diff options
| author | Stephen Rothwell <sfr@canb.auug.org.au> | 2009-08-10 17:34:43 +1000 |
|---|---|---|
| committer | Stephen Rothwell <sfr@canb.auug.org.au> | 2009-08-10 17:34:43 +1000 |
| commit | 2b734d4162abda541183e8e16e6482559a7f20d2 (patch) | |
| tree | 4d6dbf8b0be36da767c7324a7f53f08130832dbb | |
| parent | e474f06419a8ee21735b7d16caa34caa147170a3 (diff) | |
| parent | 5fe910df8a000984eb136869407483d76122c22b (diff) | |
Merge commit 'tip/auto-latest'
Conflicts:
arch/x86/include/asm/socket.h
include/linux/rcupdate.h
kernel/fork.c
219 files changed, 9049 insertions, 6898 deletions
diff --git a/Documentation/RCU/RTFP.txt b/Documentation/RCU/RTFP.txt index 9f711d2df91b..d2b85237c76e 100644 --- a/Documentation/RCU/RTFP.txt +++ b/Documentation/RCU/RTFP.txt @@ -743,3 +743,80 @@ Revised: RCU, realtime RCU, sleepable RCU, performance. " } + +@article{PaulEMcKenney2008RCUOSR +,author="Paul E. McKenney and Jonathan Walpole" +,title="Introducing technology into the {Linux} kernel: a case study" +,Year="2008" +,journal="SIGOPS Oper. Syst. Rev." +,volume="42" +,number="5" +,pages="4--17" +,issn="0163-5980" +,doi={http://doi.acm.org/10.1145/1400097.1400099} +,publisher="ACM" +,address="New York, NY, USA" +,annotation={ + Linux changed RCU to a far greater degree than RCU has changed Linux. +} +} + +@unpublished{PaulEMcKenney2008HierarchicalRCU +,Author="Paul E. McKenney" +,Title="Hierarchical {RCU}" +,month="November" +,day="3" +,year="2008" +,note="Available: +\url{http://lwn.net/Articles/305782/} +[Viewed November 6, 2008]" +,annotation=" + RCU with combining-tree-based grace-period detection, + permitting it to handle thousands of CPUs. +" +} + +@conference{PaulEMcKenney2009MaliciousURCU +,Author="Paul E. McKenney" +,Title="Using a Malicious User-Level {RCU} to Torture {RCU}-Based Algorithms" +,Booktitle="linux.conf.au 2009" +,month="January" +,year="2009" +,address="Hobart, Australia" +,note="Available: +\url{http://www.rdrop.com/users/paulmck/RCU/urcutorture.2009.01.22a.pdf} +[Viewed February 2, 2009]" +,annotation=" + Realtime RCU and torture-testing RCU uses. +" +} + +@unpublished{MathieuDesnoyers2009URCU +,Author="Mathieu Desnoyers" +,Title="[{RFC} git tree] Userspace {RCU} (urcu) for {Linux}" +,month="February" +,day="5" +,year="2009" +,note="Available: +\url{http://lkml.org/lkml/2009/2/5/572} +\url{git://lttng.org/userspace-rcu.git} +[Viewed February 20, 2009]" +,annotation=" + Mathieu Desnoyers's user-space RCU implementation. + git://lttng.org/userspace-rcu.git +" +} + +@unpublished{PaulEMcKenney2009BloatWatchRCU +,Author="Paul E. McKenney" +,Title="{RCU}: The {Bloatwatch} Edition" +,month="March" +,day="17" +,year="2009" +,note="Available: +\url{http://lwn.net/Articles/323929/} +[Viewed March 20, 2009]" +,annotation=" + Uniprocessor assumptions allow simplified RCU implementation. +" +} diff --git a/Documentation/RCU/UP.txt b/Documentation/RCU/UP.txt index aab4a9ec3931..90ec5341ee98 100644 --- a/Documentation/RCU/UP.txt +++ b/Documentation/RCU/UP.txt @@ -2,14 +2,13 @@ RCU on Uniprocessor Systems A common misconception is that, on UP systems, the call_rcu() primitive -may immediately invoke its function, and that the synchronize_rcu() -primitive may return immediately. The basis of this misconception +may immediately invoke its function. The basis of this misconception is that since there is only one CPU, it should not be necessary to wait for anything else to get done, since there are no other CPUs for anything else to be happening on. Although this approach will -sort- -of- work a surprising amount of the time, it is a very bad idea in general. -This document presents three examples that demonstrate exactly how bad an -idea this is. +This document presents three examples that demonstrate exactly how bad +an idea this is. Example 1: softirq Suicide @@ -82,11 +81,18 @@ Quick Quiz #2: What locking restriction must RCU callbacks respect? Summary -Permitting call_rcu() to immediately invoke its arguments or permitting -synchronize_rcu() to immediately return breaks RCU, even on a UP system. -So do not do it! Even on a UP system, the RCU infrastructure -must- -respect grace periods, and -must- invoke callbacks from a known environment -in which no locks are held. +Permitting call_rcu() to immediately invoke its arguments breaks RCU, +even on a UP system. So do not do it! Even on a UP system, the RCU +infrastructure -must- respect grace periods, and -must- invoke callbacks +from a known environment in which no locks are held. + +It -is- safe for synchronize_sched() and synchronize_rcu_bh() to return +immediately on an UP system. It is also safe for synchronize_rcu() +to return immediately on UP systems, except when running preemptable +RCU. + +Quick Quiz #3: Why can't synchronize_rcu() return immediately on + UP systems running preemptable RCU? Answer to Quick Quiz #1: @@ -117,3 +123,13 @@ Answer to Quick Quiz #2: callbacks acquire locks directly. However, a great many RCU callbacks do acquire locks -indirectly-, for example, via the kfree() primitive. + +Answer to Quick Quiz #3: + Why can't synchronize_rcu() return immediately on UP systems + running preemptable RCU? + + Because some other task might have been preempted in the middle + of an RCU read-side critical section. If synchronize_rcu() + simply immediately returned, it would prematurely signal the + end of the grace period, which would come as a nasty shock to + that other thread when it started running again. diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt index accfe2f5247d..51525a30e8b4 100644 --- a/Documentation/RCU/checklist.txt +++ b/Documentation/RCU/checklist.txt @@ -11,7 +11,10 @@ over a rather long period of time, but improvements are always welcome! structure is updated more than about 10% of the time, then you should strongly consider some other approach, unless detailed performance measurements show that RCU is nonetheless - the right tool for the job. + the right tool for the job. Yes, you might think of RCU + as simply cutting overhead off of the readers and imposing it + on the writers. That is exactly why normal uses of RCU will + do much more reading than updating. Another exception is where performance is not an issue, and RCU provides a simpler implementation. An example of this situation @@ -240,10 +243,11 @@ over a rather long period of time, but improvements are always welcome! instead need to use synchronize_irq() or synchronize_sched(). 12. Any lock acquired by an RCU callback must be acquired elsewhere - with irq disabled, e.g., via spin_lock_irqsave(). Failing to - disable irq on a given acquisition of that lock will result in - deadlock as soon as the RCU callback happens to interrupt that - acquisition's critical section. + with softirq disabled, e.g., via spin_lock_irqsave(), + spin_lock_bh(), etc. Failing to disable irq on a given + acquisition of that lock will result in deadlock as soon as the + RCU callback happens to interrupt that acquisition's critical + section. 13. RCU callbacks can be and are executed in parallel. In many cases, the callback code simply wrappers around kfree(), so that this @@ -310,3 +314,9 @@ over a rather long period of time, but improvements are always welcome! Because these primitives only wait for pre-existing readers, it is the caller's responsibility to guarantee safety to any subsequent readers. + +16. The various RCU read-side primitives do -not- contain memory + barriers. The CPU (and in some cases, the compiler) is free + to reorder code into and out of RCU read-side critical sections. + It is the responsibility of the RCU update-side primitives to + deal with this. diff --git a/Documentation/RCU/rcubarrier.txt b/Documentation/RCU/rcubarrier.txt index 909602d409bb..e439a0edee22 100644 --- a/Documentation/RCU/rcubarrier.txt +++ b/Documentation/RCU/rcubarrier.txt @@ -170,6 +170,13 @@ module invokes call_rcu() from timers, you will need to first cancel all the timers, and only then invoke rcu_barrier() to wait for any remaining RCU callbacks to complete. +Of course, if you module uses call_rcu_bh(), you will need to invoke +rcu_barrier_bh() before unloading. Similarly, if your module uses +call_rcu_sched(), you will need to invoke rcu_barrier_sched() before +unloading. If your module uses call_rcu(), call_rcu_bh(), -and- +call_rcu_sched(), then you will need to invoke each of rcu_barrier(), +rcu_barrier_bh(), and rcu_barrier_sched(). + Implementing rcu_barrier() diff --git a/Documentation/RCU/torture.txt b/Documentation/RCU/torture.txt index a342b6e1cc10..9dba3bb90e60 100644 --- a/Documentation/RCU/torture.txt +++ b/Documentation/RCU/torture.txt @@ -76,8 +76,10 @@ torture_type The type of RCU to test: "rcu" for the rcu_read_lock() API, "rcu_sync" for rcu_read_lock() with synchronous reclamation, "rcu_bh" for the rcu_read_lock_bh() API, "rcu_bh_sync" for rcu_read_lock_bh() with synchronous reclamation, "srcu" for - the "srcu_read_lock()" API, and "sched" for the use of - preempt_disable() together with synchronize_sched(). + the "srcu_read_lock()" API, "sched" for the use of + preempt_disable() together with synchronize_sched(), + and "sched_expedited" for the use of preempt_disable() + with synchronize_sched_expedited(). verbose Enable debug printk()s. Default is disabled. @@ -162,6 +164,23 @@ of the "old" and "current" counters for the corresponding CPU. The "idx" value maps the "old" and "current" values to the underlying array, and is useful for debugging. +Similarly, sched_expedited RCU provides the following: + + sched_expedited-torture: rtc: d0000000016c1880 ver: 1090796 tfle: 0 rta: 1090796 rtaf: 0 rtf: 1090787 rtmbe: 0 nt: 27713319 + sched_expedited-torture: Reader Pipe: 12660320201 95875 0 0 0 0 0 0 0 0 0 + sched_expedited-torture: Reader Batch: 12660424885 0 0 0 0 0 0 0 0 0 0 + sched_expedited-torture: Free-Block Circulation: 1090795 1090795 1090794 1090793 1090792 1090791 1090790 1090789 1090788 1090787 0 + state: -1 / 0:0 3:0 4:0 + +As before, the first four lines are similar to those for RCU. +The last line shows the task-migration state. The first number is +-1 if synchronize_sched_expedited() is idle, -2 if in the process of +posting wakeups to the migration kthreads, and N when waiting on CPU N. +Each of the colon-separated fields following the "/" is a CPU:state pair. +Valid states are "0" for idle, "1" for waiting for quiescent state, +"2" for passed through quiescent state, and "3" when a race with a +CPU-hotplug event forces use of the synchronize_sched() primitive. + USAGE diff --git a/Documentation/RCU/whatisRCU.txt b/Documentation/RCU/whatisRCU.txt index 96170824a717..97ded2432c59 100644 --- a/Documentation/RCU/whatisRCU.txt +++ b/Documentation/RCU/whatisRCU.txt @@ -785,6 +785,7 @@ RCU pointer/list traversal: rcu_dereference list_for_each_entry_rcu hlist_for_each_entry_rcu + hlist_nulls_for_each_entry_rcu list_for_each_continue_rcu (to be deprecated in favor of new list_for_each_entry_continue_rcu) @@ -807,19 +808,23 @@ RCU: Critical sections Grace period Barrier rcu_read_lock synchronize_net rcu_barrier rcu_read_unlock synchronize_rcu + synchronize_rcu_expedited call_rcu bh: Critical sections Grace period Barrier rcu_read_lock_bh call_rcu_bh rcu_barrier_bh - rcu_read_unlock_bh + rcu_read_unlock_bh synchronize_rcu_bh + synchronize_rcu_bh_expedited sched: Critical sections Grace period Barrier - [preempt_disable] synchronize_sched rcu_barrier_sched - [and friends] call_rcu_sched + rcu_read_lock_sched synchronize_sched rcu_barrier_sched + rcu_read_unlock_sched call_rcu_sched + [preempt_disable] synchronize_sched_expedited + [and friends] SRCU: Critical sections Grace period Barrier @@ -827,6 +832,9 @@ SRCU: Critical sections Grace period Barrier srcu_read_lock synchronize_srcu N/A srcu_read_unlock +SRCU: Initialization/cleanup + init_srcu_struct + cleanup_srcu_struct 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 ed89299d5b74..8fa078af30ef 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -376,15 +376,6 @@ Who: Thomas Gleixner <tglx@linutronix.de> ----------------------------- -What: obsolete generic irq defines and typedefs -When: 2.6.30 -Why: The defines and typedefs (hw_interrupt_type, no_irq_type, irq_desc_t) - have been kept around for migration reasons. After more than two years - it's time to remove them finally -Who: Thomas Gleixner <tglx@linutronix.de> - ---------------------------- - What: fakephp and associated sysfs files in /sys/bus/pci/slots/ When: 2011 Why: In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to @@ -440,16 +431,6 @@ Who: Eric Paris <eparis@redhat.com> ---------------------------- -What: CONFIG_X86_OLD_MCE -When: 2.6.32 -Why: Remove the old legacy 32bit machine check code. This has been - superseded by the newer machine check code from the 64bit port, - but the old version has been kept around for easier testing. Note this - doesn't impact the old P5 and WinChip machine check handlers. -Who: Andi Kleen <andi@firstfloor.org> - ----------------------------- - What: lock_policy_rwsem_* and unlock_policy_rwsem_* will not be exported interface anymore. When: 2.6.33 diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index fff9a94b3229..0eb0ffacde3a 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2541,6 +2541,11 @@ and is between 256 and 4096 characters. It is defined in the file trace_buf_size=nn[KMG] [FTRACE] will set tracing buffer size. + trace_event=[event-list] + [FTRACE] Set and start specified trace events in order + to facilitate early boot debugging. + See also Documentation/trace/events.txt + trix= [HW,OSS] MediaTrix AudioTrix Pro Format: <io>,<irq>,<dma>,<dma2>,<sb_io>,<sb_irq>,<sb_dma>,<mpu_io>,<mpu_irq> diff --git a/Documentation/trace/events.txt b/Documentation/trace/events.txt index f157d7594ea7..2bcc8d4dea29 100644 --- a/Documentation/trace/events.txt +++ b/Documentation/trace/events.txt @@ -83,6 +83,15 @@ When reading one of these enable files, there are four results: X - there is a mixture of events enabled and disabled ? - this file does not affect any event +2.3 Boot option +--------------- + +In order to facilitate early boot debugging, use boot option: + + trace_event=[event-list] + +The format of this boot option is the same as described in section 2.1. + 3. Defining an event-enabled tracepoint ======================================= diff --git a/Documentation/trace/ring-buffer-design.txt b/Documentation/trace/ring-buffer-design.txt new file mode 100644 index 000000000000..5b1d23d604c5 --- /dev/null +++ b/Documentation/trace/ring-buffer-design.txt @@ -0,0 +1,955 @@ + Lockless Ring Buffer Design + =========================== + +Copyright 2009 Red Hat Inc. + Author: Steven Rostedt <srostedt@redhat.com> + License: The GNU Free Documentation License, Version 1.2 + (dual licensed under the GPL v2) +Reviewers: Mathieu Desnoyers, Huang Ying, Hidetoshi Seto, + and Frederic Weisbecker. + + +Written for: 2.6.31 + +Terminology used in this Document +--------------------------------- + +tail - where new writes happen in the ring buffer. + +head - where new reads happen in the ring buffer. + +producer - the task that writes into the ring buffer (same as writer) + +writer - same as producer + +consumer - the task that reads from the buffer (same as reader) + +reader - same as consumer. + +reader_page - A page outside the ring buffer used solely (for the most part) + by the reader. + +head_page - a pointer to the page that the reader will use next + +tail_page - a pointer to the page that will be written to next + +commit_page - a pointer to the page with the last finished non nested write. + +cmpxchg - hardware assisted atomic transaction that performs the following: + + A = B iff previous A == C + + R = cmpxchg(A, C, B) is saying that we replace A with B if and only if + current A is equal to C, and we put the old (current) A into R + + R gets the previous A regardless if A is updated with B or not. + + To see if the update was successful a compare of R == C may be used. + +The Generic Ring Buffer +----------------------- + +The ring buffer can be used in either an overwrite mode or in +producer/consumer mode. + +Producer/consumer mode is where the producer were to fill up the +buffer before the consumer could free up anything, the producer +will stop writing to the buffer. This will lose most recent events. + +Overwrite mode is where the produce were to fill up the buffer +before the consumer could free up anything, the producer will +overwrite the older data. This will lose the oldest events. + +No two writers can write at the same time (on the same per cpu buffer), +but a writer may interrupt another writer, but it must finish writing +before the previous writer may continue. This is very important to the +algorithm. The writers act like a "stack". The way interrupts works +enforces this behavior. + + + writer1 start + <preempted> writer2 start + <preempted> writer3 start + writer3 finishes + writer2 finishes + writer1 finishes + +This is very much like a writer being preempted by an interrupt and +the interrupt doing a write as well. + +Readers can happen at any time. But no two readers may run at the +same time, nor can a reader preempt/interrupt another reader. A reader +can not preempt/interrupt a writer, but it may read/consume from the +buffer at the same time as a writer is writing, but the reader must be +on another processor to do so. A reader may read on its own processor +and can be preempted by a writer. + +A writer can preempt a reader, but a reader can not preempt a writer. +But a reader can read the buffer at the same time (on another processor) +as a writer. + +The ring buffer is made up of a list of pages held together by a link list. + +At initialization a reader page is allocated for the reader that is not +part of the ring buffer. + +The head_page, tail_page and commit_page are all initialized to point +to the same page. + +The reader page is initialized to have its next pointer pointing to +the head page, and its previous pointer pointing to a page before +the head page. + +The reader has its own page to use. At start up time, this page is +allocated but is not attached to the list. When the reader wants +to read from the buffer, if its page is empty (like it is on start up) +it will swap its page with the head_page. The old reader page will +become part of the ring buffer and the head_page will be removed. +The page after the inserted page (old reader_page) will become the +new head page. + +Once the new page is given to the reader, the reader could do what +it wants with it, as long as a writer has left that page. + +A sample of how the reader page is swapped: Note this does not +show the head page in the buffer, it is for demonstrating a swap +only. + + +------+ + |reader| RING BUFFER + |page | + +------+ + +---+ +---+ +---+ + | |-->| |-->| | + | |<--| |<--| | + +---+ +---+ +---+ + ^ | ^ | + | +-------------+ | + +-----------------+ + + + +------+ + |reader| RING BUFFER + |page |-------------------+ + +------+ v + | +---+ +---+ +---+ + | | |-->| |-->| | + | | |<--| |<--| |<-+ + | +---+ +---+ +---+ | + | ^ | ^ | | + | | +-------------+ | | + | +-----------------+ | + +------------------------------------+ + + +------+ + |reader| RING BUFFER + |page |-------------------+ + +------+ <---------------+ v + | ^ +---+ +---+ +---+ + | | | |-->| |-->| | + | | | | | |<--| |<-+ + | | +---+ +---+ +---+ | + | | | ^ | | + | | +-------------+ | | + | +-----------------------------+ | + +------------------------------------+ + + +------+ + |buffer| RING BUFFER + |page |-------------------+ + +------+ <---------------+ v + | ^ +---+ +---+ +---+ + | | | | | |-->| | + | | New | | | |<--| |<-+ + | | Reader +---+ +---+ +---+ | + | | page ----^ | | + | | | | + | +-----------------------------+ | + +------------------------------------+ + + + +It is possible that the page swapped is the commit page and the tail page, +if what is in the ring buffer is less than what is held in a buffer page. + + + reader page commit page tail page + | | | + v | | + +---+ | | + | |<----------+ | + | |<------------------------+ + | |------+ + +---+ | + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +This case is still valid for this algorithm. +When the writer leaves the page, it simply goes into the ring buffer +since the reader page still points to the next location in the ring +buffer. + + +The main pointers: + + reader page - The page used solely by the reader and is not part + of the ring buffer (may be swapped in) + + head page - the next page in the ring buffer that will be swapped + with the reader page. + + tail page - the page where the next write will take place. + + commit page - the page that last finished a write. + +The commit page only is updated by the outer most writer in the +writer stack. A writer that preempts another writer will not move the +commit page. + +When data is written into the ring buffer, a position is reserved +in the ring buffer and passed back to the writer. When the writer +is finished writing data into that position, it commits the write. + +Another write (or a read) may take place at anytime during this +transaction. If another write happens it must finish before continuing +with the previous write. + + + Write reserve: + + Buffer page + +---------+ + |written | + +---------+ <--- given back to writer (current commit) + |reserved | + +---------+ <--- tail pointer + | empty | + +---------+ + + Write commit: + + Buffer page + +---------+ + |written | + +---------+ + |written | + +---------+ <--- next positon for write (current commit) + | empty | + +---------+ + + + If a write happens after the first reserve: + + Buffer page + +---------+ + |written | + +---------+ <-- current commit + |reserved | + +---------+ <--- given back to second writer + |reserved | + +---------+ <--- tail pointer + + After second writer commits: + + + Buffer page + +---------+ + |written | + +---------+ <--(last full commit) + |reserved | + +---------+ + |pending | + |commit | + +---------+ <--- tail pointer + + When the first writer commits: + + Buffer page + +---------+ + |written | + +---------+ + |written | + +---------+ + |written | + +---------+ <--(last full commit and tail pointer) + + +The commit pointer points to the last write location that was +committed without preempting another write. When a write that +preempted another write is committed, it only becomes a pending commit +and will not be a full commit till all writes have been committed. + +The commit page points to the page that has the last full commit. +The tail page points to the page with the last write (before +committing). + +The tail page is always equal to or after the commit page. It may +be several pages ahead. If the tail page catches up to the commit +page then no more writes may take place (regardless of the mode +of the ring buffer: overwrite and produce/consumer). + +The order of pages are: + + head page + commit page + tail page + +Possible scenario: + tail page + head page commit page | + | | | + v v v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +There is a special case that the head page is after either the commit page +and possibly the tail page. That is when the commit (and tail) page has been +swapped with the reader page. This is because the head page is always +part of the ring buffer, but the reader page is not. When ever there +has been less than a full page that has been committed inside the ring buffer, +and a reader swaps out a page, it will be swapping out the commit page. + + + reader page commit page tail page + | | | + v | | + +---+ | | + | |<----------+ | + | |<------------------------+ + | |------+ + +---+ | + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + ^ + | + head page + + +In this case, the head page will not move when the tail and commit +move back into the ring buffer. + +The reader can not swap a page into the ring buffer if the commit page +is still on that page. If the read meets the last commit (real commit +not pending or reserved), then there is nothing more to read. +The buffer is considered empty until another full commit finishes. + +When the tail meets the head page, if the buffer is in overwrite mode, +the head page will be pushed ahead one. If the buffer is in producer/consumer +mode, the write will fail. + +Overwrite mode: + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + ^ + | + head page + + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + ^ + | + head page + + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + ^ + | + head page + +Note, the reader page will still point to the previous head page. +But when a swap takes place, it will use the most recent head page. + + +Making the Ring Buffer Lockless: +-------------------------------- + +The main idea behind the lockless algorithm is to combine the moving +of the head_page pointer with the swapping of pages with the reader. +State flags are placed inside the pointer to the page. To do this, +each page must be aligned in memory by 4 bytes. This will allow the 2 +least significant bits of the address to be used as flags. Since +they will always be zero for the address. To get the address, +simply mask out the flags. + + MASK = ~3 + + address & MASK + +Two flags will be kept by these two bits: + + HEADER - the page being pointed to is a head page + + UPDATE - the page being pointed to is being updated by a writer + and was or is about to be a head page. + + + reader page + | + v + +---+ + | |------+ + +---+ | + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + +The above pointer "-H->" would have the HEADER flag set. That is +the next page is the next page to be swapped out by the reader. +This pointer means the next page is the head page. + +When the tail page meets the head pointer, it will use cmpxchg to +change the pointer to the UPDATE state: + + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +"-U->" represents a pointer in the UPDATE state. + +Any access to the reader will need to take some sort of lock to serialize +the readers. But the writers will never take a lock to write to the +ring buffer. This means we only need to worry about a single reader, +and writes only preempt in "stack" formation. + +When the reader tries to swap the page with the ring buffer, it +will also use cmpxchg. If the flag bit in the pointer to the +head page does not have the HEADER flag set, the compare will fail +and the reader will need to look for the new head page and try again. +Note, the flag UPDATE and HEADER are never set at the same time. + +The reader swaps the reader page as follows: + + +------+ + |reader| RING BUFFER + |page | + +------+ + +---+ +---+ +---+ + | |--->| |--->| | + | |<---| |<---| | + +---+ +---+ +---+ + ^ | ^ | + | +---------------+ | + +-----H-------------+ + +The reader sets the reader page next pointer as HEADER to the page after +the head page. + + + +------+ + |reader| RING BUFFER + |page |-------H-----------+ + +------+ v + | +---+ +---+ +---+ + | | |--->| |--->| | + | | |<---| |<---| |<-+ + | +---+ +---+ +---+ | + | ^ | ^ | | + | | +---------------+ | | + | +-----H-------------+ | + +--------------------------------------+ + +It does a cmpxchg with the pointer to the previous head page to make it +point to the reader page. Note that the new pointer does not have the HEADER +flag set. This action atomically moves the head page forward. + + +------+ + |reader| RING BUFFER + |page |-------H-----------+ + +------+ v + | ^ +---+ +---+ +---+ + | | | |-->| |-->| | + | | | |<--| |<--| |<-+ + | | +---+ +---+ +---+ | + | | | ^ | | + | | +-------------+ | | + | +-----------------------------+ | + +------------------------------------+ + +After the new head page is set, the previous pointer of the head page is +updated to the reader page. + + +------+ + |reader| RING BUFFER + |page |-------H-----------+ + +------+ <---------------+ v + | ^ +---+ +---+ +---+ + | | | |-->| |-->| | + | | | | | |<--| |<-+ + | | +---+ +---+ +---+ | + | | | ^ | | + | | +-------------+ | | + | +-----------------------------+ | + +------------------------------------+ + + +------+ + |buffer| RING BUFFER + |page |-------H-----------+ <--- New head page + +------+ <---------------+ v + | ^ +---+ +---+ +---+ + | | | | | |-->| | + | | New | | | |<--| |<-+ + | | Reader +---+ +---+ +---+ | + | | page ----^ | | + | | | | + | +-----------------------------+ | + +------------------------------------+ + +Another important point. The page that the reader page points back to +by its previous pointer (the one that now points to the new head page) +never points back to the reader page. That is because the reader page is +not part of the ring buffer. Traversing the ring buffer via the next pointers +will always stay in the ring buffer. Traversing the ring buffer via the +prev pointers may not. + +Note, the way to determine a reader page is simply by examining the previous +pointer of the page. If the next pointer of the previous page does not +point back to the original page, then the original page is a reader page: + + + +--------+ + | reader | next +----+ + | page |-------->| |<====== (buffer page) + +--------+ +----+ + | | ^ + | v | next + prev | +----+ + +------------->| | + +----+ + +The way the head page moves forward: + +When the tail page meets the head page and the buffer is in overwrite mode +and more writes take place, the head page must be moved forward before the +writer may move the tail page. The way this is done is that the writer +performs a cmpxchg to convert the pointer to the head page from the HEADER +flag to have the UPDATE flag set. Once this is done, the reader will +not be able to swap the head page from the buffer, nor will it be able to +move the head page, until the writer is finished with the move. + +This eliminates any races that the reader can have on the writer. The reader +must spin, and this is why the reader can not preempt the writer. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The following page will be made into the new head page. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +After the new head page has been set, we can set the old head page +pointer back to NORMAL. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +After the head page has been moved, the tail page may now move forward. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + +The above are the trivial updates. Now for the more complex scenarios. + + +As stated before, if enough writes preempt the first write, the +tail page may make it all the way around the buffer and meet the commit +page. At this time, we must start dropping writes (usually with some kind +of warning to the user). But what happens if the commit was still on the +reader page? The commit page is not part of the ring buffer. The tail page +must account for this. + + + reader page commit page + | | + v | + +---+ | + | |<----------+ + | | + | |------+ + +---+ | + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + ^ + | + tail page + +If the tail page were to simply push the head page forward, the commit when +leaving the reader page would not be pointing to the correct page. + +The solution to this is to test if the commit page is on the reader page +before pushing the head page. If it is, then it can be assumed that the +tail page wrapped the buffer, and we must drop new writes. + +This is not a race condition, because the commit page can only be moved +by the outter most writer (the writer that was preempted). +This means that the commit will not move while a writer is moving the +tail page. The reader can not swap the reader page if it is also being +used as the commit page. The reader can simply check that the commit +is off the reader page. Once the commit page leaves the reader page +it will never go back on it unless a reader does another swap with the +buffer page that is also the commit page. + + +Nested writes +------------- + +In the pushing forward of the tail page we must first push forward +the head page if the head page is the next page. If the head page +is not the next page, the tail page is simply updated with a cmpxchg. + +Only writers move the tail page. This must be done atomically to protect +against nested writers. + + temp_page = tail_page + next_page = temp_page->next + cmpxchg(tail_page, temp_page, next_page) + +The above will update the tail page if it is still pointing to the expected +page. If this fails, a nested write pushed it forward, the the current write +does not need to push it. + + + temp page + | + v + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +Nested write comes in and moves the tail page forward: + + tail page (moved by nested writer) + temp page | + | | + v v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The above would fail the cmpxchg, but since the tail page has already +been moved forward, the writer will just try again to reserve storage +on the new tail page. + +But the moving of the head page is a bit more complex. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The write converts the head page pointer to UPDATE. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +But if a nested writer preempts here. It will see that the next +page is a head page, but it is also nested. It will detect that +it is nested and will save that information. The detection is the +fact that it sees the UPDATE flag instead of a HEADER or NORMAL +pointer. + +The nested writer will set the new head page pointer. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +But it will not reset the update back to normal. Only the writer +that converted a pointer from HEAD to UPDATE will convert it back +to NORMAL. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +After the nested writer finishes, the outer most writer will convert +the UPDATE pointer to NORMAL. + + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + +It can be even more complex if several nested writes came in and moved +the tail page ahead several pages: + + +(first writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-H->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The write converts the head page pointer to UPDATE. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +Next writer comes in, and sees the update and sets up the new +head page. + +(second writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The nested writer moves the tail page forward. But does not set the old +update page to NORMAL because it is not the outer most writer. + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +Another writer preempts and sees the page after the tail page is a head page. +It changes it from HEAD to UPDATE. + +(third writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-U->| |---> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The writer will move the head page forward: + + +(third writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-U->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +But now that the third writer did change the HEAD flag to UPDATE it +will convert it to normal: + + +(third writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + +Then it will move the tail page, and return back to the second writer. + + +(second writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + + +The second writer will fail to move the tail page because it was already +moved, so it will try again and add its data to the new tail page. +It will return to the first writer. + + +(first writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +The first writer can not know atomically test if the tail page moved +while it updates the HEAD page. It will then update the head page to +what it thinks is the new head page. + + +(first writer) + + tail page + | + v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +Since the cmpxchg returns the old value of the pointer the first writer +will see it succeeded in updating the pointer from NORMAL to HEAD. +But as we can see, this is not good enough. It must also check to see +if the tail page is either where it use to be or on the next page: + + +(first writer) + + A B tail page + | | | + v v v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |-H->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +If tail page != A and tail page does not equal B, then it must reset the +pointer back to NORMAL. The fact that it only needs to worry about +nested writers, it only needs to check this after setting the HEAD page. + + +(first writer) + + A B tail page + | | | + v v v + +---+ +---+ +---+ +---+ +<---| |--->| |-U->| |--->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + +Now the writer can update the head page. This is also why the head page must +remain in UPDATE and only reset by the outer most writer. This prevents +the reader from seeing the incorrect head page. + + +(first writer) + + A B tail page + | | | + v v v + +---+ +---+ +---+ +---+ +<---| |--->| |--->| |--->| |-H-> +--->| |<---| |<---| |<---| |<--- + +---+ +---+ +---+ +---+ + diff --git a/arch/Kconfig b/arch/Kconfig index 99193b160232..beea3ccebb5e 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -30,6 +30,18 @@ config OPROFILE_IBS If unsure, say N. +config OPROFILE_EVENT_MULTIPLEX + bool "OProfile multiplexing support (EXPERIMENTAL)" + default n + depends on OPROFILE && X86 + help + The number of hardware counters is limited. The multiplexing + feature enables OProfile to gather more events than counters + are provided by the hardware. This is realized by switching + between events at an user specified time interval. + + If unsure, say N. + config HAVE_OPROFILE bool diff --git a/arch/ia64/include/asm/cputime.h b/arch/ia64/include/asm/cputime.h index d20b998cb91d..7fa8a8594660 100644 --- a/arch/ia64/include/asm/cputime.h +++ b/arch/ia64/include/asm/cputime.h @@ -30,6 +30,7 @@ typedef u64 cputime_t; typedef u64 cputime64_t; #define cputime_zero ((cputime_t)0) +#define cputime_one_jiffy jiffies_to_cputime(1) #define cputime_max ((~((cputime_t)0) >> 1) - 1) #define cputime_add(__a, __b) ((__a) + (__b)) #define cputime_sub(__a, __b) ((__a) - (__b)) diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h index 5a61b5c2e18f..f91829de329f 100644 --- a/arch/ia64/include/asm/dma-mapping.h +++ b/arch/ia64/include/asm/dma-mapping.h @@ -69,6 +69,24 @@ dma_set_mask (struct device *dev, u64 mask) return 0; } +static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size) +{ + if (!dev->dma_mask) + return 0; + + return addr + size <= *dev->dma_mask; +} + +static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr) +{ + return paddr; +} + +static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr) +{ + return daddr; +} + extern int dma_get_cache_alignment(void); static inline void diff --git a/arch/powerpc/include/asm/cputime.h b/arch/powerpc/include/asm/cputime.h index f42e623030ee..fa19f3fe05ff 100644 --- a/arch/powerpc/include/asm/cputime.h +++ b/arch/powerpc/include/asm/cputime.h @@ -18,6 +18,9 @@ #ifndef CONFIG_VIRT_CPU_ACCOUNTING #include <asm-generic/cputime.h> +#ifdef __KERNEL__ +static inline void setup_cputime_one_jiffy(void) { } +#endif #else #include <linux/types.h> @@ -49,6 +52,11 @@ typedef u64 cputime64_t; #ifdef __KERNEL__ /* + * One jiffy in timebase units computed during initialization + */ +extern cputime_t cputime_one_jiffy; + +/* * Convert cputime <-> jiffies */ extern u64 __cputime_jiffies_factor; @@ -89,6 +97,11 @@ static inline cputime_t jiffies_to_cputime(const unsigned long jif) return ct; } +static inline void setup_cputime_one_jiffy(void) +{ + cputime_one_jiffy = jiffies_to_cputime(1); +} + static inline cputime64_t jiffies64_to_cputime64(const u64 jif) { cputime_t ct; diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h index b44aaabdd1a6..0c34371ec49c 100644 --- a/arch/powerpc/include/asm/dma-mapping.h +++ b/arch/powerpc/include/asm/dma-mapping.h @@ -424,6 +424,29 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) #endif } +static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size) +{ + struct dma_mapping_ops *ops = get_dma_ops(dev); + + if (ops->addr_needs_map && ops->addr_needs_map(dev, addr, size)) + return 0; + + if (!dev->dma_mask) + return 0; + + return addr + size <= *dev->dma_mask; +} + +static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr) +{ + return paddr + get_dma_direct_offset(dev); +} + +static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr) +{ + return daddr - get_dma_direct_offset(dev); +} + #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) #ifdef CONFIG_NOT_COHERENT_CACHE diff --git a/arch/powerpc/kernel/dma-swiotlb.c b/arch/powerpc/kernel/dma-swiotlb.c index 68ccf11e4f19..e8a57de85bcf 100644 --- a/arch/powerpc/kernel/dma-swiotlb.c +++ b/arch/powerpc/kernel/dma-swiotlb.c @@ -24,50 +24,12 @@ int swiotlb __read_mostly; unsigned int ppc_swiotlb_enable; -void *swiotlb_bus_to_virt(struct device *hwdev, dma_addr_t addr) -{ - unsigned long pfn = PFN_DOWN(swiotlb_bus_to_phys(hwdev, addr)); - void *pageaddr = page_address(pfn_to_page(pfn)); - - if (pageaddr != NULL) - return pageaddr + (addr % PAGE_SIZE); - return NULL; -} - -dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr) -{ - return paddr + get_dma_direct_offset(hwdev); -} - -phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr) - -{ - return baddr - get_dma_direct_offset(hwdev); -} - -/* - * Determine if an address needs bounce buffering via swiotlb. - * Going forward I expect the swiotlb code to generalize on using - * a dma_ops->addr_needs_map, and this function will move from here to the - * generic swiotlb code. - */ -int -swiotlb_arch_address_needs_mapping(struct device *hwdev, dma_addr_t addr, - size_t size) -{ - struct dma_mapping_ops *dma_ops = get_dma_ops(hwdev); - - BUG_ON(!dma_ops); - return dma_ops->addr_needs_map(hwdev, addr, size); -} - /* * Determine if an address is reachable by a pci device, or if we must bounce. */ static int swiotlb_pci_addr_needs_map(struct device *hwdev, dma_addr_t addr, size_t size) { - u64 mask = dma_get_mask(hwdev); dma_addr_t max; struct pci_controller *hose; struct pci_dev *pdev = to_pci_dev(hwdev); @@ -79,16 +41,9 @@ swiotlb_pci_addr_needs_map(struct device *hwdev, dma_addr_t addr, size_t size) if ((addr + size > max) | (addr < hose->dma_window_base_cur)) return 1; - return !is_buffer_dma_capable(mask, addr, size); -} - -static int -swiotlb_addr_needs_map(struct device *hwdev, dma_addr_t addr, size_t size) -{ - return !is_buffer_dma_capable(dma_get_mask(hwdev), addr, size); + return 0; } - /* * At the moment, all platforms that use this code only require * swiotlb to be used if we're operating on HIGHMEM. Since @@ -104,7 +59,6 @@ struct dma_mapping_ops swiotlb_dma_ops = { .dma_supported = swiotlb_dma_supported, .map_page = swiotlb_map_page, .unmap_page = swiotlb_unmap_page, - .addr_needs_map = swiotlb_addr_needs_map, .sync_single_range_for_cpu = swiotlb_sync_single_range_for_cpu, .sync_single_range_for_device = swiotlb_sync_single_range_for_device, .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu, diff --git a/arch/powerpc/kernel/perf_counter.c b/arch/powerpc/kernel/perf_counter.c index 809fdf94b95f..70e1f57f7dd8 100644 --- a/arch/powerpc/kernel/perf_counter.c +++ b/arch/powerpc/kernel/perf_counter.c @@ -518,6 +518,8 @@ void hw_perf_disable(void) struct cpu_hw_counters *cpuhw; unsigned long flags; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); @@ -572,6 +574,8 @@ void hw_perf_enable(void) int n_lim; int idx; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); if (!cpuhw->disabled) { @@ -737,6 +741,8 @@ int hw_perf_group_sched_in(struct perf_counter *group_leader, long i, n, n0; struct perf_counter *sub; + if (!ppmu) + return 0; cpuhw = &__get_cpu_var(cpu_hw_counters); n0 = cpuhw->n_counters; n = collect_events(group_leader, ppmu->n_counter - n0, @@ -1281,6 +1287,8 @@ void hw_perf_counter_setup(int cpu) { struct cpu_hw_counters *cpuhw = &per_cpu(cpu_hw_counters, cpu); + if (!ppmu) + return; memset(cpuhw, 0, sizeof(*cpuhw)); cpuhw->mmcr[0] = MMCR0_FC; } diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index eae4511ceeac..211d7b0cd370 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -193,6 +193,8 @@ EXPORT_SYMBOL(__cputime_clockt_factor); DEFINE_PER_CPU(unsigned long, cputime_last_delta); DEFINE_PER_CPU(unsigned long, cputime_scaled_last_delta); +cputime_t cputime_one_jiffy; + static void calc_cputime_factors(void) { struct div_result res; @@ -500,6 +502,7 @@ static int __init iSeries_tb_recal(void) tb_to_xs = divres.result_low; vdso_data->tb_ticks_per_sec = tb_ticks_per_sec; vdso_data->tb_to_xs = tb_to_xs; + setup_cputime_one_jiffy(); } else { printk( "Titan recalibrate: FAILED (difference > 4 percent)\n" @@ -945,6 +948,7 @@ void __init time_init(void) tb_ticks_per_usec = ppc_tb_freq / 1000000; tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000); calc_cputime_factors(); + setup_cputime_one_jiffy(); /* * Calculate the length of each tick in ns. It will not be diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h index 7a3817a656df..24b1244aadb9 100644 --- a/arch/s390/include/asm/cputime.h +++ b/arch/s390/include/asm/cputime.h @@ -42,6 +42,7 @@ __div(unsigned long long n, unsigned int base) #endif /* __s390x__ */ #define cputime_zero (0ULL) +#define cputime_one_jiffy jiffies_to_cputime(1) #define cputime_max ((~0UL >> 1) - 1) #define cputime_add(__a, __b) ((__a) + (__b)) #define cputime_sub(__a, __b) ((__a) - (__b)) diff --git a/arch/sparc/include/asm/irq_64.h b/arch/sparc/include/asm/irq_64.h index 1934f2cbf513..a0b443cb3c1f 100644 --- a/arch/sparc/include/asm/irq_64.h +++ b/arch/sparc/include/asm/irq_64.h @@ -89,8 +89,8 @@ static inline unsigned long get_softint(void) return retval; } -void __trigger_all_cpu_backtrace(void); -#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() +void arch_trigger_all_cpu_backtrace(void); +#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace extern void *hardirq_stack[NR_CPUS]; extern void *softirq_stack[NR_CPUS]; diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index 4041f94e7724..18d67854a1b8 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -251,7 +251,7 @@ static void __global_reg_poll(struct global_reg_snapshot *gp) } } -void __trigger_all_cpu_backtrace(void) +void arch_trigger_all_cpu_backtrace(void) { struct thread_info *tp = current_thread_info(); struct pt_regs *regs = get_irq_regs(); @@ -304,7 +304,7 @@ void __trigger_all_cpu_backtrace(void) static void sysrq_handle_globreg(int key, struct tty_struct *tty) { - __trigger_all_cpu_backtrace(); + arch_trigger_all_cpu_backtrace(); } static struct sysrq_key_op sparc_globalreg_op = { diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 738bdc6b0f8b..cd2788c78a8a 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -777,41 +777,17 @@ config X86_REROUTE_FOR_BROKEN_BOOT_IRQS increased on these systems. config X86_MCE - bool "Machine Check Exception" + bool "Machine Check / overheating reporting" ---help--- - Machine Check Exception support allows the processor to notify the - kernel if it detects a problem (e.g. overheating, component failure). + Machine Check support allows the processor to notify the + kernel if it detects a problem (e.g. overheating, data corruption). The action the kernel takes depends on the severity of the problem, - ranging from a warning message on the console, to halting the machine. - Your processor must be a Pentium or newer to support this - check the - flags in /proc/cpuinfo for mce. Note that some older Pentium systems - have a design flaw which leads to false MCE events - hence MCE is - disabled on all P5 processors, unless explicitly enabled with "mce" - as a boot argument. Similarly, if MCE is built in and creates a - problem on some new non-standard machine, you can boot with "nomce" - to disable it. MCE support simply ignores non-MCE processors like - the 386 and 486, so nearly everyone can say Y here. - -config X86_OLD_MCE - depends on X86_32 && X86_MCE - bool "Use legacy machine check code (will go away)" - default n - select X86_ANCIENT_MCE - ---help--- - Use the old i386 machine check code. This is merely intended for - testing in a transition period. Try this if you run into any machine - check related software problems, but report the problem to - linux-kernel. When in doubt say no. - -config X86_NEW_MCE - depends on X86_MCE - bool - default y if (!X86_OLD_MCE && X86_32) || X86_64 + ranging from warning messages to halting the machine. config X86_MCE_INTEL def_bool y prompt "Intel MCE features" - depends on X86_NEW_MCE && X86_LOCAL_APIC + depends on X86_MCE && X86_LOCAL_APIC ---help--- Additional support for intel specific MCE features such as the thermal monitor. @@ -819,14 +795,14 @@ config X86_MCE_INTEL config X86_MCE_AMD def_bool y prompt "AMD MCE features" - depends on X86_NEW_MCE && X86_LOCAL_APIC + depends on X86_MCE && X86_LOCAL_APIC ---help--- Additional support for AMD specific MCE features such as the DRAM Error Threshold. config X86_ANCIENT_MCE def_bool n - depends on X86_32 + depends on X86_32 && X86_MCE prompt "Support for old Pentium 5 / WinChip machine checks" ---help--- Include support for machine check handling on old Pentium 5 or WinChip @@ -839,36 +815,16 @@ config X86_MCE_THRESHOLD default y config X86_MCE_INJECT - depends on X86_NEW_MCE + depends on X86_MCE tristate "Machine check injector support" ---help--- Provide support for injecting machine checks for testing purposes. If you don't know what a machine check is and you don't do kernel QA it is safe to say n. -config X86_MCE_NONFATAL - tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" - depends on X86_OLD_MCE - ---help--- - Enabling this feature starts a timer that triggers every 5 seconds which - will look at the machine check registers to see if anything happened. - Non-fatal problems automatically get corrected (but still logged). - Disable this if you don't want to see these messages. - Seeing the messages this option prints out may be indicative of dying - or out-of-spec (ie, overclocked) hardware. - This option only does something on certain CPUs. - (AMD Athlon/Duron and Intel Pentium 4) - -config X86_MCE_P4THERMAL - bool "check for P4 thermal throttling interrupt." - depends on X86_OLD_MCE && X86_MCE && (X86_UP_APIC || SMP) - ---help--- - Enabling this feature will cause a message to be printed when the P4 - enters thermal throttling. - config X86_THERMAL_VECTOR def_bool y - depends on X86_MCE_P4THERMAL || X86_MCE_INTEL + depends on X86_MCE_INTEL config VM86 bool "Enable VM86 support" if EMBEDDED @@ -1247,6 +1203,11 @@ config ARCH_MEMORY_PROBE def_bool X86_64 depends on MEMORY_HOTPLUG +config ILLEGAL_POINTER_VALUE + hex + default 0 if X86_32 + default 0xdead000000000000 if X86_64 + source "mm/Kconfig" config HIGHPTE diff --git a/arch/x86/Makefile b/arch/x86/Makefile index 1b68659c41b4..1f3851a626d3 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -32,8 +32,8 @@ ifeq ($(CONFIG_X86_32),y) # Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use # a lot more stack due to the lack of sharing of stacklots: - KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \ - echo $(call cc-option,-fno-unit-at-a-time); fi ;) + KBUILD_CFLAGS += $(call cc-ifversion, -lt, 0400, \ + $(call cc-option,-fno-unit-at-a-time)) # CPU-specific tuning. Anything which can be shared with UML should go here. include $(srctree)/arch/x86/Makefile_32.cpu diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c index 275dd177f198..11e8c6eb80a1 100644 --- a/arch/x86/boot/video-vesa.c +++ b/arch/x86/boot/video-vesa.c @@ -31,7 +31,6 @@ static inline void vesa_store_mode_params_graphics(void) {} static int vesa_probe(void) { -#if defined(CONFIG_VIDEO_VESA) || defined(CONFIG_FIRMWARE_EDID) struct biosregs ireg, oreg; u16 mode; addr_t mode_ptr; @@ -49,8 +48,7 @@ static int vesa_probe(void) vginfo.signature != VESA_MAGIC || vginfo.version < 0x0102) return 0; /* Not present */ -#endif /* CONFIG_VIDEO_VESA || CONFIG_FIRMWARE_EDID */ -#ifdef CONFIG_VIDEO_VESA + set_fs(vginfo.video_mode_ptr.seg); mode_ptr = vginfo.video_mode_ptr.off; @@ -102,9 +100,6 @@ static int vesa_probe(void) } return nmodes; -#else - return 0; -#endif /* CONFIG_VIDEO_VESA */ } static int vesa_set_mode(struct mode_info *mode) diff --git a/arch/x86/boot/video-vga.c b/arch/x86/boot/video-vga.c index 8f8d827e254d..819caa1f2008 100644 --- a/arch/x86/boot/video-vga.c +++ b/arch/x86/boot/video-vga.c @@ -47,14 +47,6 @@ static u8 vga_set_basic_mode(void) initregs(&ireg); -#ifdef CONFIG_VIDEO_400_HACK - if (adapter >= ADAPTER_VGA) { - ireg.ax = 0x1202; - ireg.bx = 0x0030; - intcall(0x10, &ireg, NULL); - } -#endif - ax = 0x0f00; intcall(0x10, &ireg, &oreg); mode = oreg.al; @@ -62,11 +54,9 @@ static u8 vga_set_basic_mode(void) set_fs(0); rows = rdfs8(0x484); /* rows minus one */ -#ifndef CONFIG_VIDEO_400_HACK if ((oreg.ax == 0x5003 || oreg.ax == 0x5007) && (rows == 0 || rows == 24)) return mode; -#endif if (mode != 3 && mode != 7) mode = 3; diff --git a/arch/x86/boot/video.c b/arch/x86/boot/video.c index bad728b76fc2..d42da3802499 100644 --- a/arch/x86/boot/video.c +++ b/arch/x86/boot/video.c @@ -221,7 +221,6 @@ static unsigned int mode_menu(void) } } -#ifdef CONFIG_VIDEO_RETAIN /* Save screen content to the heap */ static struct saved_screen { int x, y; @@ -299,10 +298,6 @@ static void restore_screen(void) ireg.dl = saved.curx; intcall(0x10, &ireg, NULL); } -#else -#define save_screen() ((void)0) -#define restore_screen() ((void)0) -#endif void set_video(void) { diff --git a/arch/x86/boot/video.h b/arch/x86/boot/video.h index 5bb174a997fc..ff339c5db311 100644 --- a/arch/x86/boot/video.h +++ b/arch/x86/boot/video.h @@ -17,19 +17,8 @@ #include <linux/types.h> -/* Enable autodetection of SVGA adapters and modes. */ -#undef CONFIG_VIDEO_SVGA - -/* Enable autodetection of VESA modes */ -#define CONFIG_VIDEO_VESA - -/* Retain screen contents when switching modes */ -#define CONFIG_VIDEO_RETAIN - -/* Force 400 scan lines for standard modes (hack to fix bad BIOS behaviour */ -#undef CONFIG_VIDEO_400_HACK - -/* This code uses an extended set of video mode numbers. These include: +/* + * This code uses an extended set of video mode numbers. These include: * Aliases for standard modes * NORMAL_VGA (-1) * EXTENDED_VGA (-2) @@ -67,13 +56,8 @@ /* The "recalculate timings" flag */ #define VIDEO_RECALC 0x8000 -/* Define DO_STORE according to CONFIG_VIDEO_RETAIN */ -#ifdef CONFIG_VIDEO_RETAIN void store_screen(void); #define DO_STORE() store_screen() -#else -#define DO_STORE() ((void)0) -#endif /* CONFIG_VIDEO_RETAIN */ /* * Mode table structures diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index e590261ba059..ba331bfd1112 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -537,7 +537,7 @@ ia32_sys_call_table: .quad sys_mkdir .quad sys_rmdir /* 40 */ .quad sys_dup - .quad sys32_pipe + .quad sys_pipe .quad compat_sys_times .quad quiet_ni_syscall /* old prof syscall holder */ .quad sys_brk /* 45 */ diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index 085a8c35f149..9f5527198825 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -189,20 +189,6 @@ asmlinkage long sys32_mprotect(unsigned long start, size_t len, return sys_mprotect(start, len, prot); } -asmlinkage long sys32_pipe(int __user *fd) -{ - int retval; - int fds[2]; - - retval = do_pipe_flags(fds, 0); - if (retval) - goto out; - if (copy_to_user(fd, fds, sizeof(fds))) - retval = -EFAULT; -out: - return retval; -} - asmlinkage long sys32_rt_sigaction(int sig, struct sigaction32 __user *act, struct sigaction32 __user *oact, unsigned int sigsetsize) diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index bb7d47925847..586b7adb8e53 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -183,6 +183,10 @@ static inline int x2apic_enabled(void) } #define x2apic_supported() (cpu_has_x2apic) +static inline void x2apic_force_phys(void) +{ + x2apic_phys = 1; +} #else static inline void check_x2apic(void) { @@ -194,6 +198,9 @@ static inline int x2apic_enabled(void) { return 0; } +static inline void x2apic_force_phys(void) +{ +} #define x2apic_preenabled 0 #define x2apic_supported() 0 diff --git a/arch/x86/include/asm/current.h b/arch/x86/include/asm/current.h index c68c361697e1..4d447b732d82 100644 --- a/arch/x86/include/asm/current.h +++ b/arch/x86/include/asm/current.h @@ -11,7 +11,7 @@ DECLARE_PER_CPU(struct task_struct *, current_task); static __always_inline struct task_struct *get_current(void) { - return percpu_read(current_task); + return percpu_read_stable(current_task); } #define current get_current() diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index c993e9e0fed4..e8de2f6f5ca5 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -291,11 +291,24 @@ static inline unsigned long get_desc_base(const struct desc_struct *desc) return desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24); } +static inline void set_desc_base(struct desc_struct *desc, unsigned long base) +{ + desc->base0 = base & 0xffff; + desc->base1 = (base >> 16) & 0xff; + desc->base2 = (base >> 24) & 0xff; +} + static inline unsigned long get_desc_limit(const struct desc_struct *desc) { return desc->limit0 | (desc->limit << 16); } +static inline void set_desc_limit(struct desc_struct *desc, unsigned long limit) +{ + desc->limit0 = limit & 0xffff; + desc->limit = (limit >> 16) & 0xf; +} + static inline void _set_gate(int gate, unsigned type, void *addr, unsigned dpl, unsigned ist, unsigned seg) { diff --git a/arch/x86/include/asm/desc_defs.h b/arch/x86/include/asm/desc_defs.h index a6adefa28b94..9d6684849fd9 100644 --- a/arch/x86/include/asm/desc_defs.h +++ b/arch/x86/include/asm/desc_defs.h @@ -34,6 +34,12 @@ struct desc_struct { }; } __attribute__((packed)); +#define GDT_ENTRY_INIT(flags, base, limit) { { { \ + .a = ((limit) & 0xffff) | (((base) & 0xffff) << 16), \ + .b = (((base) & 0xff0000) >> 16) | (((flags) & 0xf0ff) << 8) | \ + ((limit) & 0xf0000) | ((base) & 0xff000000), \ + } } } + enum { GATE_INTERRUPT = 0xE, GATE_TRAP = 0xF, diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index 1c3f9435f1c9..0ee770d23d0e 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -55,6 +55,24 @@ extern int dma_set_mask(struct device *dev, u64 mask); extern void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag); +static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size) +{ + if (!dev->dma_mask) + return 0; + + return addr + size <= *dev->dma_mask; +} + +static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr) +{ + return paddr; +} + +static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr) +{ + return daddr; +} + static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size, enum dma_data_direction dir) diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h index ff8cbfa07851..5e3f2044f0d3 100644 --- a/arch/x86/include/asm/entry_arch.h +++ b/arch/x86/include/asm/entry_arch.h @@ -61,7 +61,7 @@ BUILD_INTERRUPT(thermal_interrupt,THERMAL_APIC_VECTOR) BUILD_INTERRUPT(threshold_interrupt,THRESHOLD_APIC_VECTOR) #endif -#ifdef CONFIG_X86_NEW_MCE +#ifdef CONFIG_X86_MCE BUILD_INTERRUPT(mce_self_interrupt,MCE_SELF_VECTOR) #endif diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 175adf58dd4f..2e7529295f55 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -26,6 +26,7 @@ extern void fpu_init(void); extern void mxcsr_feature_mask_init(void); extern int init_fpu(struct task_struct *child); extern asmlinkage void math_state_restore(void); +extern void __math_state_restore(void); extern void init_thread_xstate(void); extern int dump_fpu(struct pt_regs *, struct user_i387_struct *); diff --git a/arch/x86/include/asm/ioctls.h b/arch/x86/include/asm/ioctls.h index 0d5b23b7b06e..ec34c760665e 100644 --- a/arch/x86/include/asm/ioctls.h +++ b/arch/x86/include/asm/ioctls.h @@ -1,94 +1 @@ -#ifndef _ASM_X86_IOCTLS_H -#define _ASM_X86_IOCTLS_H - -#include <asm/ioctl.h> - -/* 0x54 is just a magic number to make these relatively unique ('T') */ - -#define TCGETS 0x5401 -#define TCSETS 0x5402 /* Clashes with SNDCTL_TMR_START sound ioctl */ -#define TCSETSW 0x5403 -#define TCSETSF 0x5404 -#define TCGETA 0x5405 -#define TCSETA 0x5406 -#define TCSETAW 0x5407 -#define TCSETAF 0x5408 -#define TCSBRK 0x5409 -#define TCXONC 0x540A -#define TCFLSH 0x540B -#define TIOCEXCL 0x540C -#define TIOCNXCL 0x540D -#define TIOCSCTTY 0x540E -#define TIOCGPGRP 0x540F -#define TIOCSPGRP 0x5410 -#define TIOCOUTQ 0x5411 -#define TIOCSTI 0x5412 -#define TIOCGWINSZ 0x5413 -#define TIOCSWINSZ 0x5414 -#define TIOCMGET 0x5415 -#define TIOCMBIS 0x5416 -#define TIOCMBIC 0x5417 -#define TIOCMSET 0x5418 -#define TIOCGSOFTCAR 0x5419 -#define TIOCSSOFTCAR 0x541A -#define FIONREAD 0x541B -#define TIOCINQ FIONREAD -#define TIOCLINUX 0x541C -#define TIOCCONS 0x541D -#define TIOCGSERIAL 0x541E -#define TIOCSSERIAL 0x541F -#define TIOCPKT 0x5420 -#define FIONBIO 0x5421 -#define TIOCNOTTY 0x5422 -#define TIOCSETD 0x5423 -#define TIOCGETD 0x5424 -#define TCSBRKP 0x5425 /* Needed for POSIX tcsendbreak() */ -/* #define TIOCTTYGSTRUCT 0x5426 - Former debugging-only ioctl */ -#define TIOCSBRK 0x5427 /* BSD compatibility */ -#define TIOCCBRK 0x5428 /* BSD compatibility */ -#define TIOCGSID 0x5429 /* Return the session ID of FD */ -#define TCGETS2 _IOR('T', 0x2A, struct termios2) -#define TCSETS2 _IOW('T', 0x2B, struct termios2) -#define TCSETSW2 _IOW('T', 0x2C, struct termios2) -#define TCSETSF2 _IOW('T', 0x2D, struct termios2) -#define TIOCGRS485 0x542E -#define TIOCSRS485 0x542F -#define TIOCGPTN _IOR('T', 0x30, unsigned int) - /* Get Pty Number (of pty-mux device) */ -#define TIOCSPTLCK _IOW('T', 0x31, int) /* Lock/unlock Pty */ -#define TCGETX 0x5432 /* SYS5 TCGETX compatibility */ -#define TCSETX 0x5433 -#define TCSETXF 0x5434 -#define TCSETXW 0x5435 - -#define FIONCLEX 0x5450 -#define FIOCLEX 0x5451 -#define FIOASYNC 0x5452 -#define TIOCSERCONFIG 0x5453 -#define TIOCSERGWILD 0x5454 -#define TIOCSERSWILD 0x5455 -#define TIOCGLCKTRMIOS 0x5456 -#define TIOCSLCKTRMIOS 0x5457 -#define TIOCSERGSTRUCT 0x5458 /* For debugging only */ -#define TIOCSERGETLSR 0x5459 /* Get line status register */ -#define TIOCSERGETMULTI 0x545A /* Get multiport config */ -#define TIOCSERSETMULTI 0x545B /* Set multiport config */ - -#define TIOCMIWAIT 0x545C /* wait for a change on serial input line(s) */ -#define TIOCGICOUNT 0x545D /* read serial port inline interrupt counts */ -#define TIOCGHAYESESP 0x545E /* Get Hayes ESP configuration */ -#define TIOCSHAYESESP 0x545F /* Set Hayes ESP configuration */ -#define FIOQSIZE 0x5460 - -/* Used for packet mode */ -#define TIOCPKT_DATA 0 -#define TIOCPKT_FLUSHREAD 1 -#define TIOCPKT_FLUSHWRITE 2 -#define TIOCPKT_STOP 4 -#define TIOCPKT_START 8 -#define TIOCPKT_NOSTOP 16 -#define TIOCPKT_DOSTOP 32 - -#define TIOCSER_TEMT 0x01 /* Transmitter physically empty */ - -#endif /* _ASM_X86_IOCTLS_H */ +#include <asm-generic/ioctls.h> diff --git a/arch/x86/include/asm/ipcbuf.h b/arch/x86/include/asm/ipcbuf.h index ee678fd51594..84c7e51cb6d0 100644 --- a/arch/x86/include/asm/ipcbuf.h +++ b/arch/x86/include/asm/ipcbuf.h @@ -1,28 +1 @@ -#ifndef _ASM_X86_IPCBUF_H -#define _ASM_X86_IPCBUF_H - -/* - * The ipc64_perm structure for x86 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space is left for: - * - 32-bit mode_t and seq - * - 2 miscellaneous 32-bit values - */ - -struct ipc64_perm { - __kernel_key_t key; - __kernel_uid32_t uid; - __kernel_gid32_t gid; - __kernel_uid32_t cuid; - __kernel_gid32_t cgid; - __kernel_mode_t mode; - unsigned short __pad1; - unsigned short seq; - unsigned short __pad2; - unsigned long __unused1; - unsigned long __unused2; -}; - -#endif /* _ASM_X86_IPCBUF_H */ +#include <asm-generic/ipcbuf.h> diff --git a/arch/x86/include/asm/lguest.h b/arch/x86/include/asm/lguest.h index 5136dad57cbb..0d97deba1e35 100644 --- a/arch/x86/include/asm/lguest.h +++ b/arch/x86/include/asm/lguest.h @@ -90,8 +90,9 @@ static inline void lguest_set_ts(void) } /* Full 4G segment descriptors, suitable for CS and DS. */ -#define FULL_EXEC_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9b00} } }) -#define FULL_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9300} } }) +#define FULL_EXEC_SEGMENT \ + ((struct desc_struct)GDT_ENTRY_INIT(0xc09b, 0, 0xfffff)) +#define FULL_SEGMENT ((struct desc_struct)GDT_ENTRY_INIT(0xc093, 0, 0xfffff)) #endif /* __ASSEMBLY__ */ diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 5cdd8d100ec9..ad7535372918 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -9,7 +9,7 @@ */ #define MCG_BANKCNT_MASK 0xff /* Number of Banks */ -#define MCG_CTL_P (1ULL<<8) /* MCG_CAP register available */ +#define MCG_CTL_P (1ULL<<8) /* MCG_CTL register available */ #define MCG_EXT_P (1ULL<<9) /* Extended registers available */ #define MCG_CMCI_P (1ULL<<10) /* CMCI supported */ #define MCG_EXT_CNT_MASK 0xff0000 /* Number of Extended registers */ @@ -115,13 +115,6 @@ void mcheck_init(struct cpuinfo_x86 *c); static inline void mcheck_init(struct cpuinfo_x86 *c) {} #endif -#ifdef CONFIG_X86_OLD_MCE -extern int nr_mce_banks; -void amd_mcheck_init(struct cpuinfo_x86 *c); -void intel_p4_mcheck_init(struct cpuinfo_x86 *c); -void intel_p6_mcheck_init(struct cpuinfo_x86 *c); -#endif - #ifdef CONFIG_X86_ANCIENT_MCE void intel_p5_mcheck_init(struct cpuinfo_x86 *c); void winchip_mcheck_init(struct cpuinfo_x86 *c); @@ -137,10 +130,11 @@ void mce_log(struct mce *m); DECLARE_PER_CPU(struct sys_device, mce_dev); /* - * To support more than 128 would need to escape the predefined - * Linux defined extended banks first. + * Maximum banks number. + * This is the limit of the current register layout on + * Intel CPUs. */ -#define MAX_NR_BANKS (MCE_EXTENDED_BANK - 1) +#define MAX_NR_BANKS 32 #ifdef CONFIG_X86_MCE_INTEL extern int mce_cmci_disabled; @@ -208,11 +202,7 @@ extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); void intel_init_thermal(struct