zero = false; +static bool dump_symtab = false; static bool hide_kernel_symbols = false; static bool hide_user_symbols = false; @@ -133,7 +136,7 @@ static inline struct symbol *sym_entry__symbol(struct sym_entry *self) return ((void *)self) + symbol_conf.priv_size; } -static void get_term_dimensions(struct winsize *ws) +void get_term_dimensions(struct winsize *ws) { char *s = getenv("LINES"); @@ -169,7 +172,7 @@ static void sig_winch_handler(int sig __used) update_print_entries(&winsize); } -static void parse_source(struct sym_entry *syme) +static int parse_source(struct sym_entry *syme) { struct symbol *sym; struct sym_entry_source *source; @@ -180,12 +183,21 @@ static void parse_source(struct sym_entry *syme) u64 len; if (!syme) - return; + return -1; + + sym = sym_entry__symbol(syme); + map = syme->map; + + /* + * We can't annotate with just /proc/kallsyms + */ + if (map->dso->origin == DSO__ORIG_KERNEL) + return -1; if (syme->src == NULL) { syme->src = zalloc(sizeof(*source)); if (syme->src == NULL) - return; + return -1; pthread_mutex_init(&syme->src->lock, NULL); } @@ -195,9 +207,6 @@ static void parse_source(struct sym_entry *syme) pthread_mutex_lock(&source->lock); goto out_assign; } - - sym = sym_entry__symbol(syme); - map = syme->map; path = map->dso->long_name; len = sym->end - sym->start; @@ -209,7 +218,7 @@ static void parse_source(struct sym_entry *syme) file = popen(command, "r"); if (!file) - return; + return -1; pthread_mutex_lock(&source->lock); source->lines_tail = &source->lines; @@ -245,6 +254,7 @@ static void parse_source(struct sym_entry *syme) out_assign: sym_filter_entry = syme; pthread_mutex_unlock(&source->lock); + return 0; } static void __zero_source_counters(struct sym_entry *syme) @@ -411,6 +421,7 @@ static double sym_weight(const struct sym_entry *sym) static long samples; static long userspace_samples; +static long exact_samples; static const char CONSOLE_CLEAR[] = "[H[2J"; static void __list_insert_active_sym(struct sym_entry *syme) @@ -451,6 +462,7 @@ static void print_sym_table(void) int counter, snap = !display_weighted ? sym_counter : 0; float samples_per_sec = samples/delay_secs; float ksamples_per_sec = (samples-userspace_samples)/delay_secs; + float esamples_percent = (100.0*exact_samples)/samples; float sum_ksamples = 0.0; struct sym_entry *syme, *n; struct rb_root tmp = RB_ROOT; @@ -458,7 +470,7 @@ static void print_sym_table(void) int sym_width = 0, dso_width = 0, dso_short_width = 0; const int win_width = winsize.ws_col - 1; - samples = userspace_samples = 0; + samples = userspace_samples = exact_samples = 0; /* Sort the active symbols */ pthread_mutex_lock(&active_symbols_lock); @@ -489,9 +501,10 @@ static void print_sym_table(void) puts(CONSOLE_CLEAR); printf("%-*.*s\n", win_width, win_width, graph_dotted_line); - printf( " PerfTop:%8.0f irqs/sec kernel:%4.1f%% [", + printf( " PerfTop:%8.0f irqs/sec kernel:%4.1f%% exact: %4.1f%% [", samples_per_sec, - 100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec))); + 100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec)), + esamples_percent); if (nr_counters == 1 || !display_weighted) { printf("%Ld", (u64)attrs[0].sample_period); @@ -514,13 +527,15 @@ static void print_sym_table(void) if (target_pid != -1) printf(" (target_pid: %d", target_pid); + else if (target_tid != -1) + printf(" (target_tid: %d", target_tid); else printf(" (all"); if (profile_cpu != -1) printf(", cpu: %d)\n", profile_cpu); else { - if (target_pid != -1) + if (target_tid != -1) printf(")\n"); else printf(", %d CPUs)\n", nr_cpus); @@ -839,7 +854,7 @@ static void handle_keypress(int c) display_weighted = ~display_weighted; break; case 'z': - zero = ~zero; + zero = !zero; break; default: break; @@ -960,6 +975,9 @@ static void event__process_sample(const event_t *self, return; } + if (self->header.misc & PERF_RECORD_MISC_EXACT) + exact_samples++; + if (event__preprocess_sample(self, session, &al, symbol_filter) < 0 || al.filtered) return; @@ -990,7 +1008,17 @@ static void event__process_sample(const event_t *self, if (sym_filter_entry_sched) { sym_filter_entry = sym_filter_entry_sched; sym_filter_entry_sched = NULL; - parse_source(sym_filter_entry); + if (parse_source(sym_filter_entry) < 0) { + struct symbol *sym = sym_entry__symbol(sym_filter_entry); + + pr_err("Can't annotate %s", sym->name); + if (sym_filter_entry->map->dso->origin == DSO__ORIG_KERNEL) { + pr_err(": No vmlinux file was found in the path:\n"); + vmlinux_path__fprintf(stderr); + } else + pr_err(".\n"); + exit(1); + } } syme = symbol__priv(al.sym); @@ -1106,16 +1134,21 @@ static void perf_session__mmap_read_counter(struct perf_session *self, md->prev = old; } -static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS]; -static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS]; +static struct pollfd *event_array; +static struct mmap_data *mmap_array[MAX_NR_CPUS][MAX_COUNTERS]; static void perf_session__mmap_read(struct perf_session *self) { - int i, counter; + int i, counter, thread_index; for (i = 0; i < nr_cpus; i++) { for (counter = 0; counter < nr_counters; counter++) - perf_session__mmap_read_counter(self, &mmap_array[i][counter]); + for (thread_index = 0; + thread_index < thread_num; + thread_index++) { + perf_session__mmap_read_counter(self, + &mmap_array[i][counter][thread_index]); + } } } @@ -1126,9 +1159,10 @@ static void start_counter(int i, int counter) { struct perf_event_attr *attr; int cpu; + int thread_index; cpu = profile_cpu; - if (target_pid == -1 && profile_cpu == -1) + if (target_tid == -1 && profile_cpu == -1) cpu = cpumap[i]; attr = attrs + counter; @@ -1144,55 +1178,58 @@ static void start_counter(int i, int counter) attr->inherit = (cpu < 0) && inherit; attr->mmap = 1; + for (thread_index = 0; thread_index < thread_num; thread_index++) { try_again: - fd[i][counter] = sys_perf_event_open(attr, target_pid, cpu, group_fd, 0); - - if (fd[i][counter] < 0) { - int err = errno; + fd[i][counter][thread_index] = sys_perf_event_open(attr, + all_tids[thread_index], cpu, group_fd, 0); + + if (fd[i][counter][thread_index] < 0) { + int err = errno; + + if (err == EPERM || err == EACCES) + die("No permission - are you root?\n"); + /* + * If it's cycles then fall back to hrtimer + * based cpu-clock-tick sw counter, which + * is always available even if no PMU support: + */ + if (attr->type == PERF_TYPE_HARDWARE + && attr->config == PERF_COUNT_HW_CPU_CYCLES) { + + if (verbose) + warning(" ... trying to fall back to cpu-clock-ticks\n"); + + attr->type = PERF_TYPE_SOFTWARE; + attr->config = PERF_COUNT_SW_CPU_CLOCK; + goto try_again; + } + printf("\n"); + error("perfcounter syscall returned with %d (%s)\n", + fd[i][counter][thread_index], strerror(err)); + die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); + exit(-1); + } + assert(fd[i][counter][thread_index] >= 0); + fcntl(fd[i][counter][thread_index], F_SETFL, O_NONBLOCK); - if (err == EPERM || err == EACCES) - die("No permission - are you root?\n"); /* - * If it's cycles then fall back to hrtimer - * based cpu-clock-tick sw counter, which - * is always available even if no PMU support: + * First counter acts as the group leader: */ - if (attr->type == PERF_TYPE_HARDWARE - && attr->config == PERF_COUNT_HW_CPU_CYCLES) { - - if (verbose) - warning(" ... trying to fall back to cpu-clock-ticks\n"); - - attr->type = PERF_TYPE_SOFTWARE; - attr->config = PERF_COUNT_SW_CPU_CLOCK; - goto try_again; - } - printf("\n"); - error("perfcounter syscall returned with %d (%s)\n", - fd[i][counter], strerror(err)); - die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); - exit(-1); + if (group && group_fd == -1) + group_fd = fd[i][counter][thread_index]; + + event_array[nr_poll].fd = fd[i][counter][thread_index]; + event_array[nr_poll].events = POLLIN; + nr_poll++; + + mmap_array[i][counter][thread_index].counter = counter; + mmap_array[i][counter][thread_index].prev = 0; + mmap_array[i][counter][thread_index].mask = mmap_pages*page_size - 1; + mmap_array[i][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size, + PROT_READ, MAP_SHARED, fd[i][counter][thread_index], 0); + if (mmap_array[i][counter][thread_index].base == MAP_FAILED) + die("failed to mmap with %d (%s)\n", errno, strerror(errno)); } - assert(fd[i][counter] >= 0); - fcntl(fd[i][counter], F_SETFL, O_NONBLOCK); - - /* - * First counter acts as the group leader: - */ - if (group && group_fd == -1) - group_fd = fd[i][counter]; - - event_array[nr_poll].fd = fd[i][counter]; - event_array[nr_poll].events = POLLIN; - nr_poll++; - - mmap_array[i][counter].counter = counter; - mmap_array[i][counter].prev = 0; - mmap_array[i][counter].mask = mmap_pages*page_size - 1; - mmap_array[i][counter].base = mmap(NULL, (mmap_pages+1)*page_size, - PROT_READ, MAP_SHARED, fd[i][counter], 0); - if (mmap_array[i][counter].base == MAP_FAILED) - die("failed to mmap with %d (%s)\n", errno, strerror(errno)); } static int __cmd_top(void) @@ -1208,8 +1245,8 @@ static int __cmd_top(void) if (session == NULL) return -ENOMEM; - if (target_pid != -1) - event__synthesize_thread(target_pid, event__process, session); + if (target_tid != -1) + event__synthesize_thread(target_tid, event__process, session); else event__synthesize_threads(event__process, session); @@ -1220,7 +1257,7 @@ static int __cmd_top(void) } /* Wait for a minimal set of events before starting the snapshot */ - poll(event_array, nr_poll, 100); + poll(&event_array[0], nr_poll, 100); perf_session__mmap_read(session); @@ -1263,7 +1300,9 @@ static const struct option options[] = { OPT_INTEGER('c', "count", &default_interval, "event period to sample"), OPT_INTEGER('p', "pid", &target_pid, - "profile events on existing pid"), + "profile events on existing process id"), + OPT_INTEGER('t', "tid", &target_tid, + "profile events on existing thread id"), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), OPT_INTEGER('C', "CPU", &profile_cpu, @@ -1296,7 +1335,7 @@ static const struct option options[] = { "display this many functions"), OPT_BOOLEAN('U', "hide_user_symbols", &hide_user_symbols, "hide user symbols"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show counter open errors, etc)"), OPT_END() }; @@ -1304,6 +1343,7 @@ static const struct option options[] = { int cmd_top(int argc, const char **argv, const char *prefix __used) { int counter; + int i,j; page_size = sysconf(_SC_PAGE_SIZE); @@ -1311,8 +1351,39 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) if (argc) usage_with_options(top_usage, options); + if (target_pid != -1) { + target_tid = target_pid; + thread_num = find_all_tid(target_pid, &all_tids); + if (thread_num <= 0) { + fprintf(stderr, "Can't find all threads of pid %d\n", + target_pid); + usage_with_options(top_usage, options); + } + } else { + all_tids=malloc(sizeof(pid_t)); + if (!all_tids) + return -ENOMEM; + + all_tids[0] = target_tid; + thread_num = 1; + } + + for (i = 0; i < MAX_NR_CPUS; i++) { + for (j = 0; j < MAX_COUNTERS; j++) { + fd[i][j] = malloc(sizeof(int)*thread_num); + mmap_array[i][j] = zalloc( + sizeof(struct mmap_data)*thread_num); + if (!fd[i][j] || !mmap_array[i][j]) + return -ENOMEM; + } + } + event_array = malloc( + sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num); + if (!event_array) + return -ENOMEM; + /* CPU and PID are mutually exclusive */ - if (target_pid != -1 && profile_cpu != -1) { + if (target_tid > 0 && profile_cpu != -1) { printf("WARNING: PID switch overriding CPU\n"); sleep(1); profile_cpu = -1; @@ -1353,7 +1424,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) attrs[counter].sample_period = default_interval; } - if (target_pid != -1 || profile_cpu != -1) + if (target_tid != -1 || profile_cpu != -1) nr_cpus = 1; else nr_cpus = read_cpu_map(); diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index 407041d20de0..2eefb33c9679 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -104,10 +104,23 @@ static int process_sample_event(event_t *event, struct perf_session *session) static struct perf_event_ops event_ops = { .sample = process_sample_event, .comm = event__process_comm, + .attr = event__process_attr, + .event_type = event__process_event_type, + .tracing_data = event__process_tracing_data, + .build_id = event__process_build_id, }; +extern volatile int session_done; + +static void sig_handler(int sig __unused) +{ + session_done = 1; +} + static int __cmd_trace(struct perf_session *session) { + signal(SIGINT, sig_handler); + return perf_session__process_events(session, &event_ops); } @@ -505,7 +518,7 @@ static const char * const trace_usage[] = { static const struct option options[] = { OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), - OPT_BOOLEAN('v', "verbose", &verbose, + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('L', "Latency", &latency_format, "show latency attributes (irqs/preemption disabled, etc)"), @@ -548,6 +561,65 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) suffix = REPORT_SUFFIX; } + if (!suffix && argc >= 2 && strncmp(argv[1], "-", strlen("-")) != 0) { + char *record_script_path, *report_script_path; + int live_pipe[2]; + pid_t pid; + + record_script_path = get_script_path(argv[1], RECORD_SUFFIX); + if (!record_script_path) { + fprintf(stderr, "record script not found\n"); + return -1; + } + + report_script_path = get_script_path(argv[1], REPORT_SUFFIX); + if (!report_script_path) { + fprintf(stderr, "report script not found\n"); + return -1; + } + + if (pipe(live_pipe) < 0) { + perror("failed to create pipe"); + exit(-1); + } + + pid = fork(); + if (pid < 0) { + perror("failed to fork"); + exit(-1); + } + + if (!pid) { + dup2(live_pipe[1], 1); + close(live_pipe[0]); + + __argv = malloc(5 * sizeof(const char *)); + __argv[0] = "/bin/sh"; + __argv[1] = record_script_path; + __argv[2] = "-o"; + __argv[3] = "-"; + __argv[4] = NULL; + + execvp("/bin/sh", (char **)__argv); + exit(-1); + } + + dup2(live_pipe[0], 0); + close(live_pipe[1]); + + __argv = malloc((argc + 3) * sizeof(const char *)); + __argv[0] = "/bin/sh"; + __argv[1] = report_script_path; + for (i = 2; i < argc; i++) + __argv[i] = argv[i]; + __argv[i++] = "-i"; + __argv[i++] = "-"; + __argv[i++] = NULL; + + execvp("/bin/sh", (char **)__argv); + exit(-1); + } + if (suffix) { script_path = get_script_path(argv[2], suffix); if (!script_path) { @@ -580,7 +652,8 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) if (session == NULL) return -ENOMEM; - if (!perf_session__has_traces(session, "record -R")) + if (strcmp(input_name, "-") && + !perf_session__has_traces(session, "record -R")) return -EINVAL; if (generate_script_lang) { diff --git a/tools/perf/perf-archive.sh b/tools/perf/perf-archive.sh index 910468e6e01c..2e7a4f417e20 100644 --- a/tools/perf/perf-archive.sh +++ b/tools/perf/perf-archive.sh @@ -30,4 +30,7 @@ done tar cfj $PERF_DATA.tar.bz2 -C $DEBUGDIR -T $MANIFEST rm -f $MANIFEST $BUILDIDS +echo -e "Now please run:\n" +echo -e "$ tar xvf $PERF_DATA.tar.bz2 -C ~/.debug\n" +echo "wherever you need to run 'perf report' on." exit 0 diff --git a/tools/perf/perf.c b/tools/perf/perf.c index cd32c200cdb3..d4be55b6cd34 100644 --- a/tools/perf/perf.c +++ b/tools/perf/perf.c @@ -13,9 +13,10 @@ #include "util/quote.h" #include "util/run-command.h" #include "util/parse-events.h" -#include "util/string.h" #include "util/debugfs.h" +bool use_browser; + const char perf_usage_string[] = "perf [--version] [--help] COMMAND [ARGS]"; @@ -262,6 +263,8 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv) set_debugfs_path(); status = p->fn(argc, argv, prefix); + exit_browser(status); + if (status) return status & 0xff; diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 6fb379bc1d1f..ec212748d651 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -1,6 +1,10 @@ #ifndef _PERF_PERF_H #define _PERF_PERF_H +struct winsize; + +void get_term_dimensions(struct winsize *ws); + #if defined(__i386__) #include "../../arch/x86/include/asm/unistd.h" #define rmb() asm volatile("lock; addl $0,0(%%esp)" ::: "memory") @@ -102,8 +106,6 @@ static inline unsigned long long rdclock(void) #define __user #define asmlinkage -#define __used __attribute__((__unused__)) - #define unlikely(x) __builtin_expect(!!(x), 0) #define min(x, y) ({ \ typeof(x) _min1 = (x); \ diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm index f869c48dc9b0..d94b40c8ac85 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm @@ -15,6 +15,7 @@ our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); our @EXPORT = qw( avg nsecs nsecs_secs nsecs_nsecs nsecs_usecs print_nsecs +clear_term ); our $VERSION = '0.01'; @@ -55,6 +56,11 @@ sub nsecs_str { return $str; } +sub clear_term +{ + print "\x1b[H\x1b[2J"; +} + 1; __END__ =head1 NAME diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-record b/tools/perf/scripts/perl/bin/failed-syscalls-record index f8885d389e6f..6ad9b8f5f009 100644 --- a/tools/perf/scripts/perl/bin/failed-syscalls-record +++ b/tools/perf/scripts/perl/bin/failed-syscalls-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit +perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit $@ diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-report b/tools/perf/scripts/perl/bin/failed-syscalls-report index 8bfc660e5056..f6346082a8fc 100644 --- a/tools/perf/scripts/perl/bin/failed-syscalls-report +++ b/tools/perf/scripts/perl/bin/failed-syscalls-report @@ -1,4 +1,10 @@ #!/bin/bash # description: system-wide failed syscalls # args: [comm] -perf trace -s ~/libexec/perf-core/scripts/perl/failed-syscalls.pl $1 +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" ; then + comm=$1 + shift + fi +fi +perf trace $@ -s ~/libexec/perf-core/scripts/perl/failed-syscalls.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-file-record b/tools/perf/scripts/perl/bin/rw-by-file-record index b25056ebf963..a828679837a8 100644 --- a/tools/perf/scripts/perl/bin/rw-by-file-record +++ b/tools/perf/scripts/perl/bin/rw-by-file-record @@ -1,2 +1,3 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_enter_write +perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@ + diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report index eddb9ccce6a5..d83070b7eeb5 100644 --- a/tools/perf/scripts/perl/bin/rw-by-file-report +++ b/tools/perf/scripts/perl/bin/rw-by-file-report @@ -1,7 +1,13 @@ #!/bin/bash # description: r/w activity for a program, by file # args: <comm> -perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl $1 +if [ $# -lt 1 ] ; then + echo "usage: rw-by-file <comm>" + exit +fi +comm=$1 +shift +perf trace $@ -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-record b/tools/perf/scripts/perl/bin/rw-by-pid-record index 8903979c5b6c..63976bf11e8b 100644 --- a/tools/perf/scripts/perl/bin/rw-by-pid-record +++ b/tools/perf/scripts/perl/bin/rw-by-pid-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write +perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@ diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report index 7f44c25cc857..7ef46983f62f 100644 --- a/tools/perf/scripts/perl/bin/rw-by-pid-report +++ b/tools/perf/scripts/perl/bin/rw-by-pid-report @@ -1,6 +1,6 @@ #!/bin/bash # description: system-wide r/w activity -perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-pid.pl +perf trace $@ -s ~/libexec/perf-core/scripts/perl/rw-by-pid.pl diff --git a/tools/perf/scripts/perl/bin/rwtop-record b/tools/perf/scripts/perl/bin/rwtop-record new file mode 100644 index 000000000000..63976bf11e8b --- /dev/null +++ b/tools/perf/scripts/perl/bin/rwtop-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@ diff --git a/tools/perf/scripts/perl/bin/rwtop-report b/tools/perf/scripts/perl/bin/rwtop-report new file mode 100644 index 000000000000..93e698cd3f38 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rwtop-report @@ -0,0 +1,23 @@ +#!/bin/bash +# description: system-wide r/w top +# args: [interval] +n_args=0 +for i in "$@" +do + if expr match "$i" "-" > /dev/null ; then + break + fi + n_args=$(( $n_args + 1 )) +done +if [ "$n_args" -gt 1 ] ; then + echo "usage: rwtop-report [interval]" + exit +fi +if [ "$n_args" -gt 0 ] ; then + interval=$1 + shift +fi +perf trace $@ -s ~/libexec/perf-core/scripts/perl/rwtop.pl $interval + + + diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-record b/tools/perf/scripts/perl/bin/wakeup-latency-record index 6abedda911a4..9c0cf588ff8c 100644 --- a/tools/perf/scripts/perl/bin/wakeup-latency-record +++ b/tools/perf/scripts/perl/bin/wakeup-latency-record @@ -1,5 +1,5 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e sched:sched_switch -e sched:sched_wakeup +perf record -c 1 -f -a -M -R -e sched:sched_switch -e sched:sched_wakeup $@ diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report index fce3adcb3249..a0d898f9ca1d 100644 --- a/tools/perf/scripts/perl/bin/wakeup-latency-report +++ b/tools/perf/scripts/perl/bin/wakeup-latency-report @@ -1,6 +1,6 @@ #!/bin/bash # description: system-wide min/max/avg wakeup latency -perf trace -s ~/libexec/perf-core/scripts/perl/wakeup-latency.pl +perf trace $@ -s ~/libexec/perf-core/scripts/perl/wakeup-latency.pl diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-record b/tools/perf/scripts/perl/bin/workqueue-stats-record index fce6637b19ba..c2a1a9421133 100644 --- a/tools/perf/scripts/perl/bin/workqueue-stats-record +++ b/tools/perf/scripts/perl/bin/workqueue-stats-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e workqueue:workqueue_creation -e workqueue:workqueue_destruction -e workqueue:workqueue_execution -e workqueue:workqueue_insertion +perf record -c 1 -f -a -M -R -e workqueue:workqueue_creation -e workqueue:workqueue_destruction -e workqueue:workqueue_execution -e workqueue:workqueue_insertion $@ diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-report b/tools/perf/scripts/perl/bin/workqueue-stats-report index 71cfbd182fb9..35081132ef97 100644 --- a/tools/perf/scripts/perl/bin/workqueue-stats-report +++ b/tools/perf/scripts/perl/bin/workqueue-stats-report @@ -1,6 +1,6 @@ #!/bin/bash # description: workqueue stats (ins/exe/create/destroy) -perf trace -s ~/libexec/perf-core/scripts/perl/workqueue-stats.pl +perf trace $@ -s ~/libexec/perf-core/scripts/perl/workqueue-stats.pl diff --git a/tools/perf/scripts/perl/rwtop.pl b/tools/perf/scripts/perl/rwtop.pl new file mode 100644 index 000000000000..ec2ab49a6f25 --- /dev/null +++ b/tools/perf/scripts/perl/rwtop.pl @@ -0,0 +1,177 @@ +#!/usr/bin/perl -w +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# read/write top +# +# Periodically displays system-wide r/w call activity, broken down by +# pid. If an [interval] arg is specified, the display will be +# refreshed every [interval] seconds. The default interval is 3 +# seconds. + +use 5.010000; +use strict; +use warnings; + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Util; + +my $default_interval = 3; +my $nlines = 20; +my $print_thread; + +my %reads; +my %writes; + +my $interval = shift; +if (!$interval) { + $interval = $default_interval; +} + +sub syscalls::sys_exit_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + if ($ret > 0) { + $reads{$common_pid}{bytes_read} += $ret; + } else { + if (!defined ($reads{$common_pid}{bytes_read})) { + $reads{$common_pid}{bytes_read} = 0; + } + $reads{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + $reads{$common_pid}{bytes_requested} += $count; + $reads{$common_pid}{total_reads}++; + $reads{$common_pid}{comm} = $common_comm; +} + +sub syscalls::sys_exit_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + if ($ret <= 0) { + $writes{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + $writes{$common_pid}{bytes_written} += $count; + $writes{$common_pid}{total_writes}++; + $writes{$common_pid}{comm} = $common_comm; +} + +sub trace_begin +{ + $SIG{ALRM} = \&print_totals; + alarm 1; +} + +sub trace_end +{ + print_unhandled(); + print_totals(); +} + +sub print_totals +{ + my $count; + + $count = 0; + + clear_term(); + + printf("\nread counts by pid:\n\n"); + + printf("%6s %20s %10s %10s %10s\n", "pid", "comm", + "# reads", "bytes_req", "bytes_read"); + printf("%6s %-20s %10s %10s %10s\n", "------", "--------------------", + "----------", "----------", "----------"); + + foreach my $pid (sort {$reads{$b}{bytes_read} <=> + $reads{$a}{bytes_read}} keys %reads) { + my $comm = $reads{$pid}{comm}; + my $total_reads = $reads{$pid}{total_reads}; + my $bytes_requested = $reads{$pid}{bytes_requested}; + my $bytes_read = $reads{$pid}{bytes_read}; + + printf("%6s %-20s %10s %10s %10s\n", $pid, $comm, + $total_reads, $bytes_requested, $bytes_read); + + if (++$count == $nlines) { + last; + } + } + + $count = 0; + + printf("\nwrite counts by pid:\n\n"); + + printf("%6s %20s %10s %13s\n", "pid", "comm", + "# writes", "bytes_written"); + printf("%6s %-20s %10s %13s\n", "------", "--------------------", + "----------", "-------------"); + + foreach my $pid (sort {$writes{$b}{bytes_written} <=> + $writes{$a}{bytes_written}} keys %writes) { + my $comm = $writes{$pid}{comm}; + my $total_writes = $writes{$pid}{total_writes}; + my $bytes_written = $writes{$pid}{bytes_written}; + + printf("%6s %-20s %10s %13s\n", $pid, $comm, + $total_writes, $bytes_written); + + if (++$count == $nlines) { + last; + } + } + + %reads = (); + %writes = (); + alarm $interval; +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py index 83e91435ed09..9689bc0acd9f 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py @@ -23,3 +23,6 @@ def nsecs_nsecs(nsecs): def nsecs_str(nsecs): str = "%5u.%09u" % (nsecs_secs(nsecs), nsecs_nsecs(nsecs)), return str + +def clear_term(): + print("\x1b[H\x1b[2J") diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record index f8885d389e6f..6ad9b8f5f009 100644 --- a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record +++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit +perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit $@ diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report index 1e0c0a860c87..8c128eff9c0a 100644 --- a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report +++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report @@ -1,4 +1,10 @@ #!/bin/bash # description: system-wide failed syscalls, by pid # args: [comm] -perf trace -s ~/libexec/perf-core/scripts/python/failed-syscalls-by-pid.py $1 +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" ; then + comm=$1 + shift + fi +fi +perf trace $@ -s ~/libexec/perf-core/scripts/python/failed-syscalls-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/sctop-record b/tools/perf/scripts/python/bin/sctop-record new file mode 100644 index 000000000000..27ccffa26ab4 --- /dev/null +++ b/tools/perf/scripts/python/bin/sctop-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -c 1 -f -a -M -R -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/sctop-report b/tools/perf/scripts/python/bin/sctop-report new file mode 100644 index 000000000000..b01c842ae7b4 --- /dev/null +++ b/tools/perf/scripts/python/bin/sctop-report @@ -0,0 +1,24 @@ +#!/bin/bash +# description: syscall top +# args: [comm] [interval] +n_args=0 +for i in "$@" +do + if expr match "$i" "-" > /dev/null ; then + break + fi + n_args=$(( $n_args + 1 )) +done +if [ "$n_args" -gt 2 ] ; then + echo "usage: sctop-report [comm] [interval]" + exit +fi +if [ "$n_args" -gt 1 ] ; then + comm=$1 + interval=$2 + shift 2 +elif [ "$n_args" -gt 0 ] ; then + interval=$1 + shift +fi +perf trace $@ -s ~/libexec/perf-core/scripts/python/sctop.py $comm $interval diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-record b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record index 45a8c50359da..27ccffa26ab4 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-by-pid-record +++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e raw_syscalls:sys_enter +perf record -c 1 -f -a -M -R -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report index f8044d192271..c53362e48602 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report +++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report @@ -1,4 +1,10 @@ #!/bin/bash # description: system-wide syscall counts, by pid # args: [comm] -perf trace -s ~/libexec/perf-core/scripts/python/syscall-counts-by-pid.py $1 +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" ; then + comm=$1 + shift + fi +fi +perf trace $@ -s ~/libexec/perf-core/scripts/python/syscall-counts-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/syscall-counts-record b/tools/perf/scripts/python/bin/syscall-counts-record index 45a8c50359da..27ccffa26ab4 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-record +++ b/tools/perf/scripts/python/bin/syscall-counts-record @@ -1,2 +1,2 @@ #!/bin/bash -perf record -c 1 -f -a -M -R -e raw_syscalls:sys_enter +perf record -c 1 -f -a -M -R -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/syscall-counts-report b/tools/perf/scripts/python/bin/syscall-counts-report index a366aa61612f..8c21552b3cdc 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-report +++ b/tools/perf/scripts/python/bin/syscall-counts-report @@ -1,4 +1,10 @@ #!/bin/bash # description: system-wide syscall counts # args: [comm] -perf trace -s ~/libexec/perf-core/scripts/python/syscall-counts.py $1 +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" ; then + comm=$1 + shift + fi +fi +perf trace $@ -s ~/libexec/perf-core/scripts/python/syscall-counts.py $comm diff --git a/tools/perf/scripts/python/sctop.py b/tools/perf/scripts/python/sctop.py new file mode 100644 index 000000000000..6cafad40c296 --- /dev/null +++ b/tools/perf/scripts/python/sctop.py @@ -0,0 +1,78 @@ +# system call top +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Periodically displays system-wide system call totals, broken down by +# syscall. If a [comm] arg is specified, only syscalls called by +# [comm] are displayed. If an [interval] arg is specified, the display +# will be refreshed every [interval] seconds. The default interval is +# 3 seconds. + +import thread +import time +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +usage = "perf trace -s syscall-counts.py [comm] [interval]\n"; + +for_comm = None +default_interval = 3 +interval = default_interval + +if len(sys.argv) > 3: + sys.exit(usage) + +if len(sys.argv) > 2: + for_comm = sys.argv[1] + interval = int(sys.argv[2]) +elif len(sys.argv) > 1: + try: + interval = int(sys.argv[1]) + except ValueError: + for_comm = sys.argv[1] + interval = default_interval + +syscalls = autodict() + +def trace_begin(): + thread.start_new_thread(print_syscall_totals, (interval,)) + pass + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + if for_comm is not None: + if common_comm != for_comm: + return + try: + syscalls[id] += 1 + except TypeError: + syscalls[id] = 1 + +def print_syscall_totals(interval): + while 1: + clear_term() + if for_comm is not None: + print "\nsyscall events for %s:\n\n" % (for_comm), + else: + print "\nsyscall events:\n\n", + + print "%-40s %10s\n" % ("event", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "----------"), + + for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \ + reverse = True): + try: + print "%-40d %10d\n" % (id, val), + except TypeError: + pass + syscalls.clear() + time.sleep(interval) diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN index 54552a00a117..49ece7921914 100755 --- a/tools/perf/util/PERF-VERSION-GEN +++ b/tools/perf/util/PERF-VERSION-GEN @@ -1,6 +1,10 @@ #!/bin/sh -GVF=PERF-VERSION-FILE +if [ $# -eq 1 ] ; then + OUTPUT=$1 +fi + +GVF=${OUTPUT}PERF-VERSION-FILE DEF_VER=v0.0.2.PERF LF=' diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h index 918eb376abe3..4b9aab7f0405 100644 --- a/tools/perf/util/cache.h +++ b/tools/perf/util/cache.h @@ -1,6 +1,7 @@ #ifndef __PERF_CACHE_H #define __PERF_CACHE_H +#include <stdbool.h> #include "util.h" #include "strbuf.h" #include "../perf.h" @@ -69,6 +70,19 @@ extern const char *pager_program; extern int pager_in_use(void); extern int pager_use_color; +extern bool use_browser; + +#ifdef NO_NEWT_SUPPORT +static inline void setup_browser(void) +{ + setup_pager(); +} +static inline void exit_browser(bool wait_for_ok __used) {} +#else +void setup_browser(void); +void exit_browser(bool wait_for_ok); +#endif + extern const char *editor_program; extern const char *excludes_file; diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c index b3b71258272a..db628af6d20d 100644 --- a/tools/perf/util/callchain.c +++ b/tools/perf/util/callchain.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2009, Frederic Weisbecker <fweisbec@gmail.com> + * Copyright (C) 2009-2010, Frederic Weisbecker <fweisbec@gmail.com> * * Handle the callchains from the stream in an ad-hoc radix tree and then * sort them in an rbtree. @@ -183,12 +183,23 @@ create_child(struct callchain_node *parent, bool inherit_children) return new; } + +struct resolved_ip { + u64 ip; + struct map_symbol ms; +}; + +struct resolved_chain { + u64 nr; + struct resolved_ip ips[0]; +}; + + /* * Fill the node with callchain values */ static void -fill_node(struct callchain_node *node, struct ip_callchain *chain, - int start, struct symbol **syms) +fill_node(struct callchain_node *node, struct resolved_chain *chain, int start) { unsigned int i; @@ -200,8 +211,8 @@ fill_node(struct callchain_node *node, struct ip_callchain *chain, perror("not enough memory for the code path tree"); return; } - call->ip = chain->ips[i]; - call->sym = syms[i]; + call->ip = chain->ips[i].ip; + call->ms = chain->ips[i].ms; list_add_tail(&call->list, &node->val); } node->val_nr = chain->nr - start; @@ -210,13 +221,13 @@ fill_node(struct callchain_node *node, struct ip_callchain *chain, } static void -add_child(struct callchain_node *parent, struct ip_callchain *chain, - int start, struct symbol **syms) +add_child(struct callchain_node *parent, struct resolved_chain *chain, + int start) { struct callchain_node *new; new = create_child(parent, false); - fill_node(new, chain, start, syms); + fill_node(new, chain, start); new->children_hit = 0; new->hit = 1; @@ -228,9 +239,8 @@ add_child(struct callchain_node *parent, struct ip_callchain *chain, * Then create another child to host the given callchain of new branch */ static void -split_add_child(struct callchain_node *parent, struct ip_callchain *chain, - struct callchain_list *to_split, int idx_parents, int idx_local, - struct symbol **syms) +split_add_child(struct callchain_node *parent, struct resolved_chain *chain, + struct callchain_list *to_split, int idx_parents, int idx_local) { struct callchain_node *new; struct list_head *old_tail; @@ -257,7 +267,7 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, /* create a new child for the new branch if any */ if (idx_total < chain->nr) { parent->hit = 0; - add_child(parent, chain, idx_total, syms); + add_child(parent, chain, idx_total); parent->children_hit++; } else { parent->hit = 1; @@ -265,32 +275,33 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, } static int -__append_chain(struct callchain_node *root, struct ip_callchain *chain, - unsigned int start, struct symbol **syms); +__append_chain(struct callchain_node *root, struct resolved_chain *chain, + unsigned int start); static void -__append_chain_children(struct callchain_node *root, struct ip_callchain *chain, - struct symbol **syms, unsigned int start) +__append_chain_children(struct callchain_node *root, + struct resolved_chain *chain, + unsigned int start) { struct callchain_node *rnode; /* lookup in childrens */ chain_for_each_child(rnode, root) { - unsigned int ret = __append_chain(rnode, chain, start, syms); + unsigned int ret = __append_chain(rnode, chain, start); if (!ret) goto inc_children_hit; } /* nothing in children, add to the current node */ - add_child(root, chain, start, syms); + add_child(root, chain, start); inc_children_hit: root->children_hit++; } static int -__append_chain(struct callchain_node *root, struct ip_callchain *chain, - unsigned int start, struct symbol **syms) +__append_chain(struct callchain_node *root, struct resolved_chain *chain, + unsigned int start) { struct callchain_list *cnode; unsigned int i = start; @@ -302,13 +313,19 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, * anywhere inside a function. */ list_for_each_entry(cnode, &root->val, list) { + struct symbol *sym; + if (i == chain->nr) break; - if (cnode->sym && syms[i]) { - if (cnode->sym->start != syms[i]->start) + + sym = chain->ips[i].ms.sym; + + if (cnode->ms.sym && sym) { + if (cnode->ms.sym->start != sym->start) break; - } else if (cnode->ip != chain->ips[i]) + } else if (cnode->ip != chain->ips[i].ip) break; + if (!found) found = true; i++; @@ -320,7 +337,7 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, /* we match only a part of the node. Split it and add the new chain */ if (i - start < root->val_nr) { - split_add_child(root, chain, cnode, start, i - start, syms); + split_add_child(root, chain, cnode, start, i - start); return 0; } @@ -331,15 +348,50 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, } /* We match the node and still have a part remaining */ - __append_chain_children(root, chain, syms, i); + __append_chain_children(root, chain, i); return 0; } -void append_chain(struct callchain_node *root, struct ip_callchain *chain, - struct symbol **syms) +static void filter_context(struct ip_callchain *old, struct resolved_chain *new, + struct map_symbol *syms) +{ + int i, j = 0; + + for (i = 0; i < (int)old->nr; i++) { + if (old->ips[i] >= PERF_CONTEXT_MAX) + continue; + + new->ips[j].ip = old->ips[i]; + new->ips[j].ms = syms[i]; + j++; + } + + new->nr = j; +} + + +int append_chain(struct callchain_node *root, struct ip_callchain *chain, + struct map_symbol *syms) { + struct resolved_chain *filtered; + if (!chain->nr) - return; - __append_chain_children(root, chain, syms, 0); + return 0; + + filtered = malloc(sizeof(*filtered) + + chain->nr * sizeof(struct resolved_ip)); + if (!filtered) + return -ENOMEM; + + filter_context(chain, filtered, syms); + + if (!filtered->nr) + goto end; + + __append_chain_children(root, filtered, 0); +end: + free(filtered); + + return 0; } diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h index ad4626de4c2b..8a7e8bbd0fda 100644 --- a/tools/perf/util/callchain.h +++ b/tools/perf/util/callchain.h @@ -39,7 +39,7 @@ struct callchain_param { struct callchain_list { u64 ip; - struct symbol *sym; + struct map_symbol ms; struct list_head list; }; @@ -56,6 +56,6 @@ static inline u64 cumul_hits(struct callchain_node *node) } int register_callchain_param(struct callchain_param *param); -void append_chain(struct callchain_node *root, struct ip_callchain *chain, - struct symbol **syms); +int append_chain(struct callchain_node *root, struct ip_callchain *chain, + struct map_symbol *syms); #endif /* __PERF_CALLCHAIN_H */ diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c index e88bca55a599..e191eb9a667f 100644 --- a/tools/perf/util/color.c +++ b/tools/perf/util/color.c @@ -166,6 +166,31 @@ int perf_color_default_config(const char *var, const char *value, void *cb) return perf_default_config(var, value, cb); } +static int __color_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args, const char *trail) +{ + int r = 0; + + /* + * Auto-detect: + */ + if (perf_use_color_default < 0) { + if (isatty(1) || pager_in_use()) + perf_use_color_default = 1; + else + perf_use_color_default = 0; + } + + if (perf_use_color_default && *color) + r += snprintf(bf, size, "%s", color); + r += vsnprintf(bf + r, size - r, fmt, args); + if (perf_use_color_default && *color) + r += snprintf(bf + r, size - r, "%s", PERF_COLOR_RESET); + if (trail) + r += snprintf(bf + r, size - r, "%s", trail); + return r; +} + static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args, const char *trail) { @@ -191,11 +216,28 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, return r; } +int color_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args) +{ + return __color_vsnprintf(bf, size, color, fmt, args, NULL); +} + int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args) { return __color_vfprintf(fp, color, fmt, args, NULL); } +int color_snprintf(char *bf, size_t size, const char *color, + const char *fmt, ...) +{ + va_list args; + int r; + + va_start(args, fmt); + r = color_vsnprintf(bf, size, color, fmt, args); + va_end(args); + return r; +} int color_fprintf(FILE *fp, const char *color, const char *fmt, ...) { @@ -274,3 +316,9 @@ int percent_color_fprintf(FILE *fp, const char *fmt, double percent) return r; } + +int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent) +{ + const char *color = get_percent_color(percent); + return color_snprintf(bf, size, color, fmt, percent); +} diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h index 24e8809210bb..dea082b79602 100644 --- a/tools/perf/util/color.h +++ b/tools/perf/util/color.h @@ -32,10 +32,14 @@ int perf_color_default_config(const char *var, const char *value, void *cb); int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty); void color_parse(const char *value, const char *var, char *dst); void color_parse_mem(const char *value, int len, const char *var, char *dst); +int color_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args); int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args); int color_fprintf(FILE *fp, const char *color, const char *fmt, ...); +int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...); int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...); int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf); +int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent); int percent_color_fprintf(FILE *fp, const char *fmt, double percent); const char *get_percent_color(double percent); diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c index 0905600c3851..dd824cf3b628 100644 --- a/tools/perf/util/debug.c +++ b/tools/perf/util/debug.c @@ -6,13 +6,14 @@ #include <stdarg.h> #include <stdio.h> +#include "cache.h" #include "color.h" #include "event.h" #include "debug.h" #include "util.h" int verbose = 0; -int dump_trace = 0; +bool dump_trace = false; int eprintf(int level, const char *fmt, ...) { @@ -21,7 +22,10 @@ int eprintf(int level, const char *fmt, ...) if (verbose >= level) { va_start(args, fmt); - ret = vfprintf(stderr, fmt, args); + if (use_browser) + ret = browser__show_help(fmt, args); + else + ret = vfprintf(stderr, fmt, args); va_end(args); } diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h index c6c24c522dea..047ac3324ebe 100644 --- a/tools/perf/util/debug.h +++ b/tools/perf/util/debug.h @@ -2,14 +2,38 @@ #ifndef __PERF_DEBUG_H #define __PERF_DEBUG_H +#include <stdbool.h> #include "event.h" extern int verbose; -extern int dump_trace; +extern bool dump_trace; -int eprintf(int level, - const char *fmt, ...) __attribute__((format(printf, 2, 3))); int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2))); void trace_event(event_t *event); +struct ui_progress; + +#ifdef NO_NEWT_SUPPORT +static inline int browser__show_help(const char *format __used, va_list ap __used) +{ + return 0; +} + +static inline struct ui_progress *ui_progress__new(const char *title __used, + u64 total __used) +{ + return (struct ui_progress *)1; +} + +static inline void ui_progress__update(struct ui_progress *self __used, + u64 curr __used) {} + +static inline void ui_progress__delete(struct ui_progress *self __used) {} +#else +int browser__show_help(const char *format, va_list ap); +struct ui_progress *ui_progress__new(const char *title, u64 total); +void ui_progress__update(struct ui_progress *self, u64 curr); +void ui_progress__delete(struct ui_progress *self); +#endif + #endif /* __PERF_DEBUG_H */ diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 705ec63548b4..571fb25f7eb9 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -130,6 +130,7 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid, continue; pbf += n + 3; if (*pbf == 'x') { /* vm_exec */ + u64 vm_pgoff; char *execname = strchr(bf, '/'); /* Catch VDSO */ @@ -139,6 +140,14 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid, if (execname == NULL) continue; + pbf += 3; + n = hex2u64(pbf, &vm_pgoff); + /* pgoff is in bytes, not pages */ + if (n >= 0) + ev.mmap.pgoff = vm_pgoff << getpagesize(); + else + ev.mmap.pgoff = 0; + size = strlen(execname); execname[size - 1] = '\0'; /* Remove \n */ memcpy(ev.mmap.filename, execname, size); @@ -393,7 +402,8 @@ int event__process_mmap(event_t *self, struct perf_session *session) } thread = perf_session__findnew(session, self->mmap.pid); - map = map__new(&self->mmap, MAP__FUNCTION, + map = map__new(self->mmap.start, self->mmap.len, self->mmap.pgoff, + self->mmap.pid, self->mmap.filename, MAP__FUNCTION, session->cwd, session->cwdlen); if (thread == NULL || map == NULL) @@ -513,24 +523,32 @@ int event__preprocess_sample(const event_t *self, struct perf_session *session, dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); - thread__find_addr_location(thread, session, cpumode, MAP__FUNCTION, - self->ip.ip, al, filter); + thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION, + self->ip.ip, al); dump_printf(" ...... dso: %s\n", al->map ? al->map->dso->long_name : al->level == 'H' ? "[hypervisor]" : "<not found>"); - /* - * We have to do this here as we may have a dso with no symbol hit that - * has a name longer than the ones with symbols sampled. - */ - if (al->map && !sort_dso.elide && !al->map->dso->slen_calculated) - dso__calc_col_width(al->map->dso); - - if (symbol_conf.dso_list && - (!al->map || !al->map->dso || - !(strlist__has_entry(symbol_conf.dso_list, al->map->dso->short_name) || - (al->map->dso->short_name != al->map->dso->long_name && - strlist__has_entry(symbol_conf.dso_list, al->map->dso->long_name))))) - goto out_filtered; + al->sym = NULL; + + if (al->map) { + if (symbol_conf.dso_list && + (!al->map || !al->map->dso || + !(strlist__has_entry(symbol_conf.dso_list, + al->map->dso->short_name) || + (al->map->dso->short_name != al->map->dso->long_name && + strlist__has_entry(symbol_conf.dso_list, + al->map->dso->long_name))))) + goto out_filtered; + /* + * We have to do this here as we may have a dso with no symbol + * hit that has a name longer than the ones with symbols + * sampled. + */ + if (!sort_dso.elide && !al->map->dso->slen_calculated) + dso__calc_col_width(al->map->dso); + + al->sym = map__find_symbol(al->map, al->addr, filter); + } if (symbol_conf.sym_list && al->sym && !strlist__has_entry(symbol_conf.sym_list, al->sym->name)) diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index a33b94952e34..e5740ea140ab 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -68,11 +68,11 @@ struct sample_data { u64 addr; u64 id; u64 stream_id; - u32 cpu; u64 period; - struct ip_callchain *callchain; + u32 cpu; u32 raw_size; void *raw_data; + struct ip_callchain *callchain; }; #define BUILD_ID_SIZE 20 @@ -83,6 +83,37 @@ struct build_id_event { char filename[]; }; +enum perf_header_event_type { /* above any possible kernel type */ + PERF_RECORD_HEADER_ATTR = 64, + PERF_RECORD_HEADER_EVENT_TYPE = 65, + PERF_RECORD_HEADER_TRACING_DATA = 66, + PERF_RECORD_HEADER_BUILD_ID = 67, + PERF_RECORD_HEADER_MAX +}; + +struct attr_event { + struct perf_event_header header; + struct perf_event_attr attr; + u64 id[]; +}; + +#define MAX_EVENT_NAME 64 + +struct perf_trace_event_type { + u64 event_id; + char name[MAX_EVENT_NAME]; +}; + +struct event_type_event { + struct perf_event_header header; + struct perf_trace_event_type event_type; +}; + +struct tracing_data_event { + struct perf_event_header header; + u32 size; +}; + typedef union event_union { struct perf_event_header header; struct ip_event ip; @@ -92,6 +123,10 @@ typedef union event_union { struct lost_event lost; struct read_event read; struct sample_event sample; + struct attr_event attr; + struct event_type_event event_type; + struct tracing_data_event tracing_data; + struct build_id_event build_id; } event_t; struct events_stats { diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 6c9aa16ee51f..628173ba689e 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -99,13 +99,6 @@ int perf_header__add_attr(struct perf_header *self, return 0; } -#define MAX_EVENT_NAME 64 - -struct perf_trace_event_type { - u64 event_id; - char name[MAX_EVENT_NAME]; -}; - static int event_count; static struct perf_trace_event_type *events; @@ -427,6 +420,25 @@ out_free: return err; } +int perf_header__write_pipe(int fd) +{ + struct perf_pipe_file_header f_header; + int err; + + f_header = (struct perf_pipe_file_header){ + .magic = PERF_MAGIC, + .size = sizeof(f_header), + }; + + err = do_write(fd, &f_header, sizeof(f_header)); + if (err < 0) { + pr_debug("failed to write perf pipe header\n"); + return err; + } + + return 0; +} + int perf_header__write(struct perf_header *self, int fd, bool at_exit) { struct perf_file_header f_header; @@ -518,25 +530,10 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit) return 0; } -static int do_read(int fd, void *buf, size_t size) -{ - while (size) { - int ret = read(fd, buf, size); - - if (ret <= 0) - return -1; - - size -= ret; - buf += ret; - } - - return 0; -} - static int perf_header__getbuffer64(struct perf_header *self, int fd, void *buf, size_t size) { - if (do_read(fd, buf, size)) + if (do_read(fd, buf, size) <= 0) return -1; if (self->needs_swap) @@ -592,7 +589,7 @@ int perf_file_header__read(struct perf_file_header *self, { lseek(fd, 0, SEEK_SET); - if (do_read(fd, self, sizeof(*self)) || + if (do_read(fd, self, sizeof(*self)) <= 0 || memcmp(&self->magic, __perf_magic, sizeof(self->magic))) return -1; @@ -662,13 +659,51 @@ static int perf_file_section__process(struct perf_file_section *self, return 0; } -int perf_header__read(struct perf_header *self, int fd) +static int perf_file_header__read_pipe(struct perf_pipe_file_header *self, + struct perf_header *ph, int fd) +{ + if (do_read(fd, self, sizeof(*self)) <= 0 || + memcmp(&self->magic, __perf_magic, sizeof(self->magic))) + return -1; + + if (self->size != sizeof(*self)) { + u64 size = bswap_64(self->size); + + if (size != sizeof(*self)) + return -1; + + ph->needs_swap = true; + } + + return 0; +} + +static int perf_header__read_pipe(struct perf_session *session, int fd) +{ + struct perf_header *self = &session->header; + struct perf_pipe_file_header f_header; + + if (perf_file_header__read_pipe(&f_header, self, fd) < 0) { + pr_debug("incompatible file format\n"); + return -EINVAL; + } + + session->fd = fd; + + return 0; +} + +int perf_header__read(struct perf_session *session, int fd) { + struct perf_header *self = &session->header; struct perf_file_header f_header; struct perf_file_attr f_attr; u64 f_id; int nr_attrs, nr_ids, i, j; + if (session->fd_pipe) + return perf_header__read_pipe(session, fd); + if (perf_file_header__read(&f_header, self, fd) < 0) { pr_debug("incompatible file format\n"); return -EINVAL; @@ -765,3 +800,279 @@ perf_header__find_attr(u64 id, struct perf_header *header) return NULL; } + +int event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id, + event__handler_t process, + struct perf_session *session) +{ + event_t *ev; + size_t size; + int err; + + size = sizeof(struct perf_event_attr); + size = ALIGN(size, sizeof(u64)); + size += sizeof(struct perf_event_header); + size += ids * sizeof(u64); + + ev = malloc(size); + + ev->attr.attr = *attr; + memcpy(ev->attr.id, id, ids * sizeof(u64)); + + ev->attr.header.type = PERF_RECORD_HEADER_ATTR; + ev->attr.header.size = size; + + err = process(ev, session); + + free(ev); + + return err; +} + +int event__synthesize_attrs(struct perf_header *self, + event__handler_t process, + struct perf_session *session) +{ + struct perf_header_attr *attr; + int i, err = 0; + + for (i = 0; i < self->attrs; i++) { + attr = self->attr[i]; + + err = event__synthesize_attr(&attr->attr, attr->ids, attr->id, + process, session); + if (err) { + pr_debug("failed to create perf header attribute\n"); + return err; + } + } + + return err; +} + +int event__process_attr(event_t *self, struct perf_session *session) +{ + struct perf_header_attr *attr; + unsigned int i, ids, n_ids; + + attr = perf_header_attr__new(&self->attr.attr); + if (attr == NULL) + return -ENOMEM; + + ids = self->header.size; + ids -= (void *)&self->attr.id - (void *)self; + n_ids = ids / sizeof(u64); + + for (i = 0; i < n_ids; i++) { + if (perf_header_attr__add_id(attr, self->attr.id[i]) < 0) { + perf_header_attr__delete(attr); + return -ENOMEM; + } + } + + if (perf_header__add_attr(&session->header, attr) < 0) { + perf_header_attr__delete(attr); + return -ENOMEM; + } + + perf_session__update_sample_type(session); + + return 0; +} + +int event__synthesize_event_type(u64 event_id, char *name, + event__handler_t process, + struct perf_session *session) +{ + event_t ev; + size_t size = 0; + int err = 0; + + memset(&ev, 0, sizeof(ev)); + + ev.event_type.event_type.event_id = event_id; + memset(ev.event_type.event_type.name, 0, MAX_EVENT_NAME); + strncpy(ev.event_type.event_type.name, name, MAX_EVENT_NAME - 1); + + ev.event_type.header.type = PERF_RECORD_HEADER_EVENT_TYPE; + size = strlen(name); + size = ALIGN(size, sizeof(u64)); + ev.event_type.header.size = sizeof(ev.event_type) - + (sizeof(ev.event_type.event_type.name) - size); + + err = process(&ev, session); + + return err; +} + +int event__synthesize_event_types(event__handler_t process, + struct perf_session *session) +{ + struct perf_trace_event_type *type; + int i, err = 0; + + for (i = 0; i < event_count; i++) { + type = &events[i]; + + err = event__synthesize_event_type(type->event_id, type->name, + process, session); + if (err) { + pr_debug("failed to create perf header event type\n"); + return err; + } + } + + return err; +} + +int event__process_event_type(event_t *self, + struct perf_session *session __unused) +{ + if (perf_header__push_event(self->event_type.event_type.event_id, + self->event_type.event_type.name) < 0) + return -ENOMEM; + + return 0; +} + +int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, + int nb_events, + event__handler_t process, + struct perf_session *session __unused) +{ + event_t ev; + ssize_t size = 0, aligned_size = 0, padding; + int err = 0; + + memset(&ev, 0, sizeof(ev)); + + ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA; + size = read_tracing_data_size(fd, pattrs, nb_events); + if (size <= 0) + return size; + aligned_size = ALIGN(size, sizeof(u64)); + padding = aligned_size - size; + ev.tracing_data.header.size = sizeof(ev.tracing_data); + ev.tracing_data.size = aligned_size; + + process(&ev, session); + + err = read_tracing_data(fd, pattrs, nb_events); + write_padded(fd, NULL, 0, padding); + + return aligned_size; +} + +int event__process_tracing_data(event_t *self, + struct perf_session *session) +{ + ssize_t size_read, padding, size = self->tracing_data.size; + off_t offset = lseek(session->fd, 0, SEEK_CUR); + char buf[BUFSIZ]; + + /* setup for reading amidst mmap */ + lseek(session->fd, offset + sizeof(struct tracing_data_event), + SEEK_SET); + + size_read = trace_report(session->fd); + + padding = ALIGN(size_read, sizeof(u64)) - size_read; + + if (read(session->fd, buf, padding) < 0) + die("reading input file"); + + if (size_read + padding != size) + die("tracing data size mismatch"); + + return size_read + padding; +} + +int event__synthesize_build_id(struct dso *pos, u16 misc, + event__handler_t process, + struct perf_session *session) +{ + event_t ev; + size_t len; + int err = 0; + + if (!pos->hit) + return err; + + memset(&ev, 0, sizeof(ev)); + + len = pos->long_name_len + 1; + len = ALIGN(len, NAME_ALIGN); + memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id)); + ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID; + ev.build_id.header.misc = misc; + ev.build_id.header.size = sizeof(ev.build_id) + len; + memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len); + + err = process(&ev, session); + + return err; +} + +static int __event_synthesize_build_ids(struct list_head *head, u16 misc, + event__handler_t process, + struct perf_session *session) +{ + struct dso *pos; + + dsos__for_each_with_build_id(pos, head) { + int err; + if (!pos->hit) + continue; + + err = event__synthesize_build_id(pos, misc, process, session); + if (err < 0) + return err; + } + + return 0; +} + +int event__synthesize_build_ids(event__handler_t process, + struct perf_session *session) +{ + int err; + + if (!dsos__read_build_ids(true)) + return 0; + + err = __event_synthesize_build_ids(&dsos__kernel, + PERF_RECORD_MISC_KERNEL, + process, session); + if (err == 0) + err = __event_synthesize_build_ids(&dsos__user, + PERF_RECORD_MISC_USER, + process, session); + + if (err < 0) { + pr_debug("failed to synthesize build ids\n"); + return err; + } + + dsos__cache_build_ids(); + + return 0; +} + +int event__process_build_id(event_t *self, + struct perf_session *session __unused) +{ + struct list_head *head = &dsos__user; + struct dso *dso; + + if (self->build_id.header.misc & PERF_RECORD_MISC_KERNEL) + head = &dsos__kernel; + + dso = __dsos__findnew(head, self->build_id.filename); + if (dso != NULL) { + dso__set_build_id(dso, &self->build_id.build_id); + if (head == &dsos__kernel && self->build_id.filename[0] == '[') + dso->kernel = 1; + } + + return 0; +} diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h index 82a6af72d4cc..4214e2375650 100644 --- a/tools/perf/util/header.h +++ b/tools/perf/util/header.h @@ -39,6 +39,11 @@ struct perf_file_header { DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS); }; +struct perf_pipe_file_header { + u64 magic; + u64 size; +}; + struct perf_header; int perf_file_header__read(struct perf_file_header *self, @@ -47,21 +52,22 @@ int perf_file_header__read(struct perf_file_header *self, struct perf_header { int frozen; int attrs, size; + bool needs_swap; struct perf_header_attr **attr; s64 attr_offset; u64 data_offset; u64 data_size; u64 event_offset; u64 event_size; - bool needs_swap; DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS); }; int perf_header__init(struct perf_header *self); void perf_header__exit(struct perf_header *self); -int perf_header__read(struct perf_header *self, int fd); +int perf_header__read(struct perf_session *session, int fd); int perf_header__write(struct perf_header *self, int fd, bool at_exit); +int perf_header__write_pipe(int fd); int perf_header__add_attr(struct perf_header *self, struct perf_header_attr *attr); @@ -89,4 +95,34 @@ int build_id_cache__add_s(const char *sbuild_id, const char *debugdir, const char *name, bool is_kallsyms); int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir); +int event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id, + event__handler_t process, + struct perf_session *session); +int event__synthesize_attrs(struct perf_header *self, + event__handler_t process, + struct perf_session *session); +int event__process_attr(event_t *self, struct perf_session *session); + +int event__synthesize_event_type(u64 event_id, char *name, + event__handler_t process, + struct perf_session *session); +int event__synthesize_event_types(event__handler_t process, + struct perf_session *session); +int event__process_event_type(event_t *self, + struct perf_session *session); + +int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, + int nb_events, + event__handler_t process, + struct perf_session *session); +int event__process_tracing_data(event_t *self, + struct perf_session *session); + +int event__synthesize_build_id(struct dso *pos, u16 misc, + event__handler_t process, + struct perf_session *session); +int event__synthesize_build_ids(event__handler_t process, + struct perf_session *session); +int event__process_build_id(event_t *self, struct perf_session *session); + #endif /* __PERF_HEADER_H */ diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index 2be33c7dbf03..9c2b8743cef6 100644 --- a/tools/perf/util/hist.c +++ b/tools/perf/util/hist.c @@ -22,8 +22,10 @@ struct hist_entry *__perf_session__add_hist_entry(struct rb_root *hists, struct hist_entry *he; struct hist_entry entry = { .thread = al->thread, - .map = al->map, - .sym = al->sym, + .ms = { + .map = al->map, + .sym = al->sym, + }, .ip = al->addr, .level = al->level, .count = count, @@ -48,7 +50,8 @@ struct hist_entry *__perf_session__add_hist_entry(struct rb_root *hists, p = &(*p)->rb_right; } - he = malloc(sizeof(*he)); + he = malloc(sizeof(*he) + (symbol_conf.use_callchain ? + sizeof(struct callchain_node) : 0)); if (!he) return NULL; *he = entry; @@ -65,7 +68,7 @@ hist_entry__cmp(struct hist_entry *left, struct hist_entry *right) int64_t cmp = 0; list_for_each_entry(se, &hist_entry__sort_list, list) { - cmp = se->cmp(left, right); + cmp = se->se_cmp(left, right); if (cmp) break; } @@ -82,7 +85,7 @@ hist_entry__collapse(struct hist_entry *left, struct hist_entry *right) list_for_each_entry(se, &hist_entry__sort_list, list) { int64_t (*f)(struct hist_entry *, struct hist_entry *); - f = se->collapse ?: se->cmp; + f = se->se_collapse ?: se->se_cmp; cmp = f(left, right); if (cmp) @@ -166,7 +169,7 @@ static void perf_session__insert_output_hist_entry(struct rb_root *root, struct hist_entry *iter; if (symbol_conf.use_callchain) - callchain_param.sort(&he->sorted_chain, &he->callchain, + callchain_param.sort(&he->sorted_chain, he->callchain, min_callchain_hits, &callchain_param); while (*p != NULL) { @@ -183,12 +186,13 @@ static void