cpuinfo_x86 *c); -#ifdef CONFIG_X86_NEW_MCE void mce_log_therm_throt_event(__u64 status); -#else -static inline void mce_log_therm_throt_event(__u64 status) {} -#endif #endif /* __KERNEL__ */ #endif /* _ASM_X86_MCE_H */ diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h index 751af2550ed9..593e51d4643f 100644 --- a/arch/x86/include/asm/mman.h +++ b/arch/x86/include/asm/mman.h @@ -1,20 +1,8 @@ #ifndef _ASM_X86_MMAN_H #define _ASM_X86_MMAN_H -#include <asm-generic/mman-common.h> - #define MAP_32BIT 0x40 /* only give out 32bit addresses */ -#define MAP_GROWSDOWN 0x0100 /* stack-like segment */ -#define MAP_DENYWRITE 0x0800 /* ETXTBSY */ -#define MAP_EXECUTABLE 0x1000 /* mark it as an executable */ -#define MAP_LOCKED 0x2000 /* pages are locked */ -#define MAP_NORESERVE 0x4000 /* don't check for reservations */ -#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */ -#define MAP_NONBLOCK 0x10000 /* do not block on IO */ -#define MAP_STACK 0x20000 /* give out an address that is best suited for process/thread stacks */ - -#define MCL_CURRENT 1 /* lock all current mappings */ -#define MCL_FUTURE 2 /* lock all future mappings */ +#include <asm-generic/mman.h> #endif /* _ASM_X86_MMAN_H */ diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h index 47d62743c4d5..555bc12bdcd6 100644 --- a/arch/x86/include/asm/module.h +++ b/arch/x86/include/asm/module.h @@ -1,18 +1,7 @@ #ifndef _ASM_X86_MODULE_H #define _ASM_X86_MODULE_H -/* x86_32/64 are simple */ -struct mod_arch_specific {}; - -#ifdef CONFIG_X86_32 -# define Elf_Shdr Elf32_Shdr -# define Elf_Sym Elf32_Sym -# define Elf_Ehdr Elf32_Ehdr -#else -# define Elf_Shdr Elf64_Shdr -# define Elf_Sym Elf64_Sym -# define Elf_Ehdr Elf64_Ehdr -#endif +#include <asm-generic/module.h> #ifdef CONFIG_X86_64 /* X86_64 does not define MODULE_PROC_FAMILY */ diff --git a/arch/x86/include/asm/msgbuf.h b/arch/x86/include/asm/msgbuf.h index 7e4e9481f51c..809134c644a6 100644 --- a/arch/x86/include/asm/msgbuf.h +++ b/arch/x86/include/asm/msgbuf.h @@ -1,39 +1 @@ -#ifndef _ASM_X86_MSGBUF_H -#define _ASM_X86_MSGBUF_H - -/* - * The msqid64_ds structure for i386 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space on i386 is left for: - * - 64-bit time_t to solve y2038 problem - * - 2 miscellaneous 32-bit values - * - * Pad space on x8664 is left for: - * - 2 miscellaneous 64-bit values - */ -struct msqid64_ds { - struct ipc64_perm msg_perm; - __kernel_time_t msg_stime; /* last msgsnd time */ -#ifdef __i386__ - unsigned long __unused1; -#endif - __kernel_time_t msg_rtime; /* last msgrcv time */ -#ifdef __i386__ - unsigned long __unused2; -#endif - __kernel_time_t msg_ctime; /* last change time */ -#ifdef __i386__ - unsigned long __unused3; -#endif - unsigned long msg_cbytes; /* current number of bytes on queue */ - unsigned long msg_qnum; /* number of messages in queue */ - unsigned long msg_qbytes; /* max number of bytes on queue */ - __kernel_pid_t msg_lspid; /* pid of last msgsnd */ - __kernel_pid_t msg_lrpid; /* last receive pid */ - unsigned long __unused4; - unsigned long __unused5; -}; - -#endif /* _ASM_X86_MSGBUF_H */ +#include <asm-generic/msgbuf.h> diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index bd5549034a95..4ffe09b2ad75 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -81,8 +81,15 @@ #define MSR_IA32_MC0_ADDR 0x00000402 #define MSR_IA32_MC0_MISC 0x00000403 +#define MSR_IA32_MCx_CTL(x) (MSR_IA32_MC0_CTL + 4*(x)) +#define MSR_IA32_MCx_STATUS(x) (MSR_IA32_MC0_STATUS + 4*(x)) +#define MSR_IA32_MCx_ADDR(x) (MSR_IA32_MC0_ADDR + 4*(x)) +#define MSR_IA32_MCx_MISC(x) (MSR_IA32_MC0_MISC + 4*(x)) + /* These are consecutive and not in the normal 4er MCE bank block */ #define MSR_IA32_MC0_CTL2 0x00000280 +#define MSR_IA32_MCx_CTL2(x) (MSR_IA32_MC0_CTL2 + (x)) + #define CMCI_EN (1ULL << 30) #define CMCI_THRESHOLD_MASK 0xffffULL @@ -215,6 +222,10 @@ #define THERM_STATUS_PROCHOT (1 << 0) +#define MSR_THERM2_CTL 0x0000019d + +#define MSR_THERM2_CTL_TM_SELECT (1ULL << 16) + #define MSR_IA32_MISC_ENABLE 0x000001a0 /* MISC_ENABLE bits: architectural */ diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index c86e5ed4af51..e63cf7d441e1 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -45,8 +45,8 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, void __user *, size_t *, loff_t *); extern int unknown_nmi_panic; -void __trigger_all_cpu_backtrace(void); -#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() +void arch_trigger_all_cpu_backtrace(void); +#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace static inline void localise_nmi_watchdog(void) { diff --git a/arch/x86/include/asm/param.h b/arch/x86/include/asm/param.h index 6f0d0422f4ca..965d45427975 100644 --- a/arch/x86/include/asm/param.h +++ b/arch/x86/include/asm/param.h @@ -1,22 +1 @@ -#ifndef _ASM_X86_PARAM_H -#define _ASM_X86_PARAM_H - -#ifdef __KERNEL__ -# define HZ CONFIG_HZ /* Internal kernel timer frequency */ -# define USER_HZ 100 /* some user interfaces are */ -# define CLOCKS_PER_SEC (USER_HZ) /* in "ticks" like times() */ -#endif - -#ifndef HZ -#define HZ 100 -#endif - -#define EXEC_PAGESIZE 4096 - -#ifndef NOGROUP -#define NOGROUP (-1) -#endif - -#define MAXHOSTNAMELEN 64 /* max length of hostname */ - -#endif /* _ASM_X86_PARAM_H */ +#include <asm-generic/param.h> diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 4fb37c8a0832..6a07af432c81 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -7,689 +7,11 @@ #include <asm/pgtable_types.h> #include <asm/asm.h> -/* Bitmask of what can be clobbered: usually at least eax. */ -#define CLBR_NONE 0 -#define CLBR_EAX (1 << 0) -#define CLBR_ECX (1 << 1) -#define CLBR_EDX (1 << 2) -#define CLBR_EDI (1 << 3) - -#ifdef CONFIG_X86_32 -/* CLBR_ANY should match all regs platform has. For i386, that's just it */ -#define CLBR_ANY ((1 << 4) - 1) - -#define CLBR_ARG_REGS (CLBR_EAX | CLBR_EDX | CLBR_ECX) -#define CLBR_RET_REG (CLBR_EAX | CLBR_EDX) -#define CLBR_SCRATCH (0) -#else -#define CLBR_RAX CLBR_EAX -#define CLBR_RCX CLBR_ECX -#define CLBR_RDX CLBR_EDX -#define CLBR_RDI CLBR_EDI -#define CLBR_RSI (1 << 4) -#define CLBR_R8 (1 << 5) -#define CLBR_R9 (1 << 6) -#define CLBR_R10 (1 << 7) -#define CLBR_R11 (1 << 8) - -#define CLBR_ANY ((1 << 9) - 1) - -#define CLBR_ARG_REGS (CLBR_RDI | CLBR_RSI | CLBR_RDX | \ - CLBR_RCX | CLBR_R8 | CLBR_R9) -#define CLBR_RET_REG (CLBR_RAX) -#define CLBR_SCRATCH (CLBR_R10 | CLBR_R11) - -#include <asm/desc_defs.h> -#endif /* X86_64 */ - -#define CLBR_CALLEE_SAVE ((CLBR_ARG_REGS | CLBR_SCRATCH) & ~CLBR_RET_REG) +#include <asm/paravirt_types.h> #ifndef __ASSEMBLY__ #include <linux/types.h> #include <linux/cpumask.h> -#include <asm/kmap_types.h> -#include <asm/desc_defs.h> - -struct page; -struct thread_struct; -struct desc_ptr; -struct tss_struct; -struct mm_struct; -struct desc_struct; -struct task_struct; - -/* - * Wrapper type for pointers to code which uses the non-standard - * calling convention. See PV_CALL_SAVE_REGS_THUNK below. - */ -struct paravirt_callee_save { - void *func; -}; - -/* general info */ -struct pv_info { - unsigned int kernel_rpl; - int shared_kernel_pmd; - int paravirt_enabled; - const char *name; -}; - -struct pv_init_ops { - /* - * Patch may replace one of the defined code sequences with - * arbitrary code, subject to the same register constraints. - * This generally means the code is not free to clobber any - * registers other than EAX. The patch function should return - * the number of bytes of code generated, as we nop pad the - * rest in generic code. - */ - unsigned (*patch)(u8 type, u16 clobber, void *insnbuf, - unsigned long addr, unsigned len); - - /* Basic arch-specific setup */ - void (*arch_setup)(void); - char *(*memory_setup)(void); - void (*post_allocator_init)(void); - - /* Print a banner to identify the environment */ - void (*banner)(void); -}; - - -struct pv_lazy_ops { - /* Set deferred update mode, used for batching operations. */ - void (*enter)(void); - void (*leave)(void); -}; - -struct pv_time_ops { - void (*time_init)(void); - - /* Set and set time of day */ - unsigned long (*get_wallclock)(void); - int (*set_wallclock)(unsigned long); - - unsigned long long (*sched_clock)(void); - unsigned long (*get_tsc_khz)(void); -}; - -struct pv_cpu_ops { - /* hooks for various privileged instructions */ - unsigned long (*get_debugreg)(int regno); - void (*set_debugreg)(int regno, unsigned long value); - - void (*clts)(void); - - unsigned long (*read_cr0)(void); - void (*write_cr0)(unsigned long); - - unsigned long (*read_cr4_safe)(void); - unsigned long (*read_cr4)(void); - void (*write_cr4)(unsigned long); - -#ifdef CONFIG_X86_64 - unsigned long (*read_cr8)(void); - void (*write_cr8)(unsigned long); -#endif - - /* Segment descriptor handling */ - void (*load_tr_desc)(void); - void (*load_gdt)(const struct desc_ptr *); - void (*load_idt)(const struct desc_ptr *); - void (*store_gdt)(struct desc_ptr *); - void (*store_idt)(struct desc_ptr *); - void (*set_ldt)(const void *desc, unsigned entries); - unsigned long (*store_tr)(void); - void (*load_tls)(struct thread_struct *t, unsigned int cpu); -#ifdef CONFIG_X86_64 - void (*load_gs_index)(unsigned int idx); -#endif - void (*write_ldt_entry)(struct desc_struct *ldt, int entrynum, - const void *desc); - void (*write_gdt_entry)(struct desc_struct *, - int entrynum, const void *desc, int size); - void (*write_idt_entry)(gate_desc *, - int entrynum, const gate_desc *gate); - void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries); - void (*free_ldt)(struct desc_struct *ldt, unsigned entries); - - void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t); - - void (*set_iopl_mask)(unsigned mask); - - void (*wbinvd)(void); - void (*io_delay)(void); - - /* cpuid emulation, mostly so that caps bits can be disabled */ - void (*cpuid)(unsigned int *eax, unsigned int *ebx, - unsigned int *ecx, unsigned int *edx); - - /* MSR, PMC and TSR operations. - err = 0/-EFAULT. wrmsr returns 0/-EFAULT. */ - u64 (*read_msr_amd)(unsigned int msr, int *err); - u64 (*read_msr)(unsigned int msr, int *err); - int (*write_msr)(unsigned int msr, unsigned low, unsigned high); - - u64 (*read_tsc)(void); - u64 (*read_pmc)(int counter); - unsigned long long (*read_tscp)(unsigned int *aux); - - /* - * Atomically enable interrupts and return to userspace. This - * is only ever used to return to 32-bit processes; in a - * 64-bit kernel, it's used for 32-on-64 compat processes, but - * never native 64-bit processes. (Jump, not call.) - */ - void (*irq_enable_sysexit)(void); - - /* - * Switch to usermode gs and return to 64-bit usermode using - * sysret. Only used in 64-bit kernels to return to 64-bit - * processes. Usermode register state, including %rsp, must - * already be restored. - */ - void (*usergs_sysret64)(void); - - /* - * Switch to usermode gs and return to 32-bit usermode using - * sysret. Used to return to 32-on-64 compat processes. - * Other usermode register state, including %esp, must already - * be restored. - */ - void (*usergs_sysret32)(void); - - /* Normal iret. Jump to this with the standard iret stack - frame set up. */ - void (*iret)(void); - - void (*swapgs)(void); - - void (*start_context_switch)(struct task_struct *prev); - void (*end_context_switch)(struct task_struct *next); -}; - -struct pv_irq_ops { - void (*init_IRQ)(void); - - /* - * Get/set interrupt state. save_fl and restore_fl are only - * expected to use X86_EFLAGS_IF; all other bits - * returned from save_fl are undefined, and may be ignored by - * restore_fl. - * - * NOTE: These functions callers expect the callee to preserve - * more registers than the standard C calling convention. - */ - struct paravirt_callee_save save_fl; - struct paravirt_callee_save restore_fl; - struct paravirt_callee_save irq_disable; - struct paravirt_callee_save irq_enable; - - void (*safe_halt)(void); - void (*halt)(void); - -#ifdef CONFIG_X86_64 - void (*adjust_exception_frame)(void); -#endif -}; - -struct pv_apic_ops { -#ifdef CONFIG_X86_LOCAL_APIC - void (*setup_boot_clock)(void); - void (*setup_secondary_clock)(void); - - void (*startup_ipi_hook)(int phys_apicid, - unsigned long start_eip, - unsigned long start_esp); -#endif -}; - -struct pv_mmu_ops { - /* - * Called before/after init_mm pagetable setup. setup_start - * may reset %cr3, and may pre-install parts of the pagetable; - * pagetable setup is expected to preserve any existing - * mapping. - */ - void (*pagetable_setup_start)(pgd_t *pgd_base); - void (*pagetable_setup_done)(pgd_t *pgd_base); - - unsigned long (*read_cr2)(void); - void (*write_cr2)(unsigned long); - - unsigned long (*read_cr3)(void); - void (*write_cr3)(unsigned long); - - /* - * Hooks for intercepting the creation/use/destruction of an - * mm_struct. - */ - void (*activate_mm)(struct mm_struct *prev, - struct mm_struct *next); - void (*dup_mmap)(struct mm_struct *oldmm, - struct mm_struct *mm); - void (*exit_mmap)(struct mm_struct *mm); - - - /* TLB operations */ - void (*flush_tlb_user)(void); - void (*flush_tlb_kernel)(void); - void (*flush_tlb_single)(unsigned long addr); - void (*flush_tlb_others)(const struct cpumask *cpus, - struct mm_struct *mm, - unsigned long va); - - /* Hooks for allocating and freeing a pagetable top-level */ - int (*pgd_alloc)(struct mm_struct *mm); - void (*pgd_free)(struct mm_struct *mm, pgd_t *pgd); - - /* - * Hooks for allocating/releasing pagetable pages when they're - * attached to a pagetable - */ - void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn); - void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn); - void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count); - void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn); - void (*release_pte)(unsigned long pfn); - void (*release_pmd)(unsigned long pfn); - void (*release_pud)(unsigned long pfn); - - /* Pagetable manipulation functions */ - void (*set_pte)(pte_t *ptep, pte_t pteval); - void (*set_pte_at)(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pteval); - void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval); - void (*pte_update)(struct mm_struct *mm, unsigned long addr, - pte_t *ptep); - void (*pte_update_defer)(struct mm_struct *mm, - unsigned long addr, pte_t *ptep); - - pte_t (*ptep_modify_prot_start)(struct mm_struct *mm, unsigned long addr, - pte_t *ptep); - void (*ptep_modify_prot_commit)(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pte); - - struct paravirt_callee_save pte_val; - struct paravirt_callee_save make_pte; - - struct paravirt_callee_save pgd_val; - struct paravirt_callee_save make_pgd; - -#if PAGETABLE_LEVELS >= 3 -#ifdef CONFIG_X86_PAE - void (*set_pte_atomic)(pte_t *ptep, pte_t pteval); - void (*pte_clear)(struct mm_struct *mm, unsigned long addr, - pte_t *ptep); - void (*pmd_clear)(pmd_t *pmdp); - -#endif /* CONFIG_X86_PAE */ - - void (*set_pud)(pud_t *pudp, pud_t pudval); - - struct paravirt_callee_save pmd_val; - struct paravirt_callee_save make_pmd; - -#if PAGETABLE_LEVELS == 4 - struct paravirt_callee_save pud_val; - struct paravirt_callee_save make_pud; - - void (*set_pgd)(pgd_t *pudp, pgd_t pgdval); -#endif /* PAGETABLE_LEVELS == 4 */ -#endif /* PAGETABLE_LEVELS >= 3 */ - -#ifdef CONFIG_HIGHPTE - void *(*kmap_atomic_pte)(struct page *page, enum km_type type); -#endif - - struct pv_lazy_ops lazy_mode; - - /* dom0 ops */ - - /* Sometimes the physical address is a pfn, and sometimes its - an mfn. We can tell which is which from the index. */ - void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx, - phys_addr_t phys, pgprot_t flags); -}; - -struct raw_spinlock; -struct pv_lock_ops { - int (*spin_is_locked)(struct raw_spinlock *lock); - int (*spin_is_contended)(struct raw_spinlock *lock); - void (*spin_lock)(struct raw_spinlock *lock); - void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); - int (*spin_trylock)(struct raw_spinlock *lock); - void (*spin_unlock)(struct raw_spinlock *lock); -}; - -/* This contains all the paravirt structures: we get a convenient - * number for each function using the offset which we use to indicate - * what to patch. */ -struct paravirt_patch_template { - struct pv_init_ops pv_init_ops; - struct pv_time_ops pv_time_ops; - struct pv_cpu_ops pv_cpu_ops; - struct pv_irq_ops pv_irq_ops; - struct pv_apic_ops pv_apic_ops; - struct pv_mmu_ops pv_mmu_ops; - struct pv_lock_ops pv_lock_ops; -}; - -extern struct pv_info pv_info; -extern struct pv_init_ops pv_init_ops; -extern struct pv_time_ops pv_time_ops; -extern struct pv_cpu_ops pv_cpu_ops; -extern struct pv_irq_ops pv_irq_ops; -extern struct pv_apic_ops pv_apic_ops; -extern struct pv_mmu_ops pv_mmu_ops; -extern struct pv_lock_ops pv_lock_ops; - -#define PARAVIRT_PATCH(x) \ - (offsetof(struct paravirt_patch_template, x) / sizeof(void *)) - -#define paravirt_type(op) \ - [paravirt_typenum] "i" (PARAVIRT_PATCH(op)), \ - [paravirt_opptr] "i" (&(op)) -#define paravirt_clobber(clobber) \ - [paravirt_clobber] "i" (clobber) - -/* - * Generate some code, and mark it as patchable by the - * apply_paravirt() alternate instruction patcher. - */ -#define _paravirt_alt(insn_string, type, clobber) \ - "771:\n\t" insn_string "\n" "772:\n" \ - ".pushsection .parainstructions,\"a\"\n" \ - _ASM_ALIGN "\n" \ - _ASM_PTR " 771b\n" \ - " .byte " type "\n" \ - " .byte 772b-771b\n" \ - " .short " clobber "\n" \ - ".popsection\n" - -/* Generate patchable code, with the default asm parameters. */ -#define paravirt_alt(insn_string) \ - _paravirt_alt(insn_string, "%c[paravirt_typenum]", "%c[paravirt_clobber]") - -/* Simple instruction patching code. */ -#define DEF_NATIVE(ops, name, code) \ - extern const char start_##ops##_##name[], end_##ops##_##name[]; \ - asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":") - -unsigned paravirt_patch_nop(void); -unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len); -unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len); -unsigned paravirt_patch_ignore(unsigned len); -unsigned paravirt_patch_call(void *insnbuf, - const void *target, u16 tgt_clobbers, - unsigned long addr, u16 site_clobbers, - unsigned len); -unsigned paravirt_patch_jmp(void *insnbuf, const void *target, - unsigned long addr, unsigned len); -unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, - unsigned long addr, unsigned len); - -unsigned paravirt_patch_insns(void *insnbuf, unsigned len, - const char *start, const char *end); - -unsigned native_patch(u8 type, u16 clobbers, void *ibuf, - unsigned long addr, unsigned len); - -int paravirt_disable_iospace(void); - -/* - * This generates an indirect call based on the operation type number. - * The type number, computed in PARAVIRT_PATCH, is derived from the - * offset into the paravirt_patch_template structure, and can therefore be - * freely converted back into a structure offset. - */ -#define PARAVIRT_CALL "call *%c[paravirt_opptr];" - -/* - * These macros are intended to wrap calls through one of the paravirt - * ops structs, so that they can be later identified and patched at - * runtime. - * - * Normally, a call to a pv_op function is a simple indirect call: - * (pv_op_struct.operations)(args...). - * - * Unfortunately, this is a relatively slow operation for modern CPUs, - * because it cannot necessarily determine what the destination - * address is. In this case, the address is a runtime constant, so at - * the very least we can patch the call to e a simple direct call, or - * ideally, patch an inline implementation into the callsite. (Direct - * calls are essentially free, because the call and return addresses - * are completely predictable.) - * - * For i386, these macros rely on the standard gcc "regparm(3)" calling - * convention, in which the first three arguments are placed in %eax, - * %edx, %ecx (in that order), and the remaining arguments are placed - * on the stack. All caller-save registers (eax,edx,ecx) are expected - * to be modified (either clobbered or used for return values). - * X86_64, on the other hand, already specifies a register-based calling - * conventions, returning at %rax, with parameteres going on %rdi, %rsi, - * %rdx, and %rcx. Note that for this reason, x86_64 does not need any - * special handling for dealing with 4 arguments, unlike i386. - * However, x86_64 also have to clobber all caller saved registers, which - * unfortunately, are quite a bit (r8 - r11) - * - * The call instruction itself is marked by placing its start address - * and size into the .parainstructions section, so that - * apply_paravirt() in arch/i386/kernel/alternative.c can do the - * appropriate patching under the control of the backend pv_init_ops - * implementation. - * - * Unfortunately there's no way to get gcc to generate the args setup - * for the call, and then allow the call itself to be generated by an - * inline asm. Because of this, we must do the complete arg setup and - * return value handling from within these macros. This is fairly - * cumbersome. - * - * There are 5 sets of PVOP_* macros for dealing with 0-4 arguments. - * It could be extended to more arguments, but there would be little - * to be gained from that. For each number of arguments, there are - * the two VCALL and CALL variants for void and non-void functions. - * - * When there is a return value, the invoker of the macro must specify - * the return type. The macro then uses sizeof() on that type to - * determine whether its a 32 or 64 bit value, and places the return - * in the right register(s) (just %eax for 32-bit, and %edx:%eax for - * 64-bit). For x86_64 machines, it just returns at %rax regardless of - * the return value size. - * - * 64-bit arguments are passed as a pair of adjacent 32-bit arguments - * i386 also passes 64-bit arguments as a pair of adjacent 32-bit arguments - * in low,high order - * - * Small structures are passed and returned in registers. The macro - * calling convention can't directly deal with this, so the wrapper - * functions must do this. - * - * These PVOP_* macros are only defined within this header. This - * means that all uses must be wrapped in inline functions. This also - * makes sure the incoming and outgoing types are always correct. - */ -#ifdef CONFIG_X86_32 -#define PVOP_VCALL_ARGS \ - unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx -#define PVOP_CALL_ARGS PVOP_VCALL_ARGS - -#define PVOP_CALL_ARG1(x) "a" ((unsigned long)(x)) -#define PVOP_CALL_ARG2(x) "d" ((unsigned long)(x)) -#define PVOP_CALL_ARG3(x) "c" ((unsigned long)(x)) - -#define PVOP_VCALL_CLOBBERS "=a" (__eax), "=d" (__edx), \ - "=c" (__ecx) -#define PVOP_CALL_CLOBBERS PVOP_VCALL_CLOBBERS - -#define PVOP_VCALLEE_CLOBBERS "=a" (__eax), "=d" (__edx) -#define PVOP_CALLEE_CLOBBERS PVOP_VCALLEE_CLOBBERS - -#define EXTRA_CLOBBERS -#define VEXTRA_CLOBBERS -#else /* CONFIG_X86_64 */ -#define PVOP_VCALL_ARGS \ - unsigned long __edi = __edi, __esi = __esi, \ - __edx = __edx, __ecx = __ecx -#define PVOP_CALL_ARGS PVOP_VCALL_ARGS, __eax - -#define PVOP_CALL_ARG1(x) "D" ((unsigned long)(x)) -#define PVOP_CALL_ARG2(x) "S" ((unsigned long)(x)) -#define PVOP_CALL_ARG3(x) "d" ((unsigned long)(x)) -#define PVOP_CALL_ARG4(x) "c" ((unsigned long)(x)) - -#define PVOP_VCALL_CLOBBERS "=D" (__edi), \ - "=S" (__esi), "=d" (__edx), \ - "=c" (__ecx) -#define PVOP_CALL_CLOBBERS PVOP_VCALL_CLOBBERS, "=a" (__eax) - -#define PVOP_VCALLEE_CLOBBERS "=a" (__eax) -#define PVOP_CALLEE_CLOBBERS PVOP_VCALLEE_CLOBBERS - -#define EXTRA_CLOBBERS , "r8", "r9", "r10", "r11" -#define VEXTRA_CLOBBERS , "rax", "r8", "r9", "r10", "r11" -#endif /* CONFIG_X86_32 */ - -#ifdef CONFIG_PARAVIRT_DEBUG -#define PVOP_TEST_NULL(op) BUG_ON(op == NULL) -#else -#define PVOP_TEST_NULL(op) ((void)op) -#endif - -#define ____PVOP_CALL(rettype, op, clbr, call_clbr, extra_clbr, \ - pre, post, ...) \ - ({ \ - rettype __ret; \ - PVOP_CALL_ARGS; \ - PVOP_TEST_NULL(op); \ - /* This is 32-bit specific, but is okay in 64-bit */ \ - /* since this condition will never hold */ \ - if (sizeof(rettype) > sizeof(unsigned long)) { \ - asm volatile(pre \ - paravirt_alt(PARAVIRT_CALL) \ - post \ - : call_clbr \ - : paravirt_type(op), \ - paravirt_clobber(clbr), \ - ##__VA_ARGS__ \ - : "memory", "cc" extra_clbr); \ - __ret = (rettype)((((u64)__edx) << 32) | __eax); \ - } else { \ - asm volatile(pre \ - paravirt_alt(PARAVIRT_CALL) \ - post \ - : call_clbr \ - : paravirt_type(op), \ - paravirt_clobber(clbr), \ - ##__VA_ARGS__ \ - : "memory", "cc" extra_clbr); \ - __ret = (rettype)__eax; \ - } \ - __ret; \ - }) - -#define __PVOP_CALL(rettype, op, pre, post, ...) \ - ____PVOP_CALL(rettype, op, CLBR_ANY, PVOP_CALL_CLOBBERS, \ - EXTRA_CLOBBERS, pre, post, ##__VA_ARGS__) - -#define __PVOP_CALLEESAVE(rettype, op, pre, post, ...) \ - ____PVOP_CALL(rettype, op.func, CLBR_RET_REG, \ - PVOP_CALLEE_CLOBBERS, , \ - pre, post, ##__VA_ARGS__) - - -#define ____PVOP_VCALL(op, clbr, call_clbr, extra_clbr, pre, post, ...) \ - ({ \ - PVOP_VCALL_ARGS; \ - PVOP_TEST_NULL(op); \ - asm volatile(pre \ - paravirt_alt(PARAVIRT_CALL) \ - post \ - : call_clbr \ - : paravirt_type(op), \ - paravirt_clobber(clbr), \ - ##__VA_ARGS__ \ - : "memory", "cc" extra_clbr); \ - }) - -#define __PVOP_VCALL(op, pre, post, ...) \ - ____PVOP_VCALL(op, CLBR_ANY, PVOP_VCALL_CLOBBERS, \ - VEXTRA_CLOBBERS, \ - pre, post, ##__VA_ARGS__) - -#define __PVOP_VCALLEESAVE(rettype, op, pre, post, ...) \ - ____PVOP_CALL(rettype, op.func, CLBR_RET_REG, \ - PVOP_VCALLEE_CLOBBERS, , \ - pre, post, ##__VA_ARGS__) - - - -#define PVOP_CALL0(rettype, op) \ - __PVOP_CALL(rettype, op, "", "") -#define PVOP_VCALL0(op) \ - __PVOP_VCALL(op, "", "") - -#define PVOP_CALLEE0(rettype, op) \ - __PVOP_CALLEESAVE(rettype, op, "", "") -#define PVOP_VCALLEE0(op) \ - __PVOP_VCALLEESAVE(op, "", "") - - -#define PVOP_CALL1(rettype, op, arg1) \ - __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) -#define PVOP_VCALL1(op, arg1) \ - __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1)) - -#define PVOP_CALLEE1(rettype, op, arg1) \ - __PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) -#define PVOP_VCALLEE1(op, arg1) \ - __PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1)) - - -#define PVOP_CALL2(rettype, op, arg1, arg2) \ - __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2)) -#define PVOP_VCALL2(op, arg1, arg2) \ - __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2)) - -#define PVOP_CALLEE2(rettype, op, arg1, arg2) \ - __PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2)) -#define PVOP_VCALLEE2(op, arg1, arg2) \ - __PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2)) - - -#define PVOP_CALL3(rettype, op, arg1, arg2, arg3) \ - __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3)) -#define PVOP_VCALL3(op, arg1, arg2, arg3) \ - __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1), \ - PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3)) - -/* This is the only difference in x86_64. We can make it much simpler */ -#ifdef CONFIG_X86_32 -#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4) \ - __PVOP_CALL(rettype, op, \ - "push %[_arg4];", "lea 4(%%esp),%%esp;", \ - PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ - PVOP_CALL_ARG3(arg3), [_arg4] "mr" ((u32)(arg4))) -#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4) \ - __PVOP_VCALL(op, \ - "push %[_arg4];", "lea 4(%%esp),%%esp;", \ - "0" ((u32)(arg1)), "1" ((u32)(arg2)), \ - "2" ((u32)(arg3)), [_arg4] "mr" ((u32)(arg4))) -#else -#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4) \ - __PVOP_CALL(rettype, op, "", "", \ - PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ - PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4)) -#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4) \ - __PVOP_VCALL(op, "", "", \ - PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ - PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4)) -#endif static inline int paravirt_enabled(void) { @@ -1393,20 +715,6 @@ static inline void pmd_clear(pmd_t *pmdp) } #endif /* CONFIG_X86_PAE */ -/* Lazy mode for batching updates / context switch */ -enum paravirt_lazy_mode { - PARAVIRT_LAZY_NONE, - PARAVIRT_LAZY_MMU, - PARAVIRT_LAZY_CPU, -}; - -enum paravirt_lazy_mode paravirt_get_lazy_mode(void); -void paravirt_start_context_switch(struct task_struct *prev); -void paravirt_end_context_switch(struct task_struct *next); - -void paravirt_enter_lazy_mmu(void); -void paravirt_leave_lazy_mmu(void); - #define __HAVE_ARCH_START_CONTEXT_SWITCH static inline void arch_start_context_switch(struct task_struct *prev) { @@ -1437,12 +745,6 @@ static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, pv_mmu_ops.set_fixmap(idx, phys, flags); } -void _paravirt_nop(void); -u32 _paravirt_ident_32(u32); -u64 _paravirt_ident_64(u64); - -#define paravirt_nop ((void *)_paravirt_nop) - #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) static inline int __raw_spin_is_locked(struct raw_spinlock *lock) @@ -1479,17 +781,6 @@ static __always_inline void __raw_spin_unlock(struct raw_spinlock *lock) #endif -/* These all sit in the .parainstructions section to tell us what to patch. */ -struct paravirt_patch_site { - u8 *instr; /* original instructions */ - u8 instrtype; /* type of this instruction */ - u8 len; /* length of original instruction */ - u16 clobbers; /* what registers you may clobber */ -}; - -extern struct paravirt_patch_site __parainstructions[], - __parainstructions_end[]; - #ifdef CONFIG_X86_32 #define PV_SAVE_REGS "pushl %ecx; pushl %edx;" #define PV_RESTORE_REGS "popl %edx; popl %ecx;" diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h new file mode 100644 index 000000000000..2b3371bae295 --- /dev/null +++ b/arch/x86/include/asm/paravirt_types.h @@ -0,0 +1,720 @@ +#ifndef _ASM_X86_PARAVIRT_TYPES_H +#define _ASM_X86_PARAVIRT_TYPES_H + +/* Bitmask of what can be clobbered: usually at least eax. */ +#define CLBR_NONE 0 +#define CLBR_EAX (1 << 0) +#define CLBR_ECX (1 << 1) +#define CLBR_EDX (1 << 2) +#define CLBR_EDI (1 << 3) + +#ifdef CONFIG_X86_32 +/* CLBR_ANY should match all regs platform has. For i386, that's just it */ +#define CLBR_ANY ((1 << 4) - 1) + +#define CLBR_ARG_REGS (CLBR_EAX | CLBR_EDX | CLBR_ECX) +#define CLBR_RET_REG (CLBR_EAX | CLBR_EDX) +#define CLBR_SCRATCH (0) +#else +#define CLBR_RAX CLBR_EAX +#define CLBR_RCX CLBR_ECX +#define CLBR_RDX CLBR_EDX +#define CLBR_RDI CLBR_EDI +#define CLBR_RSI (1 << 4) +#define CLBR_R8 (1 << 5) +#define CLBR_R9 (1 << 6) +#define CLBR_R10 (1 << 7) +#define CLBR_R11 (1 << 8) + +#define CLBR_ANY ((1 << 9) - 1) + +#define CLBR_ARG_REGS (CLBR_RDI | CLBR_RSI | CLBR_RDX | \ + CLBR_RCX | CLBR_R8 | CLBR_R9) +#define CLBR_RET_REG (CLBR_RAX) +#define CLBR_SCRATCH (CLBR_R10 | CLBR_R11) + +#endif /* X86_64 */ + +#define CLBR_CALLEE_SAVE ((CLBR_ARG_REGS | CLBR_SCRATCH) & ~CLBR_RET_REG) + +#ifndef __ASSEMBLY__ + +#include <asm/desc_defs.h> +#include <asm/kmap_types.h> + +struct page; +struct thread_struct; +struct desc_ptr; +struct tss_struct; +struct mm_struct; +struct desc_struct; +struct task_struct; +struct cpumask; + +/* + * Wrapper type for pointers to code which uses the non-standard + * calling convention. See PV_CALL_SAVE_REGS_THUNK below. + */ +struct paravirt_callee_save { + void *func; +}; + +/* general info */ +struct pv_info { + unsigned int kernel_rpl; + int shared_kernel_pmd; + int paravirt_enabled; + const char *name; +}; + +struct pv_init_ops { + /* + * Patch may replace one of the defined code sequences with + * arbitrary code, subject to the same register constraints. + * This generally means the code is not free to clobber any + * registers other than EAX. The patch function should return + * the number of bytes of code generated, as we nop pad the + * rest in generic code. + */ + unsigned (*patch)(u8 type, u16 clobber, void *insnbuf, + unsigned long addr, unsigned len); + + /* Basic arch-specific setup */ + void (*arch_setup)(void); + char *(*memory_setup)(void); + void (*post_allocator_init)(void); + + /* Print a banner to identify the environment */ + void (*banner)(void); +}; + + +struct pv_lazy_ops { + /* Set deferred update mode, used for batching operations. */ + void (*enter)(void); + void (*leave)(void); +}; + +struct pv_time_ops { + void (*time_init)(void); + + /* Set and set time of day */ + unsigned long (*get_wallclock)(void); + int (*set_wallclock)(unsigned long); + + unsigned long long (*sched_clock)(void); + unsigned long (*get_tsc_khz)(void); +}; + +struct pv_cpu_ops { + /* hooks for various privileged instructions */ + unsigned long (*get_debugreg)(int regno); + void (*set_debugreg)(int regno, unsigned long value); + + void (*clts)(void); + + unsigned long (*read_cr0)(void); + void (*write_cr0)(unsigned long); + + unsigned long (*read_cr4_safe)(void); + unsigned long (*read_cr4)(void); + void (*write_cr4)(unsigned long); + +#ifdef CONFIG_X86_64 + unsigned long (*read_cr8)(void); + void (*write_cr8)(unsigned long); +#endif + + /* Segment descriptor handling */ + void (*load_tr_desc)(void); + void (*load_gdt)(const struct desc_ptr *); + void (*load_idt)(const struct desc_ptr *); + void (*store_gdt)(struct desc_ptr *); + void (*store_idt)(struct desc_ptr *); + void (*set_ldt)(const void *desc, unsigned entries); + unsigned long (*store_tr)(void); + void (*load_tls)(struct thread_struct *t, unsigned int cpu); +#ifdef CONFIG_X86_64 + void (*load_gs_index)(unsigned int idx); +#endif + void (*write_ldt_entry)(struct desc_struct *ldt, int entrynum, + const void *desc); + void (*write_gdt_entry)(struct desc_struct *, + int entrynum, const void *desc, int size); + void (*write_idt_entry)(gate_desc *, + int entrynum, const gate_desc *gate); + void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries); + void (*free_ldt)(struct desc_struct *ldt, unsigned entries); + + void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t); + + void (*set_iopl_mask)(unsigned mask); + + void (*wbinvd)(void); + void (*io_delay)(void); + + /* cpuid emulation, mostly so that caps bits can be disabled */ + void (*cpuid)(unsigned int *eax, unsigned int *ebx, + unsigned int *ecx, unsigned int *edx); + + /* MSR, PMC and TSR operations. + err = 0/-EFAULT. wrmsr returns 0/-EFAULT. */ + u64 (*read_msr_amd)(unsigned int msr, int *err); + u64 (*read_msr)(unsigned int msr, int *err); + int (*write_msr)(unsigned int msr, unsigned low, unsigned high); + + u64 (*read_tsc)(void); + u64 (*read_pmc)(int counter); + unsigned long long (*read_tscp)(unsigned int *aux); + + /* + * Atomically enable interrupts and return to userspace. This + * is only ever used to return to 32-bit processes; in a + * 64-bit kernel, it's used for 32-on-64 compat processes, but + * never native 64-bit processes. (Jump, not call.) + */ + void (*irq_enable_sysexit)(void); + + /* + * Switch to usermode gs and return to 64-bit usermode using + * sysret. Only used in 64-bit kernels to return to 64-bit + * processes. Usermode register state, including %rsp, must + * already be restored. + */ + void (*usergs_sysret64)(void); + + /* + * Switch to usermode gs and return to 32-bit usermode using + * sysret. Used to return to 32-on-64 compat processes. + * Other usermode register state, including %esp, must already + * be restored. + */ + void (*usergs_sysret32)(void); + + /* Normal iret. Jump to this with the standard iret stack + frame set up. */ + void (*iret)(void); + + void (*swapgs)(void); + + void (*start_context_switch)(struct task_struct *prev); + void (*end_context_switch)(struct task_struct *next); +}; + +struct pv_irq_ops { + void (*init_IRQ)(void); + + /* + * Get/set interrupt state. save_fl and restore_fl are only + * expected to use X86_EFLAGS_IF; all other bits + * returned from save_fl are undefined, and may be ignored by + * restore_fl. + * + * NOTE: These functions callers expect the callee to preserve + * more registers than the standard C calling convention. + */ + struct paravirt_callee_save save_fl; + struct paravirt_callee_save restore_fl; + struct paravirt_callee_save irq_disable; + struct paravirt_callee_save irq_enable; + + void (*safe_halt)(void); + void (*halt)(void); + +#ifdef CONFIG_X86_64 + void (*adjust_exception_frame)(void); +#endif +}; + +struct pv_apic_ops { +#ifdef CONFIG_X86_LOCAL_APIC + void (*setup_boot_clock)(void); + void (*setup_secondary_clock)(void); + + void (*startup_ipi_hook)(int phys_apicid, + unsigned long start_eip, + unsigned long start_esp); +#endif +}; + +struct pv_mmu_ops { + /* + * Called before/after init_mm pagetable setup. setup_start + * may reset %cr3, and may pre-install parts of the pagetable; + * pagetable setup is expected to preserve any existing + * mapping. + */ + void (*pagetable_setup_start)(pgd_t *pgd_base); + void (*pagetable_setup_done)(pgd_t *pgd_base); + + unsigned long (*read_cr2)(void); + void (*write_cr2)(unsigned long); + + unsigned long (*read_cr3)(void); + void (*write_cr3)(unsigned long); + + /* + * Hooks for intercepting the creation/use/destruction of an + * mm_struct. + */ + void (*activate_mm)(struct mm_struct *prev, + struct mm_struct *next); + void (*dup_mmap)(struct mm_struct *oldmm, + struct mm_struct *mm); + void (*exit_mmap)(struct mm_struct *mm); + + + /* TLB operations */ + void (*flush_tlb_user)(void); + void (*flush_tlb_kernel)(void); + void (*flush_tlb_single)(unsigned long addr); + void (*flush_tlb_others)(const struct cpumask *cpus, + struct mm_struct *mm, + unsigned long va); + + /* Hooks for allocating and freeing a pagetable top-level */ + int (*pgd_alloc)(struct mm_struct *mm); + void (*pgd_free)(struct mm_struct *mm, pgd_t *pgd); + + /* + * Hooks for allocating/releasing pagetable pages when they're + * attached to a pagetable + */ + void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn); + void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn); + void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count); + void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn); + void (*release_pte)(unsigned long pfn); + void (*release_pmd)(unsigned long pfn); + void (*release_pud)(unsigned long pfn); + + /* Pagetable manipulation functions */ + void (*set_pte)(pte_t *ptep, pte_t pteval); + void (*set_pte_at)(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pteval); + void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval); + void (*pte_update)(struct mm_struct *mm, unsigned long addr, + pte_t *ptep); + void (*pte_update_defer)(struct mm_struct *mm, + unsigned long addr, pte_t *ptep); + + pte_t (*ptep_modify_prot_start)(struct mm_struct *mm, unsigned long addr, + pte_t *ptep); + void (*ptep_modify_prot_commit)(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte); + + struct paravirt_callee_save pte_val; + struct paravirt_callee_save make_pte; + + struct paravirt_callee_save pgd_val; + struct paravirt_callee_save make_pgd; + +#if PAGETABLE_LEVELS >= 3 +#ifdef CONFIG_X86_PAE + void (*set_pte_atomic)(pte_t *ptep, pte_t pteval); + void (*pte_clear)(struct mm_struct *mm, unsigned long addr, + pte_t *ptep); + void (*pmd_clear)(pmd_t *pmdp); + +#endif /* CONFIG_X86_PAE */ + + void (*set_pud)(pud_t *pudp, pud_t pudval); + + struct paravirt_callee_save pmd_val; + struct paravirt_callee_save make_pmd; + +#if PAGETABLE_LEVELS == 4 + struct paravirt_callee_save pud_val; + struct paravirt_callee_save make_pud; + + void (*set_pgd)(pgd_t *pudp, pgd_t pgdval); +#endif /* PAGETABLE_LEVELS == 4 */ +#endif /* PAGETABLE_LEVELS >= 3 */ + +#ifdef CONFIG_HIGHPTE + void *(*kmap_atomic_pte)(struct page *page, enum km_type type); +#endif + + struct pv_lazy_ops lazy_mode; + + /* dom0 ops */ + + /* Sometimes the physical address is a pfn, and sometimes its + an mfn. We can tell which is which from the index. */ + void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx, + phys_addr_t phys, pgprot_t flags); +}; + +struct raw_spinlock; +struct pv_lock_ops { + int (*spin_is_locked)(struct raw_spinlock *lock); + int (*spin_is_contended)(struct raw_spinlock *lock); + void (*spin_lock)(struct raw_spinlock *lock); + void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); + int (*spin_trylock)(struct raw_spinlock *lock); + void (*spin_unlock)(struct raw_spinlock *lock); +}; + +/* This contains all the paravirt structures: we get a convenient + * number for each function using the offset which we use to indicate + * what to patch. */ +struct paravirt_patch_template { + struct pv_init_ops pv_init_ops; + struct pv_time_ops pv_time_ops; + struct pv_cpu_ops pv_cpu_ops; + struct pv_irq_ops pv_irq_ops; + struct pv_apic_ops pv_apic_ops; + struct pv_mmu_ops pv_mmu_ops; + struct pv_lock_ops pv_lock_ops; +}; + +extern struct pv_info pv_info; +extern struct pv_init_ops pv_init_ops; +extern struct pv_time_ops pv_time_ops; +extern struct pv_cpu_ops pv_cpu_ops; +extern struct pv_irq_ops pv_irq_ops; +extern struct pv_apic_ops pv_apic_ops; +extern struct pv_mmu_ops pv_mmu_ops; +extern struct pv_lock_ops pv_lock_ops; + +#define PARAVIRT_PATCH(x) \ + (offsetof(struct paravirt_patch_template, x) / sizeof(void *)) + +#define paravirt_type(op) \ + [paravirt_typenum] "i" (PARAVIRT_PATCH(op)), \ + [paravirt_opptr] "i" (&(op)) +#define paravirt_clobber(clobber) \ + [paravirt_clobber] "i" (clobber) + +/* + * Generate some code, and mark it as patchable by the + * apply_paravirt() alternate instruction patcher. + */ +#define _paravirt_alt(insn_string, type, clobber) \ + "771:\n\t" insn_string "\n" "772:\n" \ + ".pushsection .parainstructions,\"a\"\n" \ + _ASM_ALIGN "\n" \ + _ASM_PTR " 771b\n" \ + " .byte " type "\n" \ + " .byte 772b-771b\n" \ + " .short " clobber "\n" \ + ".popsection\n" + +/* Generate patchable code, with the default asm parameters. */ +#define paravirt_alt(insn_string) \ + _paravirt_alt(insn_string, "%c[paravirt_typenum]", "%c[paravirt_clobber]") + +/* Simple instruction patching code. */ +#define DEF_NATIVE(ops, name, code) \ + extern const char start_##ops##_##name[], end_##ops##_##name[]; \ + asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":") + +unsigned paravirt_patch_nop(void); +unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len); +unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len); +unsigned paravirt_patch_ignore(unsigned len); +unsigned paravirt_patch_call(void *insnbuf, + const void *target, u16 tgt_clobbers, + unsigned long addr, u16 site_clobbers, + unsigned len); +unsigned paravirt_patch_jmp(void *insnbuf, const void *target, + unsigned long addr, unsigned len); +unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, + unsigned long addr, unsigned len); + +unsigned paravirt_patch_insns(void *insnbuf, unsigned len, + const char *start, const char *end); + +unsigned native_patch(u8 type, u16 clobbers, void *ibuf, + unsigned long addr, unsigned len); + +int paravirt_disable_iospace(void); + +/* + * This generates an indirect call based on the operation type number. + * The type number, computed in PARAVIRT_PATCH, is derived from the + * offset into the paravirt_patch_template structure, and can therefore be + * freely converted back into a structure offset. + */ +#define PARAVIRT_CALL "call *%c[paravirt_opptr];" + +/* + * These macros are intended to wrap calls through one of the paravirt + * ops structs, so that they can be later identified and patched at + * runtime. + * + * Normally, a call to a pv_op function is a simple indirect call: + * (pv_op_struct.operations)(args...). + * + * Unfortunately, this is a relatively slow operation for modern CPUs, + * because it cannot necessarily determine what the destination + * address is. In this case, the address is a runtime constant, so at + * the very least we can patch the call to e a simple direct call, or + * ideally, patch an inline implementation into the callsite. (Direct + * calls are essentially free, because the call and return addresses + * are completely predictable.) + * + * For i386, these macros rely on the standard gcc "regparm(3)" calling + * convention, in which the first three arguments are placed in %eax, + * %edx, %ecx (in that order), and the remaining arguments are placed + * on the stack. All caller-save registers (eax,edx,ecx) are expected + * to be modified (either clobbered or used for return values). + * X86_64, on the other hand, already specifies a register-based calling + * conventions, returning at %rax, with parameteres going on %rdi, %rsi, + * %rdx, and %rcx. Note that for this reason, x86_64 does not need any + * special handling for dealing with 4 arguments, unlike i386. + * However, x86_64 also have to clobber all caller saved registers, which + * unfortunately, are quite a bit (r8 - r11) + * + * The call instruction itself is marked by placing its start address + * and size into the .parainstructions section, so that + * apply_paravirt() in arch/i386/kernel/alternative.c can do the + * appropriate patching under the control of the backend pv_init_ops + * implementation. + * + * Unfortunately there's no way to get gcc to generate the args setup + * for the call, and then allow the call itself to be generated by an + * inline asm. Because of this, we must do the complete arg setup and + * return value handling from within these macros. This is fairly + * cumbersome. + * + * There are 5 sets of PVOP_* macros for dealing with 0-4 arguments. + * It could be extended to more arguments, but there would be little + * to be gained from that. For each number of arguments, there are + * the two VCALL and CALL variants for void and non-void functions. + * + * When there is a return value, the invoker of the macro must specify + * the return type. The macro then uses sizeof() on that type to + * determine whether its a 32 or 64 bit value, and places the return + * in the right register(s) (just %eax for 32-bit, and %edx:%eax for + * 64-bit). For x86_64 machines, it just returns at %rax regardless of + * the return value size. + * + * 64-bit arguments are passed as a pair of adjacent 32-bit arguments + * i386 also passes 64-bit arguments as a pair of adjacent 32-bit arguments + * in low,high order + * + * Small structures are passed and returned in registers. The macro + * calling convention can't directly deal with this, so the wrapper + * functions must do this. + * + * These PVOP_* macros are only defined within this header. This + * means that all uses must be wrapped in inline functions. This also + * makes sure the incoming and outgoing types are always correct. + */ +#ifdef CONFIG_X86_32 +#define PVOP_VCALL_ARGS \ + unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx +#define PVOP_CALL_ARGS PVOP_VCALL_ARGS + +#define PVOP_CALL_ARG1(x) "a" ((unsigned long)(x)) +#define PVOP_CALL_ARG2(x) "d" ((unsigned long)(x)) +#define PVOP_CALL_ARG3(x) "c" ((unsigned long)(x)) + +#define PVOP_VCALL_CLOBBERS "=a" (__eax), "=d" (__edx), \ + "=c" (__ecx) +#define PVOP_CALL_CLOBBERS PVOP_VCALL_CLOBBERS + +#define PVOP_VCALLEE_CLOBBERS "=a" (__eax), "=d" (__edx) +#define PVOP_CALLEE_CLOBBERS PVOP_VCALLEE_CLOBBERS + +#define EXTRA_CLOBBERS +#define VEXTRA_CLOBBERS +#else /* CONFIG_X86_64 */ +#define PVOP_VCALL_ARGS \ + unsigned long __edi = __edi, __esi = __esi, \ + __edx = __edx, __ecx = __ecx +#define PVOP_CALL_ARGS PVOP_VCALL_ARGS, __eax + +#define PVOP_CALL_ARG1(x) "D" ((unsigned long)(x)) +#define PVOP_CALL_ARG2(x) "S" ((unsigned long)(x)) +#define PVOP_CALL_ARG3(x) "d" ((unsigned long)(x)) +#define PVOP_CALL_ARG4(x) "c" ((unsigned long)(x)) + +#define PVOP_VCALL_CLOBBERS "=D" (__edi), \ + "=S" (__esi), "=d" (__edx), \ + "=c" (__ecx) +#define PVOP_CALL_CLOBBERS PVOP_VCALL_CLOBBERS, "=a" (__eax) + +#define PVOP_VCALLEE_CLOBBERS "=a" (__eax) +#define PVOP_CALLEE_CLOBBERS PVOP_VCALLEE_CLOBBERS + +#define EXTRA_CLOBBERS , "r8", "r9", "r10", "r11" +#define VEXTRA_CLOBBERS , "rax", "r8", "r9", "r10", "r11" +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_PARAVIRT_DEBUG +#define PVOP_TEST_NULL(op) BUG_ON(op == NULL) +#else +#define PVOP_TEST_NULL(op) ((void)op) +#endif + +#define ____PVOP_CALL(rettype, op, clbr, call_clbr, extra_clbr, \ + pre, post, ...) \ + ({ \ + rettype __ret; \ + PVOP_CALL_ARGS; \ + PVOP_TEST_NULL(op); \ + /* This is 32-bit specific, but is okay in 64-bit */ \ + /* since this condition will never hold */ \ + if (sizeof(rettype) > sizeof(unsigned long)) { \ + asm volatile(pre \ + paravirt_alt(PARAVIRT_CALL) \ + post \ + : call_clbr \ + : paravirt_type(op), \ + paravirt_clobber(clbr), \ + ##__VA_ARGS__ \ + : "memory", "cc" extra_clbr); \ + __ret = (rettype)((((u64)__edx) << 32) | __eax); \ + } else { \ + asm volatile(pre \ + paravirt_alt(PARAVIRT_CALL) \ + post \ + : call_clbr \ + : paravirt_type(op), \ + paravirt_clobber(clbr), \ + ##__VA_ARGS__ \ + : "memory", "cc" extra_clbr); \ + __ret = (rettype)__eax; \ + } \ + __ret; \ + }) + +#define __PVOP_CALL(rettype, op, pre, post, ...) \ + ____PVOP_CALL(rettype, op, CLBR_ANY, PVOP_CALL_CLOBBERS, \ + EXTRA_CLOBBERS, pre, post, ##__VA_ARGS__) + +#define __PVOP_CALLEESAVE(rettype, op, pre, post, ...) \ + ____PVOP_CALL(rettype, op.func, CLBR_RET_REG, \ + PVOP_CALLEE_CLOBBERS, , \ + pre, post, ##__VA_ARGS__) + + +#define ____PVOP_VCALL(op, clbr, call_clbr, extra_clbr, pre, post, ...) \ + ({ \ + PVOP_VCALL_ARGS; \ + PVOP_TEST_NULL(op); \ + asm volatile(pre \ + paravirt_alt(PARAVIRT_CALL) \ + post \ + : call_clbr \ + : paravirt_type(op), \ + paravirt_clobber(clbr), \ + ##__VA_ARGS__ \ + : "memory", "cc" extra_clbr); \ + }) + +#define __PVOP_VCALL(op, pre, post, ...) \ + ____PVOP_VCALL(op, CLBR_ANY, PVOP_VCALL_CLOBBERS, \ + VEXTRA_CLOBBERS, \ + pre, post, ##__VA_ARGS__) + +#define __PVOP_VCALLEESAVE(rettype, op, pre, post, ...) \ + ____PVOP_CALL(rettype, op.func, CLBR_RET_REG, \ + PVOP_VCALLEE_CLOBBERS, , \ + pre, post, ##__VA_ARGS__) + + + +#define PVOP_CALL0(rettype, op) \ + __PVOP_CALL(rettype, op, "", "") +#define PVOP_VCALL0(op) \ + __PVOP_VCALL(op, "", "") + +#define PVOP_CALLEE0(rettype, op) \ + __PVOP_CALLEESAVE(rettype, op, "", "") +#define PVOP_VCALLEE0(op) \ + __PVOP_VCALLEESAVE(op, "", "") + + +#define PVOP_CALL1(rettype, op, arg1) \ + __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) +#define PVOP_VCALL1(op, arg1) \ + __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1)) + +#define PVOP_CALLEE1(rettype, op, arg1) \ + __PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) +#define PVOP_VCALLEE1(op, arg1) \ + __PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1)) + + +#define PVOP_CALL2(rettype, op, arg1, arg2) \ + __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2)) +#define PVOP_VCALL2(op, arg1, arg2) \ + __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2)) + +#define PVOP_CALLEE2(rettype, op, arg1, arg2) \ + __PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2)) +#define PVOP_VCALLEE2(op, arg1, arg2) \ + __PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2)) + + +#define PVOP_CALL3(rettype, op, arg1, arg2, arg3) \ + __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3)) +#define PVOP_VCALL3(op, arg1, arg2, arg3) \ + __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1), \ + PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3)) + +/* This is the only difference in x86_64. We can make it much simpler */ +#ifdef CONFIG_X86_32 +#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4) \ + __PVOP_CALL(rettype, op, \ + "push %[_arg4];", "lea 4(%%esp),%%esp;", \ + PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ + PVOP_CALL_ARG3(arg3), [_arg4] "mr" ((u32)(arg4))) +#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4) \ + __PVOP_VCALL(op, \ + "push %[_arg4];", "lea 4(%%esp),%%esp;", \ + "0" ((u32)(arg1)), "1" ((u32)(arg2)), \ + "2" ((u32)(arg3)), [_arg4] "mr" ((u32)(arg4))) +#else +#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4) \ + __PVOP_CALL(rettype, op, "", "", \ + PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ + PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4)) +#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4) \ + __PVOP_VCALL(op, "", "", \ + PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ + PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4)) +#endif + +/* Lazy mode for batching updates / context switch */ +enum paravirt_lazy_mode { + PARAVIRT_LAZY_NONE, + PARAVIRT_LAZY_MMU, + PARAVIRT_LAZY_CPU, +}; + +enum paravirt_lazy_mode paravirt_get_lazy_mode(void); +void paravirt_start_context_switch(struct task_struct *prev); +void paravirt_end_context_switch(struct task_struct *next); + +void paravirt_enter_lazy_mmu(void); +void paravirt_leave_lazy_mmu(void); + +void _paravirt_nop(void); +u32 _paravirt_ident_32(u32); +u64 _paravirt_ident_64(u64); + +#define paravirt_nop ((void *)_paravirt_nop) + +/* These all sit in the .parainstructions section to tell us what to patch. */ +struct paravirt_patch_site { + u8 *instr; /* original instructions */ + u8 instrtype; /* type of this instruction */ + u8 len; /* length of original instruction */ + u16 clobbers; /* what registers you may clobber */ +}; + +extern struct paravirt_patch_site __parainstructions[], + __parainstructions_end[]; + +#endif /* __ASSEMBLY__ */ + +#endif /* _ASM_X86_PARAVIRT_TYPES_H */ diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index 103f1ddb0d85..04eacefcfd26 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -49,7 +49,7 @@ #define __percpu_arg(x) "%%"__stringify(__percpu_seg)":%P" #x #define __my_cpu_offset percpu_read(this_cpu_off) #else -#define __percpu_arg(x) "%" #x +#define __percpu_arg(x) "%P" #x #endif /* @@ -104,36 +104,48 @@ do { \ } \ } while (0) -#define percpu_from_op(op, var) \ +#define percpu_from_op(op, var, constraint) \ ({ \ typeof(var) ret__; \ switch (sizeof(var)) { \ case 1: \ asm(op "b "__percpu_arg(1)",%0" \ : "=q" (ret__) \ - : "m" (var)); \ + : constraint); \ break; \ case 2: \ asm(op "w "__percpu_arg(1)",%0" \ : "=r" (ret__) \ - : "m" (var)); \ + : constraint); \ break; \ case 4: \ asm(op "l "__percpu_arg(1)",%0" \ : "=r" (ret__) \ - : "m" (var)); \ + : constraint); \ break; \ case 8: \ asm(op "q "__percpu_arg(1)",%0" \ : "=r" (ret__) \ - : "m" (var)); \ + : constraint); \ break; \ default: __bad_percpu_size(); \ } \ ret__; \ }) -#define percpu_read(var) percpu_from_op("mov", per_cpu__##var) +/* + * percpu_read() makes gcc load the percpu variable every time it is + * accessed while percpu_read_stable() allows the value to be cached. + * percpu_read_stable() is more efficient and can be used if its value + * is guaranteed to be valid across cpus. The current users include + * get_current() and get_thread_info() both of which are actually + * per-thread variables implemented as per-cpu variables and thus + * stable for the duration of the respective task. + */ +#define percpu_read(var) percpu_from_op("mov", per_cpu__##var, \ + "m" (per_cpu__##var)) +#define percpu_read_stable(var) percpu_from_op("mov", per_cpu__##var, \ + "p" (&per_cpu__##var)) #define percpu_write(var, val) percpu_to_op("mov", per_cpu__##var, val) #define percpu_add(var, val) percpu_to_op("add", per_cpu__##var, val) #define percpu_sub(var, val) percpu_to_op("sub", per_cpu__##var, val) diff --git a/arch/x86/include/asm/perf_counter.h b/arch/x86/include/asm/perf_counter.h index fa64e401589d..e7b7c938ae27 100644 --- a/arch/x86/include/asm/perf_counter.h +++ b/arch/x86/include/asm/perf_counter.h @@ -84,6 +84,16 @@ union cpuid10_edx { #define MSR_ARCH_PERFMON_FIXED_CTR2 0x30b #define X86_PMC_IDX_FIXED_BUS_CYCLES (X86_PMC_IDX_FIXED + 2) +/* + * We model BTS tracing as another fixed-mode PMC. + * + * We choose a value in the middle of the fixed counter range, since lower + * values are used by actual fixed counters and higher values are used + * to indicate other overflow conditions in the PERF_GLOBAL_STATUS msr. + */ +#define X86_PMC_IDX_FIXED_BTS (X86_PMC_IDX_FIXED + 16) + + #ifdef CONFIG_PERF_COUNTERS extern void init_hw_perf_counters(void); extern void perf_counters_lapic_init(void); diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 3cc06e3fceb8..9de8729c1c8f 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -134,6 +134,11 @@ static inline unsigned long pte_pfn(pte_t pte) return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT; } +static inline unsigned long pmd_pfn(pmd_t pmd) +{ + return (pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT; +} + #define pte_page(pte) pfn_to_page(pte_pfn(pte)) static inline int pmd_large(pmd_t pte) @@ -351,7 +356,7 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd) * this macro returns the index of the entry in the pmd page which would * control the given virtual address */ -static inline unsigned pmd_index(unsigned long address) +static inline unsigned long pmd_index(unsigned long address) { return (address >> PMD_SHIFT) & (PTRS_PER_PMD - 1); } @@ -371,7 +376,7 @@ static inline unsigned pmd_index(unsigned long address) * this function returns the index of the entry in the pte page which would * control the given virtual address */ -static inline unsigned pte_index(unsigned long address) +static inline unsigned long pte_index(unsigned long address) { return (address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1); } @@ -422,11 +427,6 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address) return (pmd_t *)pud_page_vaddr(*pud) + pmd_index(address); } -static inline unsigned long pmd_pfn(pmd_t pmd) -{ - return (pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT; -} - static inline int pud_large(pud_t pud) { return (pud_val(pud) & (_PAGE_PSE | _PAGE_PRESENT)) == @@ -462,7 +462,7 @@ static inline unsigned long pgd_page_vaddr(pgd_t pgd) #define pgd_page(pgd) pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT) /* to find an entry in a page-table-directory. */ -static inline unsigned pud_index(unsigned long address) +static inline unsigned long pud_index(unsigned long address) { return (address >> PUD_SHIFT) & (PTRS_PER_PUD - 1); } diff --git a/arch/x86/include/asm/scatterlist.h b/arch/x86/include/asm/scatterlist.h index 263d397d2eef..75af592677ec 100644 --- a/arch/x86/include/asm/scatterlist.h +++ b/arch/x86/include/asm/scatterlist.h @@ -1,33 +1,8 @@ #ifndef _ASM_X86_SCATTERLIST_H #define _ASM_X86_SCATTERLIST_H -#include <asm/types.h> - -struct scatterlist { -#ifdef CONFIG_DEBUG_SG - unsigned long sg_magic; -#endif - unsigned long page_link; - unsigned int offset; - unsigned int length; - dma_addr_t dma_address; - unsigned int dma_length; -}; - -#define ARCH_HAS_SG_CHAIN #define ISA_DMA_THRESHOLD (0x00ffffff) -/* - * These macros should be used after a pci_map_sg call has been done - * to get bus addresses of each of the SG entries and their lengths. - * You should only work with the number of sg entries pci_map_sg - * returns. - */ -#define sg_dma_address(sg) ((sg)->dma_address) -#ifdef CONFIG_X86_32 -# define sg_dma_len(sg) ((sg)->length) -#else -# define sg_dma_len(sg) ((sg)->dma_length) -#endif +#include <asm-generic/scatterlist.h> #endif /* _ASM_X86_SCATTERLIST_H */ diff --git a/arch/x86/include/asm/shmbuf.h b/arch/x86/include/asm/shmbuf.h index b51413b74971..83c05fc2de38 100644 --- a/arch/x86/include/asm/shmbuf.h +++ b/arch/x86/include/asm/shmbuf.h @@ -1,51 +1 @@ -#ifndef _ASM_X86_SHMBUF_H -#define _ASM_X86_SHMBUF_H - -/* - * The shmid64_ds structure for x86 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space on 32 bit is left for: - * - 64-bit time_t to solve y2038 problem - * - 2 miscellaneous 32-bit values - * - * Pad space on 64 bit is left for: - * - 2 miscellaneous 64-bit values - */ - -struct shmid64_ds { - struct ipc64_perm shm_perm; /* operation perms */ - size_t shm_segsz; /* size of segment (bytes) */ - __kernel_time_t shm_atime; /* last attach time */ -#ifdef __i386__ - unsigned long __unused1; -#endif - __kernel_time_t shm_dtime; /* last detach time */ -#ifdef __i386__ - unsigned long __unused2; -#endif - __kernel_time_t shm_ctime; /* last change time */ -#ifdef __i386__ - unsigned long __unused3; -#endif - __kernel_pid_t shm_cpid; /* pid of creator */ - __kernel_pid_t shm_lpid; /* pid of last operator */ - unsigned long shm_nattch; /* no. of current attaches */ - unsigned long __unused4; - unsigned long __unused5; -}; - -struct shminfo64 { - unsigned long shmmax; - unsigned long shmmin; - unsigned long shmmni; - unsigned long shmseg; - unsigned long shmall; - unsigned long __unused1; - unsigned long __unused2; - unsigned long __unused3; - unsigned long __unused4; -}; - -#endif /* _ASM_X86_SHMBUF_H */ +#include <asm-generic/shmbuf.h> diff --git a/arch/x86/include/asm/socket.h b/arch/x86/include/asm/socket.h index b2a8c74f2d06..6b71384b9d8b 100644 --- a/arch/x86/include/asm/socket.h +++ b/arch/x86/include/asm/socket.h @@ -1,63 +1 @@ -#ifndef _ASM_X86_SOCKET_H -#define _ASM_X86_SOCKET_H - -#include <asm/sockios.h> - -/* For setsockopt(2) */ -#define SOL_SOCKET 1 - -#define SO_DEBUG 1 -#define SO_REUSEADDR 2 -#define SO_TYPE 3 -#define SO_ERROR 4 -#define SO_DONTROUTE 5 -#define SO_BROADCAST 6 -#define SO_SNDBUF 7 -#define SO_RCVBUF 8 -#define SO_SNDBUFFORCE 32 -#define SO_RCVBUFFORCE 33 -#define SO_KEEPALIVE 9 -#define SO_OOBINLINE 10 -#define SO_NO_CHECK 11 -#define SO_PRIORITY 12 -#define SO_LINGER 13 -#define SO_BSDCOMPAT 14 -/* To add :#define SO_REUSEPORT 15 */ -#define SO_PASSCRED 16 -#define SO_PEERCRED 17 -#define SO_RCVLOWAT 18 -#define SO_SNDLOWAT 19 -#define SO_RCVTIMEO 20 -#define SO_SNDTIMEO 21 - -/* Security levels - as per NRL IPv6 - don't actually do anything */ -#define SO_SECURITY_AUTHENTICATION 22 -#define SO_SECURITY_ENCRYPTION_TRANSPORT 23 -#define SO_SECURITY_ENCRYPTION_NETWORK 24 - -#define SO_BINDTODEVICE 25 - -/* Socket filtering */ -#define SO_ATTACH_FILTER 26 -#define SO_DETACH_FILTER 27 - -#define SO_PEERNAME 28 -#define SO_TIMESTAMP 29 -#define SCM_TIMESTAMP SO_TIMESTAMP - -#define SO_ACCEPTCONN 30 - -#define SO_PEERSEC 31 -#define SO_PASSSEC 34 -#define SO_TIMESTAMPNS 35 -#define SCM_TIMESTAMPNS SO_TIMESTAMPNS - -#define SO_MARK 36 - -#define SO_TIMESTAMPING 37 -#define SCM_TIMESTAMPING SO_TIMESTAMPING - -#define SO_PROTOCOL 38 -#define SO_DOMAIN 39 - -#endif /* _ASM_X86_SOCKET_H */ +#include <asm-generic/socket.h> diff --git a/arch/x86/include/asm/sockios.h b/arch/x86/include/asm/sockios.h index 49cc72b5d3c9..def6d4746ee7 100644 --- a/arch/x86/include/asm/sockios.h +++ b/arch/x86/include/asm/sockios.h @@ -1,13 +1 @@ -#ifndef _ASM_X86_SOCKIOS_H -#define _ASM_X86_SOCKIOS_H - -/* Socket-level I/O control calls. */ -#define FIOSETOWN 0x8901 -#define SIOCSPGRP 0x8902 -#define FIOGETOWN 0x8903 -#define SIOCGPGRP 0x8904 -#define SIOCATMARK 0x8905 -#define SIOCGSTAMP 0x8906 /* Get stamp (timeval) */ -#define SIOCGSTAMPNS 0x8907 /* Get stamp (timespec) */ - -#endif /* _ASM_X86_SOCKIOS_H */ +#include <asm-generic/sockios.h> diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h index c2d742c6e15f..44efdff3975d 100644 --- a/arch/x86/include/asm/stackprotector.h +++ b/arch/x86/include/asm/stackprotector.h @@ -48,7 +48,7 @@ * head_32 for boot CPU and setup_per_cpu_areas() for others. */ #define GDT_STACK_CANARY_INIT \ - [GDT_ENTRY_STACK_CANARY] = { { { 0x00000018, 0x00409000 } } }, + [GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x18), /* * Initialize the stackprotector canary value. @@ -90,9 +90,7 @@ static inline void setup_stack_canary_segment(int cpu) struct desc_struct desc; desc = gdt_table[GDT_ENTRY_STACK_CANARY]; - desc.base0 = canary & 0xffff; - desc.base1 = (canary >> 16) & 0xff; - desc.base2 = (canary >> 24) & 0xff; + set_desc_base(&desc, canary); write_gdt_entry(gdt_table, GDT_ENTRY_STACK_CANARY, &desc, DESCTYPE_S); #endif } diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h index 643c59b4bc6e..75c49c782e20 100644 --- a/arch/x86/include/asm/system.h +++ b/arch/x86/include/asm/system.h @@ -150,33 +150,6 @@ do { \ #endif #ifdef __KERNEL__ -#define _set_base(addr, base) do { unsigned long __pr; \ -__asm__ __volatile__ ("movw %%dx,%1\n\t" \ - "rorl $16,%%edx\n\t" \ - "movb %%dl,%2\n\t" \ - "movb %%dh,%3" \ - :"=&d" (__pr) \ - :"m" (*((addr)+2)), \ - "m" (*((addr)+4)), \ - "m" (*((addr)+7)), \ - "0" (base) \ - ); } while (0) - -#define _set_limit(addr, limit) do { unsigned long __lr; \ -__asm__ __volatile__ ("movw %%dx,%1\n\t" \ - "rorl $16,%%edx\n\t" \ - "movb %2,%%dh\n\t" \ - "andb $0xf0,%%dh\n\t" \ - "orb %%dh,%%dl\n\t" \ - "movb %%dl,%2" \ - :"=&d" (__lr) \ - :"m" (*(addr)), \ - "m" (*((addr)+6)), \ - "0" (limit) \ - ); } while (0) - -#define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base)) -#define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1)) extern void native_load_gs_index(unsigned); diff --git a/arch/x86/include/asm/termbits.h b/arch/x86/include/asm/termbits.h index af1b70ea440f..3935b106de79 100644 --- a/arch/x86/include/asm/termbits.h +++ b/arch/x86/include/asm/termbits.h @@ -1,198 +1 @@ -#ifndef _ASM_X86_TERMBITS_H -#define _ASM_X86_TERMBITS_H - -#include <linux/posix_types.h> - -typedef unsigned char cc_t; -typedef unsigned int speed_t; -typedef unsigned int tcflag_t; - -#define NCCS 19 -struct termios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ -}; - -struct termios2 { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -struct ktermios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -/* c_cc characters */ -#define VINTR 0 -#define VQUIT 1 -#define VERASE 2 -#define VKILL 3 -#define VEOF 4 -#define VTIME 5 -#define VMIN 6 -#define VSWTC 7 -#define VSTART 8 -#define VSTOP 9 -#define VSUSP 10 -#define VEOL 11 -#define VREPRINT 12 -#define VDISCARD 13 -#define VWERASE 14 -#define VLNEXT 15 -#define VEOL2 16 - -/* c_iflag bits */ -#define IGNBRK 0000001 -#define BRKINT 0000002 -#define IGNPAR 0000004 -#define PARMRK 0000010 -#define INPCK 0000020 -#define ISTRIP 0000040 -#define INLCR 0000100 -#define IGNCR 0000200 -#define ICRNL 0000400 -#define IUCLC 0001000 -#define IXON 0002000 -#define IXANY 0004000 -#define IXOFF 0010000 -#define IMAXBEL 0020000 -#define IUTF8 0040000 - -/* c_oflag bits */ -#define OPOST 0000001 -#define OLCUC 0000002 -#define ONLCR 0000004 -#define OCRNL 0000010 -#define ONOCR 0000020 -#define ONLRET 0000040 -#define OFILL 