perf_session__insert_output_hist_entry(struct rb_root *root, rb_insert_color(&he->rb_node, root); } -void perf_session__output_resort(struct rb_root *hists, u64 total_samples) +u64 perf_session__output_resort(struct rb_root *hists, u64 total_samples) { struct rb_root tmp; struct rb_node *next; struct hist_entry *n; u64 min_callchain_hits; + u64 nr_hists = 0; min_callchain_hits = total_samples * (callchain_param.min_percent / 100); @@ -203,9 +207,11 @@ void perf_session__output_resort(struct rb_root *hists, u64 total_samples) rb_erase(&n->rb_node, hists); perf_session__insert_output_hist_entry(&tmp, n, min_callchain_hits); + ++nr_hists; } *hists = tmp; + return nr_hists; } static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin) @@ -258,8 +264,8 @@ static size_t ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, } else ret += fprintf(fp, "%s", " "); } - if (chain->sym) - ret += fprintf(fp, "%s\n", chain->sym->name); + if (chain->ms.sym) + ret += fprintf(fp, "%s\n", chain->ms.sym->name); else ret += fprintf(fp, "%p\n", (void *)(long)chain->ip); @@ -278,7 +284,7 @@ static void init_rem_hits(void) } strcpy(rem_sq_bracket->name, "[...]"); - rem_hits.sym = rem_sq_bracket; + rem_hits.ms.sym = rem_sq_bracket; } static size_t __callchain__fprintf_graph(FILE *fp, struct callchain_node *self, @@ -328,8 +334,6 @@ static size_t __callchain__fprintf_graph(FILE *fp, struct callchain_node *self, left_margin); i = 0; list_for_each_entry(chain, &child->val, list) { - if (chain->ip >= PERF_CONTEXT_MAX) - continue; ret += ipchain__fprintf_graph(fp, chain, depth, new_depth_mask, i++, new_total, @@ -368,9 +372,6 @@ static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self, int ret = 0; list_for_each_entry(chain, &self->val, list) { - if (chain->ip >= PERF_CONTEXT_MAX) - continue; - if (!i++ && sort__first_dimension == SORT_SYM) continue; @@ -385,8 +386,8 @@ static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self, } else ret += callchain__fprintf_left_margin(fp, left_margin); - if (chain->sym) - ret += fprintf(fp, " %s\n", chain->sym->name); + if (chain->ms.sym) + ret += fprintf(fp, " %s\n", chain->ms.sym->name); else ret += fprintf(fp, " %p\n", (void *)(long)chain->ip); } @@ -411,8 +412,8 @@ static size_t callchain__fprintf_flat(FILE *fp, struct callchain_node *self, list_for_each_entry(chain, &self->val, list) { if (chain->ip >= PERF_CONTEXT_MAX) continue; - if (chain->sym) - ret += fprintf(fp, " %s\n", chain->sym->name); + if (chain->ms.sym) + ret += fprintf(fp, " %s\n", chain->ms.sym->name); else ret += fprintf(fp, " %p\n", (void *)(long)chain->ip); @@ -455,16 +456,17 @@ static size_t hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self, return ret; } -static size_t hist_entry__fprintf(struct hist_entry *self, - struct perf_session *pair_session, - bool show_displacement, - long displacement, FILE *fp, - u64 session_total) +int hist_entry__snprintf(struct hist_entry *self, + char *s, size_t size, + struct perf_session *pair_session, + bool show_displacement, + long displacement, bool color, + u64 session_total) { struct sort_entry *se; u64 count, total; const char *sep = symbol_conf.field_sep; - size_t ret; + int ret; if (symbol_conf.exclude_other && !self->parent) return 0; @@ -477,17 +479,22 @@ static size_t hist_entry__fprintf(struct hist_entry *self, total = session_total; } - if (total) - ret = percent_color_fprintf(fp, sep ? "%.2f" : " %6.2f%%", - (count * 100.0) / total); - else - ret = fprintf(fp, sep ? "%lld" : "%12lld ", count); + if (total) { + if (color) + ret = percent_color_snprintf(s, size, + sep ? "%.2f" : " %6.2f%%", + (count * 100.0) / total); + else + ret = snprintf(s, size, sep ? "%.2f" : " %6.2f%%", + (count * 100.0) / total); + } else + ret = snprintf(s, size, sep ? "%lld" : "%12lld ", count); if (symbol_conf.show_nr_samples) { if (sep) - fprintf(fp, "%c%lld", *sep, count); + ret += snprintf(s + ret, size - ret, "%c%lld", *sep, count); else - fprintf(fp, "%11lld", count); + ret += snprintf(s + ret, size - ret, "%11lld", count); } if (pair_session) { @@ -507,9 +514,9 @@ static size_t hist_entry__fprintf(struct hist_entry *self, snprintf(bf, sizeof(bf), " "); if (sep) - ret += fprintf(fp, "%c%s", *sep, bf); + ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf); else - ret += fprintf(fp, "%11.11s", bf); + ret += snprintf(s + ret, size - ret, "%11.11s", bf); if (show_displacement) { if (displacement) @@ -518,9 +525,9 @@ static size_t hist_entry__fprintf(struct hist_entry *self, snprintf(bf, sizeof(bf), " "); if (sep) - fprintf(fp, "%c%s", *sep, bf); + ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf); else - fprintf(fp, "%6.6s", bf); + ret += snprintf(s + ret, size - ret, "%6.6s", bf); } } @@ -528,27 +535,41 @@ static size_t hist_entry__fprintf(struct hist_entry *self, if (se->elide) continue; - fprintf(fp, "%s", sep ?: " "); - ret += se->print(fp, self, se->width ? *se->width : 0); + ret += snprintf(s + ret, size - ret, "%s", sep ?: " "); + ret += se->se_snprintf(self, s + ret, size - ret, + se->se_width ? *se->se_width : 0); } - ret += fprintf(fp, "\n"); + return ret; +} - if (symbol_conf.use_callchain) { - int left_margin = 0; +int hist_entry__fprintf(struct hist_entry *self, + struct perf_session *pair_session, + bool show_displacement, + long displacement, FILE *fp, + u64 session_total) +{ + char bf[512]; + hist_entry__snprintf(self, bf, sizeof(bf), pair_session, + show_displacement, displacement, + true, session_total); + return fprintf(fp, "%s\n", bf); +} - if (sort__first_dimension == SORT_COMM) { - se = list_first_entry(&hist_entry__sort_list, typeof(*se), - list); - left_margin = se->width ? *se->width : 0; - left_margin -= thread__comm_len(self->thread); - } +static size_t hist_entry__fprintf_callchain(struct hist_entry *self, FILE *fp, + u64 session_total) +{ + int left_margin = 0; - hist_entry_callchain__fprintf(fp, self, session_total, - left_margin); + if (sort__first_dimension == SORT_COMM) { + struct sort_entry *se = list_first_entry(&hist_entry__sort_list, + typeof(*se), list); + left_margin = se->se_width ? *se->se_width : 0; + left_margin -= thread__comm_len(self->thread); } - return ret; + return hist_entry_callchain__fprintf(fp, self, session_total, + left_margin); } size_t perf_session__fprintf_hists(struct rb_root *hists, @@ -594,22 +615,22 @@ size_t perf_session__fprintf_hists(struct rb_root *hists, if (se->elide) continue; if (sep) { - fprintf(fp, "%c%s", *sep, se->header); + fprintf(fp, "%c%s", *sep, se->se_header); continue; } - width = strlen(se->header); - if (se->width) { + width = strlen(se->se_header); + if (se->se_width) { if (symbol_conf.col_width_list_str) { if (col_width) { - *se->width = atoi(col_width); + *se->se_width = atoi(col_width); col_width = strchr(col_width, ','); if (col_width) ++col_width; } } - width = *se->width = max(*se->width, width); + width = *se->se_width = max(*se->se_width, width); } - fprintf(fp, " %*s", width, se->header); + fprintf(fp, " %*s", width, se->se_header); } fprintf(fp, "\n"); @@ -631,10 +652,10 @@ size_t perf_session__fprintf_hists(struct rb_root *hists, continue; fprintf(fp, " "); - if (se->width) - width = *se->width; + if (se->se_width) + width = *se->se_width; else - width = strlen(se->header); + width = strlen(se->se_header); for (i = 0; i < width; i++) fprintf(fp, "."); } @@ -655,9 +676,13 @@ print_entries: } ret += hist_entry__fprintf(h, pair, show_displacement, displacement, fp, session_total); - if (h->map == NULL && verbose > 1) { + + if (symbol_conf.use_callchain) + ret += hist_entry__fprintf_callchain(h, fp, session_total); + + if (h->ms.map == NULL && verbose > 1) { __map_groups__fprintf_maps(&h->thread->mg, - MAP__FUNCTION, fp); + MAP__FUNCTION, verbose, fp); fprintf(fp, "%.10s end\n", graph_dotted_line); } } diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h index 16f360cce5bf..ad17f0ad798b 100644 --- a/tools/perf/util/hist.h +++ b/tools/perf/util/hist.h @@ -18,9 +18,19 @@ struct hist_entry *__perf_session__add_hist_entry(struct rb_root *hists, u64 count, bool *hit); extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *); extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *); +int hist_entry__fprintf(struct hist_entry *self, + struct perf_session *pair_session, + bool show_displacement, + long displacement, FILE *fp, + u64 session_total); +int hist_entry__snprintf(struct hist_entry *self, + char *bf, size_t size, + struct perf_session *pair_session, + bool show_displacement, long displacement, + bool color, u64 session_total); void hist_entry__free(struct hist_entry *); -void perf_session__output_resort(struct rb_root *hists, u64 total_samples); +u64 perf_session__output_resort(struct rb_root *hists, u64 total_samples); void perf_session__collapse_resort(struct rb_root *hists); size_t perf_session__fprintf_hists(struct rb_root *hists, struct perf_session *pair, diff --git a/tools/perf/util/include/linux/compiler.h b/tools/perf/util/include/linux/compiler.h index dfb0713ed47f..791f9dd27ebf 100644 --- a/tools/perf/util/include/linux/compiler.h +++ b/tools/perf/util/include/linux/compiler.h @@ -7,4 +7,6 @@ #define __user #define __attribute_const__ +#define __used __attribute__((__unused__)) + #endif diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h index f2611655ab51..388ab1bfd114 100644 --- a/tools/perf/util/include/linux/kernel.h +++ b/tools/perf/util/include/linux/kernel.h @@ -85,16 +85,19 @@ simple_strtoul(const char *nptr, char **endptr, int base) return strtoul(nptr, endptr, base); } +int eprintf(int level, + const char *fmt, ...) __attribute__((format(printf, 2, 3))); + #ifndef pr_fmt #define pr_fmt(fmt) fmt #endif #define pr_err(fmt, ...) \ - do { fprintf(stderr, pr_fmt(fmt), ##__VA_ARGS__); } while (0) + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) #define pr_warning(fmt, ...) \ - do { fprintf(stderr, pr_fmt(fmt), ##__VA_ARGS__); } while (0) + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) #define pr_info(fmt, ...) \ - do { fprintf(stderr, pr_fmt(fmt), ##__VA_ARGS__); } while (0) + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) #define pr_debug(fmt, ...) \ eprintf(1, pr_fmt(fmt), ##__VA_ARGS__) #define pr_debugN(n, fmt, ...) \ diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index e509cd59c67d..37913b241bdf 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -1,9 +1,10 @@ -#include "event.h" #include "symbol.h" +#include <errno.h> +#include <limits.h> #include <stdlib.h> #include <string.h> #include <stdio.h> -#include "debug.h" +#include "map.h" const char *map_type__name[MAP__NR_TYPES] = { [MAP__FUNCTION] = "Functions", @@ -38,13 +39,12 @@ void map__init(struct map *self, enum map_type type, RB_CLEAR_NODE(&self->rb_node); } -struct map *map__new(struct mmap_event *event, enum map_type type, - char *cwd, int cwdlen) +struct map *map__new(u64 start, u64 len, u64 pgoff, u32 pid, char *filename, + enum map_type type, char *cwd, int cwdlen) { struct map *self = malloc(sizeof(*self)); if (self != NULL) { - const char *filename = event->filename; char newfilename[PATH_MAX]; struct dso *dso; int anon; @@ -62,7 +62,7 @@ struct map *map__new(struct mmap_event *event, enum map_type type, anon = is_anon_memory(filename); if (anon) { - snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", event->pid); + snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid); filename = newfilename; } @@ -70,8 +70,7 @@ struct map *map__new(struct mmap_event *event, enum map_type type, if (dso == NULL) goto out_delete; - map__init(self, type, event->start, event->start + event->len, - event->pgoff, dso); + map__init(self, type, start, start + len, pgoff, dso); if (anon) { set_identity: @@ -235,3 +234,277 @@ u64 map__objdump_2ip(struct map *map, u64 addr) map->unmap_ip(map, addr); /* RIP -> IP */ return ip; } + +void map_groups__init(struct map_groups *self) +{ + int i; + for (i = 0; i < MAP__NR_TYPES; ++i) { + self->maps[i] = RB_ROOT; + INIT_LIST_HEAD(&self->removed_maps[i]); + } +} + +void map_groups__flush(struct map_groups *self) +{ + int type; + + for (type = 0; type < MAP__NR_TYPES; type++) { + struct rb_root *root = &self->maps[type]; + struct rb_node *next = rb_first(root); + + while (next) { + struct map *pos = rb_entry(next, struct map, rb_node); + next = rb_next(&pos->rb_node); + rb_erase(&pos->rb_node, root); + /* + * We may have references to this map, for + * instance in some hist_entry instances, so + * just move them to a separate list. + */ + list_add_tail(&pos->node, &self->removed_maps[pos->type]); + } + } +} + +struct symbol *map_groups__find_symbol(struct map_groups *self, + enum map_type type, u64 addr, + struct map **mapp, + symbol_filter_t filter) +{ + struct map *map = map_groups__find(self, type, addr); + + if (map != NULL) { + if (mapp != NULL) + *mapp = map; + return map__find_symbol(map, map->map_ip(map, addr), filter); + } + + return NULL; +} + +struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, + enum map_type type, + const char *name, + struct map **mapp, + symbol_filter_t filter) +{ + struct rb_node *nd; + + for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + struct symbol *sym = map__find_symbol_by_name(pos, name, filter); + + if (sym == NULL) + continue; + if (mapp != NULL) + *mapp = pos; + return sym; + } + + return NULL; +} + +size_t __map_groups__fprintf_maps(struct map_groups *self, + enum map_type type, int verbose, FILE *fp) +{ + size_t printed = fprintf(fp, "%s:\n", map_type__name[type]); + struct rb_node *nd; + + for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + printed += fprintf(fp, "Map:"); + printed += map__fprintf(pos, fp); + if (verbose > 2) { + printed += dso__fprintf(pos->dso, type, fp); + printed += fprintf(fp, "--\n"); + } + } + + return printed; +} + +size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp) +{ + size_t printed = 0, i; + for (i = 0; i < MAP__NR_TYPES; ++i) + printed += __map_groups__fprintf_maps(self, i, verbose, fp); + return printed; +} + +static size_t __map_groups__fprintf_removed_maps(struct map_groups *self, + enum map_type type, + int verbose, FILE *fp) +{ + struct map *pos; + size_t printed = 0; + + list_for_each_entry(pos, &self->removed_maps[type], node) { + printed += fprintf(fp, "Map:"); + printed += map__fprintf(pos, fp); + if (verbose > 1) { + printed += dso__fprintf(pos->dso, type, fp); + printed += fprintf(fp, "--\n"); + } + } + return printed; +} + +static size_t map_groups__fprintf_removed_maps(struct map_groups *self, + int verbose, FILE *fp) +{ + size_t printed = 0, i; + for (i = 0; i < MAP__NR_TYPES; ++i) + printed += __map_groups__fprintf_removed_maps(self, i, verbose, fp); + return printed; +} + +size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp) +{ + size_t printed = map_groups__fprintf_maps(self, verbose, fp); + printed += fprintf(fp, "Removed maps:\n"); + return printed + map_groups__fprintf_removed_maps(self, verbose, fp); +} + +int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, + int verbose, FILE *fp) +{ + struct rb_root *root = &self->maps[map->type]; + struct rb_node *next = rb_first(root); + + while (next) { + struct map *pos = rb_entry(next, struct map, rb_node); + next = rb_next(&pos->rb_node); + + if (!map__overlap(pos, map)) + continue; + + if (verbose >= 2) { + fputs("overlapping maps:\n", fp); + map__fprintf(map, fp); + map__fprintf(pos, fp); + } + + rb_erase(&pos->rb_node, root); + /* + * We may have references to this map, for instance in some + * hist_entry instances, so just move them to a separate + * list. + */ + list_add_tail(&pos->node, &self->removed_maps[map->type]); + /* + * Now check if we need to create new maps for areas not + * overlapped by the new map: + */ + if (map->start > pos->start) { + struct map *before = map__clone(pos); + + if (before == NULL) + return -ENOMEM; + + before->end = map->start - 1; + map_groups__insert(self, before); + if (verbose >= 2) + map__fprintf(before, fp); + } + + if (map->end < pos->end) { + struct map *after = map__clone(pos); + + if (after == NULL) + return -ENOMEM; + + after->start = map->end + 1; + map_groups__insert(self, after); + if (verbose >= 2) + map__fprintf(after, fp); + } + } + + return 0; +} + +/* + * XXX This should not really _copy_ te maps, but refcount them. + */ +int map_groups__clone(struct map_groups *self, + struct map_groups *parent, enum map_type type) +{ + struct rb_node *nd; + for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) { + struct map *map = rb_entry(nd, struct map, rb_node); + struct map *new = map__clone(map); + if (new == NULL) + return -ENOMEM; + map_groups__insert(self, new); + } + return 0; +} + +static u64 map__reloc_map_ip(struct map *map, u64 ip) +{ + return ip + (s64)map->pgoff; +} + +static u64 map__reloc_unmap_ip(struct map *map, u64 ip) +{ + return ip - (s64)map->pgoff; +} + +void map__reloc_vmlinux(struct map *self) +{ + struct kmap *kmap = map__kmap(self); + s64 reloc; + + if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr) + return; + + reloc = (kmap->ref_reloc_sym->unrelocated_addr - + kmap->ref_reloc_sym->addr); + + if (!reloc) + return; + + self->map_ip = map__reloc_map_ip; + self->unmap_ip = map__reloc_unmap_ip; + self->pgoff = reloc; +} + +void maps__insert(struct rb_root *maps, struct map *map) +{ + struct rb_node **p = &maps->rb_node; + struct rb_node *parent = NULL; + const u64 ip = map->start; + struct map *m; + + while (*p != NULL) { + parent = *p; + m = rb_entry(parent, struct map, rb_node); + if (ip < m->start) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&map->rb_node, parent, p); + rb_insert_color(&map->rb_node, maps); +} + +struct map *maps__find(struct rb_root *maps, u64 ip) +{ + struct rb_node **p = &maps->rb_node; + struct rb_node *parent = NULL; + struct map *m; + + while (*p != NULL) { + parent = *p; + m = rb_entry(parent, struct map, rb_node); + if (ip < m->start) + p = &(*p)->rb_left; + else if (ip > m->end) + p = &(*p)->rb_right; + else + return m; + } + + return NULL; +} diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index b756368076c6..2031278cc06a 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -4,7 +4,8 @@ #include <linux/compiler.h> #include <linux/list.h> #include <linux/rbtree.h> -#include <linux/types.h> +#include <stdio.h> +#include "types.h" enum map_type { MAP__FUNCTION = 0, @@ -68,14 +69,13 @@ u64 map__rip_2objdump(struct map *map, u64 rip); u64 map__objdump_2ip(struct map *map, u64 addr); struct symbol; -struct mmap_event; typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym); void map__init(struct map *self, enum map_type type, u64 start, u64 end, u64 pgoff, struct dso *dso); -struct map *map__new(struct mmap_event *event, enum map_type, - char *cwd, int cwdlen); +struct map *map__new(u64 start, u64 len, u64 pgoff, u32 pid, char *filename, + enum map_type type, char *cwd, int cwdlen); void map__delete(struct map *self); struct map *map__clone(struct map *self); int map__overlap(struct map *l, struct map *r); @@ -91,4 +91,70 @@ void map__fixup_end(struct map *self); void map__reloc_vmlinux(struct map *self); +struct map_groups { + struct rb_root maps[MAP__NR_TYPES]; + struct list_head removed_maps[MAP__NR_TYPES]; +}; + +size_t __map_groups__fprintf_maps(struct map_groups *self, + enum map_type type, int verbose, FILE *fp); +void maps__insert(struct rb_root *maps, struct map *map); +struct map *maps__find(struct rb_root *maps, u64 addr); +void map_groups__init(struct map_groups *self); +int map_groups__clone(struct map_groups *self, + struct map_groups *parent, enum map_type type); +size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp); +size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp); + +static inline void map_groups__insert(struct map_groups *self, struct map *map) +{ + maps__insert(&self->maps[map->type], map); +} + +static inline struct map *map_groups__find(struct map_groups *self, + enum map_type type, u64 addr) +{ + return maps__find(&self->maps[type], addr); +} + +struct symbol *map_groups__find_symbol(struct map_groups *self, + enum map_type type, u64 addr, + struct map **mapp, + symbol_filter_t filter); + +struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, + enum map_type type, + const char *name, + struct map **mapp, + symbol_filter_t filter); + +static inline +struct symbol *map_groups__find_function(struct map_groups *self, u64 addr, + struct map **mapp, symbol_filter_t filter) +{ + return map_groups__find_symbol(self, MAP__FUNCTION, addr, mapp, filter); +} + +static inline +struct symbol *map_groups__find_function_by_name(struct map_groups *self, + const char *name, struct map **mapp, + symbol_filter_t filter) +{ + return map_groups__find_symbol_by_name(self, MAP__FUNCTION, name, mapp, filter); +} + +int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, + int verbose, FILE *fp); + +struct map *map_groups__find_by_name(struct map_groups *self, + enum map_type type, const char *name); +int __map_groups__create_kernel_maps(struct map_groups *self, + struct map *vmlinux_maps[MAP__NR_TYPES], + struct dso *kernel); +int map_groups__create_kernel_maps(struct map_groups *self, + struct map *vmlinux_maps[MAP__NR_TYPES]); +struct map *map_groups__new_module(struct map_groups *self, u64 start, + const char *filename); +void map_groups__flush(struct map_groups *self); + #endif /* __PERF_MAP_H */ diff --git a/tools/perf/util/newt.c b/tools/perf/util/newt.c new file mode 100644 index 000000000000..7a123a94e3fc --- /dev/null +++ b/tools/perf/util/newt.c @@ -0,0 +1,725 @@ +#define _GNU_SOURCE +#include <stdio.h> +#undef _GNU_SOURCE + +#include <stdlib.h> +#include <newt.h> +#include <sys/ttydefaults.h> + +#include "cache.h" +#include "hist.h" +#include "session.h" +#include "sort.h" +#include "symbol.h" + +struct ui_progress { + newtComponent form, scale; +}; + +struct ui_progress *ui_progress__new(const char *title, u64 total) +{ + struct ui_progress *self = malloc(sizeof(*self)); + + if (self != NULL) { + int cols; + newtGetScreenSize(&cols, NULL); + cols -= 4; + newtCenteredWindow(cols, 1, title); + self->form = newtForm(NULL, NULL, 0); + if (self->form == NULL) + goto out_free_self; + self->scale = newtScale(0, 0, cols, total); + if (self->scale == NULL) + goto out_free_form; + newtFormAddComponents(self->form, self->scale, NULL); + newtRefresh(); + } + + return self; + +out_free_form: + newtFormDestroy(self->form); +out_free_self: + free(self); + return NULL; +} + +void ui_progress__update(struct ui_progress *self, u64 curr) +{ + newtScaleSet(self->scale, curr); + newtRefresh(); +} + +void ui_progress__delete(struct ui_progress *self) +{ + newtFormDestroy(self->form); + newtPopWindow(); + free(self); +} + +static char browser__last_msg[1024]; + +int browser__show_help(const char *format, va_list ap) +{ + int ret; + static int backlog; + + ret = vsnprintf(browser__last_msg + backlog, + sizeof(browser__last_msg) - backlog, format, ap); + backlog += ret; + + if (browser__last_msg[backlog - 1] == '\n') { + newtPopHelpLine(); + newtPushHelpLine(browser__last_msg); + newtRefresh(); + backlog = 0; + } + + return ret; +} + +static void newt_form__set_exit_keys(newtComponent self) +{ + newtFormAddHotKey(self, NEWT_KEY_ESCAPE); + newtFormAddHotKey(self, 'Q'); + newtFormAddHotKey(self, 'q'); + newtFormAddHotKey(self, CTRL('c')); +} + +static newtComponent newt_form__new(void) +{ + newtComponent self = newtForm(NULL, NULL, 0); + if (self) + newt_form__set_exit_keys(self); + return self; +} + +static int popup_menu(int argc, char * const argv[]) +{ + struct newtExitStruct es; + int i, rc = -1, max_len = 5; + newtComponent listbox, form = newt_form__new(); + + if (form == NULL) + return -1; + + listbox = newtListbox(0, 0, argc, NEWT_FLAG_RETURNEXIT); + if (listbox == NULL) + goto out_destroy_form; + + newtFormAddComponents(form, listbox, NULL); + + for (i = 0; i < argc; ++i) { + int len = strlen(argv[i]); + if (len > max_len) + max_len = len; + if (newtListboxAddEntry(listbox, argv[i], (void *)(long)i)) + goto out_destroy_form; + } + + newtCenteredWindow(max_len, argc, NULL); + newtFormRun(form, &es); + rc = newtListboxGetCurrent(listbox) - NULL; + if (es.reason == NEWT_EXIT_HOTKEY) + rc = -1; + newtPopWindow(); +out_destroy_form: + newtFormDestroy(form); + return rc; +} + +static bool dialog_yesno(const char *msg) +{ + /* newtWinChoice should really be accepting const char pointers... */ + char yes[] = "Yes", no[] = "No"; + return newtWinChoice(NULL, yes, no, (char *)msg) == 1; +} + +/* + * When debugging newt problems it was useful to be able to "unroll" + * the calls to newtCheckBoxTreeAdd{Array,Item}, so that we can generate + * a source file with the sequence of calls to these methods, to then + * tweak the arrays to get the intended results, so I'm keeping this code + * here, may be useful again in the future. + */ +#undef NEWT_DEBUG + +static void newt_checkbox_tree__add(newtComponent tree, const char *str, + void *priv, int *indexes) +{ +#ifdef NEWT_DEBUG + /* Print the newtCheckboxTreeAddArray to tinker with its index arrays */ + int i = 0, len = 40 - strlen(str); + + fprintf(stderr, + "\tnewtCheckboxTreeAddItem(tree, %*.*s\"%s\", (void *)%p, 0, ", + len, len, " ", str, priv); + while (indexes[i] != NEWT_ARG_LAST) { + if (indexes[i] != NEWT_ARG_APPEND) + fprintf(stderr, " %d,", indexes[i]); + else + fprintf(stderr, " %s,", "NEWT_ARG_APPEND"); + ++i; + } + fprintf(stderr, " %s", " NEWT_ARG_LAST);\n"); + fflush(stderr); +#endif + newtCheckboxTreeAddArray(tree, str, priv, 0, indexes); +} + +static char *callchain_list__sym_name(struct callchain_list *self, + char *bf, size_t bfsize) +{ + if (self->ms.sym) + return self->ms.sym->name; + + snprintf(bf, bfsize, "%#Lx", self->ip); + return bf; +} + +static void __callchain__append_graph_browser(struct callchain_node *self, + newtComponent tree, u64 total, + int *indexes, int depth) +{ + struct rb_node *node; + u64 new_total, remaining; + int idx = 0; + + if (callchain_param.mode == CHAIN_GRAPH_REL) + new_total = self->children_hit; + else + new_total = total; + + remaining = new_total; + node = rb_first(&self->rb_root); + while (node) { + struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node); + struct rb_node *next = rb_next(node); + u64 cumul = cumul_hits(child); + struct callchain_list *chain; + int first = true, printed = 0; + int chain_idx = -1; + remaining -= cumul; + + indexes[depth] = NEWT_ARG_APPEND; + indexes[depth + 1] = NEWT_ARG_LAST; + + list_for_each_entry(chain, &child->val, list) { + char ipstr[BITS_PER_LONG / 4 + 1], + *alloc_str = NULL; + const char *str = callchain_list__sym_name(chain, ipstr, sizeof(ipstr)); + + if (first) { + double percent = cumul * 100.0 / new_total; + + first = false; + if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0) + str = "Not enough memory!"; + else + str = alloc_str; + } else { + indexes[depth] = idx; + indexes[depth + 1] = NEWT_ARG_APPEND; + indexes[depth + 2] = NEWT_ARG_LAST; + ++chain_idx; + } + newt_checkbox_tree__add(tree, str, &chain->ms, indexes); + free(alloc_str); + ++printed; + } + + indexes[depth] = idx; + if (chain_idx != -1) + indexes[depth + 1] = chain_idx; + if (printed != 0) + ++idx; + __callchain__append_graph_browser(child, tree, new_total, indexes, + depth + (chain_idx != -1 ? 2 : 1)); + node = next; + } +} + +static void callchain__append_graph_browser(struct callchain_node *self, + newtComponent tree, u64 total, + int *indexes, int parent_idx) +{ + struct callchain_list *chain; + int i = 0; + + indexes[1] = NEWT_ARG_APPEND; + indexes[2] = NEWT_ARG_LAST; + + list_for_each_entry(chain, &self->val, list) { + char ipstr[BITS_PER_LONG / 4 + 1], *str; + + if (chain->ip >= PERF_CONTEXT_MAX) + continue; + + if (!i++ && sort__first_dimension == SORT_SYM) + continue; + + str = callchain_list__sym_name(chain, ipstr, sizeof(ipstr)); + newt_checkbox_tree__add(tree, str, &chain->ms, indexes); + } + + indexes[1] = parent_idx; + indexes[2] = NEWT_ARG_APPEND; + indexes[3] = NEWT_ARG_LAST; + __callchain__append_graph_browser(self, tree, total, indexes, 2); +} + +static void hist_entry__append_callchain_browser(struct hist_entry *self, + newtComponent tree, u64 total, int parent_idx) +{ + struct rb_node *rb_node; + int indexes[1024] = { [0] = parent_idx, }; + int idx = 0; + struct callchain_node *chain; + + rb_node = rb_first(&self->sorted_chain); + while (rb_node) { + chain = rb_entry(rb_node, struct callchain_node, rb_node); + switch (callchain_param.mode) { + case CHAIN_FLAT: + break; + case CHAIN_GRAPH_ABS: /* falldown */ + case CHAIN_GRAPH_REL: + callchain__append_graph_browser(chain, tree, total, indexes, idx++); + break; + case CHAIN_NONE: + default: + break; + } + rb_node = rb_next(rb_node); + } +} + +static size_t hist_entry__append_browser(struct hist_entry *self, + newtComponent tree, u64 total) +{ + char s[256]; + size_t ret; + + if (symbol_conf.exclude_other && !self->parent) + return 0; + + ret = hist_entry__snprintf(self, s, sizeof(s), NULL, + false, 0, false, total); + if (symbol_conf.use_callchain) { + int indexes[2]; + + indexes[0] = NEWT_ARG_APPEND; + indexes[1] = NEWT_ARG_LAST; + newt_checkbox_tree__add(tree, s, &self->ms, indexes); + } else + newtListboxAppendEntry(tree, s, &self->ms); + + return ret; +} + +static void map_symbol__annotate_browser(const struct map_symbol *self, + const char *input_name) +{ + FILE *fp; + int cols, rows; + newtComponent form, tree; + struct newtExitStruct es; + char *str; + size_t line_len, max_line_len = 0; + size_t max_usable_width; + char *line = NULL; + + if (self->sym == NULL) + return; + + if (asprintf(&str, "perf annotate -i \"%s\" -d \"%s\" %s 2>&1 | expand", + input_name, self->map->dso->name, self->sym->name) < 0) + return; + + fp = popen(str, "r"); + if (fp == NULL) + goto out_free_str; + + newtPushHelpLine("Press ESC to exit"); + newtGetScreenSize(&cols, &rows); + tree = newtListbox(0, 0, rows - 5, NEWT_FLAG_SCROLL); + + while (!feof(fp)) { + if (getline(&line, &line_len, fp) < 0 || !line_len) + break; + while (line_len != 0 && isspace(line[line_len - 1])) + line[--line_len] = '\0'; + + if (line_len > max_line_len) + max_line_len = line_len; + newtListboxAppendEntry(tree, line, NULL); + } + fclose(fp); + free(line); + + max_usable_width = cols - 22; + if (max_line_len > max_usable_width) + max_line_len = max_usable_width; + + newtListboxSetWidth(tree, max_line_len); + + newtCenteredWindow(max_line_len + 2, rows - 5, self->sym->name); + form = newt_form__new(); + newtFormAddComponents(form, tree, NULL); + + newtFormRun(form, &es); + newtFormDestroy(form); + newtPopWindow(); + newtPopHelpLine(); +out_free_str: + free(str); +} + +static const void *newt__symbol_tree_get_current(newtComponent self) +{ + if (symbol_conf.use_callchain) + return newtCheckboxTreeGetCurrent(self); + return newtListboxGetCurrent(self); +} + +static void hist_browser__selection(newtComponent self, void *data) +{ + const struct map_symbol **symbol_ptr = data; + *symbol_ptr = newt__symbol_tree_get_current(self); +} + +struct hist_browser { + newtComponent form, tree; + const struct map_symbol *selection; +}; + +static struct hist_browser *hist_browser__new(void) +{ + struct hist_browser *self = malloc(sizeof(*self)); + + if (self != NULL) + self->form = NULL; + + return self; +} + +static void hist_browser__delete(struct hist_browser *self) +{ + newtFormDestroy(self->form); + newtPopWindow(); + free(self); +} + +static int hist_browser__populate(struct hist_browser *self, struct rb_root *hists, + u64 nr_hists, u64 session_total, const char *title) +{ + int max_len = 0, idx, cols, rows; + struct ui_progress *progress; + struct rb_node *nd; + u64 curr_hist = 0; + char seq[] = "."; + char str[256]; + + if (self->form) { + newtFormDestroy(self->form); + newtPopWindow(); + } + + snprintf(str, sizeof(str), "Samples: %Ld ", + session_total); + newtDrawRootText(0, 0, str); + + newtGetScreenSize(NULL, &rows); + + if (symbol_conf.use_callchain) + self->tree = newtCheckboxTreeMulti(0, 0, rows - 5, seq, + NEWT_FLAG_SCROLL); + else + self->tree = newtListbox(0, 0, rows - 5, + (NEWT_FLAG_SCROLL | + NEWT_FLAG_RETURNEXIT)); + + newtComponentAddCallback(self->tree, hist_browser__selection, + &self->selection); + + progress = ui_progress__new("Adding entries to the browser...", nr_hists); + if (progress == NULL) + return -1; + + idx = 0; + for (nd = rb_first(hists); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + int len; + + if (h->filtered) + continue; + + len = hist_entry__append_browser(h, self->tree, session_total); + if (len > max_len) + max_len = len; + if (symbol_conf.use_callchain) + hist_entry__append_callchain_browser(h, self->tree, + session_total, idx++); + ++curr_hist; + if (curr_hist % 5) + ui_progress__update(progress, curr_hist); + } + + ui_progress__delete(progress); + + newtGetScreenSize(&cols, &rows); + + if (max_len > cols) + max_len = cols - 3; + + if (!symbol_conf.use_callchain) + newtListboxSetWidth(self->tree, max_len); + + newtCenteredWindow(max_len + (symbol_conf.use_callchain ? 5 : 0), + rows - 5, title); + self->form = newt_form__new(); + if (self->form == NULL) + return -1; + + newtFormAddHotKey(self->form, 'A'); + newtFormAddHotKey(self->form, 'a'); + newtFormAddHotKey(self->form, NEWT_KEY_RIGHT); + newtFormAddComponents(self->form, self->tree, NULL); + self->selection = newt__symbol_tree_get_current(self->tree); + + return 0; +} + +enum hist_filter { + HIST_FILTER__DSO, + HIST_FILTER__THREAD, +}; + +static u64 hists__filter_by_dso(struct rb_root *hists, const struct dso *dso, + u64 *session_total) +{ + struct rb_node *nd; + u64 nr_hists = 0; + + *session_total = 0; + + for (nd = rb_first(hists); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (dso != NULL && (h->ms.map == NULL || h->ms.map->dso != dso)) { + h->filtered |= (1 << HIST_FILTER__DSO); + continue; + } + h->filtered &= ~(1 << HIST_FILTER__DSO); + ++nr_hists; + *session_total += h->count; + } + + return nr_hists; +} + +static u64 hists__filter_by_thread(struct rb_root *hists, const struct thread *thread, + u64 *session_total) +{ + struct rb_node *nd; + u64 nr_hists = 0; + + *session_total = 0; + + for (nd = rb_first(hists); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (thread != NULL && h->thread != thread) { + h->filtered |= (1 << HIST_FILTER__THREAD); + continue; + } + h->filtered &= ~(1 << HIST_FILTER__THREAD); + ++nr_hists; + *session_total += h->count; + } + + return nr_hists; +} + +static struct thread *hist_browser__selected_thread(struct hist_browser *self) +{ + int *indexes; + + if (!symbol_conf.use_callchain) + goto out; + + indexes = newtCheckboxTreeFindItem(self->tree, (void *)self->selection); + if (indexes) { + bool is_hist_entry = indexes[1] == NEWT_ARG_LAST; + free(indexes); + if (is_hist_entry) + goto out; + } + return NULL; +out: + return *(struct thread **)(self->selection + 1); +} + +static int hist_browser__title(char *bf, size_t size, const char *input_name, + const struct dso *dso, const struct thread *thread) +{ + int printed = 0; + + if (thread) + printed += snprintf(bf + printed, size - printed, + "Thread: %s(%d)", + (thread->comm_set ? thread->comm : ""), + thread->pid); + if (dso) + printed += snprintf(bf + printed, size - printed, + "%sDSO: %s", thread ? " " : "", + dso->short_name); + return printed ?: snprintf(bf, size, "Report: %s", input_name); +} + +int perf_session__browse_hists(struct rb_root *hists, u64 nr_hists, + u64 session_total, const char *helpline, + const char *input_name) +{ + struct hist_browser *browser = hist_browser__new(); + const struct thread *thread_filter = NULL; + const struct dso *dso_filter = NULL; + struct newtExitStruct es; + char msg[160]; + int err = -1; + + if (browser == NULL) + return -1; + + newtPushHelpLine(helpline); + + hist_browser__title(msg, sizeof(msg), input_name, + dso_filter, thread_filter); + if (hist_browser__populate(browser, hists, nr_hists, session_total, msg) < 0) + goto out; + + while (1) { + const struct thread *thread; + const struct dso *dso; + char *options[16]; + int nr_options = 0, choice = 0, i, + annotate = -2, zoom_dso = -2, zoom_thread = -2; + + newtFormRun(browser->form, &es); + if (es.reason == NEWT_EXIT_HOTKEY) { + if (toupper(es.u.key) == 'A') + goto do_annotate; + if (es.u.key == NEWT_KEY_ESCAPE || + toupper(es.u.key) == 'Q' || + es.u.key == CTRL('c')) { + if (dialog_yesno("Do you really want to exit?")) + break; + else + continue; + } + } + + if (browser->selection->sym != NULL && + asprintf(&options[nr_options], "Annotate %s", + browser->selection->sym->name) > 0) + annotate = nr_options++; + + thread = hist_browser__selected_thread(browser); + if (thread != NULL && + asprintf(&options[nr_options], "Zoom %s %s(%d) thread", + (thread_filter ? "out of" : "into"), + (thread->comm_set ? thread->comm : ""), + thread->pid) > 0) + zoom_thread = nr_options++; + + dso = browser->selection->map ? browser->selection->map->dso : NULL; + if (dso != NULL && + asprintf(&options[nr_options], "Zoom %s %s DSO", + (dso_filter ? "out of" : "into"), + (dso->kernel ? "the Kernel" : dso->short_name)) > 0) + zoom_dso = nr_options++; + + options[nr_options++] = (char *)"Exit"; + + choice = popup_menu(nr_options, options); + + for (i = 0; i < nr_options - 1; ++i) + free(options[i]); + + if (choice == nr_options - 1) + break; + + if (choice == -1) + continue; +do_annotate: + if (choice == annotate) { + if (browser->selection->map->dso->origin == DSO__ORIG_KERNEL) { + newtPopHelpLine(); + newtPushHelpLine("No vmlinux file found, can't " + "annotate with just a " + "kallsyms file"); + continue; + } + map_symbol__annotate_browser(browser->selection, input_name); + } else if (choice == zoom_dso) { + if (dso_filter) { + newtPopHelpLine(); + dso_filter = NULL; + } else { + snprintf(msg, sizeof(msg), + "To zoom out press -> + \"Zoom out of %s DSO\"", + dso->kernel ? "the Kernel" : dso->short_name); + newtPushHelpLine(msg); + dso_filter = dso; + } + nr_hists = hists__filter_by_dso(hists, dso_filter, &session_total); + hist_browser__title(msg, sizeof(msg), input_name, + dso_filter, thread_filter); + if (hist_browser__populate(browser, hists, nr_hists, session_total, msg) < 0) + goto out; + } else if (choice == zoom_thread) { + if (thread_filter) { + newtPopHelpLine(); + thread_filter = NULL; + } else { + snprintf(msg, sizeof(msg), + "To zoom out press -> + \"Zoom out of %s(%d) thread\"", + (thread->comm_set ? thread->comm : ""), + thread->pid); + newtPushHelpLine(msg); + thread_filter = thread; + } + nr_hists = hists__filter_by_thread(hists, thread_filter, &session_total); + hist_browser__title(msg, sizeof(msg), input_name, + dso_filter, thread_filter); + if (hist_browser__populate(browser, hists, nr_hists, session_total, msg) < 0) + goto out; + } + } + err = 0; +out: + hist_browser__delete(browser); + return err; +} + +void setup_browser(void) +{ + if (!isatty(1)) + return; + + use_browser = true; + newtInit(); + newtCls(); + newtPushHelpLine(" "); +} + +void exit_browser(bool wait_for_ok) +{ + if (use_browser) { + if (wait_for_ok) { + char title[] = "Fatal Error", ok[] = "Ok"; + newtWinMessage(title, ok, browser__last_msg); + } + newtFinished(); + } +} diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 05d0c5c2030c..3b4ec6797565 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -5,6 +5,7 @@ #include "parse-events.h" #include "exec_cmd.h" #include "string.h" +#include "symbol.h" #include "cache.h" #include "header.h" #include "debugfs.h" @@ -409,7 +410,6 @@ static enum event_result parse_single_tracepoint_event(char *sys_name, const char *evt_name, unsigned int evt_length, - char *flags, struct perf_event_attr *attr, const char **strp) { @@ -418,13 +418,11 @@ parse_single_tracepoint_event(char *sys_name, u64 id; int fd; - if (flags) { - if (!strncmp(flags, "record", strlen(flags))) { - attr->sample_type |= PERF_SAMPLE_RAW; - attr->sample_type |= PERF_SAMPLE_TIME; - attr->sample_type |= PERF_SAMPLE_CPU; - } - } + attr->sample_type |= PERF_SAMPLE_RAW; + attr->sample_type |= PERF_SAMPLE_TIME; + attr->sample_type |= PERF_SAMPLE_CPU; + + attr->sample_period = 1; snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path, sys_name, evt_name); @@ -532,8 +530,7 @@ static enum event_result parse_tracepoint_event(const char **strp, flags); } else return parse_single_tracepoint_event(sys_name, evt_name, - evt_length, flags, - attr, strp); + evt_length, attr, strp); } static enum event_result @@ -656,6 +653,10 @@ parse_raw_event(const char **strp, struct perf_event_attr *attr) return EVT_FAILED; n = hex2u64(str + 1, &config); if (n > 0) { + if (str[n+1] == 'p') { + attr->precise = 1; + n++; + } *strp = str + n + 1; attr->type = PERF_TYPE_RAW; attr->config = config; diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c index efebd5b476b3..ed887642460c 100644 --- a/tools/perf/util/parse-options.c +++ b/tools/perf/util/parse-options.c @@ -49,6 +49,7 @@ static int get_value(struct parse_opt_ctx_t *p, break; /* FALLTHROUGH */ case OPTION_BOOLEAN: + case OPTION_INCR: case OPTION_BIT: case OPTION_SET_INT: case OPTION_SET_PTR: @@ -73,6 +74,10 @@ static int get_value(struct parse_opt_ctx_t *p, return 0; case OPTION_BOOLEAN: + *(bool *)opt->value = unset ? false : true; + return 0; + + case OPTION_INCR: *(int *)opt->value = unset ? 0 : *(int *)opt->value + 1; return 0; @@ -478,6 +483,7 @@ int usage_with_options_internal(const char * const *usagestr, case OPTION_GROUP: case OPTION_BIT: case OPTION_BOOLEAN: + case OPTION_INCR: case OPTION_SET_INT: case OPTION_SET_PTR: case OPTION_LONG: @@ -500,6 +506,7 @@ int usage_with_options_internal(const char * const *usagestr, void usage_with_options(const char * const *usagestr, const struct option *opts) { + exit_browser(false); usage_with_options_internal(usagestr, opts, 0); exit(129); } diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h index 948805af43c2..b2da725f102a 100644 --- a/tools/perf/util/parse-options.h +++ b/tools/perf/util/parse-options.h @@ -8,7 +8,8 @@ enum parse_opt_type { OPTION_GROUP, /* options with no arguments */ OPTION_BIT, - OPTION_BOOLEAN, /* _INCR would have been a better name */ + OPTION_BOOLEAN, + OPTION_INCR, OPTION_SET_INT, OPTION_SET_PTR, /* options with arguments (usually) */ @@ -95,6 +96,7 @@ struct option { #define OPT_GROUP(h) { .type = OPTION_GROUP, .help = (h) } #define OPT_BIT(s, l, v, h, b) { .type = OPTION_BIT, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (b) } #define OPT_BOOLEAN(s, l, v, h) { .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), .value = (v), .help = (h) } +#define OPT_INCR(s, l, v, h) { .type = OPTION_INCR, .short_name = (s), .long_name = (l), .value = (v), .help = (h) } #define OPT_SET_INT(s, l, v, h, i) { .type = OPTION_SET_INT, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (i) } #define OPT_SET_PTR(s, l, v, h, p) { .type = OPTION_SET_PTR, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (p) } #define OPT_INTEGER(s, l, v, h) { .type = OPTION_INTEGER, .short_name = (s), .long_name = (l), .value = (v), .help = (h) } diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 7c004b6ef24f..5bf8ab034466 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -33,20 +33,27 @@ #include <limits.h> #undef _GNU_SOURCE +#include "util.h" #include "event.h" #include "string.h" #include "strlist.h" #include "debug.h" #include "cache.h" #include "color.h" -#include "parse-events.h" /* For debugfs_path */ +#include "symbol.h" +#include "thread.h" +#include "debugfs.h" +#include "trace-event.h" /* For __unused */ #include "probe-event.h" +#include "probe-finder.h" #define MAX_CMDLEN 256 #define MAX_PROBE_ARGS 128 #define PERFPROBE_GROUP "probe" -#define semantic_error(msg ...) die("Semantic error :" msg) +bool probe_event_dry_run; /* Dry run flag */ + +#define semantic_error(msg ...) pr_err("Semantic error :" msg) /* If there is no space to write, returns -E2BIG. */ static int e_snprintf(char *str, size_t size, const char *format, ...) @@ -64,7 +71,265 @@ static int e_snprintf(char *str, size_t size, const char *format, ...) return ret; } -void parse_line_range_desc(const char *arg, struct line_range *lr) +static char *synthesize_perf_probe_point(struct perf_probe_point *pp); +static struct map_groups kmap_groups; +static struct map *kmaps[MAP__NR_TYPES]; + +/* Initialize symbol maps and path of vmlinux */ +static int init_vmlinux(void) +{ + int ret; + + symbol_conf.sort_by_name = true; + if (symbol_conf.vmlinux_name == NULL) + symbol_conf.try_vmlinux_path = true; + else + pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name); + ret = symbol__init(); + if (ret < 0) { + pr_debug("Failed to init symbol map.\n"); + goto out; + } + + map_groups__init(&kmap_groups); + ret = map_groups__create_kernel_maps(&kmap_groups, kmaps); + if (ret < 0) + pr_debug("Failed to create kernel maps.\n"); + +out: + if (ret < 0) + pr_warning("Failed to init vmlinux path.\n"); + return ret; +} + +#ifdef DWARF_SUPPORT +static int open_vmlinux(void) +{ + if (map__load(kmaps[MAP__FUNCTION], NULL) < 0) { + pr_debug("Failed to load kernel map.\n"); + return -EINVAL; + } + pr_debug("Try to open %s\n", kmaps[MAP__FUNCTION]->dso->long_name); + return open(kmaps[MAP__FUNCTION]->dso->long_name, O_RDONLY); +} + +/* Convert trace point to probe point with debuginfo */ +static int convert_to_perf_probe_point(struct kprobe_trace_point *tp, + struct perf_probe_point *pp) +{ + struct symbol *sym; + int fd, ret = -ENOENT; + + sym = map__find_symbol_by_name(kmaps[MAP__FUNCTION], + tp->symbol, NULL); + if (sym) { + fd = open_vmlinux(); + if (fd >= 0) { + ret = find_perf_probe_point(fd, + sym->start + tp->offset, pp); + close(fd); + } + } + if (ret <= 0) { + pr_debug("Failed to find corresponding probes from " + "debuginfo. Use kprobe event information.\n"); + pp->function = strdup(tp->symbol); + if (pp->function == NULL) + return -ENOMEM; + pp->offset = tp->offset; + } + pp->retprobe = tp->retprobe; + + return 0; +} + +/* Try to find perf_probe_event with debuginfo */ +static int try_to_find_kprobe_trace_events(struct perf_probe_event *pev, + struct kprobe_trace_event **tevs) +{ + bool need_dwarf = perf_probe_event_need_dwarf(pev); + int fd, ntevs; + + fd = open_vmlinux(); + if (fd < 0) { + if (need_dwarf) { + pr_warning("Failed to open debuginfo file.\n"); + return fd; + } + pr_debug("Could not open vmlinux. Try to use symbols.\n"); + return 0; + } + + /* Searching trace events corresponding to probe event */ + ntevs = find_kprobe_trace_events(fd, pev, tevs); + close(fd); + + if (ntevs > 0) { /* Succeeded to find trace events */ + pr_debug("find %d kprobe_trace_events.\n", ntevs); + return ntevs; + } + + if (ntevs == 0) { /* No error but failed to find probe point. */ + pr_warning("Probe point '%s' not found.\n", + synthesize_perf_probe_point(&pev->point)); + return -ENOENT; + } + /* Error path : ntevs < 0 */ + if (need_dwarf) { + if (ntevs == -EBADF) + pr_warning("No dwarf info found in the vmlinux - " + "please rebuild with CONFIG_DEBUG_INFO=y.\n"); + return ntevs; + } + pr_debug("An error occurred in debuginfo analysis." + " Try to use symbols.\n"); + return 0; +} + +#define LINEBUF_SIZE 256 +#define NR_ADDITIONAL_LINES 2 + +static int show_one_line(FILE *fp, int l, bool skip, bool show_num) +{ + char buf[LINEBUF_SIZE]; + const char *color = PERF_COLOR_BLUE; + + if (fgets(buf, LINEBUF_SIZE, fp) == NULL) + goto error; + if (!skip) { + if (show_num) + fprintf(stdout, "%7d %s", l, buf); + else + color_fprintf(stdout, color, " %s", buf); + } + + while (strlen(buf) == LINEBUF_SIZE - 1 && + buf[LINEBUF_SIZE - 2] != '\n') { + if (fgets(buf, LINEBUF_SIZE, fp) == NULL) + goto error; + if (!skip) { + if (show_num) + fprintf(stdout, "%s", buf); + else + color_fprintf(stdout, color, "%s", buf); + } + } + + return 0; +error: + if (feof(fp)) + pr_warning("Source file is shorter than expected.\n"); + else + pr_warning("File read error: %s\n", strerror(errno)); + + return -1; +} + +/* + * Show line-range always requires debuginfo to find source file and + * line number. + */ +int show_line_range(struct line_range *lr) +{ + int l = 1; + struct line_node *ln; + FILE *fp; + int fd, ret; + + /* Search a line range */ + ret = init_vmlinux(); + if (ret < 0) + return ret; + + fd = open_vmlinux(); + if (fd < 0) { + pr_warning("Failed to open debuginfo file.\n"); + return fd; + } + + ret = find_line_range(fd, lr); + close(fd); + if (ret == 0) { + pr_warning("Specified source line is not found.\n"); + return -ENOENT; + } else if (ret < 0) { + pr_warning("Debuginfo analysis failed. (%d)\n", ret); + return ret; + } + + setup_pager(); + + if (lr->function) + fprintf(stdout, "<%s:%d>\n", lr->function, + lr->start - lr->offset); + else + fprintf(stdout, "<%s:%d>\n", lr->file, lr->start); + + fp = fopen(lr->path, "r"); + if (fp == NULL) { + pr_warning("Failed to open %s: %s\n", lr->path, + strerror(errno)); + return -errno; + } + /* Skip to starting line number */ + while (l < lr->start && ret >= 0) + ret = show_one_line(fp, l++, true, false); + if (ret < 0) + goto end; + + list_for_each_entry(ln, &lr->line_list, list) { + while (ln->line > l && ret >= 0) + ret = show_one_line(fp, (l++) - lr->offset, + false, false); + if (ret >= 0) + ret = show_one_line(fp, (l++) - lr->offset, + false, true); + if (ret < 0) + goto end; + } + + if (lr->end == INT_MAX) + lr->end = l + NR_ADDITIONAL_LINES; + while (l <= lr->end && !feof(fp) && ret >= 0) + ret = show_one_line(fp, (l++) - lr->offset, false, false); +end: + fclose(fp); + return ret; +} + +#else /* !DWARF_SUPPORT */ + +static int convert_to_perf_probe_point(struct kprobe_trace_point *tp, + struct perf_probe_point *pp) +{ + pp->function = strdup(tp->symbol); + if (pp->function == NULL) + return -ENOMEM; + pp->offset = tp->offset; + pp->retprobe = tp->retprobe; + + return 0; +} + +static int try_to_find_kprobe_trace_events(struct perf_probe_event *pev, + struct kprobe_trace_event **tevs __unused) +{ + if (perf_probe_event_need_dwarf(pev)) { + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; + } + return 0; +} + +int show_line_range(struct line_range *lr __unused) +{ + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; +} + +#endif + +int parse_line_range_desc(const char *arg, struct line_range *lr) { const char *ptr; char *tmp; @@ -75,29 +340,45 @@ void parse_line_range_desc(const char *arg, struct line_range *lr) */ ptr = strchr(arg, ':'); if (ptr) { - lr->start = (unsigned int)strtoul(ptr + 1, &tmp, 0); - if (*tmp == '+') - lr->end = lr->start + (unsigned int)strtoul(tmp + 1, - &tmp, 0); - else if (*tmp == '-') - lr->end = (unsigned int)strtoul(tmp + 1, &tmp, 0); + lr->start = (int)strtoul(ptr + 1, &tmp, 0); + if (*tmp == '+') { + lr->end = lr->start + (int)strtoul(tmp + 1, &tmp, 0); + lr->end--; /* + * Adjust the number of lines here. + * If the number of lines == 1, the + * the end of line should be equal to + * the start of line. + */ + } else if (*tmp == '-') + lr->end = (int)strtoul(tmp + 1, &tmp, 0); else - lr->end = 0; - pr_debug("Line range is %u to %u\n", lr->start, lr->end); - if (lr->end && lr->start > lr->end) + lr->end = INT_MAX; + pr_debug("Line range is %d to %d\n", lr->start, lr->end); + if (lr->start > lr->end) { semantic_error("Start line must be smaller" - " than end line."); - if (*tmp != '\0') - semantic_error("Tailing with invalid character '%d'.", + " than end line.\n"); + return -EINVAL; + } + if (*tmp != '\0') { + semantic_error("Tailing with invalid character '%d'.\n", *tmp); + return -EINVAL; + } tmp = strndup(arg, (ptr - arg)); - } else + } else { tmp = strdup(arg); + lr->end = INT_MAX; + } + + if (tmp == NULL) + return -ENOMEM; if (strchr(tmp, '.')) lr->file = tmp; else lr->function = tmp; + + return 0; } /* Check the name is good for event/group */ @@ -113,8 +394,9 @@ static bool check_event_name(const char *name) } /* Parse probepoint definition. */ -static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp) +static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) { + struct perf_probe_point *pp = &pev->point; char *ptr, *tmp; char c, nc = 0; /* @@ -129,13 +411,19 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp) if (ptr && *ptr == '=') { /* Event name */ *ptr = '\0'; tmp = ptr + 1; - ptr = strchr(arg, ':'); - if (ptr) /* Group name is not supported yet. */ - semantic_error("Group name is not supported yet."); - if (!check_event_name(arg)) + if (strchr(arg, ':')) { + semantic_error("Group name is not supported yet.\n"); + return -ENOTSUP; + } + if (!check_event_name(arg)) { semantic_error("%s is bad for event name -it must " - "follow C symbol-naming rule.", arg); - pp->event = strdup(arg); + "follow C symbol-naming rule.\n", arg); + return -EINVAL; + } + pev->event = strdup(arg); + if (pev->event == NULL) + return -ENOMEM; + pev->group = NULL; arg = tmp; } @@ -145,12 +433,15 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp) *ptr++ = '\0'; } + tmp = strdup(arg); + if (tmp == NULL) + return -ENOMEM; + /* Check arg is function or file and copy it */ - if (strchr(arg, '.')) /* File */ - pp->file = strdup(arg); + if (strchr(tmp, '.')) /* File */ + pp->file = tmp; else /* Function */ - pp->function = strdup(arg); - DIE_IF(pp->file == NULL && pp->function == NULL); + pp->function = tmp; /* Parse other options */ while (ptr) { @@ -158,6 +449,8 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp) c = nc; if (c == ';') { /* Lazy pattern must be the last part */ pp->lazy_line = strdup(arg); + if (pp->lazy_line == NULL) + return -ENOMEM; break; } ptr = strpbrk(arg, ";:+@%"); @@ -168,266 +461,658 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp) switch (c) { case ':': /* Line number */ pp->line = strtoul(arg, &tmp, 0); - if (*tmp != '\0') + if (*tmp != '\0') { semantic_error("There is non-digit char" - " in line number."); + " in line number.\n"); + return -EINVAL; + } break; case '+': /* Byte offset from a symbol */ pp->offset = strtoul(arg, &tmp, 0); - if (*tmp != '\0') + if (*tmp != '\0') { semantic_error("There is non-digit character" - " in offset."); + " in offset.\n"); + return -EINVAL; + } break; case '@': /* File name */ - if (pp->file) - semantic_error("SRC@SRC is not allowed."); + if (pp->file) { + semantic_error("SRC@SRC is not allowed.\n"); + return -EINVAL; + } pp->file = strdup(arg); - DIE_IF(pp->file == NULL); + if (pp->file == NULL) + return -ENOMEM; break; case '%': /* Probe places */ if (strcmp(arg, "return") == 0) { pp->retprobe = 1; - } else /* Others not supported yet */ - semantic_error("%%%s is not supported.", arg); + } else { /* Others not supported yet */ + semantic_error("%%%s is not supported.\n", arg); + return -ENOTSUP; + } break; - default: - DIE_IF("Program has a bug."); + default: /* Buggy case */ + pr_err("This program has a bug at %s:%d.\n", + __FILE__, __LINE__); + return -ENOTSUP; break; } } /* Exclusion check */ - if (pp->lazy_line && pp->line) + if (pp->lazy_line && pp->line) { semantic_error("Lazy pattern can't be used with line number."); + return -EINVAL; + } - if (pp->lazy_line && pp->offset) + if (pp->lazy_line && pp->offset) { semantic_error("Lazy pattern can't be used with offset."); + return -EINVAL; + } - if (pp->line && pp->offset) + if (pp->line && pp->offset) { semantic_error("Offset can't be used with line number."); + return -EINVAL; + } - if (!pp->line && !pp->lazy_line && pp->file && !pp->function) + if (!pp->line && !pp->lazy_line && pp->file && !pp->function) { semantic_error("File always requires line number or " "lazy pattern."); + return -EINVAL; + } - if (pp->offset && !pp->function) + if (pp->offset && !pp->function) { semantic_error("Offset requires an entry function."); + return -EINVAL; + } - if (pp->retprobe && !pp->function) + if (pp->retprobe && !pp->function) { semantic_error("Return probe requires an entry function."); + return -EINVAL; + } - if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) + if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) { semantic_error("Offset/Line/Lazy pattern can't be used with " "return probe."); + return -EINVAL; + } - pr_debug("symbol:%s file:%s line:%d offset:%d return:%d lazy:%s\n", + pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n", pp->function, pp->file, pp->line, pp->offset, pp->retprobe, pp->lazy_line); + return 0; } -/* Parse perf-probe event definition */ -void parse_perf_probe_event(const char *str, struct probe_point *pp, - bool *need_dwarf) +/* Parse perf-probe event argument */ +static int parse_perf_probe_arg(char *str, struct perf_probe_arg *arg) { - char **argv; - int argc, i; + char *tmp; + struct perf_probe_arg_field **fieldp; - *need_dwarf = false; + pr_debug("parsing arg: %s into ", str); - argv = argv_split(str, &argc); - if (!argv) - die("argv_split failed."); - if (argc > MAX_PROBE_ARGS + 1) - semantic_error("Too many arguments"); + tmp = strchr(str, '='); + if (tmp) { + arg->name = strndup(str, tmp - str); + if (arg->name == NULL) + return -ENOMEM; + pr_debug("name:%s ", arg->name); + str = tmp + 1; + } + + tmp = strchr(str, ':'); + if (tmp) { /* Type setting */ + *tmp = '\0'; + arg->type = strdup(tmp + 1); + if (arg->type == NULL) + return -ENOMEM; + pr_debug("type:%s ", arg->type); + } + + tmp = strpbrk(str, "-."); + if (!is_c_varname(str) || !tmp) { + /* A variable, register, symbol or special value */ + arg->var = strdup(str); + if (arg->var == NULL) + return -ENOMEM; + pr_debug("%s\n", arg->var); + return 0; + } + + /* Structure fields */ + arg->var = strndup(str, tmp - str); + if (arg->var == NULL) + return -ENOMEM; + pr_debug("%s, ", arg->var); + fieldp = &arg->field; + + do { + *fieldp = zalloc(sizeof(struct perf_probe_arg_field)); + if (*fieldp == NULL) + return -ENOMEM; + if (*tmp == '.') { + str = tmp + 1; + (*fieldp)->ref = false; + } else if (tmp[1] == '>') { + str = tmp + 2; + (*fieldp)->ref = true; + } else { + semantic_error("Argument parse error: %s\n", str); + return -EINVAL; + } + + tmp = strpbrk(str, "-."); + if (tmp) { + (*fieldp)->name = strndup(str, tmp - str); + if ((*fieldp)->name == NULL) + return -ENOMEM; + pr_debug("%s(%d), ", (*fieldp)->name, (*fieldp)->ref); + fieldp = &(*fieldp)->next; + } + } while (tmp); + (*fieldp)->name = strdup(str); + if ((*fieldp)->name == NULL) + return -ENOMEM; + pr_debug("%s(%d)\n", (*fieldp)->name, (*fieldp)->ref); + + /* If no name is specified, set the last field name */ + if (!arg->name) { + arg->name = strdup((*fieldp)->name); + if (arg->name == NULL) + return -ENOMEM; + } + return 0; +} + +/* Parse perf-probe event command */ +int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev) +{ + char **argv; + int argc, i, ret = 0; + argv = argv_split(cmd, &argc); + if (!argv) { + pr_debug("Failed to split arguments.\n"); + return -ENOMEM; + } + if (argc - 1 > MAX_PROBE_ARGS) { + semantic_error("Too many probe arguments (%d).\n", argc - 1); + ret = -ERANGE; + goto out; + } /* Parse probe point */ - parse_perf_probe_probepoint(argv[0], pp); - if (pp->file || pp->line || pp->lazy_line) - *need_dwarf = true; + ret = parse_perf_probe_point(argv[0], pev); + if (ret < 0) + goto out; /* Copy arguments and ensure return probe has no C argument */ - pp->nr_args = argc - 1; - pp->args = zalloc(sizeof(char *) * pp->nr_args); - for (i = 0; i < pp->nr_args; i++) { - pp->args[i] = strdup(argv[i + 1]); - if (!pp->args[i]) - die("Failed to copy argument."); - if (is_c_varname(pp->args[i])) { - if (pp->retprobe) - semantic_error("You can't specify local" - " variable for kretprobe"); - *need_dwarf = true; + pev->nargs = argc - 1; + pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs); + if (pev->args == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < pev->nargs && ret >= 0; i++) { + ret = parse_perf_probe_arg(argv[i + 1], &pev->args[i]); + if (ret >= 0 && + is_c_varname(pev->args[i].var) && pev->point.retprobe) { + semantic_error("You can't specify local variable for" + " kretprobe.\n"); + ret = -EINVAL; } } - +out: argv_free(argv); + + return ret; +} + +/* Return true if this perf_probe_event requires debuginfo */ +bool perf_probe_event_need_dwarf(struct perf_probe_event *pev) +{ + int i; + + if (pev->point.file || pev->point.line || pev->point.lazy_line) + return true; + + for (i = 0; i < pev->nargs; i++) + if (is_c_varname(pev->args[i].var)) + return true; + + return false; } /* Parse kprobe_events event into struct probe_point */ -void parse_trace_kprobe_event(const char *str, struct probe_point *pp) +int parse_kprobe_trace_command(const char *cmd, struct kprobe_trace_event *tev) { + struct kprobe_trace_point *tp = &tev->point; char pr; char *p; int ret, i, argc; char **argv; - pr_debug("Parsing kprobe_events: %s\n", str); - argv = argv_split(str, &argc); - if (!argv) - die("argv_split failed."); - if (argc < 2) - semantic_error("Too less arguments."); + pr_debug("Parsing kprobe_events: %s\n", cmd); + argv = argv_split(cmd, &argc); + if (!argv) { + pr_debug("Failed to split arguments.\n"); + return -ENOMEM; + } + if (argc < 2) { + semantic_error("Too few probe arguments.