0000100 -#define OFDEL 0000200 -#define NLDLY 0000400 -#define NL0 0000000 -#define NL1 0000400 -#define CRDLY 0003000 -#define CR0 0000000 -#define CR1 0001000 -#define CR2 0002000 -#define CR3 0003000 -#define TABDLY 0014000 -#define TAB0 0000000 -#define TAB1 0004000 -#define TAB2 0010000 -#define TAB3 0014000 -#define XTABS 0014000 -#define BSDLY 0020000 -#define BS0 0000000 -#define BS1 0020000 -#define VTDLY 0040000 -#define VT0 0000000 -#define VT1 0040000 -#define FFDLY 0100000 -#define FF0 0000000 -#define FF1 0100000 - -/* c_cflag bit meaning */ -#define CBAUD 0010017 -#define B0 0000000 /* hang up */ -#define B50 0000001 -#define B75 0000002 -#define B110 0000003 -#define B134 0000004 -#define B150 0000005 -#define B200 0000006 -#define B300 0000007 -#define B600 0000010 -#define B1200 0000011 -#define B1800 0000012 -#define B2400 0000013 -#define B4800 0000014 -#define B9600 0000015 -#define B19200 0000016 -#define B38400 0000017 -#define EXTA B19200 -#define EXTB B38400 -#define CSIZE 0000060 -#define CS5 0000000 -#define CS6 0000020 -#define CS7 0000040 -#define CS8 0000060 -#define CSTOPB 0000100 -#define CREAD 0000200 -#define PARENB 0000400 -#define PARODD 0001000 -#define HUPCL 0002000 -#define CLOCAL 0004000 -#define CBAUDEX 0010000 -#define BOTHER 0010000 /* non standard rate */ -#define B57600 0010001 -#define B115200 0010002 -#define B230400 0010003 -#define B460800 0010004 -#define B500000 0010005 -#define B576000 0010006 -#define B921600 0010007 -#define B1000000 0010010 -#define B1152000 0010011 -#define B1500000 0010012 -#define B2000000 0010013 -#define B2500000 0010014 -#define B3000000 0010015 -#define B3500000 0010016 -#define B4000000 0010017 -#define CIBAUD 002003600000 /* input baud rate */ -#define CMSPAR 010000000000 /* mark or space (stick) parity */ -#define CRTSCTS 020000000000 /* flow control */ - -#define IBSHIFT 16 /* Shift from CBAUD to CIBAUD */ - -/* c_lflag bits */ -#define ISIG 0000001 -#define ICANON 0000002 -#define XCASE 0000004 -#define ECHO 0000010 -#define ECHOE 0000020 -#define ECHOK 0000040 -#define ECHONL 0000100 -#define NOFLSH 0000200 -#define TOSTOP 0000400 -#define ECHOCTL 0001000 -#define ECHOPRT 0002000 -#define ECHOKE 0004000 -#define FLUSHO 0010000 -#define PENDIN 0040000 -#define IEXTEN 0100000 - -/* tcflow() and TCXONC use these */ -#define TCOOFF 0 -#define TCOON 1 -#define TCIOFF 2 -#define TCION 3 - -/* tcflush() and TCFLSH use these */ -#define TCIFLUSH 0 -#define TCOFLUSH 1 -#define TCIOFLUSH 2 - -/* tcsetattr uses these */ -#define TCSANOW 0 -#define TCSADRAIN 1 -#define TCSAFLUSH 2 - -#endif /* _ASM_X86_TERMBITS_H */ +#include <asm-generic/termbits.h> diff --git a/arch/x86/include/asm/termios.h b/arch/x86/include/asm/termios.h index c4ee8056baca..280d78a9d966 100644 --- a/arch/x86/include/asm/termios.h +++ b/arch/x86/include/asm/termios.h @@ -1,114 +1 @@ -#ifndef _ASM_X86_TERMIOS_H -#define _ASM_X86_TERMIOS_H - -#include <asm/termbits.h> -#include <asm/ioctls.h> - -struct winsize { - unsigned short ws_row; - unsigned short ws_col; - unsigned short ws_xpixel; - unsigned short ws_ypixel; -}; - -#define NCC 8 -struct termio { - unsigned short c_iflag; /* input mode flags */ - unsigned short c_oflag; /* output mode flags */ - unsigned short c_cflag; /* control mode flags */ - unsigned short c_lflag; /* local mode flags */ - unsigned char c_line; /* line discipline */ - unsigned char c_cc[NCC]; /* control characters */ -}; - -/* modem lines */ -#define TIOCM_LE 0x001 -#define TIOCM_DTR 0x002 -#define TIOCM_RTS 0x004 -#define TIOCM_ST 0x008 -#define TIOCM_SR 0x010 -#define TIOCM_CTS 0x020 -#define TIOCM_CAR 0x040 -#define TIOCM_RNG 0x080 -#define TIOCM_DSR 0x100 -#define TIOCM_CD TIOCM_CAR -#define TIOCM_RI TIOCM_RNG -#define TIOCM_OUT1 0x2000 -#define TIOCM_OUT2 0x4000 -#define TIOCM_LOOP 0x8000 - -/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */ - -#ifdef __KERNEL__ - -#include <asm/uaccess.h> - -/* intr=^C quit=^\ erase=del kill=^U - eof=^D vtime=\0 vmin=\1 sxtc=\0 - start=^Q stop=^S susp=^Z eol=\0 - reprint=^R discard=^U werase=^W lnext=^V - eol2=\0 -*/ -#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0" - -/* - * Translate a "termio" structure into a "termios". Ugh. - */ -#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \ - unsigned short __tmp; \ - get_user(__tmp,&(termio)->x); \ - *(unsigned short *) &(termios)->x = __tmp; \ -} - -static inline int user_termio_to_kernel_termios(struct ktermios *termios, - struct termio __user *termio) -{ - SET_LOW_TERMIOS_BITS(termios, termio, c_iflag); - SET_LOW_TERMIOS_BITS(termios, termio, c_oflag); - SET_LOW_TERMIOS_BITS(termios, termio, c_cflag); - SET_LOW_TERMIOS_BITS(termios, termio, c_lflag); - get_user(termios->c_line, &termio->c_line); - return copy_from_user(termios->c_cc, termio->c_cc, NCC); -} - -/* - * Translate a "termios" structure into a "termio". Ugh. - */ -static inline int kernel_termios_to_user_termio(struct termio __user *termio, - struct ktermios *termios) -{ - put_user((termios)->c_iflag, &(termio)->c_iflag); - put_user((termios)->c_oflag, &(termio)->c_oflag); - put_user((termios)->c_cflag, &(termio)->c_cflag); - put_user((termios)->c_lflag, &(termio)->c_lflag); - put_user((termios)->c_line, &(termio)->c_line); - return copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); -} - -static inline int user_termios_to_kernel_termios(struct ktermios *k, - struct termios2 __user *u) -{ - return copy_from_user(k, u, sizeof(struct termios2)); -} - -static inline int kernel_termios_to_user_termios(struct termios2 __user *u, - struct ktermios *k) -{ - return copy_to_user(u, k, sizeof(struct termios2)); -} - -static inline int user_termios_to_kernel_termios_1(struct ktermios *k, - struct termios __user *u) -{ - return copy_from_user(k, u, sizeof(struct termios)); -} - -static inline int kernel_termios_to_user_termios_1(struct termios __user *u, - struct ktermios *k) -{ - return copy_to_user(u, k, sizeof(struct termios)); -} - -#endif /* __KERNEL__ */ - -#endif /* _ASM_X86_TERMIOS_H */ +#include <asm-generic/termios.h> diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index f26432ac1c57..ae0f88de09ab 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -218,7 +218,7 @@ DECLARE_PER_CPU(unsigned long, kernel_stack); static inline struct thread_info *current_thread_info(void) { struct thread_info *ti; - ti = (void *)(percpu_read(kernel_stack) + + ti = (void *)(percpu_read_stable(kernel_stack) + KERNEL_STACK_OFFSET - THREAD_SIZE); return ti; } diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index bfd74c032fca..4da91ad69e0d 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -81,9 +81,7 @@ extern int panic_on_unrecovered_nmi; void math_error(void __user *); void math_emulate(struct math_emu_info *); -#ifdef CONFIG_X86_32 -unsigned long patch_espfix_desc(unsigned long, unsigned long); -#else +#ifndef CONFIG_X86_32 asmlinkage void smp_thermal_interrupt(void); asmlinkage void mce_threshold_interrupt(void); #endif diff --git a/arch/x86/include/asm/types.h b/arch/x86/include/asm/types.h index 09b97745772f..df1da20f4534 100644 --- a/arch/x86/include/asm/types.h +++ b/arch/x86/include/asm/types.h @@ -1,19 +1,11 @@ #ifndef _ASM_X86_TYPES_H #define _ASM_X86_TYPES_H -#include <asm-generic/int-ll64.h> +#define dma_addr_t dma_addr_t -#ifndef __ASSEMBLY__ - -typedef unsigned short umode_t; +#include <asm-generic/types.h> -#endif /* __ASSEMBLY__ */ - -/* - * These aren't exported outside the kernel to avoid name space clashes - */ #ifdef __KERNEL__ - #ifndef __ASSEMBLY__ typedef u64 dma64_addr_t; diff --git a/arch/x86/include/asm/ucontext.h b/arch/x86/include/asm/ucontext.h index 87324cf439d9..b7c29c8017f2 100644 --- a/arch/x86/include/asm/ucontext.h +++ b/arch/x86/include/asm/ucontext.h @@ -7,12 +7,6 @@ * sigcontext struct (uc_mcontext). */ -struct ucontext { - unsigned long uc_flags; - struct ucontext *uc_link; - stack_t uc_stack; - struct sigcontext uc_mcontext; - sigset_t uc_sigmask; /* mask last for extensibility */ -}; +#include <asm-generic/ucontext.h> #endif /* _ASM_X86_UCONTEXT_H */ diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index a7f9e89741c2..d29eba3c04d2 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -1157,9 +1157,8 @@ static int __init acpi_parse_madt_ioapic_entries(void) * If MPS is present, it will handle them, * otherwise the system will stay in PIC mode */ - if (acpi_disabled || acpi_noirq) { + if (acpi_disabled || acpi_noirq) return -ENODEV; - } if (!cpu_has_apic) return -ENODEV; diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 0a1c2830ec66..de039fcdd053 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -49,6 +49,7 @@ #include <asm/mtrr.h> #include <asm/smp.h> #include <asm/mce.h> +#include <asm/kvm_para.h> unsigned int num_processors; @@ -1361,52 +1362,76 @@ void enable_x2apic(void) } #endif /* CONFIG_X86_X2APIC */ -void __init enable_IR_x2apic(void) +int __init enable_IR(void) { #ifdef CONFIG_INTR_REMAP int ret; - unsigned long flags; - struct IO_APIC_route_entry **ioapic_entries = NULL; ret = dmar_table_init(); if (ret) { pr_debug("dmar_table_init() failed with %d:\n", ret); - goto ir_failed; + return 0; } if (!intr_remapping_supported()) { pr_debug("intr-remapping not supported\n"); - goto ir_failed; + return 0; } - if (!x2apic_preenabled && skip_ioapic_setup) { pr_info("Skipped enabling intr-remap because of skipping " "io-apic setup\n"); - return; + return 0; } + if (enable_intr_remapping(x2apic_supported())) + return 0; + + pr_info("Enabled Interrupt-remapping\n"); + + return 1; + +#endif + return 0; +} + +void __init enable_IR_x2apic(void) +{ + unsigned long flags; + struct IO_APIC_route_entry **ioapic_entries = NULL; + int ret, x2apic_enabled = 0; + ioapic_entries = alloc_ioapic_entries(); if (!ioapic_entries) { - pr_info("Allocate ioapic_entries failed: %d\n", ret); - goto end; + pr_err("Allocate ioapic_entries failed\n"); + goto out; } ret = save_IO_APIC_setup(ioapic_entries); if (ret) { pr_info("Saving IO-APIC state failed: %d\n", ret); - goto end; + goto out; } local_irq_save(flags); - mask_IO_APIC_setup(ioapic_entries); mask_8259A(); + mask_IO_APIC_setup(ioapic_entries); - ret = enable_intr_remapping(x2apic_supported()); - if (ret) - goto end_restore; + ret = enable_IR(); + if (!ret) { + /* IR is required if there is APIC ID > 255 even when running + * under KVM + */ + if (max_physical_apicid > 255 || !kvm_para_available()) + goto nox2apic; + /* + * without IR all CPUs can be addressed by IOAPIC/MSI + * only in physical mode + */ + x2apic_force_phys(); + } - pr_info("Enabled Interrupt-remapping\n"); + x2apic_enabled = 1; if (x2apic_supported() && !x2apic_mode) { x2apic_mode = 1; @@ -1414,41 +1439,25 @@ void __init enable_IR_x2apic(void) pr_info("Enabled x2apic\n"); } -end_restore: - if (ret) - /* - * IR enabling failed - */ +nox2apic: + if (!ret) /* IR enabling failed */ restore_IO_APIC_setup(ioapic_entries); - unmask_8259A(); local_irq_restore(flags); -end: +out: if (ioapic_entries) free_ioapic_entries(ioapic_entries); - if (!ret) + if (x2apic_enabled) return; -ir_failed: if (x2apic_preenabled) - panic("x2apic enabled by bios. But IR enabling failed"); + panic("x2apic: enabled by BIOS but kernel init failed."); else if (cpu_has_x2apic) - pr_info("Not enabling x2apic,Intr-remapping\n"); -#else - if (!cpu_has_x2apic) - return; - - if (x2apic_preenabled) - panic("x2apic enabled prior OS handover," - " enable CONFIG_X86_X2APIC, CONFIG_INTR_REMAP"); -#endif - - return; + pr_info("Not enabling x2apic, Intr-remapping init failed.\n"); } - #ifdef CONFIG_X86_64 /* * Detect and enable local APICs on non-SMP boards. @@ -1651,7 +1660,6 @@ int __init APIC_init_uniprocessor(void) APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { pr_err("BIOS bug, local APIC 0x%x not detected!...\n", boot_cpu_physical_apicid); - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); return -1; } #endif diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 7d52343356e8..78ca97fffa94 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -66,6 +66,8 @@ #include <asm/apic.h> #define __apicdebuginit(type) static type __init +#define for_each_irq_pin(entry, head) \ + for (entry = head; entry; entry = entry->next) /* * Is the SiS APIC rmw bug present ? @@ -116,15 +118,6 @@ static int __init parse_noapic(char *str) } early_param("noapic", parse_noapic); -struct irq_pin_list; - -/* - * This is performance-critical, we want to do it O(1) - * - * the indexing order of this array favors 1:1 mappings - * between pins and IRQs. - */ - struct irq_pin_list { int apic, pin; struct irq_pin_list *next; @@ -139,6 +132,11 @@ static struct irq_pin_list *get_one_free_irq_2_pin(int node) return pin; } +/* + * This is performance-critical, we want to do it O(1) + * + * Most irqs are mapped 1:1 with pins. + */ struct irq_cfg { struct irq_pin_list *irq_2_pin; cpumask_var_t domain; @@ -411,13 +409,10 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg) unsigned long flags; spin_lock_irqsave(&ioapic_lock, flags); - entry = cfg->irq_2_pin; - for (;;) { + for_each_irq_pin(entry, cfg->irq_2_pin) { unsigned int reg; int pin; - if (!entry) - break; pin = entry->pin; reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ @@ -425,9 +420,6 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg) spin_unlock_irqrestore(&ioapic_lock, flags); return true; } - if (!entry->next) - break; - entry = entry->next; } spin_unlock_irqrestore(&ioapic_lock, flags); @@ -495,72 +487,68 @@ static void ioapic_mask_entry(int apic, int pin) * shared ISA-space IRQs, so we have to support them. We are super * fast in the common case, and fast for shared ISA-space IRQs. */ -static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) +static int +add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) { - struct irq_pin_list *entry; - - entry = cfg->irq_2_pin; - if (!entry) { - entry = get_one_free_irq_2_pin(node); - if (!entry) { - printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n", - apic, pin); - return; - } - cfg->irq_2_pin = entry; - entry->apic = apic; - entry->pin = pin; - return; - } + struct irq_pin_list **last, *entry; - while (entry->next) { - /* not again, please */ + /* don't allow duplicates */ + last = &cfg->irq_2_pin; + for_each_irq_pin(entry, cfg->irq_2_pin) { if (entry->apic == apic && entry->pin == pin) - return; - - entry = entry->next; + return 0; + last = &entry->next; } - entry->next = get_one_free_irq_2_pin(node); - entry = entry->next; + entry = get_one_free_irq_2_pin(node); + if (!entry) { + printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n", + node, apic, pin); + return -ENOMEM; + } entry->apic = apic; entry->pin = pin; + + *last = entry; + return 0; +} + +static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) +{ + if (add_pin_to_irq_node_nopanic(cfg, node, apic, pin)) + panic("IO-APIC: failed to add irq-pin. Can not proceed\n"); } /* * Reroute an IRQ to a different pin. */ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node, - int oldapic, int oldpin, - int newapic, int newpin) + int oldapic, int oldpin, + int newapic, int newpin) { - struct irq_pin_list *entry = cfg->irq_2_pin; - int replaced = 0; + struct irq_pin_list *entry; - while (entry) { + for_each_irq_pin(entry, cfg->irq_2_pin) { if (entry->apic == oldapic && entry->pin == oldpin) { entry->apic = newapic; entry->pin = newpin; - replaced = 1; /* every one is different, right? */ - break; + return; } - entry = entry->next; } - /* why? call replace before add? */ - if (!replaced) - add_pin_to_irq_node(cfg, node, newapic, newpin); + /* old apic/pin didn't exist, so just add new ones */ + add_pin_to_irq_node(cfg, node, newapic, newpin); } -static inline void io_apic_modify_irq(struct irq_cfg *cfg, - int mask_and, int mask_or, - void (*final)(struct irq_pin_list *entry)) +static void io_apic_modify_irq(struct irq_cfg *cfg, + int mask_and, int mask_or, + void (*final)(struct irq_pin_list *entry)) { int pin; struct irq_pin_list *entry; - for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) { + for_each_irq_pin(entry, cfg->irq_2_pin) { unsigned int reg; pin = entry->pin; reg = io_apic_read(entry->apic, 0x10 + pin * 2); @@ -577,7 +565,6 @@ static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); } -#ifdef CONFIG_X86_64 static void io_apic_sync(struct irq_pin_list *entry) { /* @@ -593,11 +580,6 @@ static void __mask_IO_APIC_irq(struct irq_cfg *cfg) { io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); } -#else /* CONFIG_X86_32 */ -static void __mask_IO_APIC_irq(struct irq_cfg *cfg) -{ - io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL); -} static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg) { @@ -610,7 +592,6 @@ static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg) io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, IO_APIC_REDIR_LEVEL_TRIGGER, NULL); } -#endif /* CONFIG_X86_32 */ static void mask_IO_APIC_irq_desc(struct irq_desc *desc) { @@ -1699,12 +1680,8 @@ __apicdebuginit(void) print_IO_APIC(void) if (!entry) continue; printk(KERN_DEBUG "IRQ%d ", irq); - for (;;) { + for_each_irq_pin(entry, cfg->irq_2_pin) printk("-> %d:%d", entry->apic, entry->pin); - if (!entry->next) - break; - entry = entry->next; - } printk("\n"); } @@ -2208,7 +2185,6 @@ static unsigned int startup_ioapic_irq(unsigned int irq) return was_pending; } -#ifdef CONFIG_X86_64 static int ioapic_retrigger_irq(unsigned int irq) { @@ -2221,14 +2197,6 @@ static int ioapic_retrigger_irq(unsigned int irq) return 1; } -#else -static int ioapic_retrigger_irq(unsigned int irq) -{ - apic->send_IPI_self(irq_cfg(irq)->vector); - - return 1; -} -#endif /* * Level and edge triggered IO-APIC interrupts need different handling, @@ -2266,13 +2234,9 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq struct irq_pin_list *entry; u8 vector = cfg->vector; - entry = cfg->irq_2_pin; - for (;;) { + for_each_irq_pin(entry, cfg->irq_2_pin) { unsigned int reg; - if (!entry) - break; - apic = entry->apic; pin = entry->pin; /* @@ -2285,9 +2249,6 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; io_apic_modify(apic, 0x10 + pin*2, reg); - if (!entry->next) - break; - entry = entry->next; } } @@ -2512,11 +2473,8 @@ atomic_t irq_mis_count; static void ack_apic_level(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); - -#ifdef CONFIG_X86_32 unsigned long v; int i; -#endif struct irq_cfg *cfg; int do_unmask_irq = 0; @@ -2529,31 +2487,28 @@ static void ack_apic_level(unsigned int irq) } #endif -#ifdef CONFIG_X86_32 /* - * It appears there is an erratum which affects at least version 0x11 - * of I/O APIC (that's the 82093AA and cores integrated into various - * chipsets). Under certain conditions a level-triggered interrupt is - * erroneously delivered as edge-triggered one but the respective IRR - * bit gets set nevertheless. As a result the I/O unit expects an EOI - * message but it will never arrive and further interrupts are blocked - * from the source. The exact reason is so far unknown, but the - * phenomenon was observed when two consecutive interrupt requests - * from a given source get delivered to the same CPU and the source is - * temporarily disabled in between. - * - * A workaround is to simulate an EOI message manually. We achieve it - * by setting the trigger mode to edge and then to level when the edge - * trigger mode gets detected in the TMR of a local APIC for a - * level-triggered interrupt. We mask the source for the time of the - * operation to prevent an edge-triggered interrupt escaping meanwhile. - * The idea is from Manfred Spraul. --macro - */ + * It appears there is an erratum which affects at least version 0x11 + * of I/O APIC (that's the 82093AA and cores integrated into various + * chipsets). Under certain conditions a level-triggered interrupt is + * erroneously delivered as edge-triggered one but the respective IRR + * bit gets set nevertheless. As a result the I/O unit expects an EOI + * message but it will never arrive and further interrupts are blocked + * from the source. The exact reason is so far unknown, but the + * phenomenon was observed when two consecutive interrupt requests + * from a given source get delivered to the same CPU and the source is + * temporarily disabled in between. + * + * A workaround is to simulate an EOI message manually. We achieve it + * by setting the trigger mode to edge and then to level when the edge + * trigger mode gets detected in the TMR of a local APIC for a + * level-triggered interrupt. We mask the source for the time of the + * operation to prevent an edge-triggered interrupt escaping meanwhile. + * The idea is from Manfred Spraul. --macro + */ cfg = desc->chip_data; i = cfg->vector; - v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); -#endif /* * We must acknowledge the irq before we move it or the acknowledge will @@ -2595,7 +2550,7 @@ static void ack_apic_level(unsigned int irq) unmask_IO_APIC_irq_desc(desc); } -#ifdef CONFIG_X86_32 + /* Tail end of version 0x11 I/O APIC bug workaround */ if (!(v & (1 << (i & 0x1f)))) { atomic_inc(&irq_mis_count); spin_lock(&ioapic_lock); @@ -2603,26 +2558,15 @@ static void ack_apic_level(unsigned int irq) __unmask_and_level_IO_APIC_irq(cfg); spin_unlock(&ioapic_lock); } -#endif } #ifdef CONFIG_INTR_REMAP static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) { - int apic, pin; struct irq_pin_list *entry; - entry = cfg->irq_2_pin; - for (;;) { - - if (!entry) - break; - - apic = entry->apic; - pin = entry->pin; - io_apic_eoi(apic, pin); - entry = entry->next; - } + for_each_irq_pin(entry, cfg->irq_2_pin) + io_apic_eoi(entry->apic, entry->pin); } static void @@ -3238,8 +3182,7 @@ void destroy_irq(unsigned int irq) cfg = desc->chip_data; dynamic_irq_cleanup(irq); /* connect back irq_cfg */ - if (desc) - desc->chip_data = cfg; + desc->chip_data = cfg; free_irte(irq); spin_lock_irqsave(&vector_lock, flags); @@ -3909,7 +3852,11 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, */ if (irq >= NR_IRQS_LEGACY) { cfg = desc->chip_data; - add_pin_to_irq_node(cfg, node, ioapic, pin); + if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { + printk(KERN_INFO "can not add pin %d for irq %d\n", + pin, irq); + return 0; + } } setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity); diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c index dbf5445727a9..e6b4f517fcfe 100644 --- a/arch/x86/kernel/apic/ipi.c +++ b/arch/x86/kernel/apic/ipi.c @@ -150,7 +150,7 @@ int safe_smp_processor_id(void) { int apicid, cpuid; - if (!boot_cpu_has(X86_FEATURE_APIC)) + if (!cpu_has_apic) return 0; apicid = hard_smp_processor_id(); diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index b3025b43b63a..cb66a22d98ad 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -39,7 +39,7 @@ int unknown_nmi_panic; int nmi_watchdog_enabled; -static cpumask_var_t backtrace_mask; +static cpumask_t backtrace_mask __read_mostly; /* nmi_active: * >0: the lapic NMI watchdog is active, but can be disabled @@ -66,7 +66,7 @@ static inline unsigned int get_nmi_count(int cpu) static inline int mce_in_progress(void) { -#if defined(CONFIG_X86_NEW_MCE) +#if defined(CONFIG_X86_MCE) return atomic_read(&mce_entry) > 0; #endif return 0; @@ -138,7 +138,6 @@ int __init check_nmi_watchdog(void) if (!prev_nmi_count) goto error; - alloc_cpumask_var(&backtrace_mask, GFP_KERNEL|__GFP_ZERO); printk(KERN_INFO "Testing NMI watchdog ... "); #ifdef CONFIG_SMP @@ -415,14 +414,17 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) } /* We can be called before check_nmi_watchdog, hence NULL check. */ - if (backtrace_mask != NULL && cpumask_test_cpu(cpu, backtrace_mask)) { + if (cpumask_test_cpu(cpu, &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, backtrace_mask); + cpumask_clear_cpu(cpu, &backtrace_mask); + + rc = 1; } /* Could check oops_in_progress here too, but it's safer not to */ @@ -552,14 +554,18 @@ int do_nmi_callback(struct pt_regs *regs, int cpu) return 0; } -void __trigger_all_cpu_backtrace(void) +void arch_trigger_all_cpu_backtrace(void) { int i; - cpumask_copy(backtrace_mask, cpu_online_mask); + cpumask_copy(&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(backtrace_mask)) + if (cpumask_empty(&backtrace_mask)) break; mdelay(1); } diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index bc3e880f9b82..f3b1037076e4 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -50,11 +50,11 @@ static struct apic *apic_probe[] __initdata = { void __init default_setup_apic_routing(void) { #ifdef CONFIG_X86_X2APIC - if (x2apic_mode && (apic != &apic_x2apic_phys && + if (x2apic_mode #ifdef CONFIG_X86_UV - apic != &apic_x2apic_uv_x && + && apic != &apic_x2apic_uv_x #endif - apic != &apic_x2apic_cluster)) { + ) { if (x2apic_phys) apic = &apic_x2apic_phys; else diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 442b5508893f..bfe0815c4f96 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -403,7 +403,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue); static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue); static struct apm_user *user_list; static DEFINE_SPINLOCK(user_list_lock); -static const struct desc_struct bad_bios_desc = { { { 0, 0x00409200 } } }; +static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0); static const char driver_version[] = "1.16ac"; /* no spaces */ @@ -2337,8 +2337,8 @@ static int __init apm_init(void) * This is for buggy BIOS's that refer to (real mode) segment 0x40 * even though they are called in protected mode. */ - set_base(bad_bios_desc, __va((unsigned long)0x40 << 4)); - _set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4)); + set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4)); + set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4)); /* * Set up the long jump entry point to the APM BIOS, which is called @@ -2358,12 +2358,12 @@ static int __init apm_init(void) * code to that CPU. */ gdt = get_cpu_gdt_table(0); - set_base(gdt[APM_CS >> 3], - __va((unsigned long)apm_info.bios.cseg << 4)); - set_base(gdt[APM_CS_16 >> 3], - __va((unsigned long)apm_info.bios.cseg_16 << 4)); - set_base(gdt[APM_DS >> 3], - __va((unsigned long)apm_info.bios.dseg << 4)); + set_desc_base(&gdt[APM_CS >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.cseg << 4)); + set_desc_base(&gdt[APM_CS_16 >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.cseg_16 << 4)); + set_desc_base(&gdt[APM_DS >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.dseg << 4)); proc_create("apm", 0, NULL, &apm_file_ops); diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 82c1b37d38f4..0e05b9f7b8ca 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -2,7 +2,7 @@ #include <linux/bitops.h> #include <linux/mm.h> -#include <asm/io.h> +#include <linux/io.h> #include <asm/processor.h> #include <asm/apic.h> #include <asm/cpu.h> @@ -45,8 +45,8 @@ static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c) #define CBAR_ENB (0x80000000) #define CBAR_KEY (0X000000CB) if (c->x86_model == 9 || c->x86_model == 10) { - if (inl (CBAR) & CBAR_ENB) - outl (0 | CBAR_KEY, CBAR); + if (inl(CBAR) & CBAR_ENB) + outl(0 | CBAR_KEY, CBAR); } } @@ -87,9 +87,10 @@ static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c) d = d2-d; if (d > 20*K6_BUG_LOOP) - printk("system stability may be impaired when more than 32 MB are used.\n"); + printk(KERN_CONT + "system stability may be impaired when more than 32 MB are used.\n"); else - printk("probably OK (after B9730xxxx).\n"); + printk(KERN_CONT "probably OK (after B9730xxxx).\n"); printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n"); } @@ -219,8 +220,9 @@ static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c) if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) { rdmsr(MSR_K7_CLK_CTL, l, h); if ((l & 0xfff00000) != 0x20000000) { - printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l, - ((l & 0x000fffff)|0x20000000)); + printk(KERN_INFO + "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", + l, ((l & 0x000fffff)|0x20000000)); wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h); } } @@ -398,7 +400,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) u32 level; level = cpuid_eax(1); - if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58) + if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58) set_cpu_cap(c, X86_FEATURE_REP_GOOD); } if (c->x86 == 0x10 || c->x86 == 0x11) @@ -487,27 +489,30 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) * benefit in doing so. */ if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) { - printk(KERN_DEBUG "tseg: %010llx\n", tseg); - if ((tseg>>PMD_SHIFT) < + printk(KERN_DEBUG "tseg: %010llx\n", tseg); + if ((tseg>>PMD_SHIFT) < (max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) || - ((tseg>>PMD_SHIFT) < + ((tseg>>PMD_SHIFT) < (max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) && - (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT)))) - set_memory_4k((unsigned long)__va(tseg), 1); + (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT)))) + set_memory_4k((unsigned long)__va(tseg), 1); } } #endif } #ifdef CONFIG_X86_32 -static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size) +static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, + unsigned int size) { /* AMD errata T13 (order #21922) */ if ((c->x86 == 6)) { - if (c->x86_model == 3 && c->x86_mask == 0) /* Duron Rev A0 */ + /* Duron Rev A0 */ + if (c->x86_model == 3 && c->x86_mask == 0) size = 64; + /* Tbird rev A1/A2 */ if (c->x86_model == 4 && - (c->x86_mask == 0 || c->x86_mask == 1)) /* Tbird rev A1/A2 */ + (c->x86_mask == 0 || c->x86_mask == 1)) size = 256; } return size; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index c8e315f1aa83..01a265212395 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -81,7 +81,7 @@ static void __init check_fpu(void) boot_cpu_data.fdiv_bug = fdiv_bug; if (boot_cpu_data.fdiv_bug) - printk("Hmm, FPU with FDIV bug.\n"); + printk(KERN_WARNING "Hmm, FPU with FDIV bug.\n"); } static void __init check_hlt(void) @@ -98,7 +98,7 @@ static void __init check_hlt(void) halt(); halt(); halt(); - printk("OK.\n"); + printk(KERN_CONT "OK.\n"); } /* @@ -122,9 +122,9 @@ static void __init check_popad(void) * CPU hard. Too bad. */ if (res != 12345678) - printk("Buggy.\n"); + printk(KERN_CONT "Buggy.\n"); else - printk("OK.\n"); + printk(KERN_CONT "OK.