\n"); + ret = -ERANGE; + goto out; + } /* Scan event and group name. */ ret = sscanf(argv[0], "%c:%a[^/ \t]/%a[^ \t]", - &pr, (float *)(void *)&pp->group, - (float *)(void *)&pp->event); - if (ret != 3) - semantic_error("Failed to parse event name: %s", argv[0]); - pr_debug("Group:%s Event:%s probe:%c\n", pp->group, pp->event, pr); + &pr, (float *)(void *)&tev->group, + (float *)(void *)&tev->event); + if (ret != 3) { + semantic_error("Failed to parse event name: %s\n", argv[0]); + ret = -EINVAL; + goto out; + } + pr_debug("Group:%s Event:%s probe:%c\n", tev->group, tev->event, pr); - pp->retprobe = (pr == 'r'); + tp->retprobe = (pr == 'r'); /* Scan function name and offset */ - ret = sscanf(argv[1], "%a[^+]+%d", (float *)(void *)&pp->function, - &pp->offset); + ret = sscanf(argv[1], "%a[^+]+%lu", (float *)(void *)&tp->symbol, + &tp->offset); if (ret == 1) - pp->offset = 0; - - /* kprobe_events doesn't have this information */ - pp->line = 0; - pp->file = NULL; + tp->offset = 0; - pp->nr_args = argc - 2; - pp->args = zalloc(sizeof(char *) * pp->nr_args); - for (i = 0; i < pp->nr_args; i++) { + tev->nargs = argc - 2; + tev->args = zalloc(sizeof(struct kprobe_trace_arg) * tev->nargs); + if (tev->args == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < tev->nargs; i++) { p = strchr(argv[i + 2], '='); if (p) /* We don't need which register is assigned. */ - *p = '\0'; - pp->args[i] = strdup(argv[i + 2]); - if (!pp->args[i]) - die("Failed to copy argument."); + *p++ = '\0'; + else + p = argv[i + 2]; + tev->args[i].name = strdup(argv[i + 2]); + /* TODO: parse regs and offset */ + tev->args[i].value = strdup(p); + if (tev->args[i].name == NULL || tev->args[i].value == NULL) { + ret = -ENOMEM; + goto out; + } } - + ret = 0; +out: argv_free(argv); + return ret; } -/* Synthesize only probe point (not argument) */ -int synthesize_perf_probe_point(struct probe_point *pp) +/* Compose only probe arg */ +int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len) { - char *buf; - char offs[64] = "", line[64] = ""; + struct perf_probe_arg_field *field = pa->field; int ret; + char *tmp = buf; - pp->probes[0] = buf = zalloc(MAX_CMDLEN); - pp->found = 1; - if (!buf) - die("Failed to allocate memory by zalloc."); + if (pa->name && pa->var) + ret = e_snprintf(tmp, len, "%s=%s", pa->name, pa->var); + else + ret = e_snprintf(tmp, len, "%s", pa->name ? pa->name : pa->var); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + + while (field) { + ret = e_snprintf(tmp, len, "%s%s", field->ref ? "->" : ".", + field->name); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + field = field->next; + } + + if (pa->type) { + ret = e_snprintf(tmp, len, ":%s", pa->type); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + } + + return tmp - buf; +error: + pr_debug("Failed to synthesize perf probe argument: %s", + strerror(-ret)); + return ret; +} + +/* Compose only probe point (not argument) */ +static char *synthesize_perf_probe_point(struct perf_probe_point *pp) +{ + char *buf, *tmp; + char offs[32] = "", line[32] = "", file[32] = ""; + int ret, len; + + buf = zalloc(MAX_CMDLEN); + if (buf == NULL) { + ret = -ENOMEM; + goto error; + } if (pp->offset) { - ret = e_snprintf(offs, 64, "+%d", pp->offset); + ret = e_snprintf(offs, 32, "+%lu", pp->offset); if (ret <= 0) goto error; } if (pp->line) { - ret = e_snprintf(line, 64, ":%d", pp->line); + ret = e_snprintf(line, 32, ":%d", pp->line); + if (ret <= 0) + goto error; + } + if (pp->file) { + len = strlen(pp->file) - 31; + if (len < 0) + len = 0; + tmp = strchr(pp->file + len, '/'); + if (!tmp) + tmp = pp->file + len; + ret = e_snprintf(file, 32, "@%s", tmp + 1); if (ret <= 0) goto error; } if (pp->function) - ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s", pp->function, - offs, pp->retprobe ? "%return" : "", line); + ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s%s", pp->function, + offs, pp->retprobe ? "%return" : "", line, + file); else - ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", pp->file, line); - if (ret <= 0) { + ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", file, line); + if (ret <= 0) + goto error; + + return buf; error: - free(pp->probes[0]); - pp->probes[0] = NULL; - pp->found = 0; - } - return ret; + pr_debug("Failed to synthesize perf probe point: %s", + strerror(-ret)); + if (buf) + free(buf); + return NULL; } -int synthesize_perf_probe_event(struct probe_point *pp) +#if 0 +char *synthesize_perf_probe_command(struct perf_probe_event *pev) { char *buf; int i, len, ret; - len = synthesize_perf_probe_point(pp); - if (len < 0) - return 0; + buf = synthesize_perf_probe_point(&pev->point); + if (!buf) + return NULL; - buf = pp->probes[0]; - for (i = 0; i < pp->nr_args; i++) { + len = strlen(buf); + for (i = 0; i < pev->nargs; i++) { ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s", - pp->args[i]); - if (ret <= 0) - goto error; + pev->args[i].name); + if (ret <= 0) { + free(buf); + return NULL; + } len += ret; } - pp->found = 1; - return pp->found; -error: - free(pp->probes[0]); - pp->probes[0] = NULL; + return buf; +} +#endif + +static int __synthesize_kprobe_trace_arg_ref(struct kprobe_trace_arg_ref *ref, + char **buf, size_t *buflen, + int depth) +{ + int ret; + if (ref->next) { + depth = __synthesize_kprobe_trace_arg_ref(ref->next, buf, + buflen, depth + 1); + if (depth < 0) + goto out; + } + + ret = e_snprintf(*buf, *buflen, "%+ld(", ref->offset); + if (ret < 0) + depth = ret; + else { + *buf += ret; + *buflen -= ret; + } +out: + return depth; - return ret; } -int synthesize_trace_kprobe_event(struct probe_point *pp) +static int synthesize_kprobe_trace_arg(struct kprobe_trace_arg *arg, + char *buf, size_t buflen) { + int ret, depth = 0; + char *tmp = buf; + + /* Argument name or separator */ + if (arg->name) + ret = e_snprintf(buf, buflen, " %s=", arg->name); + else + ret = e_snprintf(buf, buflen, " "); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + + /* Dereferencing arguments */ + if (arg->ref) { + depth = __synthesize_kprobe_trace_arg_ref(arg->ref, &buf, + &buflen, 1); + if (depth < 0) + return depth; + } + + /* Print argument value */ + ret = e_snprintf(buf, buflen, "%s", arg->value); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + + /* Closing */ + while (depth--) { + ret = e_snprintf(buf, buflen, ")"); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + } + /* Print argument type */ + if (arg->type) { + ret = e_snprintf(buf, buflen, ":%s", arg->type); + if (ret <= 0) + return ret; + buf += ret; + } + + return buf - tmp; +} + +char *synthesize_kprobe_trace_command(struct kprobe_trace_event *tev) +{ + struct kprobe_trace_point *tp = &tev->point; char *buf; int i, len, ret; - pp->probes[0] = buf = zalloc(MAX_CMDLEN); - if (!buf) - die("Failed to allocate memory by zalloc."); - ret = e_snprintf(buf, MAX_CMDLEN, "%s+%d", pp->function, pp->offset); - if (ret <= 0) + buf = zalloc(MAX_CMDLEN); + if (buf == NULL) + return NULL; + + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->symbol, tp->offset); + if (len <= 0) goto error; - len = ret; - for (i = 0; i < pp->nr_args; i++) { - ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s", - pp->args[i]); + for (i = 0; i < tev->nargs; i++) { + ret = synthesize_kprobe_trace_arg(&tev->args[i], buf + len, + MAX_CMDLEN - len); if (ret <= 0) goto error; len += ret; } - pp->found = 1; - return pp->found; + return buf; error: - free(pp->probes[0]); - pp->probes[0] = NULL; + free(buf); + return NULL; +} + +int convert_to_perf_probe_event(struct kprobe_trace_event *tev, + struct perf_probe_event *pev) +{ + char buf[64] = ""; + int i, ret; + + /* Convert event/group name */ + pev->event = strdup(tev->event); + pev->group = strdup(tev->group); + if (pev->event == NULL || pev->group == NULL) + return -ENOMEM; + + /* Convert trace_point to probe_point */ + ret = convert_to_perf_probe_point(&tev->point, &pev->point); + if (ret < 0) + return ret; + + /* Convert trace_arg to probe_arg */ + pev->nargs = tev->nargs; + pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs); + if (pev->args == NULL) + return -ENOMEM; + for (i = 0; i < tev->nargs && ret >= 0; i++) { + if (tev->args[i].name) + pev->args[i].name = strdup(tev->args[i].name); + else { + ret = synthesize_kprobe_trace_arg(&tev->args[i], + buf, 64); + pev->args[i].name = strdup(buf); + } + if (pev->args[i].name == NULL && ret >= 0) + ret = -ENOMEM; + } + + if (ret < 0) + clear_perf_probe_event(pev); return ret; } -static int open_kprobe_events(int flags, int mode) +void clear_perf_probe_event(struct perf_probe_event *pev) +{ + struct perf_probe_point *pp = &pev->point; + struct perf_probe_arg_field *field, *next; + int i; + + if (pev->event) + free(pev->event); + if (pev->group) + free(pev->group); + if (pp->file) + free(pp->file); + if (pp->function) + free(pp->function); + if (pp->lazy_line) + free(pp->lazy_line); + for (i = 0; i < pev->nargs; i++) { + if (pev->args[i].name) + free(pev->args[i].name); + if (pev->args[i].var) + free(pev->args[i].var); + if (pev->args[i].type) + free(pev->args[i].type); + field = pev->args[i].field; + while (field) { + next = field->next; + if (field->name) + free(field->name); + free(field); + field = next; + } + } + if (pev->args) + free(pev->args); + memset(pev, 0, sizeof(*pev)); +} + +void clear_kprobe_trace_event(struct kprobe_trace_event *tev) +{ + struct kprobe_trace_arg_ref *ref, *next; + int i; + + if (tev->event) + free(tev->event); + if (tev->group) + free(tev->group); + if (tev->point.symbol) + free(tev->point.symbol); + for (i = 0; i < tev->nargs; i++) { + if (tev->args[i].name) + free(tev->args[i].name); + if (tev->args[i].value) + free(tev->args[i].value); + if (tev->args[i].type) + free(tev->args[i].type); + ref = tev->args[i].ref; + while (ref) { + next = ref->next; + free(ref); + ref = next; + } + } + if (tev->args) + free(tev->args); + memset(tev, 0, sizeof(*tev)); +} + +static int open_kprobe_events(bool readwrite) { char buf[PATH_MAX]; + const char *__debugfs; int ret; - ret = e_snprintf(buf, PATH_MAX, "%s/../kprobe_events", debugfs_path); - if (ret < 0) - die("Failed to make kprobe_events path."); + __debugfs = debugfs_find_mountpoint(); + if (__debugfs == NULL) { + pr_warning("Debugfs is not mounted.\n"); + return -ENOENT; + } + + ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", __debugfs); + if (ret >= 0) { + pr_debug("Opening %s write=%d\n", buf, readwrite); + if (readwrite && !probe_event_dry_run) + ret = open(buf, O_RDWR, O_APPEND); + else + ret = open(buf, O_RDONLY, 0); + } - ret = open(buf, flags, mode); if (ret < 0) { if (errno == ENOENT) - die("kprobe_events file does not exist -" - " please rebuild with CONFIG_KPROBE_EVENT."); + pr_warning("kprobe_events file does not exist - please" + " rebuild kernel with CONFIG_KPROBE_EVENT.\n"); else - die("Could not open kprobe_events file: %s", - strerror(errno)); + pr_warning("Failed to open kprobe_events file: %s\n", + strerror(errno)); } return ret; } /* Get raw string list of current kprobe_events */ -static struct strlist *get_trace_kprobe_event_rawlist(int fd) +static struct strlist *get_kprobe_trace_command_rawlist(int fd) { int ret, idx; FILE *fp; @@ -447,271 +1132,485 @@ static struct strlist *get_trace_kprobe_event_rawlist(int fd) if (p[idx] == '\n') p[idx] = '\0'; ret = strlist__add(sl, buf); - if (ret < 0) - die("strlist__add failed: %s", strerror(-ret)); + if (ret < 0) { + pr_debug("strlist__add failed: %s\n", strerror(-ret)); + strlist__delete(sl); + return NULL; + } } fclose(fp); return sl; } -/* Free and zero clear probe_point */ -static void clear_probe_point(struct probe_point *pp) -{ - int i; - - if (pp->event) - free(pp->event); - if (pp->group) - free(pp->group); - if (pp->function) - free(pp->function); - if (pp->file) - free(pp->file); - if (pp->lazy_line) - free(pp->lazy_line); - for (i = 0; i < pp->nr_args; i++) - free(pp->args[i]); - if (pp->args) - free(pp->args); - for (i = 0; i < pp->found; i++) - free(pp->probes[i]); - memset(pp, 0, sizeof(*pp)); -} - /* Show an event */ -static void show_perf_probe_event(const char *event, const char *place, - struct probe_point *pp) +static int show_perf_probe_event(struct perf_probe_event *pev) { int i, ret; char buf[128]; + char *place; - ret = e_snprintf(buf, 128, "%s:%s", pp->group, event); + /* Synthesize only event probe point */ + place = synthesize_perf_probe_point(&pev->point); + if (!place) + return -EINVAL; + + ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event); if (ret < 0) - die("Failed to copy event: %s", strerror(-ret)); - printf(" %-40s (on %s", buf, place); + return ret; + + printf(" %-20s (on %s", buf, place); - if (pp->nr_args > 0) { + if (pev->nargs > 0) { printf(" with"); - for (i = 0; i < pp->nr_args; i++) - printf(" %s", pp->args[i]); + for (i = 0; i < pev->nargs; i++) { + ret = synthesize_perf_probe_arg(&pev->args[i], + buf, 128); + if (ret < 0) + break; + printf(" %s", buf); + } } printf(")\n"); + free(place); + return ret; } /* List up current perf-probe events */ -void show_perf_probe_events(void) +int show_perf_probe_events(void) { - int fd; - struct probe_point pp; + int fd, ret; + struct kprobe_trace_event tev; + struct perf_probe_event pev; struct strlist *rawlist; struct str_node *ent; setup_pager(); - memset(&pp, 0, sizeof(pp)); + ret = init_vmlinux(); + if (ret < 0) + return ret; + + memset(&tev, 0, sizeof(tev)); + memset(&pev, 0, sizeof(pev)); + + fd = open_kprobe_events(false); + if (fd < 0) + return fd; - fd = open_kprobe_events(O_RDONLY, 0); - rawlist = get_trace_kprobe_event_rawlist(fd); + rawlist = get_kprobe_trace_command_rawlist(fd); close(fd); + if (!rawlist) + return -ENOENT; strlist__for_each(ent, rawlist) { - parse_trace_kprobe_event(ent->s, &pp); - /* Synthesize only event probe point */ - synthesize_perf_probe_point(&pp); - /* Show an event */ - show_perf_probe_event(pp.event, pp.probes[0], &pp); - clear_probe_point(&pp); + ret = parse_kprobe_trace_command(ent->s, &tev); + if (ret >= 0) { + ret = convert_to_perf_probe_event(&tev, &pev); + if (ret >= 0) + ret = show_perf_probe_event(&pev); + } + clear_perf_probe_event(&pev); + clear_kprobe_trace_event(&tev); + if (ret < 0) + break; } - strlist__delete(rawlist); + + return ret; } /* Get current perf-probe event names */ -static struct strlist *get_perf_event_names(int fd, bool include_group) +static struct strlist *get_kprobe_trace_event_names(int fd, bool include_group) { char buf[128]; struct strlist *sl, *rawlist; struct str_node *ent; - struct probe_point pp; + struct kprobe_trace_event tev; + int ret = 0; - memset(&pp, 0, sizeof(pp)); - rawlist = get_trace_kprobe_event_rawlist(fd); + memset(&tev, 0, sizeof(tev)); + rawlist = get_kprobe_trace_command_rawlist(fd); sl = strlist__new(true, NULL); strlist__for_each(ent, rawlist) { - parse_trace_kprobe_event(ent->s, &pp); + ret = parse_kprobe_trace_command(ent->s, &tev); + if (ret < 0) + break; if (include_group) { - if (e_snprintf(buf, 128, "%s:%s", pp.group, - pp.event) < 0) - die("Failed to copy group:event name."); - strlist__add(sl, buf); + ret = e_snprintf(buf, 128, "%s:%s", tev.group, + tev.event); + if (ret >= 0) + ret = strlist__add(sl, buf); } else - strlist__add(sl, pp.event); - clear_probe_point(&pp); + ret = strlist__add(sl, tev.event); + clear_kprobe_trace_event(&tev); + if (ret < 0) + break; } - strlist__delete(rawlist); + if (ret < 0) { + strlist__delete(sl); + return NULL; + } return sl; } -static void write_trace_kprobe_event(int fd, const char *buf) +static int write_kprobe_trace_event(int fd, struct kprobe_trace_event *tev) { int ret; + char *buf = synthesize_kprobe_trace_command(tev); + + if (!buf) { + pr_debug("Failed to synthesize kprobe trace event.\n"); + return -EINVAL; + } pr_debug("Writing event: %s\n", buf); - ret = write(fd, buf, strlen(buf)); - if (ret <= 0) - die("Failed to write event: %s", strerror(errno)); + if (!probe_event_dry_run) { + ret = write(fd, buf, strlen(buf)); + if (ret <= 0) + pr_warning("Failed to write event: %s\n", + strerror(errno)); + } + free(buf); + return ret; } -static void get_new_event_name(char *buf, size_t len, const char *base, - struct strlist *namelist, bool allow_suffix) +static int get_new_event_name(char *buf, size_t len, const char *base, + struct strlist *namelist, bool allow_suffix) { int i, ret; /* Try no suffix */ ret = e_snprintf(buf, len, "%s", base); - if (ret < 0) - die("snprintf() failed: %s", strerror(-ret)); + if (ret < 0) { + pr_debug("snprintf() failed: %s\n", strerror(-ret)); + return ret; + } if (!strlist__has_entry(namelist, buf)) - return; + return 0; if (!allow_suffix) { pr_warning("Error: event \"%s\" already exists. " "(Use -f to force duplicates.)\n", base); - die("Can't add new event."); + return -EEXIST; } /* Try to add suffix */ for (i = 1; i < MAX_EVENT_INDEX; i++) { ret = e_snprintf(buf, len, "%s_%d", base, i); - if (ret < 0) - die("snprintf() failed: %s", strerror(-ret)); + if (ret < 0) { + pr_debug("snprintf() failed: %s\n", strerror(-ret)); + return ret; + } if (!strlist__has_entry(namelist, buf)) break; } - if (i == MAX_EVENT_INDEX) - die("Too many events are on the same function."); + if (i == MAX_EVENT_INDEX) { + pr_warning("Too many events are on the same function.\n"); + ret = -ERANGE; + } + + return ret; } -void add_trace_kprobe_events(struct probe_point *probes, int nr_probes, - bool force_add) +static int __add_kprobe_trace_events(struct perf_probe_event *pev, + struct kprobe_trace_event *tevs, + int ntevs, bool allow_suffix) { - int i, j, fd; - struct probe_point *pp; - char buf[MAX_CMDLEN]; - char event[64]; + int i, fd, ret; + struct kprobe_trace_event *tev = NULL; + char buf[64]; + const char *event, *group; struct strlist *namelist; - bool allow_suffix; - fd = open_kprobe_events(O_RDWR, O_APPEND); + fd = open_kprobe_events(true); + if (fd < 0) + return fd; /* Get current event names */ - namelist = get_perf_event_names(fd, false); - - for (j = 0; j < nr_probes; j++) { - pp = probes + j; - if (!pp->event) - pp->event = strdup(pp->function); - if (!pp->group) - pp->group = strdup(PERFPROBE_GROUP); - DIE_IF(!pp->event || !pp->group); - /* If force_add is true, suffix search is allowed */ - allow_suffix = force_add; - for (i = 0; i < pp->found; i++) { - /* Get an unused new event name */ - get_new_event_name(event, 64, pp->event, namelist, - allow_suffix); - snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s\n", - pp->retprobe ? 'r' : 'p', - pp->group, event, - pp->probes[i]); - write_trace_kprobe_event(fd, buf); - printf("Added new event:\n"); - /* Get the first parameter (probe-point) */ - sscanf(pp->probes[i], "%s", buf); - show_perf_probe_event(event, buf, pp); - /* Add added event name to namelist */ - strlist__add(namelist, event); - /* - * Probes after the first probe which comes from same - * user input are always allowed to add suffix, because - * there might be several addresses corresponding to - * one code line. - */ - allow_suffix = true; + namelist = get_kprobe_trace_event_names(fd, false); + if (!namelist) { + pr_debug("Failed to get current event list.\n"); + return -EIO; + } + + ret = 0; + printf("Add new event%s\n", (ntevs > 1) ? "s:" : ":"); + for (i = 0; i < ntevs; i++) { + tev = &tevs[i]; + if (pev->event) + event = pev->event; + else + if (pev->point.function) + event = pev->point.function; + else + event = tev->point.symbol; + if (pev->group) + group = pev->group; + else + group = PERFPROBE_GROUP; + + /* Get an unused new event name */ + ret = get_new_event_name(buf, 64, event, + namelist, allow_suffix); + if (ret < 0) + break; + event = buf; + + tev->event = strdup(event); + tev->group = strdup(group); + if (tev->event == NULL || tev->group == NULL) { + ret = -ENOMEM; + break; } + ret = write_kprobe_trace_event(fd, tev); + if (ret < 0) + break; + /* Add added event name to namelist */ + strlist__add(namelist, event); + + /* Trick here - save current event/group */ + event = pev->event; + group = pev->group; + pev->event = tev->event; + pev->group = tev->group; + show_perf_probe_event(pev); + /* Trick here - restore current event/group */ + pev->event = (char *)event; + pev->group = (char *)group; + + /* + * Probes after the first probe which comes from same + * user input are always allowed to add suffix, because + * there might be several addresses corresponding to + * one code line. + */ + allow_suffix = true; + } + + if (ret >= 0) { + /* Show how to use the event. */ + printf("\nYou can now use it on all perf tools, such as:\n\n"); + printf("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group, + tev->event); } - /* Show how to use the event. */ - printf("\nYou can now use it on all perf tools, such as:\n\n"); - printf("\tperf record -e %s:%s -a sleep 1\n\n", PERFPROBE_GROUP, event); strlist__delete(namelist); close(fd); + return ret; } -static void __del_trace_kprobe_event(int fd, struct str_node *ent) +static int convert_to_kprobe_trace_events(struct perf_probe_event *pev, + struct kprobe_trace_event **tevs) +{ + struct symbol *sym; + int ret = 0, i; + struct kprobe_trace_event *tev; + + /* Convert perf_probe_event with debuginfo */ + ret = try_to_find_kprobe_trace_events(pev, tevs); + if (ret != 0) + return ret; + + /* Allocate trace event buffer */ + tev = *tevs = zalloc(sizeof(struct kprobe_trace_event)); + if (tev == NULL) + return -ENOMEM; + + /* Copy parameters */ + tev->point.symbol = strdup(pev->point.function); + if (tev->point.symbol == NULL) { + ret = -ENOMEM; + goto error; + } + tev->point.offset = pev->point.offset; + tev->nargs = pev->nargs; + if (tev->nargs) { + tev->args = zalloc(sizeof(struct kprobe_trace_arg) + * tev->nargs); + if (tev->args == NULL) { + ret = -ENOMEM; + goto error; + } + for (i = 0; i < tev->nargs; i++) { + if (pev->args[i].name) { + tev->args[i].name = strdup(pev->args[i].name); + if (tev->args[i].name == NULL) { + ret = -ENOMEM; + goto error; + } + } + tev->args[i].value = strdup(pev->args[i].var); + if (tev->args[i].value == NULL) { + ret = -ENOMEM; + goto error; + } + if (pev->args[i].type) { + tev->args[i].type = strdup(pev->args[i].type); + if (tev->args[i].type == NULL) { + ret = -ENOMEM; + goto error; + } + } + } + } + + /* Currently just checking function name from symbol map */ + sym = map__find_symbol_by_name(kmaps[MAP__FUNCTION], + tev->point.symbol, NULL); + if (!sym) { + pr_warning("Kernel symbol \'%s\' not found.\n", + tev->point.symbol); + ret = -ENOENT; + goto error; + } + + return 1; +error: + clear_kprobe_trace_event(tev); + free(tev); + *tevs = NULL; + return ret; +} + +struct __event_package { + struct perf_probe_event *pev; + struct kprobe_trace_event *tevs; + int ntevs; +}; + +int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, + bool force_add) +{ + int i, j, ret; + struct __event_package *pkgs; + + pkgs = zalloc(sizeof(struct __event_package) * npevs); + if (pkgs == NULL) + return -ENOMEM; + + /* Init vmlinux path */ + ret = init_vmlinux(); + if (ret < 0) + return ret; + + /* Loop 1: convert all events */ + for (i = 0; i < npevs; i++) { + pkgs[i].pev = &pevs[i]; + /* Convert with or without debuginfo */ + ret = convert_to_kprobe_trace_events(pkgs[i].pev, + &pkgs[i].tevs); + if (ret < 0) + goto end; + pkgs[i].ntevs = ret; + } + + /* Loop 2: add all events */ + for (i = 0; i < npevs && ret >= 0; i++) + ret = __add_kprobe_trace_events(pkgs[i].pev, pkgs[i].tevs, + pkgs[i].ntevs, force_add); +end: + /* Loop 3: cleanup trace events */ + for (i = 0; i < npevs; i++) + for (j = 0; j < pkgs[i].ntevs; j++) + clear_kprobe_trace_event(&pkgs[i].tevs[j]); + + return ret; +} + +static int __del_trace_kprobe_event(int fd, struct str_node *ent) { char *p; char buf[128]; + int ret; /* Convert from perf-probe event to trace-kprobe event */ - if (e_snprintf(buf, 128, "-:%s", ent->s) < 0) - die("Failed to copy event."); + ret = e_snprintf(buf, 128, "-:%s", ent->s); + if (ret < 0) + goto error; + p = strchr(buf + 2, ':'); - if (!p) - die("Internal error: %s should have ':' but not.", ent->s); + if (!p) { + pr_debug("Internal error: %s should have ':' but not.\n", + ent->s); + ret = -ENOTSUP; + goto error; + } *p = '/'; - write_trace_kprobe_event(fd, buf); + pr_debug("Writing event: %s\n", buf); + ret = write(fd, buf, strlen(buf)); + if (ret < 0) + goto error; + printf("Remove event: %s\n", ent->s); + return 0; +error: + pr_warning("Failed to delete event: %s\n", strerror(-ret)); + return ret; } -static void del_trace_kprobe_event(int fd, const char *group, - const char *event, struct strlist *namelist) +static int del_trace_kprobe_event(int fd, const char *group, + const char *event, struct strlist *namelist) { char buf[128]; struct str_node *ent, *n; - int found = 0; + int found = 0, ret = 0; - if (e_snprintf(buf, 128, "%s:%s", group, event) < 0) - die("Failed to copy event."); + ret = e_snprintf(buf, 128, "%s:%s", group, event); + if (ret < 0) { + pr_err("Failed to copy event."); + return ret; + } if (strpbrk(buf, "*?")) { /* Glob-exp */ strlist__for_each_safe(ent, n, namelist) if (strglobmatch(ent->s, buf)) { found++; - __del_trace_kprobe_event(fd, ent); + ret = __del_trace_kprobe_event(fd, ent); + if (ret < 0) + break; strlist__remove(namelist, ent); } } else { ent = strlist__find(namelist, buf); if (ent) { found++; - __del_trace_kprobe_event(fd, ent); - strlist__remove(namelist, ent); + ret = __del_trace_kprobe_event(fd, ent); + if (ret >= 0) + strlist__remove(namelist, ent); } } - if (found == 0) - pr_info("Info: event \"%s\" does not exist, could not remove it.\n", buf); + if (found == 0 && ret >= 0) + pr_info("Info: Event \"%s\" does not exist.\n", buf); + + return ret; } -void del_trace_kprobe_events(struct strlist *dellist) +int del_perf_probe_events(struct strlist *dellist) { - int fd; + int fd, ret = 0; const char *group, *event; char *p, *str; struct str_node *ent; struct strlist *namelist; - fd = open_kprobe_events(O_RDWR, O_APPEND); + fd = open_kprobe_events(true); + if (fd < 0) + return fd; + /* Get current event names */ - namelist = get_perf_event_names(fd, true); + namelist = get_kprobe_trace_event_names(fd, true); + if (namelist == NULL) + return -EINVAL; strlist__for_each(ent, dellist) { str = strdup(ent->s); - if (!str) - die("Failed to copy event."); + if (str == NULL) { + ret = -ENOMEM; + break; + } pr_debug("Parsing: %s\n", str); p = strchr(str, ':'); if (p) { @@ -723,80 +1622,14 @@ void del_trace_kprobe_events(struct strlist *dellist) event = str; } pr_debug("Group: %s, Event: %s\n", group, event); - del_trace_kprobe_event(fd, group, event, namelist); + ret = del_trace_kprobe_event(fd, group, event, namelist); free(str); + if (ret < 0) + break; } strlist__delete(namelist); close(fd); -} -#define LINEBUF_SIZE 256 -#define NR_ADDITIONAL_LINES 2 - -static void show_one_line(FILE *fp, unsigned int l, bool skip, bool show_num) -{ - char buf[LINEBUF_SIZE]; - const char *color = PERF_COLOR_BLUE; - - if (fgets(buf, LINEBUF_SIZE, fp) == NULL) - goto error; - if (!skip) { - if (show_num) - fprintf(stdout, "%7u %s", l, buf); - else - color_fprintf(stdout, color, " %s", buf); - } - - while (strlen(buf) == LINEBUF_SIZE - 1 && - buf[LINEBUF_SIZE - 2] != '\n') { - if (fgets(buf, LINEBUF_SIZE, fp) == NULL) - goto error; - if (!skip) { - if (show_num) - fprintf(stdout, "%s", buf); - else - color_fprintf(stdout, color, "%s", buf); - } - } - return; -error: - if (feof(fp)) - die("Source file is shorter than expected."); - else - die("File read error: %s", strerror(errno)); + return ret; } -void show_line_range(struct line_range *lr) -{ - unsigned int l = 1; - struct line_node *ln; - FILE *fp; - - setup_pager(); - - if (lr->function) - fprintf(stdout, "<%s:%d>\n", lr->function, - lr->start - lr->offset); - else - fprintf(stdout, "<%s:%d>\n", lr->file, lr->start); - - fp = fopen(lr->path, "r"); - if (fp == NULL) - die("Failed to open %s: %s", lr->path, strerror(errno)); - /* Skip to starting line number */ - while (l < lr->start) - show_one_line(fp, l++, true, false); - - list_for_each_entry(ln, &lr->line_list, list) { - while (ln->line > l) - show_one_line(fp, (l++) - lr->offset, false, false); - show_one_line(fp, (l++) - lr->offset, false, true); - } - - if (lr->end == INT_MAX) - lr->end = l + NR_ADDITIONAL_LINES; - while (l < lr->end && !feof(fp)) - show_one_line(fp, (l++) - lr->offset, false, false); - - fclose(fp); -} diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h index 711287d4baea..e7ff0d02c0d4 100644 --- a/tools/perf/util/probe-event.h +++ b/tools/perf/util/probe-event.h @@ -2,21 +2,125 @@ #define _PROBE_EVENT_H #include <stdbool.h> -#include "probe-finder.h" #include "strlist.h" -extern void parse_line_range_desc(const char *arg, struct line_range *lr); -extern void parse_perf_probe_event(const char *str, struct probe_point *pp, - bool *need_dwarf); -extern int synthesize_perf_probe_point(struct probe_point *pp); -extern int synthesize_perf_probe_event(struct probe_point *pp); -extern void parse_trace_kprobe_event(const char *str, struct probe_point *pp); -extern int synthesize_trace_kprobe_event(struct probe_point *pp); -extern void add_trace_kprobe_events(struct probe_point *probes, int nr_probes, - bool force_add); -extern void del_trace_kprobe_events(struct strlist *dellist); -extern void show_perf_probe_events(void); -extern void show_line_range(struct line_range *lr); +extern bool probe_event_dry_run; + +/* kprobe-tracer tracing point */ +struct kprobe_trace_point { + char *symbol; /* Base symbol */ + unsigned long offset; /* Offset from symbol */ + bool retprobe; /* Return probe flag */ +}; + +/* kprobe-tracer tracing argument referencing offset */ +struct kprobe_trace_arg_ref { + struct kprobe_trace_arg_ref *next; /* Next reference */ + long offset; /* Offset value */ +}; + +/* kprobe-tracer tracing argument */ +struct kprobe_trace_arg { + char *name; /* Argument name */ + char *value; /* Base value */ + char *type; /* Type name */ + struct kprobe_trace_arg_ref *ref; /* Referencing offset */ +}; + +/* kprobe-tracer tracing event (point + arg) */ +struct kprobe_trace_event { + char *event; /* Event name */ + char *group; /* Group name */ + struct kprobe_trace_point point; /* Trace point */ + int nargs; /* Number of args */ + struct kprobe_trace_arg *args; /* Arguments */ +}; + +/* Perf probe probing point */ +struct perf_probe_point { + char *file; /* File path */ + char *function; /* Function name */ + int line; /* Line number */ + bool retprobe; /* Return probe flag */ + char *lazy_line; /* Lazy matching pattern */ + unsigned long offset; /* Offset from function entry */ +}; + +/* Perf probe probing argument field chain */ +struct perf_probe_arg_field { + struct perf_probe_arg_field *next; /* Next field */ + char *name; /* Name of the field */ + bool ref; /* Referencing flag */ +}; + +/* Perf probe probing argument */ +struct perf_probe_arg { + char *name; /* Argument name */ + char *var; /* Variable name */ + char *type; /* Type name */ + struct perf_probe_arg_field *field; /* Structure fields */ +}; + +/* Perf probe probing event (point + arg) */ +struct perf_probe_event { + char *event; /* Event name */ + char *group; /* Group name */ + struct perf_probe_point point; /* Probe point */ + int nargs; /* Number of arguments */ + struct perf_probe_arg *args; /* Arguments */ +}; + + +/* Line number container */ +struct line_node { + struct list_head list; + int line; +}; + +/* Line range */ +struct line_range { + char *file; /* File name */ + char *function; /* Function name */ + int start; /* Start line number */ + int end; /* End line number */ + int offset; /* Start line offset */ + char *path; /* Real path name */ + struct list_head line_list; /* Visible lines */ +}; + +/* Command string to events */ +extern int parse_perf_probe_command(const char *cmd, + struct perf_probe_event *pev); +extern int parse_kprobe_trace_command(const char *cmd, + struct kprobe_trace_event *tev); + +/* Events to command string */ +extern char *synthesize_perf_probe_command(struct perf_probe_event *pev); +extern char *synthesize_kprobe_trace_command(struct kprobe_trace_event *tev); +extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, + size_t len); + +/* Check the perf_probe_event needs debuginfo */ +extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev); + +/* Convert from kprobe_trace_event to perf_probe_event */ +extern int convert_to_perf_probe_event(struct kprobe_trace_event *tev, + struct perf_probe_event *pev); + +/* Release event contents */ +extern void clear_perf_probe_event(struct perf_probe_event *pev); +extern void clear_kprobe_trace_event(struct kprobe_trace_event *tev); + +/* Command string to line-range */ +extern int parse_line_range_desc(const char *cmd, struct line_range *lr); + + +extern int add_perf_probe_events(struct perf_probe_event *pevs, int ntevs, + bool force_add); +extern int del_perf_probe_events(struct strlist *dellist); +extern int show_perf_probe_events(void); +extern int show_line_range(struct line_range *lr); + /* Maximum index number of event-name postfix */ #define MAX_EVENT_INDEX 1024 diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index c171a243d05b..3e7977560be5 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -84,6 +84,9 @@ const char *x86_64_regs_table[X86_64_MAX_REGS] = { #define arch_regs_table x86_32_regs_table #endif +/* Kprobe tracer basic type is up to u64 */ +#define MAX_BASIC_TYPE_BITS 64 + /* Return architecture dependent register string (for kprobe-tracer) */ static const char *get_arch_regstr(unsigned int n) { @@ -108,7 +111,7 @@ static int strtailcmp(const char *s1, const char *s2) /* Line number list operations */ /* Add a line to line number list */ -static void line_list__add_line(struct list_head *head, unsigned int line) +static int line_list__add_line(struct list_head *head, int line) { struct line_node *ln; struct list_head *p; @@ -119,21 +122,23 @@ static void line_list__add_line(struct list_head *head, unsigned int line) p = &ln->list; goto found; } else if (ln->line == line) /* Already exist */ - return ; + return 1; } /* List is empty, or the smallest entry */ p = head; found: pr_debug("line list: add a line %u\n", line); ln = zalloc(sizeof(struct line_node)); - DIE_IF(ln == NULL); + if (ln == NULL) + return -ENOMEM; ln->line = line; INIT_LIST_HEAD(&ln->list); list_add(&ln->list, p); + return 0; } /* Check if the line in line number list */ -static int line_list__has_line(struct list_head *head, unsigned int line) +static int line_list__has_line(struct list_head *head, int line) { struct line_node *ln; @@ -184,9 +189,129 @@ static const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname) if (strtailcmp(src, fname) == 0) break; } + if (i == nfiles) + return NULL; return src; } +/* Compare diename and tname */ +static bool die_compare_name(Dwarf_Die *dw_die, const char *tname) +{ + const char *name; + name = dwarf_diename(dw_die); + return name ? strcmp(tname, name) : -1; +} + +/* Get type die, but skip qualifiers and typedef */ +static Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem) +{ + Dwarf_Attribute attr; + int tag; + + do { + if (dwarf_attr(vr_die, DW_AT_type, &attr) == NULL || + dwarf_formref_die(&attr, die_mem) == NULL) + return NULL; + + tag = dwarf_tag(die_mem); + vr_die = die_mem; + } while (tag == DW_TAG_const_type || + tag == DW_TAG_restrict_type || + tag == DW_TAG_volatile_type || + tag == DW_TAG_shared_type || + tag == DW_TAG_typedef); + + return die_mem; +} + +static bool die_is_signed_type(Dwarf_Die *tp_die) +{ + Dwarf_Attribute attr; + Dwarf_Word ret; + + if (dwarf_attr(tp_die, DW_AT_encoding, &attr) == NULL || + dwarf_formudata(&attr, &ret) != 0) + return false; + + return (ret == DW_ATE_signed_char || ret == DW_ATE_signed || + ret == DW_ATE_signed_fixed); +} + +static int die_get_byte_size(Dwarf_Die *tp_die) +{ + Dwarf_Attribute attr; + Dwarf_Word ret; + + if (dwarf_attr(tp_die, DW_AT_byte_size, &attr) == NULL || + dwarf_formudata(&attr, &ret) != 0) + return 0; + + return (int)ret; +} + +/* Get data_member_location offset */ +static int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs) +{ + Dwarf_Attribute attr; + Dwarf_Op *expr; + size_t nexpr; + int ret; + + if (dwarf_attr(mb_die, DW_AT_data_member_location, &attr) == NULL) + return -ENOENT; + + if (dwarf_formudata(&attr, offs) != 0) { + /* DW_AT_data_member_location should be DW_OP_plus_uconst */ + ret = dwarf_getlocation(&attr, &expr, &nexpr); + if (ret < 0 || nexpr == 0) + return -ENOENT; + + if (expr[0].atom != DW_OP_plus_uconst || nexpr != 1) { + pr_debug("Unable to get offset:Unexpected OP %x (%zd)\n", + expr[0].atom, nexpr); + return -ENOTSUP; + } + *offs = (Dwarf_Word)expr[0].number; + } + return 0; +} + +/* Return values for die_find callbacks */ +enum { + DIE_FIND_CB_FOUND = 0, /* End of Search */ + DIE_FIND_CB_CHILD = 1, /* Search only children */ + DIE_FIND_CB_SIBLING = 2, /* Search only siblings */ + DIE_FIND_CB_CONTINUE = 3, /* Search children and siblings */ +}; + +/* Search a child die */ +static Dwarf_Die *die_find_child(Dwarf_Die *rt_die, + int (*callback)(Dwarf_Die *, void *), + void *data, Dwarf_Die *die_mem) +{ + Dwarf_Die child_die; + int ret; + + ret = dwarf_child(rt_die, die_mem); + if (ret != 0) + return NULL; + + do { + ret = callback(die_mem, data); + if (ret == DIE_FIND_CB_FOUND) + return die_mem; + + if ((ret & DIE_FIND_CB_CHILD) && + die_find_child(die_mem, callback, data, &child_die)) { + memcpy(die_mem, &child_die, sizeof(Dwarf_Die)); + return die_mem; + } + } while ((ret & DIE_FIND_CB_SIBLING) && + dwarf_siblingof(die_mem, die_mem) == 0); + + return NULL; +} + struct __addr_die_search_param { Dwarf_Addr addr; Dwarf_Die *die_mem; @@ -205,8 +330,8 @@ static int __die_search_func_cb(Dwarf_Die *fn_die, void *data) } /* Search a real subprogram including this line, */ -static Dwarf_Die *die_get_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr, - Dwarf_Die *die_mem) +static Dwarf_Die *die_find_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr, + Dwarf_Die *die_mem) { struct __addr_die_search_param ad; ad.addr = addr; @@ -218,77 +343,64 @@ static Dwarf_Die *die_get_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr, return die_mem; } -/* Similar to dwarf_getfuncs, but returns inlined_subroutine if exists. */ -static Dwarf_Die *die_get_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, - Dwarf_Die *die_mem) +/* die_find callback for inline function search */ +static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data) { - Dwarf_Die child_die; - int ret; + Dwarf_Addr *addr = data; - ret = dwarf_child(sp_die, die_mem); - if (ret != 0) - return NULL; + if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine && + dwarf_haspc(die_mem, *addr)) + return DIE_FIND_CB_FOUND; - do { - if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine && - dwarf_haspc(die_mem, addr)) - return die_mem; - - if (die_get_inlinefunc(die_mem, addr, &child_die)) { - memcpy(die_mem, &child_die, sizeof(Dwarf_Die)); - return die_mem; - } - } while (dwarf_siblingof(die_mem, die_mem) == 0); - - return NULL; + return DIE_FIND_CB_CONTINUE; } -/* Compare diename and tname */ -static bool die_compare_name(Dwarf_Die *dw_die, const char *tname) +/* Similar to dwarf_getfuncs, but returns inlined_subroutine if exists. */ +static Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, + Dwarf_Die *die_mem) { - const char *name; - name = dwarf_diename(dw_die); - DIE_IF(name == NULL); - return strcmp(tname, name); + return die_find_child(sp_die, __die_find_inline_cb, &addr, die_mem); } -/* Get entry pc(or low pc, 1st entry of ranges) of the die */ -static Dwarf_Addr die_get_entrypc(Dwarf_Die *dw_die) +static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data) { - Dwarf_Addr epc; - int ret; + const char *name = data; + int tag; - ret = dwarf_entrypc(dw_die, &epc); - DIE_IF(ret == -1); - return epc; + tag = dwarf_tag(die_mem); + if ((tag == DW_TAG_formal_parameter || + tag == DW_TAG_variable) && + (die_compare_name(die_mem, name) == 0)) + return DIE_FIND_CB_FOUND; + + return DIE_FIND_CB_CONTINUE; } -/* Get a variable die */ +/* Find a variable called 'name' */ static Dwarf_Die *die_find_variable(Dwarf_Die *sp_die, const char *name, Dwarf_Die *die_mem) { - Dwarf_Die child_die; - int tag; - int ret; + return die_find_child(sp_die, __die_find_variable_cb, (void *)name, + die_mem); +} - ret = dwarf_child(sp_die, die_mem); - if (ret != 0) - return NULL; +static int __die_find_member_cb(Dwarf_Die *die_mem, void *data) +{ + const char *name = data; - do { - tag = dwarf_tag(die_mem); - if ((tag == DW_TAG_formal_parameter || - tag == DW_TAG_variable) && - (die_compare_name(die_mem, name) == 0)) - return die_mem; + if ((dwarf_tag(die_mem) == DW_TAG_member) && + (die_compare_name(die_mem, name) == 0)) + return DIE_FIND_CB_FOUND; - if (die_find_variable(die_mem, name, &child_die)) { - memcpy(die_mem, &child_die, sizeof(Dwarf_Die)); - return die_mem; - } - } while (dwarf_siblingof(die_mem, die_mem) == 0); + return DIE_FIND_CB_SIBLING; +} - return NULL; +/* Find a member called 'name' */ +static Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name, + Dwarf_Die *die_mem) +{ + return die_find_child(st_die, __die_find_member_cb, (void *)name, + die_mem); } /* @@ -296,19 +408,22 @@ static Dwarf_Die *die_find_variable(Dwarf_Die *sp_die, const char *name, */ /* Show a location */ -static void show_location(Dwarf_Op *op, struct probe_finder *pf) +static int convert_location(Dwarf_Op *op, struct probe_finder *pf) { unsigned int regn; Dwarf_Word offs = 0; - int deref = 0, ret; + bool ref = false; const char *regs; + struct kprobe_trace_arg *tvar = pf->tvar; - /* TODO: support CFA */ /* If this is based on frame buffer, set the offset */ if (op->atom == DW_OP_fbreg) { - if (pf->fb_ops == NULL) - die("The attribute of frame base is not supported.\n"); - deref = 1; + if (pf->fb_ops == NULL) { + pr_warning("The attribute of frame base is not " + "supported.\n"); + return -ENOTSUP; + } + ref = true; offs = op->number; op = &pf->fb_ops[0]; } @@ -316,35 +431,164 @@ static void show_location(Dwarf_Op *op, struct probe_finder *pf) if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) { regn = op->atom - DW_OP_breg0; offs += op->number; - deref = 1; + ref = true; } else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) { regn = op->atom - DW_OP_reg0; } else if (op->atom == DW_OP_bregx) { regn = op->number; offs += op->number2; - deref = 1; + ref = true; } else if (op->atom == DW_OP_regx) { regn = op->number; - } else - die("DW_OP %d is not supported.", op->atom); + } else { + pr_warning("DW_OP %x is not supported.\n", op->atom); + return -ENOTSUP; + } regs = get_arch_regstr(regn); - if (!regs) - die("%u exceeds max register number.", regn); + if (!regs) { + pr_warning("%u exceeds max register number.\n", regn); + return -ERANGE; + } + + tvar->value = strdup(regs); + if (tvar->value == NULL) + return -ENOMEM; + + if (ref) { + tvar->ref = zalloc(sizeof(struct kprobe_trace_arg_ref)); + if (tvar->ref == NULL) + return -ENOMEM; + tvar->ref->offset = (long)offs; + } + return 0; +} + +static int convert_variable_type(Dwarf_Die *vr_die, + struct kprobe_trace_arg *targ) +{ + Dwarf_Die type; + char buf[16]; + int ret; + + if (die_get_real_type(vr_die, &type) == NULL) { + pr_warning("Failed to get a type information of %s.\n", + dwarf_diename(vr_die)); + return -ENOENT; + } + + ret = die_get_byte_size(&type) * 8; + if (ret) { + /* Check the bitwidth */ + if (ret > MAX_BASIC_TYPE_BITS) { + pr_info("%s exceeds max-bitwidth." + " Cut down to %d bits.\n", + dwarf_diename(&type), MAX_BASIC_TYPE_BITS); + ret = MAX_BASIC_TYPE_BITS; + } + + ret = snprintf(buf, 16, "%c%d", + die_is_signed_type(&type) ? 's' : 'u', ret); + if (ret < 0 || ret >= 16) { + if (ret >= 16) + ret = -E2BIG; + pr_warning("Failed to convert variable type: %s\n", + strerror(-ret)); + return ret; + } + targ->type = strdup(buf); + if (targ->type == NULL) + return -ENOMEM; + } + return 0; +} + +static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, + struct perf_probe_arg_field *field, + struct kprobe_trace_arg_ref **ref_ptr, + Dwarf_Die *die_mem) +{ + struct kprobe_trace_arg_ref *ref = *ref_ptr; + Dwarf_Die type; + Dwarf_Word offs; + int ret; + + pr_debug("converting %s in %s\n", field->name, varname); + if (die_get_real_type(vr_die, &type) == NULL) { + pr_warning("Failed to get the type of %s.\n", varname); + return -ENOENT; + } - if (deref) - ret = snprintf(pf->buf, pf->len, " %s=%+jd(%s)", - pf->var, (intmax_t)offs, regs); + /* Check the pointer and dereference */ + if (dwarf_tag(&type) == DW_TAG_pointer_type) { + if (!field->ref) { + pr_err("Semantic error: %s must be referred by '->'\n", + field->name); + return -EINVAL; + } + /* Get the type pointed by this pointer */ + if (die_get_real_type(&type, &type) == NULL) { + pr_warning("Failed to get the type of %s.\n", varname); + return -ENOENT; + } + /* Verify it is a data structure */ + if (dwarf_tag(&type) != DW_TAG_structure_type) { + pr_warning("%s is not a data structure.\n", varname); + return -EINVAL; + } + + ref = zalloc(sizeof(struct kprobe_trace_arg_ref)); + if (ref == NULL) + return -ENOMEM; + if (*ref_ptr) + (*ref_ptr)->next = ref; + else + *ref_ptr = ref; + } else { + /* Verify it is a data structure */ + if (dwarf_tag(&type) != DW_TAG_structure_type) { + pr_warning("%s is not a data structure.\n", varname); + return -EINVAL; + } + if (field->ref) { + pr_err("Semantic error: %s must be referred by '.'\n", + field->name); + return -EINVAL; + } + if (!ref) { + pr_warning("Structure on a register is not " + "supported yet.\n"); + return -ENOTSUP; + } + } + + if (die_find_member(&type, field->name, die_mem) == NULL) { + pr_warning("%s(tyep:%s) has no member %s.\n", varname, + dwarf_diename(&type), field->name); + return -EINVAL; + } + + /* Get the offset of the field */ + ret = die_get_data_member_location(die_mem, &offs); + if (ret < 0) { + pr_warning("Failed to get the offset of %s.\n", field->name); + return ret; + } + ref->offset += (long)offs; + + /* Converting next field */ + if (field->next) + return convert_variable_fields(die_mem, field->name, + field->next, &ref, die_mem); else - ret = snprintf(pf->buf, pf->len, " %s=%s", pf->var, regs); - DIE_IF(ret < 0); - DIE_IF(ret >= pf->len); + return 0; } /* Show a variables in kprobe event format */ -static void show_variable(Dwarf_Die *vr_die, struct probe_finder *pf) +static int convert_variable(Dwarf_Die *vr_die, struct probe_finder *pf) { Dwarf_Attribute attr; + Dwarf_Die die_mem; Dwarf_Op *expr; size_t nexpr; int ret; @@ -356,142 +600,190 @@ static void show_variable(Dwarf_Die *vr_die, struct probe_finder *pf) if (ret <= 0 || nexpr == 0) goto error; - show_location(expr, pf); + ret = convert_location(expr, pf); + if (ret == 0 && pf->pvar->field) { + ret = convert_variable_fields(vr_die, pf->pvar->var, + pf->pvar->field, &pf->tvar->ref, + &die_mem); + vr_die = &die_mem; + } + if (ret == 0) { + if (pf->pvar->type) { + pf->tvar->type = strdup(pf->pvar->type); + if (pf->tvar->type == NULL) + ret = -ENOMEM; + } else + ret = convert_variable_type(vr_die, pf->tvar); + } /* *expr will be cached in libdw. Don't free it. */ - return ; + return ret; error: /* TODO: Support const_value */ - die("Failed to find the location of %s at this address.\n" - " Perhaps, it has been optimized out.", pf->var); + pr_err("Failed to find the location of %s at this address.\n" + " Perhaps, it has been optimized out.\n", pf->pvar->var); + return -ENOENT; } /* Find a variable in a subprogram die */ -static void find_variable(Dwarf_Die *sp_die, struct probe_finder *pf) +static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf) { - int ret; Dwarf_Die vr_die; + char buf[32], *ptr; + int ret; - /* TODO: Support struct members and arrays */ - if (!is_c_varname(pf->var)) { - /* Output raw parameters */ - ret = snprintf(pf->buf, pf->len, " %s", pf->var); - DIE_IF(ret < 0); - DIE_IF(ret >= pf->len); - return ; + /* TODO: Support arrays */ + if (pf->pvar->name) + pf->tvar->name = strdup(pf->pvar->name); + else { + ret = synthesize_perf_probe_arg(pf->pvar, buf, 32); + if (ret < 0) + return ret; + ptr = strchr(buf, ':'); /* Change type separator to _ */ + if (ptr) + *ptr = '_'; + pf->tvar->name = strdup(buf); + } + if (pf->tvar->name == NULL) + return -ENOMEM; + + if (!is_c_varname(pf->pvar->var)) { + /* Copy raw parameters */ + pf->tvar->value = strdup(pf->pvar->var); + if (pf->tvar->value == NULL) + return -ENOMEM; + else + return 0; } - pr_debug("Searching '%s' variable in context.\n", pf->var); + pr_debug("Searching '%s' variable in context.\n", + pf->pvar->var); /* Search child die for local variables and parameters. */ - if (!die_find_variable(sp_die, pf->var, &vr_die)) - die("Failed to find '%s' in this function.", pf->var); - - show_variable(&vr_die, pf); + if (!die_find_variable(sp_die, pf->pvar->var, &vr_die)) { + pr_warning("Failed to find '%s' in this function.\n", + pf->pvar->var); + return -ENOENT; + } + return convert_variable(&vr_die, pf); } /* Show a probe point to output buffer */ -static void show_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf) +static int convert_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf) { - struct probe_point *pp = pf->pp; + struct kprobe_trace_event *tev; Dwarf_Addr eaddr; Dwarf_Die die_mem; const char *name; - char tmp[MAX_PROBE_BUFFER]; - int ret, i, len; + int ret, i; Dwarf_Attribute fb_attr; size_t nops; + if (pf->ntevs == MAX_PROBES) { + pr_warning("Too many( > %d) probe point found.\n", MAX_PROBES); + return -ERANGE; + } + tev = &pf->tevs[pf->ntevs++]; + /* If no real subprogram, find a real one */ if (!sp_die || dwarf_tag(sp_die) != DW_TAG_subprogram) { - sp_die = die_get_real_subprogram(&pf->cu_die, + sp_die = die_find_real_subprogram(&pf->cu_die, pf->addr, &die_mem); - if (!sp_die) - die("Probe point is not found in subprograms."); + if (!sp_die) { + pr_warning("Failed to find probe point in any " + "functions.\n"); + return -ENOENT; + } } - /* Output name of probe point */ + /* Copy the name of probe point */ name = dwarf_diename(sp_die); if (name) { - dwarf_entrypc(sp_die, &eaddr); - ret = snprintf(tmp, MAX_PROBE_BUFFER, "%s+%lu", name, - (unsigned long)(pf->addr - eaddr)); - /* Copy the function name if possible */ - if (!pp->function) { - pp->function = strdup(name); - pp->offset = (size_t)(pf->addr - eaddr); + if (dwarf_entrypc(sp_die, &eaddr) != 0) { + pr_warning("Failed to get entry pc of %s\n", + dwarf_diename(sp_die)); + return -ENOENT; } - } else { + tev->point.symbol = strdup(name); + if (tev->point.symbol == NULL) + return -ENOMEM; + tev->point.offset = (unsigned long)(pf->addr - eaddr); + } else /* This function has no name. */ - ret = snprintf(tmp, MAX_PROBE_BUFFER, "0x%jx", - (uintmax_t)pf->addr); - if (!pp->function) { - /* TODO: Use _stext */ - pp->function = strdup(""); - pp->offset = (size_t)pf->addr; - } - } - DIE_IF(ret < 0); - DIE_IF(ret >= MAX_PROBE_BUFFER); - len = ret; - pr_debug("Probe point found: %s\n", tmp); + tev->point.offset = (unsigned long)pf->addr; + + pr_debug("Probe point found: %s+%lu\n", tev->point.symbol, + tev->point.offset); /* Get the frame base attribute/ops */ dwarf_attr(sp_die, DW_AT_frame_base, &fb_attr); ret = dwarf_getlocation_addr(&fb_attr, pf->addr, &pf->fb_ops, &nops, 1); - if (ret <= 0 || nops == 0) + if (ret <= 0 || nops == 0) { pf->fb_ops = NULL; + } else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa && + pf->cfi != NULL) { + Dwarf_Frame *frame; + if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 || + dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) { + pr_warning("Failed to get CFA on 0x%jx\n", + (uintmax_t)pf->addr); + return -ENOENT; + } + } /* Find each argument */ - /* TODO: use dwarf_cfi_addrframe */ - for (i = 0; i < pp->nr_args; i++) { - pf->var = pp->args[i]; - pf->buf = &tmp[len]; - pf->len = MAX_PROBE_BUFFER - len; - find_variable(sp_die, pf); - len += strlen(pf->buf); + tev->nargs = pf->pev->nargs; + tev->args = zalloc(sizeof(struct kprobe_trace_arg) * tev->nargs); + if (tev->args == NULL) + return -ENOMEM; + for (i = 0; i < pf->pev->nargs; i++) { + pf->pvar = &pf->pev->args[i]; + pf->tvar = &tev->args[i]; + ret = find_variable(sp_die, pf); + if (ret != 0) + return ret; } /* *pf->fb_ops will be cached in libdw. Don't free it. */ pf->fb_ops = NULL; - - if (pp->found == MAX_PROBES) - die("Too many( > %d) probe point found.\n", MAX_PROBES); - - pp->probes[pp->found] = strdup(tmp); - pp->found++; + return 0; } /* Find probe point from its line number */ -static void find_probe_point_by_line(struct probe_finder *pf) +static int find_probe_point_by_line(struct probe_finder *pf) { Dwarf_Lines *lines; Dwarf_Line *line; size_t nlines, i; Dwarf_Addr addr; int lineno; - int ret; + int ret = 0; - ret = dwarf_getsrclines(&pf->cu_die, &lines, &nlines); - DIE_IF(ret != 0); + if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) { + pr_warning("No source lines found in this CU.\n"); + return -ENOENT; + } - for (i = 0; i < nlines; i++) { + for (i = 0; i < nlines && ret == 0; i++) { line = dwarf_onesrcline(lines, i); - dwarf_lineno(line, &lineno); - if (lineno != pf->lno) + if (dwarf_lineno(line, &lineno) != 0 || + lineno != pf->lno) continue; /* TODO: Get fileno from line, but how? */ if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0) continue; - ret = dwarf_lineaddr(line, &addr); - DIE_IF(ret != 0); + if (dwarf_lineaddr(line, &addr) != 0) { + pr_warning("Failed to get the address of the line.\n"); + return -ENOENT; + } pr_debug("Probe line found: line[%d]:%d addr:0x%jx\n", (int)i, lineno, (uintmax_t)addr); pf->addr = addr; - show_probe_point(NULL, pf); + ret = convert_probe_point(NULL, pf); /* Continuing, because target line might be inlined. */ } + return ret; } /* Find lines which match lazy pattern */ @@ -499,16 +791,27 @@ static int find_lazy_match_lines(struct list_head *head, const char *fname, const char *pat) { char *fbuf, *p1, *p2; - int fd, line, nlines = 0; + int fd, ret, line, nlines = 0; struct stat st; fd = open(fname, O_RDONLY); - if (fd < 0) - die("failed to open %s", fname); - DIE_IF(fstat(fd, &st) < 0); - fbuf = malloc(st.st_size + 2); - DIE_IF(fbuf == NULL); - DIE_IF(read(fd, fbuf, st.st_size) < 0); + if (fd < 0) { + pr_warning("Failed to open %s: %s\n", fname, strerror(-fd)); + return fd; + } + + ret = fstat(fd, &st); + if (ret < 0) { + pr_warning("Failed to get the size of %s: %s\n", + fname, strerror(errno)); + return ret; + } + fbuf = xmalloc(st.st_size + 2); + ret = read(fd, fbuf, st.st_size); + if (ret < 0) { + pr_warning("Failed to read %s: %s\n", fname, strerror(errno)); + return ret; + } close(fd); fbuf[st.st_size] = '\n'; /* Dummy line */ fbuf[st.st_size + 1] = '\0'; @@ -528,7 +831,7 @@ static int find_lazy_match_lines(struct list_head *head, } /* Find probe points from lazy pattern */ -static void find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) +static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) { Dwarf_Lines *lines; Dwarf_Line *line; @@ -536,37 +839,46 @@ static void find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) Dwarf_Addr addr; Dwarf_Die die_mem; int lineno; - int ret; + int ret = 0; if (list_empty(&pf->lcache)) { /* Matching lazy line pattern */ ret = find_lazy_match_lines(&pf->lcache, pf->fname, - pf->pp->lazy_line); - if (ret <= 0) - die("No matched lines found in %s.", pf->fname); + pf->pev->point.lazy_line); + if (ret == 0) { + pr_debug("No matched lines found in %s.\n", pf->fname); + return 0; + } else if (ret < 0) + return ret; } - ret = dwarf_getsrclines(&pf->cu_die, &lines, &nlines); - DIE_IF(ret != 0); - for (i = 0; i < nlines; i++) { + if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) { + pr_warning("No source lines found in this CU.\n"); + return -ENOENT; + } + + for (i = 0; i < nlines && ret >= 0; i++) { line = dwarf_onesrcline(lines, i); - dwarf_lineno(line, &lineno); - if (!line_list__has_line(&pf->lcache, lineno)) + if (dwarf_lineno(line, &lineno) != 0 || + !line_list__has_line(&pf->lcache, lineno)) continue; /* TODO: Get fileno from line, but how? */ if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0) continue; - ret = dwarf_lineaddr(line, &addr); - DIE_IF(ret != 0); + if (dwarf_lineaddr(line, &addr) != 0) { + pr_debug("Failed to get the address of line %d.\n", + lineno); + continue; + } if (sp_die) { /* Address filtering 1: does sp_die include addr? */ if (!dwarf_haspc(sp_die, addr)) continue; /* Address filtering 2: No child include addr? */ - if (die_get_inlinefunc(sp_die, addr, &die_mem)) + if (die_find_inlinefunc(sp_die, addr, &die_mem)) continue; } @@ -574,27 +886,42 @@ static void find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) (int)i, lineno, (unsigned long long)addr); pf->addr = addr; - show_probe_point(sp_die, pf); + ret = convert_probe_point(sp_die, pf); /* Continuing, because target line might be inlined. */ } /* TODO: deallocate lines, but how? */ + return ret; } +/* Callback parameter with return value */ +struct dwarf_callback_param { + void *data; + int retval; +}; + static int probe_point_inline_cb(Dwarf_Die *in_die, void *data) { - struct probe_finder *pf = (struct probe_finder *)data; - struct probe_point *pp = pf->pp; + struct dwarf_callback_param *param = data; + struct probe_finder *pf = param->data; + struct perf_probe_point *pp = &pf->pev->point; + Dwarf_Addr addr; if (pp->lazy_line) - find_probe_point_lazy(in_die, pf); + param->retval = find_probe_point_lazy(in_die, pf); else { /* Get probe address */ - pf->addr = die_get_entrypc(in_die); + if (dwarf_entrypc(in_die, &addr) != 0) { + pr_warning("Failed to get entry pc of %s.\n", + dwarf_diename(in_die)); + param->retval = -ENOENT; + return DWARF_CB_ABORT; + } + pf->addr = addr; pf->addr += pp->offset; pr_debug("found inline addr: 0x%jx\n", (uintmax_t)pf->addr); - show_probe_point(in_die, pf); + param->retval = convert_probe_point(in_die, pf); } return DWARF_CB_OK; @@ -603,59 +930,88 @@ static int probe_point_inline_cb(Dwarf_Die *in_die, void *data) /* Search function from function name */ static int probe_point_search_cb(Dwarf_Die *sp_die, void *data) { - struct probe_finder *pf = (struct probe_finder *)data; - struct probe_point *pp = pf->pp; + struct dwarf_callback_param *param = data; + struct probe_finder *pf = param->data; + struct perf_probe_point *pp = &pf->pev->point; /* Check tag and diename */ if (dwarf_tag(sp_die) != DW_TAG_subprogram || die_compare_name(sp_die, pp->function) != 0) - return 0; + return DWARF_CB_OK; pf->fname = dwarf_decl_file(sp_die); if (pp->line) { /* Function relative line */ dwarf_decl_line(sp_die, &pf->lno); pf->lno += pp->line; - find_probe_point_by_line(pf); + param->retval = find_probe_point_by_line(pf); } else if (!dwarf_func_inline(sp_die)) { /* Real function */ if (pp->lazy_line) - find_probe_point_lazy(sp_die, pf); + param->retval = find_probe_point_lazy(sp_die, pf); else { - pf->addr = die_get_entrypc(sp_die); + if (dwarf_entrypc(sp_die, &pf->addr) != 0) { + pr_warning("Failed to get entry pc of %s.\n", + dwarf_diename(sp_die)); + param->retval = -ENOENT; + return DWARF_CB_ABORT; + } pf->addr += pp->offset; /* TODO: Check the address in this function */ - show_probe_point(sp_die, pf); + param->retval = convert_probe_point(sp_die, pf); } - } else + } else { + struct dwarf_callback_param _param = {.data = (void *)pf, + .retval = 0}; /* Inlined function: search instances */ - dwarf_func_inline_instances(sp_die, probe_point_inline_cb, pf); + dwarf_func_inline_instances(sp_die, probe_point_inline_cb, + &_param); + param->retval = _param.retval; + } - return 1; /* Exit; no same symbol in this CU. */ + return DWARF_CB_ABORT; /* Exit; no same symbol in this CU. */ } -static void find_probe_point_by_func(struct probe_finder *pf) +static int find_probe_point_by_func(struct probe_finder *pf) { - dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, pf, 0); + struct dwarf_callback_param _param = {.data = (void *)pf, + .retval = 0}; + dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, &_param, 0); + return _param.retval; } -/* Find a probe point */ -int find_probe_point(int fd, struct probe_point *pp) +/* Find kprobe_trace_events specified by perf_probe_event from debuginfo */ +int find_kprobe_trace_events(int fd, struct perf_probe_event *pev, + struct kprobe_trace_event **tevs) { - struct probe_finder pf = {.pp = pp}; + struct probe_finder pf = {.pev = pev}; + struct perf_probe_point *pp = &pev->point; Dwarf_Off off, noff; size_t cuhl; Dwarf_Die *diep; Dwarf *dbg; + int ret = 0; + + pf.tevs = zalloc(sizeof(struct kprobe_trace_event) * MAX_PROBES); + if (pf.tevs == NULL) + return -ENOMEM; + *tevs = pf.tevs; + pf.ntevs = 0; dbg = dwarf_begin(fd, DWARF_C_READ); - if (!dbg) - return -ENOENT; + if (!dbg) { + pr_warning("No dwarf info found in the vmlinux - " + "please rebuild with CONFIG_DEBUG_INFO=y.