\n"); #endif } @@ -156,7 +156,7 @@ void __init check_bugs(void) { identify_boot_cpu(); #ifndef CONFIG_SMP - printk("CPU: "); + printk(KERN_INFO "CPU: "); print_cpu_info(&boot_cpu_data); #endif check_config(); diff --git a/arch/x86/kernel/cpu/bugs_64.c b/arch/x86/kernel/cpu/bugs_64.c index 9a3ed0649d4e..04f0fe5af83e 100644 --- a/arch/x86/kernel/cpu/bugs_64.c +++ b/arch/x86/kernel/cpu/bugs_64.c @@ -15,7 +15,7 @@ void __init check_bugs(void) { identify_boot_cpu(); #if !defined(CONFIG_SMP) - printk("CPU: "); + printk(KERN_INFO "CPU: "); print_cpu_info(&boot_cpu_data); #endif alternative_instructions(); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index f1961c07af9a..54a3ead5cfb2 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -18,8 +18,8 @@ #include <asm/hypervisor.h> #include <asm/processor.h> #include <asm/sections.h> -#include <asm/topology.h> -#include <asm/cpumask.h> +#include <linux/topology.h> +#include <linux/cpumask.h> #include <asm/pgtable.h> #include <asm/atomic.h> #include <asm/proto.h> @@ -28,13 +28,13 @@ #include <asm/desc.h> #include <asm/i387.h> #include <asm/mtrr.h> -#include <asm/numa.h> +#include <linux/numa.h> #include <asm/asm.h> #include <asm/cpu.h> #include <asm/mce.h> #include <asm/msr.h> #include <asm/pat.h> -#include <asm/smp.h> +#include <linux/smp.h> #ifdef CONFIG_X86_LOCAL_APIC #include <asm/uv/uv.h> @@ -71,45 +71,45 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { * TLS descriptors are currently at a different place compared to i386. * Hopefully nobody expects them at a fixed place (Wine?) */ - [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } }, - [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } }, - [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } }, - [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } }, - [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } }, - [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } }, + [GDT_ENTRY_KERNEL32_CS] = GDT_ENTRY_INIT(0xc09b, 0, 0xfffff), + [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xa09b, 0, 0xfffff), + [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc093, 0, 0xfffff), + [GDT_ENTRY_DEFAULT_USER32_CS] = GDT_ENTRY_INIT(0xc0fb, 0, 0xfffff), + [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f3, 0, 0xfffff), + [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xa0fb, 0, 0xfffff), #else - [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } }, - [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } }, - [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } }, - [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } }, + [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xc09a, 0, 0xfffff), + [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), + [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xc0fa, 0, 0xfffff), + [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f2, 0, 0xfffff), /* * Segments used for calling PnP BIOS have byte granularity. * They code segments and data segments have fixed 64k limits, * the transfer segment sizes are set at run time. */ /* 32-bit code */ - [GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } }, + [GDT_ENTRY_PNPBIOS_CS32] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), /* 16-bit code */ - [GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } }, + [GDT_ENTRY_PNPBIOS_CS16] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_DS] = GDT_ENTRY_INIT(0x0092, 0, 0xffff), /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_TS1] = GDT_ENTRY_INIT(0x0092, 0, 0), /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } }, + [GDT_ENTRY_PNPBIOS_TS2] = GDT_ENTRY_INIT(0x0092, 0, 0), /* * The APM segments have byte granularity and their bases * are set at run time. All have 64k limits. */ /* 32-bit code */ - [GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } }, + [GDT_ENTRY_APMBIOS_BASE] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), /* 16-bit code */ - [GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } }, + [GDT_ENTRY_APMBIOS_BASE+1] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), /* data */ - [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, + [GDT_ENTRY_APMBIOS_BASE+2] = GDT_ENTRY_INIT(0x4092, 0, 0xffff), - [GDT_ENTRY_ESPFIX_SS] = { { { 0x0000ffff, 0x00cf9200 } } }, - [GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } }, + [GDT_ENTRY_ESPFIX_SS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), + [GDT_ENTRY_PERCPU] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), GDT_STACK_CANARY_INIT #endif } }; @@ -982,18 +982,26 @@ static __init int setup_disablecpuid(char *arg) __setup("clearcpuid=", setup_disablecpuid); #ifdef CONFIG_X86_64 -struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; +struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table }; DEFINE_PER_CPU_FIRST(union irq_stack_union, irq_stack_union) __aligned(PAGE_SIZE); -DEFINE_PER_CPU(char *, irq_stack_ptr) = - init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64; +/* + * The following four percpu variables are hot. Align current_task to + * cacheline size such that all four fall in the same cacheline. + */ +DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned = + &init_task; +EXPORT_PER_CPU_SYMBOL(current_task); DEFINE_PER_CPU(unsigned long, kernel_stack) = (unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE; EXPORT_PER_CPU_SYMBOL(kernel_stack); +DEFINE_PER_CPU(char *, irq_stack_ptr) = + init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64; + DEFINE_PER_CPU(unsigned int, irq_count) = -1; /* @@ -1008,8 +1016,7 @@ static const unsigned int exception_stack_sizes[N_EXCEPTION_STACKS] = { }; static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks - [(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]) - __aligned(PAGE_SIZE); + [(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]); /* May not be marked __init: used by software suspend */ void syscall_init(void) @@ -1042,6 +1049,9 @@ DEFINE_PER_CPU(struct orig_ist, orig_ist); #else /* CONFIG_X86_64 */ +DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; +EXPORT_PER_CPU_SYMBOL(current_task); + #ifdef CONFIG_CC_STACKPROTECTOR DEFINE_PER_CPU(unsigned long, stack_canary); #endif diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index 593171e967ef..19807b89f058 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -3,10 +3,10 @@ #include <linux/delay.h> #include <linux/pci.h> #include <asm/dma.h> -#include <asm/io.h> +#include <linux/io.h> #include <asm/processor-cyrix.h> #include <asm/processor-flags.h> -#include <asm/timer.h> +#include <linux/timer.h> #include <asm/pci-direct.h> #include <asm/tsc.h> @@ -282,7 +282,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) * The 5510/5520 companion chips have a funky PIT. */ if (vendor == PCI_VENDOR_ID_CYRIX && - (device == PCI_DEVICE_ID_CYRIX_5510 || device == PCI_DEVICE_ID_CYRIX_5520)) + (device == PCI_DEVICE_ID_CYRIX_5510 || + device == PCI_DEVICE_ID_CYRIX_5520)) mark_tsc_unstable("cyrix 5510/5520 detected"); } #endif @@ -299,7 +300,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c) * ? : 0x7x * GX1 : 0x8x GX1 datasheet 56 */ - if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f)) + if ((0x30 <= dir1 && dir1 <= 0x6f) || + (0x80 <= dir1 && dir1 <= 0x8f)) geode_configure(); return; } else { /* MediaGX */ @@ -427,9 +429,12 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c) printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n"); local_irq_save(flags); ccr3 = getCx86(CX86_CCR3); - setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ - setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80); /* enable cpuid */ - setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ + /* enable MAPEN */ + setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); + /* enable cpuid */ + setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80); + /* disable MAPEN */ + setCx86(CX86_CCR3, ccr3); local_irq_restore(flags); } } diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c index fb5b86af0b01..93ba8eeb100a 100644 --- a/arch/x86/kernel/cpu/hypervisor.c +++ b/arch/x86/kernel/cpu/hypervisor.c @@ -28,11 +28,10 @@ static inline void __cpuinit detect_hypervisor_vendor(struct cpuinfo_x86 *c) { - if (vmware_platform()) { + if (vmware_platform()) c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE; - } else { + else c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE; - } } unsigned long get_hypervisor_tsc_freq(void) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 3260ab044996..80a722a071b5 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -7,17 +7,17 @@ #include <linux/sched.h> #include <linux/thread_info.h> #include <linux/module.h> +#include <linux/uaccess.h> #include <asm/processor.h> #include <asm/pgtable.h> #include <asm/msr.h> -#include <asm/uaccess.h> #include <asm/ds.h> #include <asm/bugs.h> #include <asm/cpu.h> #ifdef CONFIG_X86_64 -#include <asm/topology.h> +#include <linux/topology.h> #include <asm/numa_64.h> #endif @@ -174,7 +174,8 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) #ifdef CONFIG_X86_F00F_BUG /* * All current models of Pentium and Pentium with MMX technology CPUs - * have the F0 0F bug, which lets nonprivileged users lock up the system. + * have the F0 0F bug, which lets nonprivileged users lock up the + * system. * Note that the workaround only should be initialized once... */ c->f00f_bug = 0; @@ -207,7 +208,7 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c) printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n"); printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n"); lo |= MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE; - wrmsr (MSR_IA32_MISC_ENABLE, lo, hi); + wrmsr(MSR_IA32_MISC_ENABLE, lo, hi); } } @@ -283,7 +284,7 @@ static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) /* Intel has a non-standard dependency on %ecx for this CPUID level. */ cpuid_count(4, 0, &eax, &ebx, &ecx, &edx); if (eax & 0x1f) - return ((eax >> 26) + 1); + return (eax >> 26) + 1; else return 1; } diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 789efe217e1a..306bf0dca061 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -3,7 +3,7 @@ * * Changes: * Venkatesh Pallipadi : Adding cache identification through cpuid(4) - * Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure. + * Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure. * Andi Kleen / Andreas Herrmann : CPUID4 emulation on AMD. */ @@ -16,7 +16,7 @@ #include <linux/pci.h> #include <asm/processor.h> -#include <asm/smp.h> +#include <linux/smp.h> #include <asm/k8.h> #define LVL_1_INST 1 @@ -25,14 +25,15 @@ #define LVL_3 4 #define LVL_TRACE 5 -struct _cache_table -{ +struct _cache_table { unsigned char descriptor; char cache_type; short size; }; -/* all the cache descriptor types we care about (no TLB or trace cache entries) */ +/* All the cache descriptor types we care about (no TLB or + trace cache entries) */ + static const struct _cache_table __cpuinitconst cache_table[] = { { 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */ @@ -105,8 +106,7 @@ static const struct _cache_table __cpuinitconst cache_table[] = }; -enum _cache_type -{ +enum _cache_type { CACHE_TYPE_NULL = 0, CACHE_TYPE_DATA = 1, CACHE_TYPE_INST = 2, @@ -170,31 +170,31 @@ unsigned short num_cache_leaves; Maybe later */ union l1_cache { struct { - unsigned line_size : 8; - unsigned lines_per_tag : 8; - unsigned assoc : 8; - unsigned size_in_kb : 8; + unsigned line_size:8; + unsigned lines_per_tag:8; + unsigned assoc:8; + unsigned size_in_kb:8; }; unsigned val; }; union l2_cache { struct { - unsigned line_size : 8; - unsigned lines_per_tag : 4; - unsigned assoc : 4; - unsigned size_in_kb : 16; + unsigned line_size:8; + unsigned lines_per_tag:4; + unsigned assoc:4; + unsigned size_in_kb:16; }; unsigned val; }; union l3_cache { struct { - unsigned line_size : 8; - unsigned lines_per_tag : 4; - unsigned assoc : 4; - unsigned res : 2; - unsigned size_encoded : 14; + unsigned line_size:8; + unsigned lines_per_tag:4; + unsigned assoc:4; + unsigned res:2; + unsigned size_encoded:14; }; unsigned val; }; @@ -350,7 +350,8 @@ static int __cpuinit find_num_cache_leaves(void) unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) { - unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */ + /* Cache sizes */ + unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */ unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */ unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb; @@ -377,8 +378,8 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) retval = cpuid4_cache_lookup_regs(i, &this_leaf); if (retval >= 0) { - switch(this_leaf.eax.split.level) { - case 1: + switch (this_leaf.eax.split.level) { + case 1: if (this_leaf.eax.split.type == CACHE_TYPE_DATA) new_l1d = this_leaf.size/1024; @@ -386,19 +387,20 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) CACHE_TYPE_INST) new_l1i = this_leaf.size/1024; break; - case 2: + case 2: new_l2 = this_leaf.size/1024; num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing; index_msb = get_count_order(num_threads_sharing); l2_id = c->apicid >> index_msb; break; - case 3: + case 3: new_l3 = this_leaf.size/1024; num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing; - index_msb = get_count_order(num_threads_sharing); + index_msb = get_count_order( + num_threads_sharing); l3_id = c->apicid >> index_msb; break; - default: + default: break; } } @@ -421,22 +423,21 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) /* Number of times to iterate */ n = cpuid_eax(2) & 0xFF; - for ( i = 0 ; i < n ; i++ ) { + for (i = 0 ; i < n ; i++) { cpuid(2, ®s[0], ®s[1], ®s[2], ®s[3]); /* If bit 31 is set, this is an unknown format */ - for ( j = 0 ; j < 3 ; j++ ) { - if (regs[j] & (1 << 31)) regs[j] = 0; - } + for (j = 0 ; j < 3 ; j++) + if (regs[j] & (1 << 31)) + regs[j] = 0; /* Byte 0 is level count, not a descriptor */ - for ( j = 1 ; j < 16 ; j++ ) { + for (j = 1 ; j < 16 ; j++) { unsigned char des = dp[j]; unsigned char k = 0; /* look up this descriptor in the table */ - while (cache_table[k].descriptor != 0) - { + while (cache_table[k].descriptor != 0) { if (cache_table[k].descriptor == des) { if (only_trace && cache_table[k].cache_type != LVL_TRACE) break; @@ -488,14 +489,14 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) } if (trace) - printk (KERN_INFO "CPU: Trace cache: %dK uops", trace); - else if ( l1i ) - printk (KERN_INFO "CPU: L1 I cache: %dK", l1i); + printk(KERN_INFO "CPU: Trace cache: %dK uops", trace); + else if (l1i) + printk(KERN_INFO "CPU: L1 I cache: %dK", l1i); if (l1d) - printk(", L1 D cache: %dK\n", l1d); + printk(KERN_CONT ", L1 D cache: %dK\n", l1d); else - printk("\n"); + printk(KERN_CONT "\n"); if (l2) printk(KERN_INFO "CPU: L2 cache: %dK\n", l2); @@ -558,8 +559,13 @@ static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) } } #else -static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) {} -static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) {} +static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) +{ +} + +static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) +{ +} #endif static void __cpuinit free_cache_attributes(unsigned int cpu) @@ -645,7 +651,7 @@ static DEFINE_PER_CPU(struct _index_kobject *, index_kobject); static ssize_t show_##file_name \ (struct _cpuid4_info *this_leaf, char *buf) \ { \ - return sprintf (buf, "%lu\n", (unsigned long)this_leaf->object + val); \ + return sprintf(buf, "%lu\n", (unsigned long)this_leaf->object + val); \ } show_one_plus(level, eax.split.level, 0); @@ -656,7 +662,7 @@ show_one_plus(number_of_sets, ecx.split.number_of_sets, 1); static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf) { - return sprintf (buf, "%luK\n", this_leaf->size / 1024); + return sprintf(buf, "%luK\n", this_leaf->size / 1024); } static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf, @@ -669,7 +675,7 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf, const struct cpumask *mask; mask = to_cpumask(this_leaf->shared_cpu_map); - n = type? + n = type ? cpulist_scnprintf(buf, len-2, mask) : cpumask_scnprintf(buf, len-2, mask); buf[n++] = '\n'; @@ -800,7 +806,7 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644, show_cache_disable_1, store_cache_disable_1); -static struct attribute * default_attrs[] = { +static struct attribute *default_attrs[] = { &type.attr, &level.attr, &coherency_line_size.attr, @@ -815,7 +821,7 @@ static struct attribute * default_attrs[] = { NULL }; -static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf) +static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf) { struct _cache_attr *fattr = to_attr(attr); struct _index_kobject *this_leaf = to_object(kobj); @@ -828,8 +834,8 @@ static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf) return ret; } -static ssize_t store(struct kobject * kobj, struct attribute * attr, - const char * buf, size_t count) +static ssize_t store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t count) { struct _cache_attr *fattr = to_attr(attr); struct _index_kobject *this_leaf = to_object(kobj); @@ -883,7 +889,7 @@ static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu) goto err_out; per_cpu(index_kobject, cpu) = kzalloc( - sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL); + sizeof(struct _index_kobject) * num_cache_leaves, GFP_KERNEL); if (unlikely(per_cpu(index_kobject, cpu) == NULL)) goto err_out; @@ -917,7 +923,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) } for (i = 0; i < num_cache_leaves; i++) { - this_object = INDEX_KOBJECT_PTR(cpu,i); + this_object = INDEX_KOBJECT_PTR(cpu, i); this_object->cpu = cpu; this_object->index = i; retval = kobject_init_and_add(&(this_object->kobj), @@ -925,9 +931,8 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) per_cpu(cache_kobject, cpu), "index%1lu", i); if (unlikely(retval)) { - for (j = 0; j < i; j++) { - kobject_put(&(INDEX_KOBJECT_PTR(cpu,j)->kobj)); - } + for (j = 0; j < i; j++) + kobject_put(&(INDEX_KOBJECT_PTR(cpu, j)->kobj)); kobject_put(per_cpu(cache_kobject, cpu)); cpuid4_cache_sysfs_exit(cpu); return retval; @@ -952,7 +957,7 @@ static void __cpuinit cache_remove_dev(struct sys_device * sys_dev) cpumask_clear_cpu(cpu, to_cpumask(cache_dev_map)); for (i = 0; i < num_cache_leaves; i++) - kobject_put(&(INDEX_KOBJECT_PTR(cpu,i)->kobj)); + kobject_put(&(INDEX_KOBJECT_PTR(cpu, i)->kobj)); kobject_put(per_cpu(cache_kobject, cpu)); cpuid4_cache_sysfs_exit(cpu); } @@ -977,8 +982,7 @@ static int __cpuinit cacheinfo_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier = -{ +static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier = { .notifier_call = cacheinfo_cpu_callback, }; diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index 188a1ca5ad2b..4ac6d48fe11b 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -1,11 +1,8 @@ -obj-y = mce.o +obj-y = mce.o mce-severity.o -obj-$(CONFIG_X86_NEW_MCE) += mce-severity.o -obj-$(CONFIG_X86_OLD_MCE) += k7.o p4.o p6.o obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o obj-$(CONFIG_X86_MCE_INTEL) += mce_intel.o obj-$(CONFIG_X86_MCE_AMD) += mce_amd.o -obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o obj-$(CONFIG_X86_MCE_INJECT) += mce-inject.o diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c deleted file mode 100644 index b945d5dbc609..000000000000 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ /dev/null @@ -1,116 +0,0 @@ -/* - * Athlon specific Machine Check Exception Reporting - * (C) Copyright 2002 Dave Jones <davej@redhat.com> - */ -#include <linux/interrupt.h> -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/processor.h> -#include <asm/system.h> -#include <asm/mce.h> -#include <asm/msr.h> - -/* Machine Check Handler For AMD Athlon/Duron: */ -static void k7_machine_check(struct pt_regs *regs, long error_code) -{ - u32 alow, ahigh, high, low; - u32 mcgstl, mcgsth; - int recover = 1; - int i; - - rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); - if (mcgstl & (1<<0)) /* Recoverable ? */ - recover = 0; - - printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", - smp_processor_id(), mcgsth, mcgstl); - - for (i = 1; i < nr_mce_banks; i++) { - rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - if (high & (1<<31)) { - char misc[20]; - char addr[24]; - - misc[0] = '\0'; - addr[0] = '\0'; - - if (high & (1<<29)) - recover |= 1; - if (high & (1<<25)) - recover |= 2; - high &= ~(1<<31); - - if (high & (1<<27)) { - rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); - snprintf(misc, 20, "[%08x%08x]", ahigh, alow); - } - if (high & (1<<26)) { - rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); - snprintf(addr, 24, " at %08x%08x", ahigh, alow); - } - - printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n", - smp_processor_id(), i, high, low, misc, addr); - - /* Clear it: */ - wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); - /* Serialize: */ - wmb(); - add_taint(TAINT_MACHINE_CHECK); - } - } - - if (recover & 2) - panic("CPU context corrupt"); - if (recover & 1) - panic("Unable to continue"); - - printk(KERN_EMERG "Attempting to continue.\n"); - - mcgstl &= ~(1<<2); - wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); -} - - -/* AMD K7 machine check is Intel like: */ -void amd_mcheck_init(struct cpuinfo_x86 *c) -{ - u32 l, h; - int i; - - if (!cpu_has(c, X86_FEATURE_MCE)) - return; - - machine_check_vector = k7_machine_check; - /* Make sure the vector pointer is visible before we enable MCEs: */ - wmb(); - - printk(KERN_INFO "Intel machine check architecture supported.\n"); - - rdmsr(MSR_IA32_MCG_CAP, l, h); - if (l & (1<<8)) /* Control register present ? */ - wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); - nr_mce_banks = l & 0xff; - - /* - * Clear status for MC index 0 separately, we don't touch CTL, - * as some K7 Athlons cause spurious MCEs when its enabled: - */ - if (boot_cpu_data.x86 == 6) { - wrmsr(MSR_IA32_MC0_STATUS, 0x0, 0x0); - i = 1; - } else - i = 0; - - for (; i < nr_mce_banks; i++) { - wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); - wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); - } - - set_in_cr4(X86_CR4_MCE); - printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", - smp_processor_id()); -} diff --git a/arch/x86/kernel/cpu/mcheck/mce-internal.h b/arch/x86/kernel/cpu/mcheck/mce-internal.h index 54dcb8ff12e5..6bd51e7ba87b 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-internal.h +++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h @@ -1,3 +1,4 @@ +#include <linux/sysdev.h> #include <asm/mce.h> enum severity_level { @@ -10,6 +11,19 @@ enum severity_level { MCE_PANIC_SEVERITY, }; +#define ATTR_LEN 16 + +/* One object for each MCE bank, shared by all CPUs */ +struct mce_bank { + u64 ctl; /* subevents to enable */ + unsigned char init; /* initialise bank? */ + struct sysdev_attribute attr; /* sysdev attribute */ + char attrname[ATTR_LEN]; /* attribute name */ +}; + int mce_severity(struct mce *a, int tolerant, char **msg); extern int mce_ser; + +extern struct mce_bank *mce_banks; + diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c index ff0807f97056..51f7c725dab5 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-severity.c +++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c @@ -209,8 +209,6 @@ static int __init severities_debugfs_init(void) return 0; err_out: - if (fseverities_coverage) - debugfs_remove(fseverities_coverage); if (dmce) debugfs_remove(dmce); return -ENOMEM; diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 1cfb623ce11c..97cd5b0baab4 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -45,21 +45,8 @@ #include "mce-internal.h" -/* Handle unconfigured int18 (should never happen) */ -static void unexpected_machine_check(struct pt_regs *regs, long error_code) -{ - printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n", - smp_processor_id()); -} - -/* Call the installed machine check handler for this CPU setup. */ -void (*machine_check_vector)(struct pt_regs *, long error_code) = - unexpected_machine_check; - int mce_disabled __read_mostly; -#ifdef CONFIG_X86_NEW_MCE - #define MISC_MCELOG_MINOR 227 #define SPINUNIT 100 /* 100ns */ @@ -77,7 +64,6 @@ DEFINE_PER_CPU(unsigned, mce_exception_count); */ static int tolerant __read_mostly = 1; static int banks __read_mostly; -static u64 *bank __read_mostly; static int rip_msr __read_mostly; static int mce_bootlog __read_mostly = -1; static int monarch_timeout __read_mostly = -1; @@ -87,13 +73,13 @@ int mce_cmci_disabled __read_mostly; int mce_ignore_ce __read_mostly; int mce_ser __read_mostly; +struct mce_bank *mce_banks __read_mostly; + /* User mode helper program triggered by machine check event */ static unsigned long mce_need_notify; static char mce_helper[128]; static char *mce_helper_argv[2] = { mce_helper, NULL }; -static unsigned long dont_init_banks; - static DECLARE_WAIT_QUEUE_HEAD(mce_wait); static DEFINE_PER_CPU(struct mce, mces_seen); static int cpu_missing; @@ -104,11 +90,6 @@ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL }; -static inline int skip_bank_init(int i) -{ - return i < BITS_PER_LONG && test_bit(i, &dont_init_banks); -} - static DEFINE_PER_CPU(struct work_struct, mce_work); /* Do initial initialization of a struct mce */ @@ -242,7 +223,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp) /* * Make sure only one CPU runs in machine check panic */ - if (atomic_add_return(1, &mce_paniced) > 1) + if (atomic_inc_return(&mce_paniced) > 1) wait_for_panic(); barrier(); @@ -286,11 +267,11 @@ static int msr_to_offset(u32 msr) unsigned bank = __get_cpu_var(injectm.bank); if (msr == rip_msr) return offsetof(struct mce, ip); - if (msr == MSR_IA32_MC0_STATUS + bank*4) + if (msr == MSR_IA32_MCx_STATUS(bank)) return offsetof(struct mce, status); - if (msr == MSR_IA32_MC0_ADDR + bank*4) + if (msr == MSR_IA32_MCx_ADDR(bank)) return offsetof(struct mce, addr); - if (msr == MSR_IA32_MC0_MISC + bank*4) + if (msr == MSR_IA32_MCx_MISC(bank)) return offsetof(struct mce, misc); if (msr == MSR_IA32_MCG_STATUS) return offsetof(struct mce, mcgstatus); @@ -495,7 +476,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); for (i = 0; i < banks; i++) { - if (!bank[i] || !test_bit(i, *b)) + if (!mce_banks[i].ctl || !test_bit(i, *b)) continue; m.misc = 0; @@ -504,7 +485,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) m.tsc = 0; barrier(); - m.status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4); + m.status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i)); if (!(m.status & MCI_STATUS_VAL)) continue; @@ -519,9 +500,9 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) continue; if (m.status & MCI_STATUS_MISCV) - m.misc = mce_rdmsrl(MSR_IA32_MC0_MISC + i*4); + m.misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i)); if (m.status & MCI_STATUS_ADDRV) - m.addr = mce_rdmsrl(MSR_IA32_MC0_ADDR + i*4); + m.addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i)); if (!(flags & MCP_TIMESTAMP)) m.tsc = 0; @@ -537,7 +518,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) /* * Clear state for this bank. */ - mce_wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + mce_wrmsrl(MSR_IA32_MCx_STATUS(i), 0); } /* @@ -558,7 +539,7 @@ static int mce_no_way_out(struct mce *m, char **msg) int i; for (i = 0; i < banks; i++) { - m->status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4); + m->status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i)); if (mce_severity(m, tolerant, msg) >= MCE_PANIC_SEVERITY) return 1; } @@ -705,7 +686,7 @@ static int mce_start(int *no_way_out) * global_nwo should be updated before mce_callin */ smp_wmb(); - order = atomic_add_return(1, &mce_callin); + order = atomic_inc_return(&mce_callin); /* * Wait for everyone. @@ -842,7 +823,7 @@ static void mce_clear_state(unsigned long *toclear) for (i = 0; i < banks; i++) { if (test_bit(i, toclear)) - mce_wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + mce_wrmsrl(MSR_IA32_MCx_STATUS(i), 0); } } @@ -916,14 +897,14 @@ void do_machine_check(struct pt_regs *regs, long error_code) order = mce_start(&no_way_out); for (i = 0; i < banks; i++) { __clear_bit(i, toclear); - if (!bank[i]) + if (!mce_banks[i].ctl) continue; m.misc = 0; m.addr = 0; m.bank = i; - m.status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4); + m.status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i)); if ((m.status & MCI_STATUS_VAL) == 0) continue; @@ -964,9 +945,9 @@ void do_machine_check(struct pt_regs *regs, long error_code) kill_it = 1; if (m.status & MCI_STATUS_MISCV) - m.misc = mce_rdmsrl(MSR_IA32_MC0_MISC + i*4); + m.misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i)); if (m.status & MCI_STATUS_ADDRV) - m.addr = mce_rdmsrl(MSR_IA32_MC0_ADDR + i*4); + m.addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i)); /* * Action optional error. Queue address for later processing. @@ -1159,10 +1140,25 @@ int mce_notify_irq(void) } EXPORT_SYMBOL_GPL(mce_notify_irq); +static int mce_banks_init(void) +{ + int i; + + mce_banks = kzalloc(banks * sizeof(struct mce_bank), GFP_KERNEL); + if (!mce_banks) + return -ENOMEM; + for (i = 0; i < banks; i++) { + struct mce_bank *b = &mce_banks[i]; + b->ctl = -1ULL; + b->init = 1; + } + return 0; +} + /* * Initialize Machine Checks for a CPU. */ -static int mce_cap_init(void) +static int __cpuinit mce_cap_init(void) { unsigned b; u64 cap; @@ -1182,11 +1178,10 @@ static int mce_cap_init(void) /* Don't support asymmetric configurations today */ WARN_ON(banks != 0 && b != banks); banks = b; - if (!bank) { - bank = kmalloc(banks * sizeof(u64), GFP_KERNEL); - if (!bank) - return -ENOMEM; - memset(bank, 0xff, banks * sizeof(u64)); + if (!mce_banks) { + int err = mce_banks_init(); + if (err) + return err; } /* Use accurate RIP reporting if available. */ @@ -1218,15 +1213,16 @@ static void mce_init(void) wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); for (i = 0; i < banks; i++) { - if (skip_bank_init(i)) + struct mce_bank *b = &mce_banks[i]; + if (!b->init) continue; - wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); - wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + wrmsrl(MSR_IA32_MCx_CTL(i), b->ctl); + wrmsrl(MSR_IA32_MCx_STATUS(i), 0); } } /* Add per CPU specific workarounds here */ -static void mce_cpu_quirks(struct cpuinfo_x86 *c) +static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD) { @@ -1236,7 +1232,7 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * trips off incorrectly with the IOMMU & 3ware * & Cerberus: */ - clear_bit(10, (unsigned long *)&bank[4]); + clear_bit(10, (unsigned long *)&mce_banks[4].ctl); } if (c->x86 <= 17 && mce_bootlog < 0) { /* @@ -1250,7 +1246,7 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * by default. */ if (c->x86 == 6 && banks > 0) - bank[0] = 0; + mce_banks[0].ctl = 0; } if (c->x86_vendor == X86_VENDOR_INTEL) { @@ -1263,8 +1259,8 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * valid event later, merely don't write CTL0. */ - if (c->x86 == 6 && c->x86_model < 0x1A) - __set_bit(0, &dont_init_banks); + if (c->x86 == 6 && c->x86_model < 0x1A && banks > 0) + mce_banks[0].init = 0; /* * All newer Intel systems support MCE broadcasting. Enable @@ -1273,6 +1269,10 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) if ((c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xe)) && monarch_timeout < 0) monarch_timeout = USEC_PER_SEC; + + /* There are also broken BIOSes on some Pentium M systems. */ + if (c->x86 == 6 && c->x86_model == 13 && mce_bootlog < 0) + mce_bootlog = 0; } if (monarch_timeout < 0) monarch_timeout = 0; @@ -1324,6 +1324,17 @@ static void mce_init_timer(void) add_timer_on(t, smp_processor_id()); } +/* Handle unconfigured int18 (should never happen) */ +static void unexpected_machine_check(struct pt_regs *regs, long error_code) +{ + printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n", + smp_processor_id()); +} + +/* Call the installed machine check handler for this CPU setup. */ +void (*machine_check_vector)(struct pt_regs *, long error_code) = + unexpected_machine_check; + /* * Called for each booted CPU to set up machine checks. * Must be called with preempt off: @@ -1538,8 +1549,10 @@ static struct miscdevice mce_log_device = { */ static int __init mcheck_enable(char *str) { - if (*str == 0) + if (*str == 0) { enable_p5_mce(); + return 1; + } if (*str == '=') str++; if (!strcmp(str, "off")) @@ -1580,8 +1593,9 @@ static int mce_disable(void) int i; for (i = 0; i < banks; i++) { - if (!skip_bank_init(i)) - wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + struct mce_bank *b = &mce_banks[i]; + if (b->init) + wrmsrl(MSR_IA32_MCx_CTL(i), 0); } return 0; } @@ -1656,14 +1670,15 @@ DEFINE_PER_CPU(struct sys_device, mce_dev); __cpuinitdata void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); -static struct sysdev_attribute *bank_attrs; +static inline struct mce_bank *attr_to_bank(struct sysdev_attribute *attr) +{ + return container_of(attr, struct mce_bank, attr); +} static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { - u64 b = bank[attr - bank_attrs]; - - return sprintf(buf, "%llx\n", b); + return sprintf(buf, "%llx\n", attr_to_bank(attr)->ctl); } static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, @@ -1674,7 +1689,7 @@ static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, if (strict_strtoull(buf, 0, &new) < 0) return -EINVAL; - bank[attr - bank_attrs] = new; + attr_to_bank(attr)->ctl = new; mce_restart(); return size; @@ -1816,7 +1831,7 @@ static __cpuinit int mce_create_device(unsigned int cpu) } for (j = 0; j < banks; j++) { err = sysdev_create_file(&per_cpu(mce_dev, cpu), - &bank_attrs[j]); + &mce_banks[j].attr); if (err) goto error2; } @@ -1825,10 +1840,10 @@ static __cpuinit int mce_create_device(unsigned int cpu) return 0; error2: while (--j >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[j]); + sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[j].attr); error: while (--i >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); + sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr); sysdev_unregister(&per_cpu(mce_dev, cpu)); @@ -1846,7 +1861,7 @@ static __cpuinit void mce_remove_device(unsigned int cpu) sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); for (i = 0; i < banks; i++) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); + sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr); sysdev_unregister(&per_cpu(mce_dev, cpu)); cpumask_clear_cpu(cpu, mce_dev_initialized); @@ -1863,8 +1878,9 @@ static void mce_disable_cpu(void *h) if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); for (i = 0; i < banks; i++) { - if (!skip_bank_init(i)) - wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + struct mce_bank *b = &mce_banks[i]; + if (b->init) + wrmsrl(MSR_IA32_MCx_CTL(i), 0); } } @@ -1879,8 +1895,9 @@ static void mce_reenable_cpu(void *h) if (!