\n"); + return -EBADF; + } + + /* Get the call frame information from this dwarf */ + pf.cfi = dwarf_getcfi(dbg); - pp->found = 0; off = 0; line_list__init(&pf.lcache); /* Loop on CUs (Compilation Unit) */ - while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL)) { + while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) && + ret >= 0) { /* Get the DIE(Debugging Information Entry) of this CU */ diep = dwarf_offdie(dbg, off + cuhl, &pf.cu_die); if (!diep) @@ -669,12 +1025,12 @@ int find_probe_point(int fd, struct probe_point *pp) if (!pp->file || pf.fname) { if (pp->function) - find_probe_point_by_func(&pf); + ret = find_probe_point_by_func(&pf); else if (pp->lazy_line) - find_probe_point_lazy(NULL, &pf); + ret = find_probe_point_lazy(NULL, &pf); else { pf.lno = pp->line; - find_probe_point_by_line(&pf); + ret = find_probe_point_by_line(&pf); } } off = noff; @@ -682,41 +1038,167 @@ int find_probe_point(int fd, struct probe_point *pp) line_list__free(&pf.lcache); dwarf_end(dbg); - return pp->found; + return (ret < 0) ? ret : pf.ntevs; +} + +/* Reverse search */ +int find_perf_probe_point(int fd, unsigned long addr, + struct perf_probe_point *ppt) +{ + Dwarf_Die cudie, spdie, indie; + Dwarf *dbg; + Dwarf_Line *line; + Dwarf_Addr laddr, eaddr; + const char *tmp; + int lineno, ret = 0; + bool found = false; + + dbg = dwarf_begin(fd, DWARF_C_READ); + if (!dbg) + return -EBADF; + + /* Find cu die */ + if (!dwarf_addrdie(dbg, (Dwarf_Addr)addr, &cudie)) { + ret = -EINVAL; + goto end; + } + + /* Find a corresponding line */ + line = dwarf_getsrc_die(&cudie, (Dwarf_Addr)addr); + if (line) { + if (dwarf_lineaddr(line, &laddr) == 0 && + (Dwarf_Addr)addr == laddr && + dwarf_lineno(line, &lineno) == 0) { + tmp = dwarf_linesrc(line, NULL, NULL); + if (tmp) { + ppt->line = lineno; + ppt->file = strdup(tmp); + if (ppt->file == NULL) { + ret = -ENOMEM; + goto end; + } + found = true; + } + } + } + + /* Find a corresponding function */ + if (die_find_real_subprogram(&cudie, (Dwarf_Addr)addr, &spdie)) { + tmp = dwarf_diename(&spdie); + if (!tmp || dwarf_entrypc(&spdie, &eaddr) != 0) + goto end; + + if (ppt->line) { + if (die_find_inlinefunc(&spdie, (Dwarf_Addr)addr, + &indie)) { + /* addr in an inline function */ + tmp = dwarf_diename(&indie); + if (!tmp) + goto end; + ret = dwarf_decl_line(&indie, &lineno); + } else { + if (eaddr == addr) { /* Function entry */ + lineno = ppt->line; + ret = 0; + } else + ret = dwarf_decl_line(&spdie, &lineno); + } + if (ret == 0) { + /* Make a relative line number */ + ppt->line -= lineno; + goto found; + } + } + /* We don't have a line number, let's use offset */ + ppt->offset = addr - (unsigned long)eaddr; +found: + ppt->function = strdup(tmp); + if (ppt->function == NULL) { + ret = -ENOMEM; + goto end; + } + found = true; + } + +end: + dwarf_end(dbg); + if (ret >= 0) + ret = found ? 1 : 0; + return ret; +} + +/* Add a line and store the src path */ +static int line_range_add_line(const char *src, unsigned int lineno, + struct line_range *lr) +{ + /* Copy real path */ + if (!lr->path) { + lr->path = strdup(src); + if (lr->path == NULL) + return -ENOMEM; + } + return line_list__add_line(&lr->line_list, lineno); +} + +/* Search function declaration lines */ +static int line_range_funcdecl_cb(Dwarf_Die *sp_die, void *data) +{ + struct dwarf_callback_param *param = data; + struct line_finder *lf = param->data; + const char *src; + int lineno; + + src = dwarf_decl_file(sp_die); + if (src && strtailcmp(src, lf->fname) != 0) + return DWARF_CB_OK; + + if (dwarf_decl_line(sp_die, &lineno) != 0 || + (lf->lno_s > lineno || lf->lno_e < lineno)) + return DWARF_CB_OK; + + param->retval = line_range_add_line(src, lineno, lf->lr); + return DWARF_CB_OK; +} + +static int find_line_range_func_decl_lines(struct line_finder *lf) +{ + struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0}; + dwarf_getfuncs(&lf->cu_die, line_range_funcdecl_cb, ¶m, 0); + return param.retval; } /* Find line range from its line number */ -static void find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf) +static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf) { Dwarf_Lines *lines; Dwarf_Line *line; size_t nlines, i; Dwarf_Addr addr; - int lineno; - int ret; + int lineno, ret = 0; const char *src; Dwarf_Die die_mem; line_list__init(&lf->lr->line_list); - ret = dwarf_getsrclines(&lf->cu_die, &lines, &nlines); - DIE_IF(ret != 0); + if (dwarf_getsrclines(&lf->cu_die, &lines, &nlines) != 0) { + pr_warning("No source lines found in this CU.\n"); + return -ENOENT; + } + /* Search probable lines on lines list */ for (i = 0; i < nlines; i++) { line = dwarf_onesrcline(lines, i); - ret = dwarf_lineno(line, &lineno); - DIE_IF(ret != 0); - if (lf->lno_s > lineno || lf->lno_e < lineno) + if (dwarf_lineno(line, &lineno) != 0 || + (lf->lno_s > lineno || lf->lno_e < lineno)) continue; if (sp_die) { /* Address filtering 1: does sp_die include addr? */ - ret = dwarf_lineaddr(line, &addr); - DIE_IF(ret != 0); - if (!dwarf_haspc(sp_die, addr)) + if (dwarf_lineaddr(line, &addr) != 0 || + !dwarf_haspc(sp_die, addr)) continue; /* Address filtering 2: No child include addr? */ - if (die_get_inlinefunc(sp_die, addr, &die_mem)) + if (die_find_inlinefunc(sp_die, addr, &die_mem)) continue; } @@ -725,30 +1207,49 @@ static void find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf) if (strtailcmp(src, lf->fname) != 0) continue; - /* Copy real path */ - if (!lf->lr->path) - lf->lr->path = strdup(src); - line_list__add_line(&lf->lr->line_list, (unsigned int)lineno); + ret = line_range_add_line(src, lineno, lf->lr); + if (ret < 0) + return ret; } + + /* + * Dwarf lines doesn't include function declarations. We have to + * check functions list or given function. + */ + if (sp_die) { + src = dwarf_decl_file(sp_die); + if (src && dwarf_decl_line(sp_die, &lineno) == 0 && + (lf->lno_s <= lineno && lf->lno_e >= lineno)) + ret = line_range_add_line(src, lineno, lf->lr); + } else + ret = find_line_range_func_decl_lines(lf); + /* Update status */ - if (!list_empty(&lf->lr->line_list)) - lf->found = 1; + if (ret >= 0) + if (!list_empty(&lf->lr->line_list)) + ret = lf->found = 1; + else + ret = 0; /* Lines are not found */ else { free(lf->lr->path); lf->lr->path = NULL; } + return ret; } static int line_range_inline_cb(Dwarf_Die *in_die, void *data) { - find_line_range_by_line(in_die, (struct line_finder *)data); + struct dwarf_callback_param *param = data; + + param->retval = find_line_range_by_line(in_die, param->data); return DWARF_CB_ABORT; /* No need to find other instances */ } /* Search function from function name */ static int line_range_search_cb(Dwarf_Die *sp_die, void *data) { - struct line_finder *lf = (struct line_finder *)data; + struct dwarf_callback_param *param = data; + struct line_finder *lf = param->data; struct line_range *lr = lf->lr; if (dwarf_tag(sp_die) == DW_TAG_subprogram && @@ -757,44 +1258,55 @@ static int line_range_search_cb(Dwarf_Die *sp_die, void *data) dwarf_decl_line(sp_die, &lr->offset); pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset); lf->lno_s = lr->offset + lr->start; - if (!lr->end) + if (lf->lno_s < 0) /* Overflow */ + lf->lno_s = INT_MAX; + lf->lno_e = lr->offset + lr->end; + if (lf->lno_e < 0) /* Overflow */ lf->lno_e = INT_MAX; - else - lf->lno_e = lr->offset + lr->end; + pr_debug("New line range: %d to %d\n", lf->lno_s, lf->lno_e); lr->start = lf->lno_s; lr->end = lf->lno_e; - if (dwarf_func_inline(sp_die)) + if (dwarf_func_inline(sp_die)) { + struct dwarf_callback_param _param; + _param.data = (void *)lf; + _param.retval = 0; dwarf_func_inline_instances(sp_die, - line_range_inline_cb, lf); - else - find_line_range_by_line(sp_die, lf); - return 1; + line_range_inline_cb, + &_param); + param->retval = _param.retval; + } else + param->retval = find_line_range_by_line(sp_die, lf); + return DWARF_CB_ABORT; } - return 0; + return DWARF_CB_OK; } -static void find_line_range_by_func(struct line_finder *lf) +static int find_line_range_by_func(struct line_finder *lf) { - dwarf_getfuncs(&lf->cu_die, line_range_search_cb, lf, 0); + struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0}; + dwarf_getfuncs(&lf->cu_die, line_range_search_cb, ¶m, 0); + return param.retval; } int find_line_range(int fd, struct line_range *lr) { struct line_finder lf = {.lr = lr, .found = 0}; - int ret; + int ret = 0; Dwarf_Off off = 0, noff; size_t cuhl; Dwarf_Die *diep; Dwarf *dbg; dbg = dwarf_begin(fd, DWARF_C_READ); - if (!dbg) - return -ENOENT; + if (!dbg) { + pr_warning("No dwarf info found in the vmlinux - " + "please rebuild with CONFIG_DEBUG_INFO=y.\n"); + return -EBADF; + } /* Loop on CUs (Compilation Unit) */ - while (!lf.found) { - ret = dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL); - if (ret != 0) + while (!lf.found && ret >= 0) { + if (dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) != 0) break; /* Get the DIE(Debugging Information Entry) of this CU */ @@ -810,20 +1322,18 @@ int find_line_range(int fd, struct line_range *lr) if (!lr->file || lf.fname) { if (lr->function) - find_line_range_by_func(&lf); + ret = find_line_range_by_func(&lf); else { lf.lno_s = lr->start; - if (!lr->end) - lf.lno_e = INT_MAX; - else - lf.lno_e = lr->end; - find_line_range_by_line(NULL, &lf); + lf.lno_e = lr->end; + ret = find_line_range_by_line(NULL, &lf); } } off = noff; } pr_debug("path: %lx\n", (unsigned long)lr->path); dwarf_end(dbg); - return lf.found; + + return (ret < 0) ? ret : lf.found; } diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h index 21f7354397b4..310ce897229c 100644 --- a/tools/perf/util/probe-finder.h +++ b/tools/perf/util/probe-finder.h @@ -3,6 +3,7 @@ #include <stdbool.h> #include "util.h" +#include "probe-event.h" #define MAX_PATH_LEN 256 #define MAX_PROBE_BUFFER 1024 @@ -14,67 +15,37 @@ static inline int is_c_varname(const char *name) return isalpha(name[0]) || name[0] == '_'; } -struct probe_point { - char *event; /* Event name */ - char *group; /* Event group */ +#ifdef DWARF_SUPPORT +/* Find kprobe_trace_events specified by perf_probe_event from debuginfo */ +extern int find_kprobe_trace_events(int fd, struct perf_probe_event *pev, + struct kprobe_trace_event **tevs); - /* Inputs */ - char *file; /* File name */ - int line; /* Line number */ - char *lazy_line; /* Lazy line pattern */ +/* Find a perf_probe_point from debuginfo */ +extern int find_perf_probe_point(int fd, unsigned long addr, + struct perf_probe_point *ppt); - char *function; /* Function name */ - int offset; /* Offset bytes */ - - int nr_args; /* Number of arguments */ - char **args; /* Arguments */ - - int retprobe; /* Return probe */ - - /* Output */ - int found; /* Number of found probe points */ - char *probes[MAX_PROBES]; /* Output buffers (will be allocated)*/ -}; - -/* Line number container */ -struct line_node { - struct list_head list; - unsigned int line; -}; - -/* Line range */ -struct line_range { - char *file; /* File name */ - char *function; /* Function name */ - unsigned int start; /* Start line number */ - unsigned int end; /* End line number */ - int offset; /* Start line offset */ - char *path; /* Real path name */ - struct list_head line_list; /* Visible lines */ -}; - -#ifndef NO_DWARF_SUPPORT -extern int find_probe_point(int fd, struct probe_point *pp); extern int find_line_range(int fd, struct line_range *lr); #include <dwarf.h> #include <libdw.h> struct probe_finder { - struct probe_point *pp; /* Target probe point */ + struct perf_probe_event *pev; /* Target probe event */ + struct kprobe_trace_event *tevs; /* Result trace events */ + int ntevs; /* number of trace events */ /* For function searching */ - Dwarf_Addr addr; /* Address */ - const char *fname; /* File name */ int lno; /* Line number */ + Dwarf_Addr addr; /* Address */ + const char *fname; /* Real file name */ Dwarf_Die cu_die; /* Current CU */ + struct list_head lcache; /* Line cache for lazy match */ /* For variable searching */ + Dwarf_CFI *cfi; /* Call Frame Information */ Dwarf_Op *fb_ops; /* Frame base attribute */ - const char *var; /* Current variable name */ - char *buf; /* Current output buffer */ - int len; /* Length of output buffer */ - struct list_head lcache; /* Line cache for lazy match */ + struct perf_probe_arg *pvar; /* Current target variable */ + struct kprobe_trace_arg *tvar; /* Current result variable */ }; struct line_finder { @@ -87,6 +58,6 @@ struct line_finder { int found; }; -#endif /* NO_DWARF_SUPPORT */ +#endif /* DWARF_SUPPORT */ #endif /*_PROBE_FINDER_H */ diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index eed1cb889008..0fdf3ebef1e9 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -14,6 +14,16 @@ static int perf_session__open(struct perf_session *self, bool force) { struct stat input_stat; + if (!strcmp(self->filename, "-")) { + self->fd_pipe = true; + self->fd = STDIN_FILENO; + + if (perf_header__read(self, self->fd) < 0) + pr_err("incompatible file format"); + + return 0; + } + self->fd = open(self->filename, O_RDONLY); if (self->fd < 0) { pr_err("failed to open file: %s", self->filename); @@ -38,7 +48,7 @@ static int perf_session__open(struct perf_session *self, bool force) goto out_close; } - if (perf_header__read(&self->header, self->fd) < 0) { + if (perf_header__read(self, self->fd) < 0) { pr_err("incompatible file format"); goto out_close; } @@ -52,9 +62,9 @@ out_close: return -1; } -static inline int perf_session__create_kernel_maps(struct perf_session *self) +void perf_session__update_sample_type(struct perf_session *self) { - return map_groups__create_kernel_maps(&self->kmaps, self->vmlinux_maps); + self->sample_type = perf_header__sample_type(&self->header); } struct perf_session *perf_session__new(const char *filename, int mode, bool force) @@ -90,7 +100,7 @@ struct perf_session *perf_session__new(const char *filename, int mode, bool forc goto out_delete; } - self->sample_type = perf_header__sample_type(&self->header); + perf_session__update_sample_type(self); out: return self; out_free: @@ -117,22 +127,17 @@ static bool symbol__match_parent_regex(struct symbol *sym) return 0; } -struct symbol **perf_session__resolve_callchain(struct perf_session *self, - struct thread *thread, - struct ip_callchain *chain, - struct symbol **parent) +struct map_symbol *perf_session__resolve_callchain(struct perf_session *self, + struct thread *thread, + struct ip_callchain *chain, + struct symbol **parent) { u8 cpumode = PERF_RECORD_MISC_USER; - struct symbol **syms = NULL; unsigned int i; + struct map_symbol *syms = calloc(chain->nr, sizeof(*syms)); - if (symbol_conf.use_callchain) { - syms = calloc(chain->nr, sizeof(*syms)); - if (!syms) { - fprintf(stderr, "Can't allocate memory for symbols\n"); - exit(-1); - } - } + if (!syms) + return NULL; for (i = 0; i < chain->nr; i++) { u64 ip = chain->ips[i]; @@ -160,7 +165,8 @@ struct symbol **perf_session__resolve_callchain(struct perf_session *self, *parent = al.sym; if (!symbol_conf.use_callchain) break; - syms[i] = al.sym; + syms[i].map = al.map; + syms[i].sym = al.sym; } } @@ -194,6 +200,14 @@ static void perf_event_ops__fill_defaults(struct perf_event_ops *handler) handler->throttle = process_event_stub; if (handler->unthrottle == NULL) handler->unthrottle = process_event_stub; + if (handler->attr == NULL) + handler->attr = process_event_stub; + if (handler->event_type == NULL) + handler->event_type = process_event_stub; + if (handler->tracing_data == NULL) + handler->tracing_data = process_event_stub; + if (handler->build_id == NULL) + handler->build_id = process_event_stub; } static const char *event__name[] = { @@ -207,16 +221,23 @@ static const char *event__name[] = { [PERF_RECORD_FORK] = "FORK", [PERF_RECORD_READ] = "READ", [PERF_RECORD_SAMPLE] = "SAMPLE", + [PERF_RECORD_HEADER_ATTR] = "ATTR", + [PERF_RECORD_HEADER_EVENT_TYPE] = "EVENT_TYPE", + [PERF_RECORD_HEADER_TRACING_DATA] = "TRACING_DATA", + [PERF_RECORD_HEADER_BUILD_ID] = "BUILD_ID", }; -unsigned long event__total[PERF_RECORD_MAX]; +unsigned long event__total[PERF_RECORD_HEADER_MAX]; void event__print_totals(void) { int i; - for (i = 0; i < PERF_RECORD_MAX; ++i) + for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) { + if (!event__name[i]) + continue; pr_info("%10s events: %10ld\n", event__name[i], event__total[i]); + } } void mem_bswap_64(void *src, int byte_size) @@ -270,6 +291,37 @@ static void event__read_swap(event_t *self) self->read.id = bswap_64(self->read.id); } +static void event__attr_swap(event_t *self) +{ + size_t size; + + self->attr.attr.type = bswap_32(self->attr.attr.type); + self->attr.attr.size = bswap_32(self->attr.attr.size); + self->attr.attr.config = bswap_64(self->attr.attr.config); + self->attr.attr.sample_period = bswap_64(self->attr.attr.sample_period); + self->attr.attr.sample_type = bswap_64(self->attr.attr.sample_type); + self->attr.attr.read_format = bswap_64(self->attr.attr.read_format); + self->attr.attr.wakeup_events = bswap_32(self->attr.attr.wakeup_events); + self->attr.attr.bp_type = bswap_32(self->attr.attr.bp_type); + self->attr.attr.bp_addr = bswap_64(self->attr.attr.bp_addr); + self->attr.attr.bp_len = bswap_64(self->attr.attr.bp_len); + + size = self->header.size; + size -= (void *)&self->attr.id - (void *)self; + mem_bswap_64(self->attr.id, size); +} + +static void event__event_type_swap(event_t *self) +{ + self->event_type.event_type.event_id = + bswap_64(self->event_type.event_type.event_id); +} + +static void event__tracing_data_swap(event_t *self) +{ + self->tracing_data.size = bswap_32(self->tracing_data.size); +} + typedef void (*event__swap_op)(event_t *self); static event__swap_op event__swap_ops[] = { @@ -280,7 +332,11 @@ static event__swap_op event__swap_ops[] = { [PERF_RECORD_LOST] = event__all64_swap, [PERF_RECORD_READ] = event__read_swap, [PERF_RECORD_SAMPLE] = event__all64_swap, - [PERF_RECORD_MAX] = NULL, + [PERF_RECORD_HEADER_ATTR] = event__attr_swap, + [PERF_RECORD_HEADER_EVENT_TYPE] = event__event_type_swap, + [PERF_RECORD_HEADER_TRACING_DATA] = event__tracing_data_swap, + [PERF_RECORD_HEADER_BUILD_ID] = NULL, + [PERF_RECORD_HEADER_MAX] = NULL, }; static int perf_session__process_event(struct perf_session *self, @@ -290,7 +346,7 @@ static int perf_session__process_event(struct perf_session *self, { trace_event(event); - if (event->header.type < PERF_RECORD_MAX) { + if (event->header.type < PERF_RECORD_HEADER_MAX) { dump_printf("%#Lx [%#x]: PERF_RECORD_%s", offset + head, event->header.size, event__name[event->header.type]); @@ -320,6 +376,16 @@ static int perf_session__process_event(struct perf_session *self, return ops->throttle(event, self); case PERF_RECORD_UNTHROTTLE: return ops->unthrottle(event, self); + case PERF_RECORD_HEADER_ATTR: + return ops->attr(event, self); + case PERF_RECORD_HEADER_EVENT_TYPE: + return ops->event_type(event, self); + case PERF_RECORD_HEADER_TRACING_DATA: + /* setup for reading amidst mmap */ + lseek(self->fd, offset + head, SEEK_SET); + return ops->tracing_data(event, self); + case PERF_RECORD_HEADER_BUILD_ID: + return ops->build_id(event, self); default: self->unknown_events++; return -1; @@ -385,6 +451,101 @@ static struct thread *perf_session__register_idle_thread(struct perf_session *se return thread; } +int do_read(int fd, void *buf, size_t size) +{ + void *buf_start = buf; + + while (size) { + int ret = read(fd, buf, size); + + if (ret <= 0) + return ret; + + size -= ret; + buf += ret; + } + + return buf - buf_start; +} + +#define session_done() (*(volatile int *)(&session_done)) +volatile int session_done; + +static int __perf_session__process_pipe_events(struct perf_session *self, + struct perf_event_ops *ops) +{ + event_t event; + uint32_t size; + int skip = 0; + u64 head; + int err; + void *p; + + perf_event_ops__fill_defaults(ops); + + head = 0; +more: + err = do_read(self->fd, &event, sizeof(struct perf_event_header)); + if (err <= 0) { + if (err == 0) + goto done; + + pr_err("failed to read event header\n"); + goto out_err; + } + + if (self->header.needs_swap) + perf_event_header__bswap(&event.header); + + size = event.header.size; + if (size == 0) + size = 8; + + p = &event; + p += sizeof(struct perf_event_header); + + err = do_read(self->fd, p, size - sizeof(struct perf_event_header)); + if (err <= 0) { + if (err == 0) { + pr_err("unexpected end of event stream\n"); + goto done; + } + + pr_err("failed to read event data\n"); + goto out_err; + } + + if (size == 0 || + (skip = perf_session__process_event(self, &event, ops, + 0, head)) < 0) { + dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n", + head, event.header.size, event.header.type); + /* + * assume we lost track of the stream, check alignment, and + * increment a single u64 in the hope to catch on again 'soon'. + */ + if (unlikely(head & 7)) + head &= ~7ULL; + + size = 8; + } + + head += size; + + dump_printf("\n%#Lx [%#x]: event: %d\n", + head, event.header.size, event.header.type); + + if (skip > 0) + head += skip; + + if (!session_done()) + goto more; +done: + err = 0; +out_err: + return err; +} + int __perf_session__process_events(struct perf_session *self, u64 data_offset, u64 data_size, u64 file_size, struct perf_event_ops *ops) @@ -396,6 +557,10 @@ int __perf_session__process_events(struct perf_session *self, event_t *event; uint32_t size; char *buf; + struct ui_progress *progress = ui_progress__new("Processing events...", + self->size); + if (progress == NULL) + return -1; perf_event_ops__fill_defaults(ops); @@ -424,6 +589,7 @@ remap: more: event = (event_t *)(buf + head); + ui_progress__update(progress, offset); if (self->header.needs_swap) perf_event_header__bswap(&event->header); @@ -474,6 +640,7 @@ more: done: err = 0; out_err: + ui_progress__delete(progress); return err; } @@ -502,9 +669,13 @@ out_getcwd_err: self->cwdlen = strlen(self->cwd); } - err = __perf_session__process_events(self, self->header.data_offset, - self->header.data_size, - self->size, ops); + if (!self->fd_pipe) + err = __perf_session__process_events(self, + self->header.data_offset, + self->header.data_size, + self->size, ops); + else + err = __perf_session__process_pipe_events(self, ops); out_err: return err; } @@ -543,32 +714,3 @@ int perf_session__set_kallsyms_ref_reloc_sym(struct perf_session *self, return 0; } - -static u64 map__reloc_map_ip(struct map *map, u64 ip) -{ - return ip + (s64)map->pgoff; -} - -static u64 map__reloc_unmap_ip(struct map *map, u64 ip) -{ - return ip - (s64)map->pgoff; -} - -void map__reloc_vmlinux(struct map *self) -{ - struct kmap *kmap = map__kmap(self); - s64 reloc; - - if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr) - return; - - reloc = (kmap->ref_reloc_sym->unrelocated_addr - - kmap->ref_reloc_sym->addr); - - if (!reloc) - return; - - self->map_ip = map__reloc_map_ip; - self->unmap_ip = map__reloc_unmap_ip; - self->pgoff = reloc; -} diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h index 5c33417eebb3..0ac14d42dc28 100644 --- a/tools/perf/util/session.h +++ b/tools/perf/util/session.h @@ -27,6 +27,7 @@ struct perf_session { u64 sample_type; struct ref_reloc_sym ref_reloc_sym; int fd; + bool fd_pipe; int cwdlen; char *cwd; char filename[0]; @@ -43,7 +44,11 @@ struct perf_event_ops { lost, read, throttle, - unthrottle; + unthrottle, + attr, + event_type, + tracing_data, + build_id; }; struct perf_session *perf_session__new(const char *filename, int mode, bool force); @@ -57,10 +62,10 @@ int __perf_session__process_events(struct perf_session *self, int perf_session__process_events(struct perf_session *self, struct perf_event_ops *event_ops); -struct symbol **perf_session__resolve_callchain(struct perf_session *self, - struct thread *thread, - struct ip_callchain *chain, - struct symbol **parent); +struct map_symbol *perf_session__resolve_callchain(struct perf_session *self, + struct thread *thread, + struct ip_callchain *chain, + struct symbol **parent); bool perf_session__has_traces(struct perf_session *self, const char *msg); @@ -80,10 +85,33 @@ static inline int __perf_session__create_kernel_maps(struct perf_session *self, self->vmlinux_maps, kernel); } +static inline int perf_session__create_kernel_maps(struct perf_session *self) +{ + return map_groups__create_kernel_maps(&self->kmaps, self->vmlinux_maps); +} + static inline struct map * perf_session__new_module_map(struct perf_session *self, u64 start, const char *filename) { return map_groups__new_module(&self->kmaps, start, filename); } + +int do_read(int fd, void *buf, size_t size); +void perf_session__update_sample_type(struct perf_session *self); + +#ifdef NO_NEWT_SUPPORT +static inline int perf_session__browse_hists(struct rb_root *hists __used, + u64 nr_hists __used, + u64 session_total __used, + const char *helpline __used, + const char *input_name __used) +{ + return 0; +} +#else +int perf_session__browse_hists(struct rb_root *hists, u64 nr_hists, + u64 session_total, const char *helpline, + const char *input_name); +#endif #endif /* __PERF_SESSION_H */ diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index cb0f327de9e8..da30b305fba0 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -18,39 +18,50 @@ char * field_sep; LIST_HEAD(hist_entry__sort_list); +static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width); +static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width); +static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width); +static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width); +static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width); + struct sort_entry sort_thread = { - .header = "Command: Pid", - .cmp = sort__thread_cmp, - .print = sort__thread_print, - .width = &threads__col_width, + .se_header = "Command: Pid", + .se_cmp = sort__thread_cmp, + .se_snprintf = hist_entry__thread_snprintf, + .se_width = &threads__col_width, }; struct sort_entry sort_comm = { - .header = "Command", - .cmp = sort__comm_cmp, - .collapse = sort__comm_collapse, - .print = sort__comm_print, - .width = &comms__col_width, + .se_header = "Command", + .se_cmp = sort__comm_cmp, + .se_collapse = sort__comm_collapse, + .se_snprintf = hist_entry__comm_snprintf, + .se_width = &comms__col_width, }; struct sort_entry sort_dso = { - .header = "Shared Object", - .cmp = sort__dso_cmp, - .print = sort__dso_print, - .width = &dsos__col_width, + .se_header = "Shared Object", + .se_cmp = sort__dso_cmp, + .se_snprintf = hist_entry__dso_snprintf, + .se_width = &dsos__col_width, }; struct sort_entry sort_sym = { - .header = "Symbol", - .cmp = sort__sym_cmp, - .print = sort__sym_print, + .se_header = "Symbol", + .se_cmp = sort__sym_cmp, + .se_snprintf = hist_entry__sym_snprintf, }; struct sort_entry sort_parent = { - .header = "Parent symbol", - .cmp = sort__parent_cmp, - .print = sort__parent_print, - .width = &parent_symbol__col_width, + .se_header = "Parent symbol", + .se_cmp = sort__parent_cmp, + .se_snprintf = hist_entry__parent_snprintf, + .se_width = &parent_symbol__col_width, }; struct sort_dimension { @@ -85,45 +96,38 @@ sort__thread_cmp(struct hist_entry *left, struct hist_entry *right) return right->thread->pid - left->thread->pid; } -int repsep_fprintf(FILE *fp, const char *fmt, ...) +static int repsep_snprintf(char *bf, size_t size, const char *fmt, ...) { int n; va_list ap; va_start(ap, fmt); - if (!field_sep) - n = vfprintf(fp, fmt, ap); - else { - char *bf = NULL; - n = vasprintf(&bf, fmt, ap); - if (n > 0) { - char *sep = bf; - - while (1) { - sep = strchr(sep, *field_sep); - if (sep == NULL) - break; - *sep = '.'; - } + n = vsnprintf(bf, size, fmt, ap); + if (field_sep && n > 0) { + char *sep = bf; + + while (1) { + sep = strchr(sep, *field_sep); + if (sep == NULL) + break; + *sep = '.'; } - fputs(bf, fp); - free(bf); } va_end(ap); return n; } -size_t -sort__thread_print(FILE *fp, struct hist_entry *self, unsigned int width) +static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) { - return repsep_fprintf(fp, "%*s:%5d", width - 6, + return repsep_snprintf(bf, size, "%*s:%5d", width, self->thread->comm ?: "", self->thread->pid); } -size_t -sort__comm_print(FILE *fp, struct hist_entry *self, unsigned int width) +static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) { - return repsep_fprintf(fp, "%*s", width, self->thread->comm); + return repsep_snprintf(bf, size, "%*s", width, self->thread->comm); } /* --sort dso */ @@ -131,8 +135,8 @@ sort__comm_print(FILE *fp, struct hist_entry *self, unsigned int width) int64_t sort__dso_cmp(struct hist_entry *left, struct hist_entry *right) { - struct dso *dso_l = left->map ? left->map->dso : NULL; - struct dso *dso_r = right->map ? right->map->dso : NULL; + struct dso *dso_l = left->ms.map ? left->ms.map->dso : NULL; + struct dso *dso_r = right->ms.map ? right->ms.map->dso : NULL; const char *dso_name_l, *dso_name_r; if (!dso_l || !dso_r) @@ -149,16 +153,16 @@ sort__dso_cmp(struct hist_entry *left, struct hist_entry *right) return strcmp(dso_name_l, dso_name_r); } -size_t -sort__dso_print(FILE *fp, struct hist_entry *self, unsigned int width) +static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) { - if (self->map && self->map->dso) { - const char *dso_name = !verbose ? self->map->dso->short_name : - self->map->dso->long_name; - return repsep_fprintf(fp, "%-*s", width, dso_name); + if (self->ms.map && self->ms.map->dso) { + const char *dso_name = !verbose ? self->ms.map->dso->short_name : + self->ms.map->dso->long_name; + return repsep_snprintf(bf, size, "%-*s", width, dso_name); } - return repsep_fprintf(fp, "%*llx", width, (u64)self->ip); + return repsep_snprintf(bf, size, "%*Lx", width, self->ip); } /* --sort symbol */ @@ -168,31 +172,31 @@ sort__sym_cmp(struct hist_entry *left, struct hist_entry *right) { u64 ip_l, ip_r; - if (left->sym == right->sym) + if (left->ms.sym == right->ms.sym) return 0; - ip_l = left->sym ? left->sym->start : left->ip; - ip_r = right->sym ? right->sym->start : right->ip; + ip_l = left->ms.sym ? left->ms.sym->start : left->ip; + ip_r = right->ms.sym ? right->ms.sym->start : right->ip; return (int64_t)(ip_r - ip_l); } - -size_t -sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used) +static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width __used) { size_t ret = 0; if (verbose) { - char o = self->map ? dso__symtab_origin(self->map->dso) : '!'; - ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip, o); + char o = self->ms.map ? dso__symtab_origin(self->ms.map->dso) : '!'; + ret += repsep_snprintf(bf, size, "%#018llx %c ", self->ip, o); } - ret += repsep_fprintf(fp, "[%c] ", self->level); - if (self->sym) - ret += repsep_fprintf(fp, "%s", self->sym->name); + ret += repsep_snprintf(bf + ret, size - ret, "[%c] ", self->level); + if (self->ms.sym) + ret += repsep_snprintf(bf + ret, size - ret, "%s", + self->ms.sym->name); else - ret += repsep_fprintf(fp, "%#016llx", (u64)self->ip); + ret += repsep_snprintf(bf + ret, size - ret, "%#016llx", self->ip); return ret; } @@ -231,10 +235,10 @@ sort__parent_cmp(struct hist_entry *left, struct hist_entry *right) return strcmp(sym_l->name, sym_r->name); } -size_t -sort__parent_print(FILE *fp, struct hist_entry *self, unsigned int width) +static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) { - return repsep_fprintf(fp, "%-*s", width, + return repsep_snprintf(bf, size, "%-*s", width, self->parent ? self->parent->name : "[other]"); } @@ -251,7 +255,7 @@ int sort_dimension__add(const char *tok) if (strncasecmp(tok, sd->name, strlen(tok))) continue; - if (sd->entry->collapse) + if (sd->entry->se_collapse) sort__need_collapse = 1; if (sd->entry == &sort_parent) { @@ -260,9 +264,8 @@ int sort_dimension__add(const char *tok) char err[BUFSIZ]; regerror(ret, &parent_regex, err, sizeof(err)); - fprintf(stderr, "Invalid regex: %s\n%s", - parent_pattern, err); - exit(-1); + pr_err("Invalid regex: %s\n%s", parent_pattern, err); + return -EINVAL; } sort__has_parent = 1; } diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h index 753f9ea99fb0..1d857aa2c01f 100644 --- a/tools/perf/util/sort.h +++ b/tools/perf/util/sort.h @@ -44,18 +44,23 @@ extern enum sort_type sort__first_dimension; struct hist_entry { struct rb_node rb_node; u64 count; + /* + * XXX WARNING! + * thread _has_ to come after ms, see + * hist_browser__selected_thread in util/newt.c + */ + struct map_symbol ms; struct thread *thread; - struct map *map; - struct symbol *sym; u64 ip; char level; - struct symbol *parent; - struct callchain_node callchain; + u8 filtered; + struct symbol *parent; union { unsigned long position; struct hist_entry *pair; struct rb_root sorted_chain; }; + struct callchain_node callchain[0]; }; enum sort_type { @@ -73,12 +78,13 @@ enum sort_type { struct sort_entry { struct list_head list; - const char *header; + const char *se_header; - int64_t (*cmp)(struct hist_entry *, struct hist_entry *); - int64_t (*collapse)(struct hist_entry *, struct hist_entry *); - size_t (*print)(FILE *fp, struct hist_entry *, unsigned int width); - unsigned int *width; + int64_t (*se_cmp)(struct hist_entry *, struct hist_entry *); + int64_t (*se_collapse)(struct hist_entry *, struct hist_entry *); + int (*se_snprintf)(struct hist_entry *self, char *bf, size_t size, + unsigned int width); + unsigned int *se_width; bool elide; }; @@ -87,7 +93,6 @@ extern struct list_head hist_entry__sort_list; void setup_sorting(const char * const usagestr[], const struct option *opts); -extern int repsep_fprintf(FILE *fp, const char *fmt, ...); extern size_t sort__thread_print(FILE *, struct hist_entry *, unsigned int); extern size_t sort__comm_print(FILE *, struct hist_entry *, unsigned int); extern size_t sort__dso_print(FILE *, struct hist_entry *, unsigned int); diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c index a175949ed216..0409fc7c0058 100644 --- a/tools/perf/util/string.c +++ b/tools/perf/util/string.c @@ -1,48 +1,5 @@ -#include "string.h" #include "util.h" - -static int hex(char ch) -{ - if ((ch >= '0') && (ch <= '9')) - return ch - '0'; - if ((ch >= 'a') && (ch <= 'f')) - return ch - 'a' + 10; - if ((ch >= 'A') && (ch <= 'F')) - return ch - 'A' + 10; - return -1; -} - -/* - * While we find nice hex chars, build a long_val. - * Return number of chars processed. - */ -int hex2u64(const char *ptr, u64 *long_val) -{ - const char *p = ptr; - *long_val = 0; - - while (*p) { - const int hex_val = hex(*p); - - if (hex_val < 0) - break; - - *long_val = (*long_val << 4) | hex_val; - p++; - } - - return p - ptr; -} - -char *strxfrchar(char *s, char from, char to) -{ - char *p = s; - - while ((p = strchr(p, from)) != NULL) - *p++ = to; - - return s; -} +#include "string.h" #define K 1024LL /* diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h deleted file mode 100644 index 542e44de3719..000000000000 --- a/tools/perf/util/string.h +++ /dev/null @@ -1,18 +0,0 @@ -#ifndef __PERF_STRING_H_ -#define __PERF_STRING_H_ - -#include <stdbool.h> -#include "types.h" - -int hex2u64(const char *ptr, u64 *val); -char *strxfrchar(char *s, char from, char to); -s64 perf_atoll(const char *str); -char **argv_split(const char *str, int *argcp); -void argv_free(char **argv); -bool strglobmatch(const char *str, const char *pat); -bool strlazymatch(const char *str, const char *pat); - -#define _STR(x) #x -#define STR(x) _STR(x) - -#endif /* __PERF_STRING_H */ diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index c458c4a371d1..f3d4151e46a1 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -1,13 +1,19 @@ -#include "util.h" -#include "../perf.h" -#include "sort.h" -#include "string.h" +#define _GNU_SOURCE +#include <ctype.h> +#include <dirent.h> +#include <errno.h> +#include <libgen.h> +#include <stdlib.h> +#include <stdio.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/param.h> +#include <fcntl.h> +#include <unistd.h> #include "symbol.h" -#include "thread.h" +#include "strlist.h" -#include "debug.h" - -#include <asm/bug.h> #include <libelf.h> #include <gelf.h> #include <elf.h> @@ -18,18 +24,6 @@ #define NT_GNU_BUILD_ID 3 #endif -enum dso_origin { - DSO__ORIG_KERNEL = 0, - DSO__ORIG_JAVA_JIT, - DSO__ORIG_BUILD_ID_CACHE, - DSO__ORIG_FEDORA, - DSO__ORIG_UBUNTU, - DSO__ORIG_BUILDID, - DSO__ORIG_DSO, - DSO__ORIG_KMODULE, - DSO__ORIG_NOT_FOUND, -}; - static void dsos__add(struct list_head *head, struct dso *dso); static struct map *map__new2(u64 start, struct dso *dso, enum map_type type); static int dso__load_kernel_sym(struct dso *self, struct map *map, @@ -126,8 +120,8 @@ static void map_groups__fixup_end(struct map_groups *self) static struct symbol *symbol__new(u64 start, u64 len, const char *name) { size_t namelen = strlen(name) + 1; - struct symbol *self = zalloc(symbol_conf.priv_size + - sizeof(*self) + namelen); + struct symbol *self = calloc(1, (symbol_conf.priv_size + + sizeof(*self) + namelen)); if (self == NULL) return NULL; @@ -178,7 +172,7 @@ static void dso__set_basename(struct dso *self) struct dso *dso__new(const char *name) { - struct dso *self = zalloc(sizeof(*self) + strlen(name) + 1); + struct dso *self = calloc(1, sizeof(*self) + strlen(name) + 1); if (self != NULL) { int i; @@ -870,8 +864,8 @@ out_close: if (err == 0) return nr; out: - pr_warning("%s: problems reading %s PLT info.\n", - __func__, self->long_name); + pr_debug("%s: problems reading %s PLT info.\n", + __func__, self->long_name); return 0; } @@ -1025,7 +1019,7 @@ static int dso__load_sym(struct dso *self, struct map *map, const char *name, } curr_map->map_ip = identity__map_ip; curr_map->unmap_ip = identity__map_ip; - curr_dso->origin = DSO__ORIG_KERNEL; + curr_dso->origin = self->origin; map_groups__insert(kmap->kmaps, curr_map); dsos__add(&dsos__kernel, curr_dso); dso__set_loaded(curr_dso, map->type); @@ -1394,13 +1388,13 @@ static int dso__kernel_module_get_build_id(struct dso *self) return 0; } -static int map_groups__set_modules_path_dir(struct map_groups *self, char *dirname) +static int map_groups__set_modules_path_dir(struct map_groups *self, char *dir_name) { struct dirent *dent; - DIR *dir = opendir(dirname); + DIR *dir = opendir(dir_name); if (!dir) { - pr_debug("%s: cannot open %s dir\n", __func__, dirname); + pr_debug("%s: cannot open %s dir\n", __func__, dir_name); return -1; } @@ -1413,7 +1407,7 @@ static int map_groups__set_modules_path_dir(struct map_groups *self, char *dirna continue; snprintf(path, sizeof(path), "%s/%s", - dirname, dent->d_name); + dir_name, dent->d_name); if (map_groups__set_modules_path_dir(self, path) < 0) goto failure; } else { @@ -1433,7 +1427,7 @@ static int map_groups__set_modules_path_dir(struct map_groups *self, char *dirna continue; snprintf(path, sizeof(path), "%s/%s", - dirname, dent->d_name); + dir_name, dent->d_name); long_name = strdup(path); if (long_name == NULL) @@ -1470,8 +1464,8 @@ static int map_groups__set_modules_path(struct map_groups *self) */ static struct map *map__new2(u64 start, struct dso *dso, enum map_type type) { - struct map *self = zalloc(sizeof(*self) + - (dso->kernel ? sizeof(struct kmap) : 0)); + struct map *self = calloc(1, (sizeof(*self) + + (dso->kernel ? sizeof(struct kmap) : 0))); if (self != NULL) { /* * ->end will be filled after we load all the symbols @@ -1895,6 +1889,17 @@ out_fail: return -1; } +size_t vmlinux_path__fprintf(FILE *fp) +{ + int i; + size_t printed = 0; + + for (i = 0; i < vmlinux_path__nr_entries; ++i) + printed += fprintf(fp, "[%d] %s\n", i, vmlinux_path[i]); + + return printed; +} + static int setup_list(struct strlist **list, const char *list_str, const char *list_name) { @@ -1964,3 +1969,46 @@ int map_groups__create_kernel_maps(struct map_groups *self, map_groups__fixup_end(self); return 0; } + +static int hex(char ch) +{ + if ((ch >= '0') && (ch <= '9')) + return ch - '0'; + if ((ch >= 'a') && (ch <= 'f')) + return ch - 'a' + 10; + if ((ch >= 'A') && (ch <= 'F')) + return ch - 'A' + 10; + return -1; +} + +/* + * While we find nice hex chars, build a long_val. + * Return number of chars processed. + */ +int hex2u64(const char *ptr, u64 *long_val) +{ + const char *p = ptr; + *long_val = 0; + + while (*p) { + const int hex_val = hex(*p); + + if (hex_val < 0) + break; + + *long_val = (*long_val << 4) | hex_val; + p++; + } + + return p - ptr; +} + +char *strxfrchar(char *s, char from, char to) +{ + char *p = s; + + while ((p = strchr(p, from)) != NULL) + *p++ = to; + + return s; +} diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index f30a37428919..757fae3f5ee0 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -3,10 +3,11 @@ #include <linux/types.h> #include <stdbool.h> -#include "types.h" +#include <stdint.h> +#include "map.h" #include <linux/list.h> #include <linux/rbtree.h> -#include "event.h" +#include <stdio.h> #define DEBUG_CACHE_DIR ".debug" @@ -29,6 +30,9 @@ static inline char *bfd_demangle(void __used *v, const char __used *c, #endif #endif +int hex2u64(const char *ptr, u64 *val); +char *strxfrchar(char *s, char from, char to); + /* * libelf 0.8.x and earlier do not support ELF_C_READ_MMAP; * for newer versions we can use mmap to reduce memory usage: @@ -44,6 +48,8 @@ static inline char *bfd_demangle(void __used *v, const char __used *c, #define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ #endif +#define BUILD_ID_SIZE 20 + struct symbol { struct rb_node rb_node; u64 start; @@ -88,6 +94,11 @@ struct ref_reloc_sym { u64 unrelocated_addr; }; +struct map_symbol { + struct map *map; + struct symbol *sym; +}; + struct addr_location { struct thread *thread; struct map *map; @@ -106,6 +117,7 @@ struct dso { u8 has_build_id:1; u8 kernel:1; u8 hit:1; + u8 annotate_warned:1; unsigned char origin; u8 sorted_by_name; u8 loaded; @@ -150,6 +162,19 @@ size_t dsos__fprintf_buildid(FILE *fp, bool with_hits); size_t dso__fprintf_buildid(struct dso *self, FILE *fp); size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp); + +enum dso_origin { + DSO__ORIG_KERNEL = 0, + DSO__ORIG_JAVA_JIT, + DSO__ORIG_BUILD_ID_CACHE, + DSO__ORIG_FEDORA, + DSO__ORIG_UBUNTU, + DSO__ORIG_BUILDID, + DSO__ORIG_DSO, + DSO__ORIG_KMODULE, + DSO__ORIG_NOT_FOUND, +}; + char dso__symtab_origin(const struct dso *self); void dso__set_long_name(struct dso *self, char *name); void dso__set_build_id(struct dso *self, void *build_id); @@ -169,4 +194,6 @@ int kallsyms__parse(const char *filename, void *arg, int symbol__init(void); bool symbol_type__is_a(char symbol_type, enum map_type map_type); +size_t vmlinux_path__fprintf(FILE *fp); + #endif /* __PERF_SYMBOL */ diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index fa968312ee7d..1f7ecd47f499 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -7,13 +7,35 @@ #include "util.h" #include "debug.h" -void map_groups__init(struct map_groups *self) +int find_all_tid(int pid, pid_t ** all_tid) { + char name[256]; + int items; + struct dirent **namelist = NULL; + int ret = 0; int i; - for (i = 0; i < MAP__NR_TYPES; ++i) { - self->maps[i] = RB_ROOT; - INIT_LIST_HEAD(&self->removed_maps[i]); + + sprintf(name, "/proc/%d/task", pid); + items = scandir(name, &namelist, NULL, NULL); + if (items <= 0) + return -ENOENT; + *all_tid = malloc(sizeof(pid_t) * items); + if (!*all_tid) { + ret = -ENOMEM; + goto failure; } + + for (i = 0; i < items; i++) + (*all_tid)[i] = atoi(namelist[i]->d_name); + + ret = items; + +failure: + for (i=0; i<items; i++) + free(namelist[i]); + free(namelist); + + return ret; } static struct thread *thread__new(pid_t pid) @@ -31,28 +53,6 @@ static struct thread *thread__new(pid_t pid) return self; } -static void map_groups__flush(struct map_groups *self) -{ - int type; - - for (type = 0; type < MAP__NR_TYPES; type++) { - struct rb_root *root = &self->maps[type]; - struct rb_node *next = rb_first(root); - - while (next) { - struct map *pos = rb_entry(next, struct map, rb_node); - next = rb_next(&pos->rb_node); - rb_erase(&pos->rb_node, root); - /* - * We may have references to this map, for - * instance in some hist_entry instances, so - * just move them to a separate list. - */ - list_add_tail(&pos->node, &self->removed_maps[pos->type]); - } - } -} - int thread__set_comm(struct thread *self, const char *comm) { int err; @@ -79,69 +79,10 @@ int thread__comm_len(struct thread *self) return self->comm_len; } -size_t __map_groups__fprintf_maps(struct map_groups *self, - enum map_type type, FILE *fp) -{ - size_t printed = fprintf(fp, "%s:\n", map_type__name[type]); - struct rb_node *nd; - - for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) { - struct map *pos = rb_entry(nd, struct map, rb_node); - printed += fprintf(fp, "Map:"); - printed += map__fprintf(pos, fp); - if (verbose > 2) { - printed += dso__fprintf(pos->dso, type, fp); - printed += fprintf(fp, "--\n"); - } - } - - return printed; -} - -size_t map_groups__fprintf_maps(struct map_groups *self, FILE *fp) -{ - size_t printed = 0, i; - for (i = 0; i < MAP__NR_TYPES; ++i) - printed += __map_groups__fprintf_maps(self, i, fp); - return printed; -} - -static size_t __map_groups__fprintf_removed_maps(struct map_groups *self, - enum map_type type, FILE *fp) -{ - struct map *pos; - size_t printed = 0; - - list_for_each_entry(pos, &self->removed_maps[type], node) { - printed += fprintf(fp, "Map:"); - printed += map__fprintf(pos, fp); - if (verbose > 1) { - printed += dso__fprintf(pos->dso, type, fp); - printed += fprintf(fp, "--\n"); - } - } - return printed; -} - -static size_t map_groups__fprintf_removed_maps(struct map_groups *self, FILE *fp) -{ - size_t printed = 0, i; - for (i = 0; i < MAP__NR_TYPES; ++i) - printed += __map_groups__fprintf_removed_maps(self, i, fp); - return printed; -} - -static size_t map_groups__fprintf(struct map_groups *self, FILE *fp) -{ - size_t printed = map_groups__fprintf_maps(self, fp); - printed += fprintf(fp, "Removed maps:\n"); - return printed + map_groups__fprintf_removed_maps(self, fp); -} - static size_t thread__fprintf(struct thread *self, FILE *fp) { return fprintf(fp, "Thread %d %s\n", self->pid, self->comm) + - map_groups__fprintf(&self->mg, fp); + map_groups__fprintf(&self->mg, verbose, fp); } struct thread *perf_session__findnew(struct perf_session *self, pid_t pid) @@ -183,127 +124,12 @@ struct thread *perf_session__findnew(struct perf_session *self, pid_t pid) return th; } -static int map_groups__fixup_overlappings(struct map_groups *self, - struct map *map) -{ - struct rb_root *root = &self->maps[map->type]; - struct rb_node *next = rb_first(root); - - while (next) { - struct map *pos = rb_entry(next, struct map, rb_node); - next = rb_next(&pos->rb_node); - - if (!map__overlap(pos, map)) - continue; - - if (verbose >= 2) { - fputs("overlapping maps:\n", stderr); - map__fprintf(map, stderr); - map__fprintf(pos, stderr); - } - - rb_erase(&pos->rb_node, root); - /* - * We may have references to this map, for instance in some - * hist_entry instances, so just move them to a separate - * list. - */ - list_add_tail(&pos->node, &self->removed_maps[map->type]); - /* - * Now check if we need to create new maps for areas not - * overlapped by the new map: - */ - if (map->start > pos->start) { - struct map *before = map__clone(pos); - - if (before == NULL) - return -ENOMEM; - - before->end = map->start - 1; - map_groups__insert(self, before); - if (verbose >= 2) - map__fprintf(before, stderr); - } - - if (map->end < pos->end) { - struct map *after = map__clone(pos); - - if (after == NULL) - return -ENOMEM; - - after->start = map->end + 1; - map_groups__insert(self, after); - if (verbose >= 2) - map__fprintf(after, stderr); - } - } - - return 0; -} - -void maps__insert(struct rb_root *maps, struct map *map) -{ - struct rb_node **p = &maps->rb_node; - struct rb_node *parent = NULL; - const u64 ip = map->start; - struct map *m; - - while (*p != NULL) { - parent = *p; - m = rb_entry(parent, struct map, rb_node); - if (ip < m->start) - p = &(*p)->rb_left; - else - p = &(*p)->rb_right; - } - - rb_link_node(&map->rb_node, parent, p); - rb_insert_color(&map->rb_node, maps); -} - -struct map *maps__find(struct rb_root *maps, u64 ip) -{ - struct rb_node **p = &maps->rb_node; - struct rb_node *parent = NULL; - struct map *m; - - while (*p != NULL) { - parent = *p; - m = rb_entry(parent, struct map, rb_node); - if (ip < m->start) - p = &(*p)->rb_left; - else if (ip > m->end) - p = &(*p)->rb_right; - else - return m; - } - - return NULL; -} - void thread__insert_map(struct thread *self, struct map *map) { - map_groups__fixup_overlappings(&self->mg, map); + map_groups__fixup_overlappings(&self->mg, map, verbose, stderr); map_groups__insert(&self->mg, map); } -/* - * XXX This should not really _copy_ te maps, but refcount them. - */ -static int map_groups__clone(struct map_groups *self, - struct map_groups *parent, enum map_type type) -{ - struct rb_node *nd; - for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) { - struct map *map = rb_entry(nd, struct map, rb_node); - struct map *new = map__clone(map); - if (new == NULL) - return -ENOMEM; - map_groups__insert(self, new); - } - return 0; -} - int thread__fork(struct thread *self, struct thread *parent) { int i; @@ -336,15 +162,3 @@ size_t perf_session__fprintf(struct perf_session *self, FILE *fp) return ret; } - -struct symbol *map_groups__find_symbol(struct map_groups *self, - enum map_type type, u64 addr, - symbol_filter_t filter) -{ - struct map *map = map_groups__find(self, type, addr); - - if (map != NULL) - return map__find_symbol(map, map->map_ip(map, addr), filter); - - return NULL; -} diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index dcf70303e58e..9c488fcadec9 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -5,14 +5,6 @@ #include <unistd.h> #include "symbol.h" -struct map_groups { - struct rb_root maps[MAP__NR_TYPES]; - struct list_head removed_maps[MAP__NR_TYPES]; -}; - -size_t __map_groups__fprintf_maps(struct map_groups *self, - enum map_type type, FILE *fp); - struct thread { struct rb_node rb_node; struct map_groups mg; @@ -23,29 +15,16 @@ struct thread { int comm_len; }; -void map_groups__init(struct map_groups *self); +struct perf_session; + +int find_all_tid(int pid, pid_t ** all_tid); int thread__set_comm(struct thread *self, const char *comm); int thread__comm_len(struct thread *self); struct thread *perf_session__findnew(struct perf_session *self, pid_t pid); void thread__insert_map(struct thread *self, struct map *map); int thread__fork(struct thread *self, struct thread *parent); -size_t map_groups__fprintf_maps(struct map_groups *self, FILE *fp); size_t perf_session__fprintf(struct perf_session *self, FILE *fp); -void maps__insert(struct rb_root *maps, struct map *map); -struct map *maps__find(struct rb_root *maps, u64 addr); - -static inline void map_groups__insert(struct map_groups *self, struct map *map) -{ - maps__insert(&self->maps[map->type], map); -} - -static inline struct map *map_groups__find(struct map_groups *self, - enum map_type type, u64 addr) -{ - return maps__find(&self->maps[type], addr); -} - static inline struct map *thread__find_map(struct thread *self, enum map_type type, u64 addr) { @@ -62,26 +41,4 @@ void thread__find_addr_location(struct thread *self, enum map_type type, u64 addr, struct addr_location *al, symbol_filter_t filter); -struct symbol *map_groups__find_symbol(struct map_groups *self, - enum map_type type, u64 addr, - symbol_filter_t filter); - -static inline struct symbol *map_groups__find_function(struct map_groups *self, - u64 addr, - symbol_filter_t filter) -{ - return map_groups__find_symbol(self, MAP__FUNCTION, addr, filter); -} - -struct map *map_groups__find_by_name(struct map_groups *self, - enum map_type type, const char *name); - -int __map_groups__create_kernel_maps(struct map_groups *self, - struct map *vmlinux_maps[MAP__NR_TYPES], - struct dso *kernel); -int map_groups__create_kernel_maps(struct map_groups *self, - struct map *vmlinux_maps[MAP__NR_TYPES]); - -struct map *map_groups__new_module(struct map_groups *self, u64 start, - const char *filename); #endif /* __PERF_THREAD_H */ diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c index 5ea8973ad331..30cd9b575953 100644 --- a/tools/perf/util/trace-event-info.c +++ b/tools/perf/util/trace-event-info.c @@ -154,10 +154,17 @@ static void put_tracing_file(char *file) free(file); } +static ssize_t calc_data_size; + static ssize_t write_or_die(const void *buf, size_t len) { int ret; + if (calc_data_size) { + calc_data_size += len; + return len; + } + ret = write(output_fd, buf, len); if (ret < 0) die("writing to '%s'", output_file); @@ -526,3 +533,20 @@ int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events) return 0; } + +ssize_t read_tracing_data_size(int fd, struct perf_event_attr *pattrs, + int nb_events) +{ + ssize_t size; + int err = 0; + + calc_data_size = 1; + err = read_tracing_data(fd, pattrs, nb_events); + size = calc_data_size - 1; + calc_data_size = 0; + + if (err < 0) + return err; + + return size; +} diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index 9b3c20f42f98..d6ef414075a6 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -37,10 +37,12 @@ int header_page_ts_offset; int header_page_ts_size; int header_page_size_offset; int header_page_size_size; +int header_page_overwrite_offset; +int header_page_overwrite_size; int header_page_data_offset; int header_page_data_size; -int latency_format; +bool latency_format; static char *input_buf; static unsigned long long input_buf_ptr; @@ -628,23 +630,32 @@ static int test_type(enum event_type type, enum event_type expect) return 0; } -static int test_type_token(enum event_type type, char *token, - enum event_type expect, const char *expect_tok) +static int __test_type_token(enum event_type type, char *token, + enum event_type expect, const char *expect_tok, + bool warn) { if (type != expect) { - warning("Error: expected type %d but read %d", - expect, type); + if (warn) + warning("Error: expected type %d but read %d", + expect, type); return -1; } if (strcmp(token, expect_tok) != 0) { - warning("Error: expected '%s' but read '%s'", - expect_tok, token); + if (warn) + warning("Error: expected '%s' but read '%s'", + expect_tok, token); return -1; } return 0; } +static int test_type_token(enum event_type type, char *token, + enum event_type expect, const char *expect_tok) +{ + return __test_type_token(type, token, expect, expect_tok, true); +} + static int __read_expect_type(enum event_type expect, char **tok, int newline_ok) { enum event_type type; @@ -661,7 +672,8 @@ static int read_expect_type(enum event_type expect, char **tok) return __read_expect_type(expect, tok, 1); } -static int __read_expected(enum event_type expect, const char *str, int newline_ok) +static int __read_expected(enum event_type expect, const char *str, + int newline_ok, bool warn) { enum event_type type; char *token; @@ -672,21 +684,26 @@ static int __read_expected(enum event_type expect, const char *str, int newline_ else type = read_token_item(&token); - ret = test_type_token(type, token, expect, str); + ret = __test_type_token(type, token, expect, str, warn); free_token(token); return ret; } +static int read_expected_warn(enum event_type expect, const char *str, bool warn) +{ + return __read_expected(expect, str, 1, warn); +} + static int read_expected(enum event_type expect, const char *str) { - return __read_expected(expect, str, 1); + return __read_expected(expect, str, 1, true); } static int read_expected_item(enum event_type expect, const char *str) { - return __read_expected(expect, str, 0); + return __read_expected(expect, str, 0, true); } static char *event_read_name(void) @@ -744,7 +761,7 @@ static int field_is_string(struct format_field *field) static int field_is_dynamic(struct format_field *field) { - if (!strcmp(field->type, "__data_loc")) + if (!strncmp(field->type, "__data_loc", 10)) return 1; return 0; @@ -3088,7 +3105,7 @@ static void print_args(struct print_arg *args) } static void parse_header_field(const char *field, - int *offset, int *size) + int *offset, int *size, bool warn) { char *token; int type; @@ -3103,7 +3120,7 @@ static void parse_header_field(const char *field, goto fail; free_token(token); - if (read_expected(EVENT_ITEM, field) < 0) + if (read_expected_warn(EVENT_ITEM, field, warn) < 0) return; if (read_expected(EVENT_OP, ";") < 0) return; @@ -3160,11 +3177,13 @@ int parse_header_page(char *buf, unsigned long size) init_input_buf(buf, size); parse_header_field("timestamp", &header_page_ts_offset, - &header_page_ts_size); + &header_page_ts_size, true); parse_header_field("commit", &header_page_size_offset, - &header_page_size_size); + &header_page_size_size, true); + parse_header_field("overwrite", &header_page_overwrite_offset, + &header_page_overwrite_size, false); parse_header_field("data", &header_page_data_offset, - &header_page_data_size); + &header_page_data_size, true); return 0; } diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c index 7cd1193918c7..44889c9b5630 100644 --- a/tools/perf/util/trace-event-read.c +++ b/tools/perf/util/trace-event-read.c @@ -50,14 +50,37 @@ static int long_size; static unsigned long page_size; +static ssize_t calc_data_size; + +static int do_read(int fd, void *buf, int size) +{ + int rsize = size; + + while (size) { + int ret = read(fd, buf, size); + + if (ret <= 0) + return -1; + + size -= ret; + buf += ret; + } + + return rsize; +} + static int read_or_die(void *data, int size) { int r; - r = read(input_fd, data, size); - if (r != size) + r = do_read(input_fd, data, size); + if (r <= 0) die("reading input file (size expected=%d received=%d)", size, r); + + if (calc_data_size) + calc_data_size += r; + return r; } @@ -82,56 +105,28 @@ static char *read_string(void) char buf[BUFSIZ]; char *str = NULL; int size = 0; - int i; off_t r; + char c; for (;;) { - r = read(input_fd, buf, BUFSIZ); + r = read(input_fd, &c, 1); if (r < 0) die("reading input file"); if (!r) die("no data"); - for (i = 0; i < r; i++) { - if (!buf[i]) - break; - } - if (i < r) - break; + buf[size++] = c; - if (str) { - size += BUFSIZ; - str = realloc(str, size); - if (!str) - die("malloc of size %d", size); - memcpy(str + (size - BUFSIZ), buf, BUFSIZ); - } else { - size = BUFSIZ; - str = malloc_or_die(size); - memcpy(str, buf, size); - } + if (!c) + break; } - /* trailing \0: */ - i++; - - /* move the file descriptor to the end of the string */ - r = lseek(input_fd, -(r - i), SEEK_CUR); - if (r == (off_t)-1) - die("lseek"); - - if (str) { - size += i; - str = realloc(str, size); - if (!str) - die("malloc of size %d", size); - memcpy(str + (size - i), buf, i); - } else { - size = i; - str = malloc_or_die(i); - memcpy(str, buf, i); - } + if (calc_data_size) + calc_data_size += size; + + str = malloc_or_die(size); + memcpy(str, buf, size); return str; } @@ -459,7 +454,7 @@ struct record *trace_read_data(int cpu) return data; } -void trace_report(int fd) +ssize_t trace_report(int fd) { char buf[BUFSIZ]; char test[] = { 23, 8, 68 }; @@ -467,6 +462,9 @@ void trace_report(int fd) int show_version = 0; int show_funcs = 0; int show_printk = 0; + ssize_t size; + + calc_data_size = 1; input_fd = fd; @@ -499,14 +497,17 @@ void trace_report(int fd) read_proc_kallsyms(); read_ftrace_printk(); + size = calc_data_size - 1; + calc_data_size = 0; + if (show_funcs) { print_funcs(); - return; + return size; } if (show_printk) { print_printk(); - return; + return size; } - return; + return size; } diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h index c3269b937db4..1f45d468fd9a 100644 --- a/tools/perf/util/trace-event.h +++ b/tools/perf/util/trace-event.h @@ -1,6 +1,7 @@ #ifndef __PERF_TRACE_EVENTS_H #define __PERF_TRACE_EVENTS_H +#include <stdbool.h> #include "parse-events.h" #define __unused __attribute__((unused)) @@ -162,7 +163,7 @@ struct record *trace_read_data(int cpu); void parse_set_info(int nr_cpus, int long_sz); -void trace_report(int fd); +ssize_t trace_report(int fd); void *malloc_or_die(unsigned int size); @@ -241,7 +242,7 @@ extern int header_page_size_size; extern int header_page_data_offset; extern int header_page_data_size; -extern int latency_format; +extern bool latency_format; int parse_header_page(char *buf, unsigned long size); int trace_parse_common_type(void *data); @@ -258,6 +259,8 @@ void *raw_field_ptr(struct event *event, const char *name, void *data); unsigned long long eval_flag(const char *flag); int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events); +ssize_t read_tracing_data_size(int fd, struct perf_event_attr *pattrs, + int nb_events); /* taken from kernel/trace/trace.h */ enum trace_flag_type { diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h index 0f5b2a6f1080..fbf45d1b26f7 100644 --- a/tools/perf/util/util.h +++ b/tools/perf/util/util.h @@ -42,12 +42,14 @@ #define _ALL_SOURCE 1 #define _GNU_SOURCE 1 #define _BSD_SOURCE 1 +#define HAS_BOOL #include <unistd.h> #include <stdio.h> #include <sys/stat.h> #include <sys/statfs.h> #include <fcntl.h> +#include <stdbool.h> #include <stddef.h> #include <stdlib.h> #include <stdarg.h> @@ -78,6 +80,7 @@ #include <pwd.h> #include <inttypes.h> #include "../../../include/linux/magic.h" +#include "types.h" #ifndef NO_ICONV @@ -295,6 +298,13 @@ extern void *xmemdupz(const void *data, size_t len); extern char *xstrndup(const char *str, size_t len); extern void *xrealloc(void *ptr, size_t size) __attribute__((weak)); +static inline void *xzalloc(size_t size) +{ + void *buf = xmalloc(size); + + return memset(buf, 0, size); +} + static inline void *zalloc(size_t size) { return calloc(1, size); @@ -309,6 +319,7 @@ static inline int has_extension(const char *filename, const char *ext) { size_t len = strlen(filename); size_t extlen = strlen(ext); + return len > extlen && !memcmp(filename + len - extlen, ext, extlen); } @@ -322,6 +333,7 @@ static inline int has_extension(const char *filename, const char *ext) #undef isalnum #undef tolower #undef toupper + extern unsigned char sane_ctype[256]; #define GIT_SPACE 0x01 #define GIT_DIGIT 0x02 @@ -406,4 +418,13 @@ void git_qsort(void *base, size_t nmemb, size_t size, int mkdir_p(char *path, mode_t mode); int copyfile(const char *from, const char *to); +s64 perf_atoll(const char *str); +char **argv_split(const char *str, int *argcp); +void argv_free(char **argv); +bool strglobmatch(const char *str, const char *pat); +bool strlazymatch(const char *str, const char *pat); + +#define _STR(x) #x +#define STR(x) _STR(x) + #endif |