(action & CPU_TASKS_FROZEN)) cmci_reenable(); for (i = 0; i < banks; i++) { - if (!skip_bank_init(i)) - wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]); + struct mce_bank *b = &mce_banks[i]; + if (b->init) + wrmsrl(MSR_IA32_MCx_CTL(i), b->ctl); } } @@ -1928,35 +1945,21 @@ static struct notifier_block mce_cpu_notifier __cpuinitdata = { .notifier_call = mce_cpu_callback, }; -static __init int mce_init_banks(void) +static __init void mce_init_banks(void) { int i; - bank_attrs = kzalloc(sizeof(struct sysdev_attribute) * banks, - GFP_KERNEL); - if (!bank_attrs) - return -ENOMEM; - for (i = 0; i < banks; i++) { - struct sysdev_attribute *a = &bank_attrs[i]; + struct mce_bank *b = &mce_banks[i]; + struct sysdev_attribute *a = &b->attr; - a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i); - if (!a->attr.name) - goto nomem; + a->attr.name = b->attrname; + snprintf(b->attrname, ATTR_LEN, "bank%d", i); a->attr.mode = 0644; a->show = show_bank; a->store = set_bank; } - return 0; - -nomem: - while (--i >= 0) - kfree(bank_attrs[i].attr.name); - kfree(bank_attrs); - bank_attrs = NULL; - - return -ENOMEM; } static __init int mce_init_device(void) @@ -1969,9 +1972,7 @@ static __init int mce_init_device(void) zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); - err = mce_init_banks(); - if (err) - return err; + mce_init_banks(); err = sysdev_class_register(&mce_sysclass); if (err) @@ -1991,51 +1992,6 @@ static __init int mce_init_device(void) device_initcall(mce_init_device); -#else /* CONFIG_X86_OLD_MCE: */ - -int nr_mce_banks; -EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ - -/* This has to be run for each processor */ -void mcheck_init(struct cpuinfo_x86 *c) -{ - if (mce_disabled) - return; - - switch (c->x86_vendor) { - case X86_VENDOR_AMD: - amd_mcheck_init(c); - break; - - case X86_VENDOR_INTEL: - if (c->x86 == 5) - intel_p5_mcheck_init(c); - if (c->x86 == 6) - intel_p6_mcheck_init(c); - if (c->x86 == 15) - intel_p4_mcheck_init(c); - break; - - case X86_VENDOR_CENTAUR: - if (c->x86 == 5) - winchip_mcheck_init(c); - break; - - default: - break; - } - printk(KERN_INFO "mce: CPU supports %d MCE banks\n", nr_mce_banks); -} - -static int __init mcheck_enable(char *str) -{ - mce_p5_enabled = 1; - return 1; -} -__setup("mce", mcheck_enable); - -#endif /* CONFIG_X86_OLD_MCE */ - /* * Old style boot options parsing. Only for compatibility. */ diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index e1acec0f7a32..889f665fe93d 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -90,7 +90,7 @@ static void cmci_discover(int banks, int boot) if (test_bit(i, owned)) continue; - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MCx_CTL2(i), val); /* Already owned by someone else? */ if (val & CMCI_EN) { @@ -101,8 +101,8 @@ static void cmci_discover(int banks, int boot) } val |= CMCI_EN | CMCI_THRESHOLD; - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + wrmsrl(MSR_IA32_MCx_CTL2(i), val); + rdmsrl(MSR_IA32_MCx_CTL2(i), val); /* Did the enable bit stick? -- the bank supports CMCI */ if (val & CMCI_EN) { @@ -152,9 +152,9 @@ void cmci_clear(void) if (!test_bit(i, __get_cpu_var(mce_banks_owned))) continue; /* Disable CMCI */ - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MCx_CTL2(i), val); val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + wrmsrl(MSR_IA32_MCx_CTL2(i), val); __clear_bit(i, __get_cpu_var(mce_banks_owned)); } spin_unlock_irqrestore(&cmci_discover_lock, flags); diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c deleted file mode 100644 index f5f2d6f71fb6..000000000000 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ /dev/null @@ -1,94 +0,0 @@ -/* - * Non Fatal Machine Check Exception Reporting - * - * (C) Copyright 2002 Dave Jones. <davej@redhat.com> - * - * This file contains routines to check for non-fatal MCEs every 15s - * - */ -#include <linux/interrupt.h> -#include <linux/workqueue.h> -#include <linux/jiffies.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/processor.h> -#include <asm/system.h> -#include <asm/mce.h> -#include <asm/msr.h> - -static int firstbank; - -#define MCE_RATE (15*HZ) /* timer rate is 15s */ - -static void mce_checkregs(void *info) -{ - u32 low, high; - int i; - - for (i = firstbank; i < nr_mce_banks; i++) { - rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - - if (!(high & (1<<31))) - continue; - - printk(KERN_INFO "MCE: The hardware reports a non fatal, " - "correctable incident occurred on CPU %d.\n", - smp_processor_id()); - - printk(KERN_INFO "Bank %d: %08x%08x\n", i, high, low); - - /* - * Scrub the error so we don't pick it up in MCE_RATE - * seconds time: - */ - wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); - - /* Serialize: */ - wmb(); - add_taint(TAINT_MACHINE_CHECK); - } -} - -static void mce_work_fn(struct work_struct *work); -static DECLARE_DELAYED_WORK(mce_work, mce_work_fn); - -static void mce_work_fn(struct work_struct *work) -{ - on_each_cpu(mce_checkregs, NULL, 1); - schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE)); -} - -static int __init init_nonfatal_mce_checker(void) -{ - struct cpuinfo_x86 *c = &boot_cpu_data; - - /* Check for MCE support */ - if (!cpu_has(c, X86_FEATURE_MCE)) - return -ENODEV; - - /* Check for PPro style MCA */ - if (!cpu_has(c, X86_FEATURE_MCA)) - return -ENODEV; - - /* Some Athlons misbehave when we frob bank 0 */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 == 6) - firstbank = 1; - else - firstbank = 0; - - /* - * Check for non-fatal errors every MCE_RATE s - */ - schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE)); - printk(KERN_INFO "Machine check exception polling timer started.\n"); - - return 0; -} -module_init(init_nonfatal_mce_checker); - -MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/mcheck/p4.c b/arch/x86/kernel/cpu/mcheck/p4.c deleted file mode 100644 index 4482aea9aa2e..000000000000 --- a/arch/x86/kernel/cpu/mcheck/p4.c +++ /dev/null @@ -1,163 +0,0 @@ -/* - * P4 specific Machine Check Exception Reporting - */ -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/processor.h> -#include <asm/mce.h> -#include <asm/msr.h> - -/* as supported by the P4/Xeon family */ -struct intel_mce_extended_msrs { - u32 eax; - u32 ebx; - u32 ecx; - u32 edx; - u32 esi; - u32 edi; - u32 ebp; - u32 esp; - u32 eflags; - u32 eip; - /* u32 *reserved[]; */ -}; - -static int mce_num_extended_msrs; - -/* P4/Xeon Extended MCE MSR retrieval, return 0 if unsupported */ -static void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) -{ - u32 h; - - rdmsr(MSR_IA32_MCG_EAX, r->eax, h); - rdmsr(MSR_IA32_MCG_EBX, r->ebx, h); - rdmsr(MSR_IA32_MCG_ECX, r->ecx, h); - rdmsr(MSR_IA32_MCG_EDX, r->edx, h); - rdmsr(MSR_IA32_MCG_ESI, r->esi, h); - rdmsr(MSR_IA32_MCG_EDI, r->edi, h); - rdmsr(MSR_IA32_MCG_EBP, r->ebp, h); - rdmsr(MSR_IA32_MCG_ESP, r->esp, h); - rdmsr(MSR_IA32_MCG_EFLAGS, r->eflags, h); - rdmsr(MSR_IA32_MCG_EIP, r->eip, h); -} - -static void intel_machine_check(struct pt_regs *regs, long error_code) -{ - u32 alow, ahigh, high, low; - u32 mcgstl, mcgsth; - int recover = 1; - int i; - - rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); - if (mcgstl & (1<<0)) /* Recoverable ? */ - recover = 0; - - printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", - smp_processor_id(), mcgsth, mcgstl); - - if (mce_num_extended_msrs > 0) { - struct intel_mce_extended_msrs dbg; - - intel_get_extended_msrs(&dbg); - - printk(KERN_DEBUG "CPU %d: EIP: %08x EFLAGS: %08x\n" - "\teax: %08x ebx: %08x ecx: %08x edx: %08x\n" - "\tesi: %08x edi: %08x ebp: %08x esp: %08x\n", - smp_processor_id(), dbg.eip, dbg.eflags, - dbg.eax, dbg.ebx, dbg.ecx, dbg.edx, - dbg.esi, dbg.edi, dbg.ebp, dbg.esp); - } - - for (i = 0; i < nr_mce_banks; i++) { - rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - if (high & (1<<31)) { - char misc[20]; - char addr[24]; - - misc[0] = addr[0] = '\0'; - if (high & (1<<29)) - recover |= 1; - if (high & (1<<25)) - recover |= 2; - high &= ~(1<<31); - if (high & (1<<27)) { - rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); - snprintf(misc, 20, "[%08x%08x]", ahigh, alow); - } - if (high & (1<<26)) { - rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); - snprintf(addr, 24, " at %08x%08x", ahigh, alow); - } - printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n", - smp_processor_id(), i, high, low, misc, addr); - } - } - - if (recover & 2) - panic("CPU context corrupt"); - if (recover & 1) - panic("Unable to continue"); - - printk(KERN_EMERG "Attempting to continue.\n"); - - /* - * Do not clear the MSR_IA32_MCi_STATUS if the error is not - * recoverable/continuable.This will allow BIOS to look at the MSRs - * for errors if the OS could not log the error. - */ - for (i = 0; i < nr_mce_banks; i++) { - u32 msr; - msr = MSR_IA32_MC0_STATUS+i*4; - rdmsr(msr, low, high); - if (high&(1<<31)) { - /* Clear it */ - wrmsr(msr, 0UL, 0UL); - /* Serialize */ - wmb(); - add_taint(TAINT_MACHINE_CHECK); - } - } - mcgstl &= ~(1<<2); - wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); -} - -void intel_p4_mcheck_init(struct cpuinfo_x86 *c) -{ - u32 l, h; - int i; - - machine_check_vector = intel_machine_check; - wmb(); - - printk(KERN_INFO "Intel machine check architecture supported.\n"); - rdmsr(MSR_IA32_MCG_CAP, l, h); - if (l & (1<<8)) /* Control register present ? */ - wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); - nr_mce_banks = l & 0xff; - - for (i = 0; i < nr_mce_banks; i++) { - wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); - wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); - } - - set_in_cr4(X86_CR4_MCE); - printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", - smp_processor_id()); - - /* Check for P4/Xeon extended MCE MSRs */ - rdmsr(MSR_IA32_MCG_CAP, l, h); - if (l & (1<<9)) {/* MCG_EXT_P */ - mce_num_extended_msrs = (l >> 16) & 0xff; - printk(KERN_INFO "CPU%d: Intel P4/Xeon Extended MCE MSRs (%d)" - " available\n", - smp_processor_id(), mce_num_extended_msrs); - -#ifdef CONFIG_X86_MCE_P4THERMAL - /* Check for P4/Xeon Thermal monitor */ - intel_init_thermal(c); -#endif - } -} diff --git a/arch/x86/kernel/cpu/mcheck/p6.c b/arch/x86/kernel/cpu/mcheck/p6.c deleted file mode 100644 index 01e4f8178183..000000000000 --- a/arch/x86/kernel/cpu/mcheck/p6.c +++ /dev/null @@ -1,127 +0,0 @@ -/* - * P6 specific Machine Check Exception Reporting - * (C) Copyright 2002 Alan Cox <alan@lxorguk.ukuu.org.uk> - */ -#include <linux/interrupt.h> -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/processor.h> -#include <asm/system.h> -#include <asm/mce.h> -#include <asm/msr.h> - -/* Machine Check Handler For PII/PIII */ -static void intel_machine_check(struct pt_regs *regs, long error_code) -{ - u32 alow, ahigh, high, low; - u32 mcgstl, mcgsth; - int recover = 1; - int i; - - rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); - if (mcgstl & (1<<0)) /* Recoverable ? */ - recover = 0; - - printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", - smp_processor_id(), mcgsth, mcgstl); - - for (i = 0; i < nr_mce_banks; i++) { - rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - if (high & (1<<31)) { - char misc[20]; - char addr[24]; - - misc[0] = '\0'; - addr[0] = '\0'; - - if (high & (1<<29)) - recover |= 1; - if (high & (1<<25)) - recover |= 2; - high &= ~(1<<31); - - if (high & (1<<27)) { - rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); - snprintf(misc, 20, "[%08x%08x]", ahigh, alow); - } - if (high & (1<<26)) { - rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); - snprintf(addr, 24, " at %08x%08x", ahigh, alow); - } - - printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n", - smp_processor_id(), i, high, low, misc, addr); - } - } - - if (recover & 2) - panic("CPU context corrupt"); - if (recover & 1) - panic("Unable to continue"); - - printk(KERN_EMERG "Attempting to continue.\n"); - /* - * Do not clear the MSR_IA32_MCi_STATUS if the error is not - * recoverable/continuable.This will allow BIOS to look at the MSRs - * for errors if the OS could not log the error: - */ - for (i = 0; i < nr_mce_banks; i++) { - unsigned int msr; - - msr = MSR_IA32_MC0_STATUS+i*4; - rdmsr(msr, low, high); - if (high & (1<<31)) { - /* Clear it: */ - wrmsr(msr, 0UL, 0UL); - /* Serialize: */ - wmb(); - add_taint(TAINT_MACHINE_CHECK); - } - } - mcgstl &= ~(1<<2); - wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); -} - -/* Set up machine check reporting for processors with Intel style MCE: */ -void intel_p6_mcheck_init(struct cpuinfo_x86 *c) -{ - u32 l, h; - int i; - - /* Check for MCE support */ - if (!cpu_has(c, X86_FEATURE_MCE)) - return; - - /* Check for PPro style MCA */ - if (!cpu_has(c, X86_FEATURE_MCA)) - return; - - /* Ok machine check is available */ - machine_check_vector = intel_machine_check; - /* Make sure the vector pointer is visible before we enable MCEs: */ - wmb(); - - printk(KERN_INFO "Intel machine check architecture supported.\n"); - rdmsr(MSR_IA32_MCG_CAP, l, h); - if (l & (1<<8)) /* Control register present ? */ - wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); - nr_mce_banks = l & 0xff; - - /* - * Following the example in IA-32 SDM Vol 3: - * - MC0_CTL should not be written - * - Status registers on all banks should be cleared on reset - */ - for (i = 1; i < nr_mce_banks; i++) - wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); - - for (i = 0; i < nr_mce_banks; i++) - wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); - - set_in_cr4(X86_CR4_MCE); - printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", - smp_processor_id()); -} diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index bff8dd191dd5..15f2bc07bb60 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -253,9 +253,6 @@ void intel_init_thermal(struct cpuinfo_x86 *c) return; } - if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) - tm2 = 1; - /* Check whether a vector already exists */ if (h & APIC_VECTOR_MASK) { printk(KERN_DEBUG @@ -264,6 +261,16 @@ void intel_init_thermal(struct cpuinfo_x86 *c) return; } + /* early Pentium M models use different method for enabling TM2 */ + if (cpu_has(c, X86_FEATURE_TM2)) { + if (c->x86 == 6 && (c->x86_model == 9 || c->x86_model == 13)) { + rdmsr(MSR_THERM2_CTL, l, h); + if (l & MSR_THERM2_CTL_TM_SELECT) + tm2 = 1; + } else if (l & MSR_IA32_MISC_ENABLE_TM2) + tm2 = 1; + } + /* We'll mask the thermal vector in the lapic till we're ready: */ h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; apic_write(APIC_LVTTHMR, h); diff --git a/arch/x86/kernel/cpu/mtrr/amd.c b/arch/x86/kernel/cpu/mtrr/amd.c index ee2331b0e58f..33af14110dfd 100644 --- a/arch/x86/kernel/cpu/mtrr/amd.c +++ b/arch/x86/kernel/cpu/mtrr/amd.c @@ -7,15 +7,15 @@ static void amd_get_mtrr(unsigned int reg, unsigned long *base, - unsigned long *size, mtrr_type * type) + unsigned long *size, mtrr_type *type) { unsigned long low, high; rdmsr(MSR_K6_UWCCR, low, high); - /* Upper dword is region 1, lower is region 0 */ + /* Upper dword is region 1, lower is region 0 */ if (reg == 1) low = high; - /* The base masks off on the right alignment */ + /* The base masks off on the right alignment */ *base = (low & 0xFFFE0000) >> PAGE_SHIFT; *type = 0; if (low & 1) @@ -27,74 +27,81 @@ amd_get_mtrr(unsigned int reg, unsigned long *base, return; } /* - * This needs a little explaining. The size is stored as an - * inverted mask of bits of 128K granularity 15 bits long offset - * 2 bits + * This needs a little explaining. The size is stored as an + * inverted mask of bits of 128K granularity 15 bits long offset + * 2 bits. * - * So to get a size we do invert the mask and add 1 to the lowest - * mask bit (4 as its 2 bits in). This gives us a size we then shift - * to turn into 128K blocks + * So to get a size we do invert the mask and add 1 to the lowest + * mask bit (4 as its 2 bits in). This gives us a size we then shift + * to turn into 128K blocks. * - * eg 111 1111 1111 1100 is 512K + * eg 111 1111 1111 1100 is 512K * - * invert 000 0000 0000 0011 - * +1 000 0000 0000 0100 - * *128K ... + * invert 000 0000 0000 0011 + * +1 000 0000 0000 0100 + * *128K ... */ low = (~low) & 0x1FFFC; *size = (low + 4) << (15 - PAGE_SHIFT); - return; } -static void amd_set_mtrr(unsigned int reg, unsigned long base, - unsigned long size, mtrr_type type) -/* [SUMMARY] Set variable MTRR register on the local CPU. - <reg> The register to set. - <base> The base address of the region. - <size> The size of the region. If this is 0 the region is disabled. - <type> The type of the region. - [RETURNS] Nothing. -*/ +/** + * amd_set_mtrr - Set variable MTRR register on the local CPU. + * + * @reg The register to set. + * @base The base address of the region. + * @size The size of the region. If this is 0 the region is disabled. + * @type The type of the region. + * + * Returns nothing. + */ +static void +amd_set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type) { u32 regs[2]; /* - * Low is MTRR0 , High MTRR 1 + * Low is MTRR0, High MTRR 1 */ rdmsr(MSR_K6_UWCCR, regs[0], regs[1]); /* - * Blank to disable + * Blank to disable */ - if (size == 0) + if (size == 0) { regs[reg] = 0; - else - /* Set the register to the base, the type (off by one) and an - inverted bitmask of the size The size is the only odd - bit. We are fed say 512K We invert this and we get 111 1111 - 1111 1011 but if you subtract one and invert you get the - desired 111 1111 1111 1100 mask - - But ~(x - 1) == ~x + 1 == -x. Two's complement rocks! */ + } else { + /* + * Set the register to the base, the type (off by one) and an + * inverted bitmask of the size The size is the only odd + * bit. We are fed say 512K We invert this and we get 111 1111 + * 1111 1011 but if you subtract one and invert you get the + * desired 111 1111 1111 1100 mask + * + * But ~(x - 1) == ~x + 1 == -x. Two's complement rocks! + */ regs[reg] = (-size >> (15 - PAGE_SHIFT) & 0x0001FFFC) | (base << PAGE_SHIFT) | (type + 1); + } /* - * The writeback rule is quite specific. See the manual. Its - * disable local interrupts, write back the cache, set the mtrr + * The writeback rule is quite specific. See the manual. Its + * disable local interrupts, write back the cache, set the mtrr */ wbinvd(); wrmsr(MSR_K6_UWCCR, regs[0], regs[1]); } -static int amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type) +static int +amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type) { - /* Apply the K6 block alignment and size rules - In order - o Uncached or gathering only - o 128K or bigger block - o Power of 2 block - o base suitably aligned to the power - */ + /* + * Apply the K6 block alignment and size rules + * In order + * o Uncached or gathering only + * o 128K or bigger block + * o Power of 2 block + * o base suitably aligned to the power + */ if (type > MTRR_TYPE_WRCOMB || size < (1 << (17 - PAGE_SHIFT)) || (size & ~(size - 1)) - size || (base & (size - 1))) return -EINVAL; @@ -115,5 +122,3 @@ int __init amd_init_mtrr(void) set_mtrr_ops(&amd_mtrr_ops); return 0; } - -//arch_initcall(amd_mtrr_init); diff --git a/arch/x86/kernel/cpu/mtrr/centaur.c b/arch/x86/kernel/cpu/mtrr/centaur.c index cb9aa3a7a7ab..de89f14eff3a 100644 --- a/arch/x86/kernel/cpu/mtrr/centaur.c +++ b/arch/x86/kernel/cpu/mtrr/centaur.c @@ -1,7 +1,9 @@ #include <linux/init.h> #include <linux/mm.h> + #include <asm/mtrr.h> #include <asm/msr.h> + #include "mtrr.h" static struct { @@ -12,25 +14,25 @@ static struct { static u8 centaur_mcr_reserved; static u8 centaur_mcr_type; /* 0 for winchip, 1 for winchip2 */ -/* - * Report boot time MCR setups +/** + * centaur_get_free_region - Get a free MTRR. + * + * @base: The starting (base) address of the region. + * @size: The size (in bytes) of the region. + * + * Returns: the index of the region on success, else -1 on error. */ - static int centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg) -/* [SUMMARY] Get a free MTRR. - <base> The starting (base) address of the region. - <size> The size (in bytes) of the region. - [RETURNS] The index of the region on success, else -1 on error. -*/ { - int i, max; - mtrr_type ltype; unsigned long lbase, lsize; + mtrr_type ltype; + int i, max; max = num_var_ranges; if (replace_reg >= 0 && replace_reg < max) return replace_reg; + for (i = 0; i < max; ++i) { if (centaur_mcr_reserved & (1 << i)) continue; @@ -38,11 +40,14 @@ centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg) if (lsize == 0) return i; } + return -ENOSPC; } -void -mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) +/* + * Report boot time MCR setups + */ +void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) { centaur_mcr[mcr].low = lo; centaur_mcr[mcr].high = hi; @@ -54,33 +59,35 @@ centaur_get_mcr(unsigned int reg, unsigned long *base, { *base = centaur_mcr[reg].high >> PAGE_SHIFT; *size = -(centaur_mcr[reg].low & 0xfffff000) >> PAGE_SHIFT; - *type = MTRR_TYPE_WRCOMB; /* If it is there, it is write-combining */ + *type = MTRR_TYPE_WRCOMB; /* write-combining */ + if (centaur_mcr_type == 1 && ((centaur_mcr[reg].low & 31) & 2)) *type = MTRR_TYPE_UNCACHABLE; if (centaur_mcr_type == 1 && (centaur_mcr[reg].low & 31) == 25) *type = MTRR_TYPE_WRBACK; if (centaur_mcr_type == 0 && (centaur_mcr[reg].low & 31) == 31) *type = MTRR_TYPE_WRBACK; - } -static void centaur_set_mcr(unsigned int reg, unsigned long base, - unsigned long size, mtrr_type type) +static void +centaur_set_mcr(unsigned int reg, unsigned long base, + unsigned long size, mtrr_type type) { unsigned long low, high; if (size == 0) { - /* Disable */ + /* Disable */ high = low = 0; } else { high = base << PAGE_SHIFT; - if (centaur_mcr_type == 0) - low = -size << PAGE_SHIFT | 0x1f; /* only support write-combining... */ - else { + if (centaur_mcr_type == 0) { + /* Only support write-combining... */ + low = -size << PAGE_SHIFT | 0x1f; + } else { if (type == MTRR_TYPE_UNCACHABLE) - low = -size << PAGE_SHIFT | 0x02; /* NC */ + low = -size << PAGE_SHIFT | 0x02; /* NC */ else - low = -size << PAGE_SHIFT | 0x09; /* WWO,WC */ + low = -size << PAGE_SHIFT | 0x09; /* WWO, WC */ } } centaur_mcr[reg].high = high; @@ -88,118 +95,16 @@ static void centaur_set_mcr(unsigned int reg, unsigned long base, wrmsr(MSR_IDT_MCR0 + reg, low, high); } -#if 0 -/* - * Initialise the later (saner) Winchip MCR variant. In this version - * the BIOS can pass us the registers it has used (but not their values) - * and the control register is read/write - */ - -static void __init -centaur_mcr1_init(void) -{ - unsigned i; - u32 lo, hi; - - /* Unfortunately, MCR's are read-only, so there is no way to - * find out what the bios might have done. - */ - - rdmsr(MSR_IDT_MCR_CTRL, lo, hi); - if (((lo >> 17) & 7) == 1) { /* Type 1 Winchip2 MCR */ - lo &= ~0x1C0; /* clear key */ - lo |= 0x040; /* set key to 1 */ - wrmsr(MSR_IDT_MCR_CTRL, lo, hi); /* unlock MCR */ - } - - centaur_mcr_type = 1; - - /* - * Clear any unconfigured MCR's. - */ - - for (i = 0; i < 8; ++i) { - if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0) { - if (!(lo & (1 << (9 + i)))) - wrmsr(MSR_IDT_MCR0 + i, 0, 0); - else - /* - * If the BIOS set up an MCR we cannot see it - * but we don't wish to obliterate it - */ - centaur_mcr_reserved |= (1 << i); - } - } - /* - * Throw the main write-combining switch... - * However if OOSTORE is enabled then people have already done far - * cleverer things and we should behave. - */ - - lo |= 15; /* Write combine enables */ - wrmsr(MSR_IDT_MCR_CTRL, lo, hi); -} - -/* - * Initialise the original winchip with read only MCR registers - * no used bitmask for the BIOS to pass on and write only control - */ - -static void __init -centaur_mcr0_init(void) -{ - unsigned i; - - /* Unfortunately, MCR's are read-only, so there is no way to - * find out what the bios might have done. - */ - - /* Clear any unconfigured MCR's. - * This way we are sure that the centaur_mcr array contains the actual - * values. The disadvantage is that any BIOS tweaks are thus undone. - * - */ - for (i = 0; i < 8; ++i) { - if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0) - wrmsr(MSR_IDT_MCR0 + i, 0, 0); - } - - wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0); /* Write only */ -} - -/* - * Initialise Winchip series MCR registers - */ - -static void __init -centaur_mcr_init(void) -{ - struct set_mtrr_context ctxt; - - set_mtrr_prepare_save(&ctxt); - set_mtrr_cache_disable(&ctxt); - - if (boot_cpu_data.x86_model == 4) - centaur_mcr0_init(); - else if (boot_cpu_data.x86_model == 8 || boot_cpu_data.x86_model == 9) - centaur_mcr1_init(); - - set_mtrr_done(&ctxt); -} -#endif - -static int centaur_validate_add_page(unsigned long base, - unsigned long size, unsigned int type) +static int +centaur_validate_add_page(unsigned long base, unsigned long size, unsigned int type) { /* - * FIXME: Winchip2 supports uncached + * FIXME: Winchip2 supports uncached */ - if (type != MTRR_TYPE_WRCOMB && + if (type != MTRR_TYPE_WRCOMB && (centaur_mcr_type == 0 || type != MTRR_TYPE_UNCACHABLE)) { - printk(KERN_WARNING - "mtrr: only write-combining%s supported\n", - centaur_mcr_type ? " and uncacheable are" - : " is"); + pr_warning("mtrr: only write-combining%s supported\n", + centaur_mcr_type ? " and uncacheable are" : " is"); return -EINVAL; } return 0; @@ -207,7 +112,6 @@ static int centaur_validate_add_page(unsigned long base, static struct mtrr_ops centaur_mtrr_ops = { .vendor = X86_VENDOR_CENTAUR, -// .init = centaur_mcr_init, .set = centaur_set_mcr, .get = centaur_get_mcr, .get_free_region = centaur_get_free_region, @@ -220,5 +124,3 @@ int __init centaur_init_mtrr(void) set_mtrr_ops(¢aur_mtrr_ops); return 0; } - -//arch_initcall(centaur_init_mtrr); diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c index 1d584a18a50d..315738c74aad 100644 --- a/arch/x86/kernel/cpu/mtrr/cleanup.c +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -1,51 +1,75 @@ -/* MTRR (Memory Type Range Register) cleanup - - Copyright (C) 2009 Yinghai Lu - - This library is free software; you can redistribute it and/or - modify it under the terms of the GNU Library General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later version. - - This library 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 - Library General Public License for more details. - - You should have received a copy of the GNU Library General Public - License along with this library; if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -*/ - +/* + * MTRR (Memory Type Range Register) cleanup + * + * Copyright (C) 2009 Yinghai Lu + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library 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 + * Library General Public License for more details. + * + * You should have received a copy of the GNU Library General Public + * License along with this library; if not, write to the Free + * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ #include <linux/module.h> #include <linux/init.h> #include <linux/pci.h> #include <linux/smp.h> #include <linux/cpu.h> -#include <linux/mutex.h> #include <linux/sort.h> +#include <linux/mutex.h> +#include <linux/uaccess.h> +#include <linux/kvm_para.h> +#include <asm/processor.h> #include <asm/e820.h> #include <asm/mtrr.h> -#include <asm/uaccess.h> -#include <asm/processor.h> #include <asm/msr.h> -#include <asm/kvm_para.h> -#include "mtrr.h" -/* should be related to MTRR_VAR_RANGES nums */ -#define RANGE_NUM 256 +#include "mtrr.h" struct res_range { - unsigned long start; - unsigned long end; + unsigned long start; + unsigned long end; +}; + +struct var_mtrr_range_state { + unsigned long base_pfn; + unsigned long size_pfn; + mtrr_type type; +}; + +struct var_mtrr_state { + unsigned long range_startk; + unsigned long range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + unsigned int reg; }; +/* Should be related to MTRR_VAR_RANGES nums */ +#define RANGE_NUM 256 + +static struct res_range __initdata range[RANGE_NUM]; +static int __initdata nr_range; + +static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; + +static int __initdata debug_print; +#define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0) + + static int __init -add_range(struct res_range *range, int nr_range, unsigned long start, - unsigned long end) +add_range(struct res_range *range, int nr_range, + unsigned long start, unsigned long end) { - /* out of slots */ + /* Out of slots: */ if (nr_range >= RANGE_NUM) return nr_range; @@ -58,12 +82,12 @@ add_range(struct res_range *range, int nr_range, unsigned long start, } static int __init -add_range_with_merge(struct res_range *range, int nr_range, unsigned long start, - unsigned long end) +add_range_with_merge(struct res_range *range, int nr_range, + unsigned long start, unsigned long end) { int i; - /* try to merge it with old one */ + /* Try to merge it with old one: */ for (i = 0; i < nr_range; i++) { unsigned long final_start, final_end; unsigned long common_start, common_end; @@ -84,7 +108,7 @@ add_range_with_merge(struct res_range *range, int nr_range, unsigned long start, return nr_range; } - /* need to add that */ + /* Need to add it: */ return add_range(range, nr_range, start, end); } @@ -117,7 +141,7 @@ subtract_range(struct res_range *range, unsigned long start, unsigned long end) } if (start > range[j].start && end < range[j].end) { - /* find the new spare */ + /* Find the new spare: */ for (i = 0; i < RANGE_NUM; i++) { if (range[i].end == 0) break; @@ -146,14 +170,8 @@ static int __init cmp_range(const void *x1, const void *x2) return start1 - start2; } -struct var_mtrr_range_state { - unsigned long base_pfn; - unsigned long size_pfn; - mtrr_type type; -}; - -static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; -static int __initdata debug_print; +#define BIOS_BUG_MSG KERN_WARNING \ + "WARNING: BIOS bug: VAR MTRR %d contains strange UC entry under 1M, check with your system vendor!\n" static int __init x86_get_mtrr_mem_range(struct res_range *range, int nr_range, @@ -180,7 +198,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, range[i].start, range[i].end + 1); } - /* take out UC ranges */ + /* Take out UC ranges: */ for (i = 0; i < num_var_ranges; i++) { type = range_state[i].type; if (type != MTRR_TYPE_UNCACHABLE && @@ -193,9 +211,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed && (mtrr_state.enabled & 1)) { /* Var MTRR contains UC entry below 1M? Skip it: */ - printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d " - "contains strange UC entry under 1M, check " - "with your system vendor!\n", i); + printk(BIOS_BUG_MSG, i); if (base + size <= (1<<(20-PAGE_SHIFT))) continue; size -= (1<<(20-PAGE_SHIFT)) - base; @@ -237,17 +253,13 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range, return nr_range; } -static struct res_range __initdata range[RANGE_NUM]; -static int __initdata nr_range; - #ifdef CONFIG_MTRR_SANITIZER static unsigned long __init sum_ranges(struct res_range *range, int nr_range) { - unsigned long sum; + unsigned long sum = 0; int i; - sum = 0; for (i = 0; i < nr_range; i++) sum += range[i].end + 1 - range[i].start; @@ -278,17 +290,9 @@ static int __init mtrr_cleanup_debug_setup(char *str) } early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); -struct var_mtrr_state { - unsigned long range_startk; - unsigned long range_sizek; - unsigned long chunk_sizek; - unsigned long gran_sizek; - unsigned int reg; -}; - static void __init set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type, unsigned int address_bits) + unsigned char type, unsigned int address_bits) { u32 base_lo, base_hi, mask_lo, mask_hi; u64 base, mask; @@ -301,7 +305,7 @@ set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, mask = (1ULL << address_bits) - 1; mask &= ~((((u64)sizek) << 10) - 1); - base = ((u64)basek) << 10; + base = ((u64)basek) << 10; base |= type; mask |= 0x800; @@ -317,15 +321,14 @@ set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, static void __init save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type) + unsigned char type) { range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); range_state[reg].type = type; } -static void __init -set_var_mtrr_all(unsigned int address_bits) +static void __init set_var_mtrr_all(unsigned int address_bits) { unsigned long basek, sizek; unsigned char type; @@ -342,11 +345,11 @@ set_var_mtrr_all(unsigned int address_bits) static unsigned long to_size_factor(unsigned long sizek, char *factorp) { - char factor; unsigned long base = sizek; + char factor; if (base & ((1<<10) - 1)) { - /* not MB alignment */ + /* Not MB-aligned: */ factor = 'K'; } else if (base & ((1<<20) - 1)) { factor = 'M'; @@ -372,11 +375,12 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk, unsigned long max_align, align; unsigned long sizek; - /* Compute the maximum size I can make a range */ + /* Compute the maximum size with which we can make a range: */ if (range_startk) max_align = ffs(range_startk) - 1; else max_align = 32; + align = fls(range_sizek) - 1; if (align > max_align) align = max_align; @@ -386,11 +390,10 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk, char start_factor = 'K', size_factor = 'K'; unsigned long start_base, size_base; - start_base = to_size_factor(range_startk, - &start_factor), - size_base = to_size_factor(sizek, &size_factor), + start_base = to_size_factor(range_startk, &start_factor); + size_base = to_size_factor(sizek, &size_factor); - printk(KERN_DEBUG "Setting variable MTRR %d, " + Dprintk("Setting variable MTRR %d, " "base: %ld%cB, range: %ld%cB, type %s\n", reg, start_base, start_factor, size_base, size_factor, @@ -425,10 +428,11 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, chunk_sizek = state->chunk_sizek; gran_sizek = state->gran_sizek; - /* align with gran size, prevent small block used up MTRRs */ + /* Align with gran size, prevent small block used up MTRRs: */ range_basek = ALIGN(state->range_startk, gran_sizek); if ((range_basek > basek) && basek) return second_sizek; + state->range_sizek -= (range_basek - state->range_startk); range_sizek = ALIGN(state->range_sizek, gran_sizek); @@ -439,22 +443,21 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, } state->range_sizek = range_sizek; - /* try to append some small hole */ + /* Try to append some small hole: */ range0_basek = state->range_startk; range0_sizek = ALIGN(state->range_sizek, chunk_sizek); - /* no increase */ + /* No increase: */ if (range0_sizek == state->range_sizek) { - if (debug_print) - printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + state->range_sizek)<<10); + Dprintk("rangeX: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + state->range_sizek)<<10); state->reg = range_to_mtrr(state->reg, range0_basek, state->range_sizek, MTRR_TYPE_WRBACK); return 0; } - /* only cut back, when it is not the last */ + /* Only cut back when it is not the last: */ if (sizek) { while (range0_basek + range0_sizek > (basek + sizek)) { if (range0_sizek >= chunk_sizek) @@ -470,16 +473,16 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, second_try: range_basek = range0_basek + range0_sizek; - /* one hole in the middle */ + /* One hole in the middle: */ if (range_basek > basek && range_basek <= (basek + sizek)) second_sizek = range_basek - basek; if (range0_sizek > state->range_sizek) { - /* one hole in middle or at end */ + /* One hole in middle or at the end: */ hole_sizek = range0_sizek - state->range_sizek - second_sizek; - /* hole size should be less than half of range0 size */ + /* Hole size should be less than half of range0 size: */ if (hole_sizek >= (range0_sizek >> 1) && range0_sizek >= chunk_sizek) { range0_sizek -= chunk_sizek; @@ -491,32 +494,30 @@ second_try: } if (range0_sizek) { - if (debug_print) - printk(KERN_DEBUG "range0: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + range0_sizek)<<10); + Dprintk("range0: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + range0_sizek)<<10); state->reg = range_to_mtrr(state->reg, range0_basek, range0_sizek, MTRR_TYPE_WRBACK); } if (range0_sizek < state->range_sizek) { - /* need to handle left over */ + /* Need to handle left over range: */ range_sizek = state->range_sizek - range0_sizek; - if (debug_print) - printk(KERN_DEBUG "range: %016lx - %016lx\n", - range_basek<<10, - (range_basek + range_sizek)<<10); + Dprintk("range: %016lx - %016lx\n", + range_basek<<10, + (range_basek + range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range_basek, range_sizek, MTRR_TYPE_WRBACK); } if (hole_sizek) { hole_basek = range_basek - hole_sizek - second_sizek; - if (debug_print) - printk(KERN_DEBUG "hole: %016lx - %016lx\n", - hole_basek<<10, - (hole_basek + hole_sizek)<<10); + Dprintk("hole: %016lx - %016lx\n", + hole_basek<<10, + (hole_basek + hole_sizek)<<10); state->reg = range_to_mtrr(state->reg, hole_basek, hole_sizek, MTRR_TYPE_UNCACHABLE); } @@ -537,23 +538,23 @@ set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, basek = base_pfn << (PAGE_SHIFT - 10); sizek = size_pfn << (PAGE_SHIFT - 10); - /* See if I can merge with the last range */ + /* See if I can merge with the last range: */ if ((basek <= 1024) || (state->range_startk + state->range_sizek == basek)) { unsigned long endk = basek + sizek; state->range_sizek = endk - state->range_startk; return; } - /* Write the range mtrrs */ + /* Write the range mtrrs: */ if (state->range_sizek != 0) second_sizek = range_to_mtrr_with_hole(state, basek, sizek); - /* Allocate an msr */ + /* Allocate an msr: */ state->range_startk = basek + second_sizek; state->range_sizek = sizek - second_sizek; } -/* mininum size of mtrr block that can take hole */ +/* Mininum size of mtrr block that can take hole: */ static u64 mtrr_chunk_size __initdata = (256ULL<<20); static int __init parse_mtrr_chunk_size_opt(char *p) @@ -565,7 +566,7 @@ static int __init parse_mtrr_chunk_size_opt(char *p) } early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); -/* granity of mtrr of block */ +/* Granularity of mtrr of block: */ static u64 mtrr_gran_size __initdata; static int __init parse_mtrr_gran_size_opt(char *p) @@ -577,7 +578,7 @@ static int __init parse_mtrr_gran_size_opt(char *p) } early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); -static int nr_mtrr_spare_reg __initdata = +static unsigned long nr_mtrr_spare_reg __initdata = CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; static int __init parse_mtrr_spare_reg(char *arg) @@ -586,7 +587,6 @@ static int __init parse_mtrr_spare_reg(char *arg) nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); return 0; } - early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); static int __init @@ -594,8 +594,8 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range, u64 chunk_size, u64 gran_size) { struct var_mtrr_state var_state; - int i; int num_reg; + int i; var_state.range_startk = 0; var_state.range_sizek = 0; @@ -605,17 +605,18 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range, memset(range_state, 0, sizeof(range_state)); - /* Write the range etc */ - for (i = 0; i < nr_range; i++) + /* Write the range: */ + for (i = 0; i < nr_range; i++) { set_var_mtrr_range(&var_state, range[i].start, range[i].end - range[i].start + 1); + } - /* Write the last range */ + /* Write the last range: */ if (var_state.range_sizek != 0) range_to_mtrr_with_hole(&var_state, 0, 0); num_reg = var_state.reg; - /* Clear out the extra MTRR's */ + /* Clear out the extra MTRR's: */ while (var_state.reg < num_var_ranges) { save_var_mtrr(var_state.reg, 0, 0, 0); var_state.reg++; @@ -625,11 +626,11 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range, } struct mtrr_cleanup_result { - unsigned long gran_sizek; - unsigned long chunk_sizek; - unsigned long lose_cover_sizek; - unsigned int num_reg; - int bad; + unsigned long gran_sizek; + unsigned long chunk_sizek; + unsigned long lose_cover_sizek; + unsigned int num_reg; + int bad; }; /* @@ -645,10 +646,10 @@ static unsigned long __initdata min_loss_pfn[RANGE_NUM]; static void __init print_out_mtrr_range_state(void) { - int i; char start_factor = 'K', size_factor = 'K'; unsigned long start_base, size_base; mtrr_type type; + int i; for (i = 0; i < num_var_ranges; i++) { @@ -676,10 +677,10 @@ static int __init mtrr_need_cleanup(void) int i; mtrr_type type; unsigned long size; - /* extra one for all 0 */ + /* Extra one for all 0: */ int num[MTRR_NUM_TYPES + 1]; - /* check entries number */ + /* Check entries number: */ memset(num, 0, sizeof(num)); for (i = 0; i < num_var_ranges; i++) { type = range_state[i].type; @@ -693,88 +694,86 @@ static int __init mtrr_need_cleanup(void) num[type]++; } - /* check if we got UC entries */ + /* Check if we got UC entries: */ if (!num[MTRR_TYPE_UNCACHABLE]) return 0; - /* check if we only had WB and UC */ + /* Check if we only had WB and UC */ if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != - num_var_ranges - num[MTRR_NUM_TYPES]) + num_var_ranges - num[MTRR_NUM_TYPES]) return 0; return 1; } static unsigned long __initdata range_sums; -static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, - unsigned long extra_remove_base, - unsigned long extra_remove_size, - int i) + +static void __init +mtrr_calc_range_state(u64 chunk_size, u64 gran_size, + unsigned long x_remove_base, + unsigned long x_remove_size, int i) { - int num_reg; static struct res_range range_new[RANGE_NUM]; - static int nr_range_new; unsigned long range_sums_new; + static int nr_range_new; + int num_reg; - /* convert ranges to var ranges state */ - num_reg = x86_setup_var_mtrrs(range, nr_range, - chunk_size, gran_size); + /* Convert ranges to var ranges state: */ + num_reg = x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); - /* we got new setting in range_state, check it */ + /* We got new setting in range_state, check it: */ memset(range_new, 0, sizeof(range_new)); nr_range_new = x86_get_mtrr_mem_range(range_new, 0, - extra_remove_base, extra_remove_size); + x_remove_base, x_remove_size); range_sums_new = sum_ranges(range_new, nr_range_new); result[i].chunk_sizek = chunk_size >> 10; result[i].gran_sizek = gran_size >> 10; result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { - result[i].lose_cover_sizek = - (range_sums_new - range_sums) << PSHIFT; + result[i].lose_cover_sizek = (range_sums_new - range_sums) << PSHIFT; result[i].bad = 1; - } else - result[i].lose_cover_sizek = - (range_sums - range_sums_new) << PSHIFT; + } else { + result[i].lose_cover_sizek = (range_sums - range_sums_new) << PSHIFT; + } - /* double check it */ + /* Double check it: */ if (!result[i].bad && !result[i].lose_cover_sizek) { - if (nr_range_new != nr_range || - memcmp(range, range_new, sizeof(range))) - result[i].bad = 1; + if (nr_range_new != nr_range || memcmp(range, range_new, sizeof(range))) + result[i].bad = 1; } - if (!result[i].bad && (range_sums - range_sums_new < - min_loss_pfn[num_reg])) { - min_loss_pfn[num_reg] = - range_sums - range_sums_new; - } + if (!result[i].bad && (range_sums - range_sums_new < min_loss_pfn[num_reg])) + min_loss_pfn[num_reg] = range_sums - range_sums_new; } static void __init mtrr_print_out_one_result(int i) { - char gran_factor, chunk_factor, lose_factor; unsigned long gran_base, chunk_base, lose_base; + char gran_factor, chunk_factor, lose_factor; gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", - result[i].bad ? "*BAD*" : " ", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", - result[i].num_reg, result[i].bad ? "-" : "", - lose_base, lose_factor); + + pr_info("%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad ? "*BAD*" : " ", + gran_base, gran_factor, chunk_base, chunk_factor); + pr_cont("num_reg: %d \tlose cover RAM: %s%ld%c\n", + result[i].num_reg, result[i].bad ? "-" : "", + lose_base, lose_factor); } static int __init mtrr_search_optimal_index(void) { - int i; int num_reg_good; int index_good; + int i; if (nr_mtrr_spare_reg >= num_var_ranges) nr_mtrr_spare_reg = num_var_ranges - 1; + num_reg_good = -1; for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { if (!min_loss_pfn[i]) @@ -796,24 +795,24 @@ static int __init mtrr_search_optimal_index(void) return index_good; } - int __init mtrr_cleanup(unsigned address_bits) { - unsigned long extra_remove_base, extra_remove_size; + unsigned long x_remove_base, x_remove_size; unsigned long base, size, def, dummy; - mtrr_type type; u64 chunk_size, gran_size; + mtrr_type type; int index_good; int i; if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) return 0; + rdmsr(MSR_MTRRdefType, def, dummy); def &= 0xff; if (def != MTRR_TYPE_UNCACHABLE) return 0; - /* get it and store it aside */ + /* Get it and store it aside: */ memset(range_state, 0, sizeof(range_state)); for (i = 0; i < num_var_ranges; i++) { mtrr_if->get(i, &base, &size, &type); @@ -822,29 +821,28 @@ int __init mtrr_cleanup(unsigned address_bits) range_state[i].type = type; } - /* check if we need handle it and can handle it */ + /* Check if we need handle it and can handle it: */ if (!mtrr_need_cleanup()) return 0; - /* print original var MTRRs at first, for debugging: */ + /* Print original var MTRRs at first, for debugging: */ printk(KERN_DEBUG "original variable MTRRs\n"); print_out_mtrr_range_state(); memset(range, 0, sizeof(range)); - extra_remove_size = 0; - extra_remove_base = 1 << (32 - PAGE_SHIFT); + x_remove_size = 0; + x_remove_base = 1 << (32 - PAGE_SHIFT); if (mtrr_tom2) - extra_remove_size = - (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; - nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, - extra_remove_size); + x_remove_size = (mtrr_tom2 >> PAGE_SHIFT) - x_remove_base; + + nr_range = x86_get_mtrr_mem_range(range, 0, x_remove_base, x_remove_size); /* - * [0, 1M) should always be coverred by var mtrr with WB - * and fixed mtrrs should take effective before var mtrr for it + * [0, 1M) should always be covered by var mtrr with WB + * and fixed mtrrs should take effect before var mtrr for it: */ nr_range = add_range_with_merge(range, nr_range, 0, (1ULL<<(20 - PAGE_SHIFT)) - 1); - /* sort the ranges */ + /* Sort the ranges: */ sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); range_sums = sum_ranges(range, nr_range); @@ -854,7 +852,7 @@ int __init mtrr_cleanup(unsigned address_bits) if (mtrr_chunk_size && mtrr_gran_size) { i = 0; mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, - extra_remove_base, extra_remove_size, i); + x_remove_base, x_remove_size, i); mtrr_print_out_one_result(i); @@ -880,7 +878,7 @@ int __init mtrr_cleanup(unsigned address_bits) continue; mtrr_calc_range_state(chunk_size, gran_size, - extra_remove_base, extra_remove_size, i); + x_remove_base, x_remove_size, i); if (debug_print) { mtrr_print_out_one_result(i); printk(KERN_INFO "\n"); @@ -890,7 +888,7 @@ int __init mtrr_cleanup(unsigned address_bits) } } - /* try to find the optimal index */ + /* Try to find the optimal index: */ index_good = mtrr_search_optimal_index(); if (index_good != -1) { @@ -898,7 +896,7 @@ int __init mtrr_cleanup(unsigned address_bits) i = index_good; mtrr_print_out_one_result(i); - /* convert ranges to var ranges state */ + /* Convert ranges to var ranges state: */ chunk_size = result[i].chunk_sizek; chunk_size <<= 10; gran_size = result[i].gran_sizek; @@ -941,8 +939,8 @@ early_param("disable_mtrr_trim", disable_mtrr_trim_setup); * Note this won't check if the MTRRs < 4GB where the magic bit doesn't * apply to are wrong, but so far we don't know of any such case in the wild. */ -#define Tom2Enabled (1U << 21) -#define Tom2ForceMemTypeWB (1U << 22) +#define Tom2Enabled (1U << 21) +#define Tom2ForceMemTypeWB (1U << 22) int __init amd_special_default_mtrr(void) { @@ -952,7 +950,7 @@ int __init amd_special_default_mtrr(void) return 0; if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) return 0; - /* In case some hypervisor doesn't pass SYSCFG through */ + /* In case some hypervisor doesn't pass SYSCFG through: */ if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) return 0; /* @@ -965,19 +963,21 @@ int __init amd_special_default_mtrr(void) return 0; } -static u64 __init real_trim_memory(unsigned long start_pfn, - unsigned long limit_pfn) +static u64 __init +real_trim_memory(unsigned long start_pfn, unsigned long limit_pfn) { u64 trim_start, trim_size; + trim_start = start_pfn; trim_start <<= PAGE_SHIFT; + trim_size = limit_pfn; trim_size <<= PAGE_SHIFT; trim_size -= trim_start; - return e820_update_range(trim_start, trim_size, E820_RAM, - E820_RESERVED); + return e820_update_range(trim_start, trim_size, E820_RAM, E820_RESERVED); } + /** * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs * @end_pfn: ending page frame number @@ -985,7 +985,7 @@ static u64 __init real_trim_memory(unsigned long start_pfn, * Some buggy BIOSes don't setup the MTRRs properly for systems with certain * memory configurations. This routine checks that the highest MTRR matches * the end of memory, to make sure the MTRRs having a write back type cover - * all of the memory the kernel is intending to use. If not, it'll trim any + * all of the memory the kernel is intending to use. If not, it'll trim any * memory off the end by adjusting end_pfn, removing it from the kernel's * allocation pools, warning the user with an obnoxious message. */ @@ -994,21 +994,22 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) unsigned long i, base, size, highest_pfn = 0, def, dummy; mtrr_type type; u64 total_trim_size; - /* extra one for all 0 */ int num[MTRR_NUM_TYPES + 1]; + /* * Make sure we only trim uncachable memory on machines that * support the Intel MTRR architecture: */ if (!is_cpu(INTEL) || disable_mtrr_trim) return 0; + rdmsr(MSR_MTRRdefType, def, dummy); def &= 0xff; if (def != MTRR_TYPE_UNCACHABLE) return 0; - /* get it and store it aside */ + /* Get it and store it aside: */ memset(range_state, 0, sizeof(range_state)); for (i = 0; i < num_var_ranges; i++) { mtrr_if->get(i, &base, &size, &type); @@ -1017,7 +1018,7 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) range_state[i].type = type; } - /* Find highest cached pfn */ + /* Find highest cached pfn: */ for (i = 0; i < num_var_ranges; i++) { type = range_state[i].type; if (type != MTRR_TYPE_WRBACK) @@ -1028,13 +1029,13 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) highest_pfn = base + size; } - /* kvm/qemu doesn't have mtrr set right, don't trim them all */ + /* kvm/qemu doesn't have mtrr set right, don't trim them all: */ if (!highest_pfn) { printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); return 0; } - /* check entries number */ + /* Check entries number: */ memset(num, 0, sizeof(num)); for (i = 0; i < num_var_ranges; i++) { type = range_state[i].type; @@ -1046,11 +1047,11 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) num[type]++; } - /* no entry for WB? */ + /* No entry for WB? */ if (!num[MTRR_TYPE_WRBACK]) return 0; - /* check if we only had WB and UC */ + /* Check if we only had WB and UC: */ if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != num_var_ranges - num[MTRR_NUM_TYPES]) return 0; @@ -1066,31 +1067,31 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) } nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); + /* Check the head: */ total_trim_size = 0; - /* check the head */ if (range[0].start) total_trim_size += real_trim_memory(0, range[0].start); - /* check the holes */ + + /* Check the holes: */ for (i = 0; i < nr_range - 1; i++) { if (range[i].end + 1 < range[i+1].start) total_trim_size += real_trim_memory(range[i].end + 1, range[i+1].start); } - /* check the top */ + + /* Check the top: */ i = nr_range - 1; if (range[i].end + 1 < end_pfn) total_trim_size += real_trim_memory(range[i].end + 1, end_pfn); if (total_trim_size) { - printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" - " all of memory, losing %lluMB of RAM.\n", - total_trim_size >> 20); + pr_warning("WARNING: BIOS bug: CPU MTRRs don't cover all of memory, losing %lluMB of RAM.\n", total_trim_size >> 20); if (!changed_by_mtrr_cleanup) WARN_ON(1); - printk(KERN_INFO "update e820 for mtrr\n"); + pr_info("update e820 for mtrr\n"); update_e820(); return 1; @@ -1098,4 +1099,3 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) return 0; } - diff --git a/arch/x86/kernel/cpu/mtrr/cyrix.c b/arch/x86/kernel/cpu/mtrr/cyrix.c index ff14c320040c..228d982ce09c 100644 --- a/arch/x86/kernel/cpu/mtrr/cyrix.c +++ b/arch/x86/kernel/cpu/mtrr/cyrix.c @@ -1,38 +1,40 @@ #include <linux/init.h> +#include <linux/io.h> #include <linux/mm.h> -#include <asm/mtrr.h> -#include <asm/msr.h> -#include <asm/io.h> + #include <asm/processor-cyrix.h> #include <asm/processor-flags.h> +#include <asm/mtrr.h> +#include <asm/msr.h> + #include "mtrr.h" static void cyrix_get_arr(unsigned int reg, unsigned long *base, unsigned long *size, mtrr_type * type) { - unsigned long flags; unsigned char arr, ccr3, rcr, shift; + unsigned long flags; arr = CX86_ARR_BASE + (reg << 1) + reg; /* avoid multiplication by 3 */ - /* Save flags and disable interrupts */ local_irq_save(flags); ccr3 = getCx86(CX86_CCR3); setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ - ((unsigned char *) base)[3] = getCx86(arr); - ((unsigned char *) base)[2] = getCx86(arr + 1); - ((unsigned char *) base)[1] = getCx86(arr + 2); + ((unsigned char *)base)[3] = getCx86(arr); + ((unsigned char *)base)[2] = getCx86(arr + 1); + ((unsigned char *)base)[1] = getCx86(arr + 2); rcr = getCx86(CX86_RCR_BASE + reg); - setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ + setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ - /* Enable interrupts if it was enabled previously */ local_irq_restore(flags); + shift = ((unsigned char *) base)[1] & 0x0f; *base >>= PAGE_SHIFT; - /* Power of two, at least 4K on ARR0-ARR6, 256K on ARR7 + /* + * Power of two, at least 4K on ARR0-ARR6, 256K on ARR7 * Note: shift==0xf means 4G, this is unsupported. */ if (shift) @@ -76,17 +78,20 @@ cyrix_get_arr(unsigned int reg, unsigned long *base, } } +/* + * cyrix_get_free_region - get a free ARR. + * + * @base: the starting (base) address of the region. + * @size: the size (in bytes) of the region. + * + * Returns: the index of the region on success, else -1 on error. +*/ static int cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg) -/* [SUMMARY] Get a free ARR. - <base> The starting (base) address of the region. - <size> The size (in bytes) of the region. - [RETURNS] The index of the region on success, else -1 on error. -*/ { - int i; - mtrr_type ltype; unsigned long lbase, lsize; + mtrr_type ltype; + int i; switch (replace_reg) { case 7: @@ -107,14 +112,17 @@ cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg) cyrix_get_arr(7, &lbase, &lsize, <ype); if (lsize == 0) return 7; - /* Else try ARR0-ARR6 first */ + /* Else try ARR0-ARR6 first */ } else { for (i = 0; i < 7; i++) { cyrix_get_arr(i, &lbase, &lsize, <ype); if (lsize == 0) return i; } - /* ARR0-ARR6 isn't free, try ARR7 but its size must be at least 256K */ + /* + * ARR0-ARR6 isn't free + * try ARR7 but its size must be at least 256K + */ cyrix_get_arr(i, &lbase, &lsize, <ype); if ((lsize == 0) && (size >= 0x40)) return i; @@ -122,21 +130,22 @@ cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg) return -ENOSPC; } -static u32 cr4 = 0; -static u32 ccr3; +static u32 cr4, ccr3; static void prepare_set(void) { u32 cr0; /* Save value of CR4 and clear Page Global Enable (bit 7) */ - if ( cpu_has_pge ) { + if (cpu_has_pge) { cr4 = read_cr4(); write_cr4(cr4 & ~X86_CR4_PGE); } - /* Disable and flush caches. Note that wbinvd flushes the TLBs as - a side-effect */ + /* + * Disable and flush caches. + * Note that wbinvd flushes the TLBs as a side-effect + */ cr0 = read_cr0() | X86_CR0_CD; wbinvd(); write_cr0(cr0); @@ -147,22 +156,21 @@ static void prepare_set(void) /* Cyrix ARRs - everything else was excluded at the top */ setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); - } static void post_set(void) { - /* Flush caches and TLBs */ + /* Flush caches and TLBs */ wbinvd(); /* Cyrix ARRs - everything else was excluded at the top */ setCx86(CX86_CCR3, ccr3); - - /* Enable caches */ + + /* Enable caches */ write_cr0(read_cr0() & 0xbfffffff); - /* Restore value of CR4 */ - if ( cpu_has_pge ) + /* Restore value of CR4 */ + if (cpu_has_pge) write_cr4(cr4); } @@ -178,7 +186,8 @@ static void cyrix_set_arr(unsigned int reg, unsigned long base, size >>= 6; size &= 0x7fff; /* make sure arr_size <= 14 */ - for (arr_size = 0; size; arr_size++, size >>= 1) ; + for (arr_size = 0; size; arr_size++, size >>= 1) + ; if (reg < 7) { switch (type) { @@ -215,18 +224,18 @@ static void cyrix_set_arr(unsigned int reg, unsigned long base, prepare_set(); base <<= PAGE_SHIFT; - setCx86(arr, ((unsigned char *) &base)[3]); - setCx86(arr + 1, ((unsigned char *) &base)[2]); - setCx86(arr + 2, (((unsigned char *) &base)[1]) | arr_size); + setCx86(arr + 0, ((unsigned char *)&base)[3]); + setCx86(arr + 1, ((unsigned char *)&base)[2]); + setCx86(arr + 2, (((unsigned char *)&base)[1]) | arr_size); setCx86(CX86_RCR_BASE + reg, arr_type); post_set(); } typedef struct { - unsigned long base; - unsigned long size; - mtrr_type type; + unsigned long base; + unsigned long size; + mtrr_type type; } arr_state_t; static arr_state_t arr_state[8] = { @@ -247,16 +256,17 @@ static void cyrix_set_all(void) setCx86(CX86_CCR0 + i, ccr_state[i]); for (; i < 7; i++) setCx86(CX86_CCR4 + i, ccr_state[i]); - for (i = 0; i < 8; i++) - cyrix_set_arr(i, arr_state[i].base, + + for (i = 0; i < 8; i++) { + cyrix_set_arr(i, arr_state[i].base, arr_state[i].size, arr_state[i].type); + } post_set(); } static struct mtrr_ops cyrix_mtrr_ops = { .vendor = X86_VENDOR_CYRIX, -// .init = cyrix_arr_init, .set_all = cyrix_set_all, .set = cyrix_set_arr, .get = cyrix_get_arr, @@ -270,5 +280,3 @@ int __init cyrix_init_mtrr(void) set_mtrr_ops(&cyrix_mtrr_ops); return 0; } - -//arch_initcall(cyrix_init_mtrr); diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 0543f69f0b27..55da0c5f68dd 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -1,28 +1,34 @@ -/* This only handles 32bit MTRR on 32bit hosts. This is strictly wrong - because MTRRs can span upto 40 bits (36bits on most modern x86) */ +/* + * This only handles 32bit MTRR on 32bit hosts. This is strictly wrong + * because MTRRs can span upto 40 bits (36bits on most modern x86) + */ +#define DEBUG + +#include <linux/module.h> #include <linux/init.h> #include <linux/slab.h> +#include <linux/io.h> #include <linux/mm.h> -#include <linux/module.h> -#include <asm/io.h> -#include <asm/mtrr.h> -#include <asm/msr.h> -#include <asm/system.h> -#include <asm/cpufeature.h> + #include <asm/processor-flags.h> +#include <asm/cpufeature.h> #include <asm/tlbflush.h> +#include <asm/system.h> +#include <asm/mtrr.h> +#include <asm/msr.h> #include <asm/pat.h> + #include "mtrr.h" struct fixed_range_block { - int base_msr; /* start address of an MTRR block */ - int ranges; /* number of MTRRs in this block */ + int base_msr; /* start address of an MTRR block */ + int ranges; /* number of MTRRs in this block */ }; static struct fixed_range_block fixed_range_blocks[] = { - { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */ - { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */ - { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */ + { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */ + { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */ + { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */ {} }; @@ -30,10 +36,10 @@ static unsigned long smp_changes_mask; static int mtrr_state_set; u64 mtrr_tom2; -struct mtrr_state_type mtrr_state = {}; +struct mtrr_state_type mtrr_state; EXPORT_SYMBOL_GPL(mtrr_state); -/** +/* * BIOS is expected to clear MtrrFixDramModEn bit, see for example * "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD * Opteron Processors" (26094 Rev. 3.30 February 2006), section @@ -104,9 +110,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) * Look of multiple ranges matching this address and pick type * as per MTRR precedence */ - if (!(mtrr_state.enabled & 2)) { + if (!(mtrr_state.enabled & 2)) return mtrr_state.def_type; - } prev_match = 0xFF; for (i = 0; i < num_var_ranges; ++i) { @@ -125,9 +130,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) if (start_state != end_state) return 0xFE; - if ((start & mask) != (base & mask)) { + if ((start & mask) != (base & mask)) continue; - } curr_match = mtrr_state.var_ranges[i].base_lo & 0xff; if (prev_match == 0xFF) { @@ -148,9 +152,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) curr_match = MTRR_TYPE_WRTHROUGH; } - if (prev_match != curr_match) { + if (prev_match != curr_match) return MTRR_TYPE_UNCACHABLE; - } } if (mtrr_tom2) { @@ -164,7 +167,7 @@ u8 mtrr_type_lookup(u64 start, u64 end) return mtrr_state.def_type; } -/* Get the MSR pair relating to a var range */ +/* Get the MSR pair relating to a var range */ static void get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) { @@ -172,7 +175,7 @@ get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) rdmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi); } -/* fill the MSR pair relating to a var range */ +/* Fill the MSR pair relating to a var range */ void fill_mtrr_var_range(unsigned int index, u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi) { @@ -186,10 +189,9 @@ void fill_mtrr_var_range(unsigned int index, vr[index].mask_hi = mask_hi; } -static void -get_fixed_ranges(mtrr_type * frs) +static void get_fixed_ranges(mtrr_type *frs) { - unsigned int *p = (unsigned int *) frs; + unsigned int *p = (unsigned int *)frs; int i; k8_check_syscfg_dram_mod_en(); @@ -217,22 +219,22 @@ static void __init print_fixed_last(void) if (!last_fixed_end) return; - printk(KERN_DEBUG " %05X-%05X %s\n", last_fixed_start, - last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type)); + pr_debug(" %05X-%05X %s\n", last_fixed_start, + last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type)); last_fixed_end = 0; } static void __init update_fixed_last(unsigned base, unsigned end, - mtrr_type type) + mtrr_type type) { last_fixed_start = base; last_fixed_end = end; last_fixed_type = type; } -static void __init print_fixed(unsigned base, unsigned step, - const mtrr_type *types) +static void __init +print_fixed(unsigned base, unsigned step, const mtrr_type *types) { unsigned i; @@ -259,54 +261,55 @@ static void __init print_mtrr_state(void) unsigned int i; int high_width; - printk(KERN_DEBUG "MTRR default type: %s\n", - mtrr_attrib_to_str(mtrr_state.def_type)); + pr_debug("MTRR default type: %s\n", + mtrr_attrib_to_str(mtrr_state.def_type)); if (mtrr_state.have_fixed) { - printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n", - mtrr_state.enabled & 1 ? "en" : "dis"); + pr_debug("MTRR fixed ranges %sabled:\n", + mtrr_state.enabled & 1 ? "en" : "dis"); print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); for (i = 0; i < 2; ++i) - print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); + print_fixed(0x80000 + i * 0x20000, 0x04000, + mtrr_state.fixed_ranges + (i + 1) * 8); for (i = 0; i < 8; ++i) - print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); + print_fixed(0xC0000 + i * 0x08000, 0x01000, + mtrr_state.fixed_ranges + (i + 3) * 8); /* tail */ print_fixed_last(); } - printk(KERN_DEBUG "MTRR variable ranges %sabled:\n", - mtrr_state.enabled & 2 ? "en" : "dis"); + pr_debug("MTRR variable ranges %sabled:\n", + mtrr_state.enabled & 2 ? "en" : "dis"); if (size_or_mask & 0xffffffffUL) high_width = ffs(size_or_mask & 0xffffffffUL) - 1; else high_width = ffs(size_or_mask>>32) + 32 - 1; high_width = (high_width - (32 - PAGE_SHIFT) + 3) / 4; + for (i = 0; i < num_var_ranges; ++i) { if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) - printk(KERN_DEBUG " %u base %0*X%05X000 mask %0*X%05X000 %s\n", - i, - high_width, - mtrr_state.var_ranges[i].base_hi, - mtrr_state.var_ranges[i].base_lo >> 12, - high_width, - mtrr_state.var_ranges[i].mask_hi, - mtrr_state.var_ranges[i].mask_lo >> 12, - mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); + pr_debug(" %u base %0*X%05X000 mask %0*X%05X000 %s\n", + i, + high_width, + mtrr_state.var_ranges[i].base_hi, + mtrr_state.var_ranges[i].base_lo >> 12, + high_width, + mtrr_state.var_ranges[i].mask_hi, + mtrr_state.var_ranges[i].mask_lo >> 12, + mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); else - printk(KERN_DEBUG " %u disabled\n", i); - } - if (mtrr_tom2) { - printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n", - mtrr_tom2, mtrr_tom2>>20); + pr_debug(" %u disabled\n", i); } + if (mtrr_tom2) + pr_debug("TOM2: %016llx aka %lldM\n", mtrr_tom2, mtrr_tom2>>20); } -/* Grab all of the MTRR state for this CPU into *state */ +/* Grab all of the MTRR state for this CPU into *state */ void __init get_mtrr_state(void) { - unsigned int i; struct mtrr_var_range *vrs; - unsigned lo, dummy; unsigned long flags; + unsigned lo, dummy; + unsigned int i; vrs = mtrr_state.var_ranges; @@ -324,6 +327,7 @@ void __init get_mtrr_state(void) if (amd_special_default_mtrr()) { unsigned low, high; + /* TOP_MEM2 */ rdmsr(MSR_K8_TOP_MEM2, low, high); mtrr_tom2 = high; @@ -344,10 +348,9 @@ void __init get_mtrr_state(void) post_set(); local_irq_restore(flags); - } -/* Some BIOS's are fucked and don't set all MTRRs the same! */ +/* Some BIOS's are messed up and don't set all MTRRs the same! */ void __init mtrr_state_warn(void) { unsigned long mask = smp_changes_mask; @@ -355,28 +358,33 @@ void __init mtrr_state_warn(void) if (!mask) return; if (mask & MTRR_CHANGE_MASK_FIXED) - printk(KERN_WARNING "mtrr: your CPUs had inconsistent fixed MTRR settings\n"); + pr_warning("mtrr: your CPUs had inconsistent fixed MTRR settings\n"); if (mask & MTRR_CHANGE_MASK_VARIABLE) - printk(KERN_WARNING "mtrr: your CPUs had inconsistent variable MTRR settings\n"); + pr_warning("mtrr: your CPUs had inconsistent variable MTRR settings\n"); if (mask & MTRR_CHANGE_MASK_DEFTYPE) - printk(KERN_WARNING "mtrr: your CPUs had inconsistent MTRRdefType settings\n"); + pr_warning("mtrr: your CPUs had inconsistent MTRRdefType settings\n"); + printk(KERN_INFO "mtrr: probably your BIOS does not setup all CPUs.\n"); printk(KERN_INFO "mtrr: corrected configuration.\n"); } -/* Doesn't attempt to pass an error out to MTRR users - because it's quite complicated in some cases and probably not - worth it because the best error handling is to ignore it. */ +/* + * Doesn't attempt to pass an error out to MTRR users + * because it's quite complicated in some cases and probably not + * worth it because the best error handling is to ignore it. + */ void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b) { - if (wrmsr_safe(msr, a, b) < 0) + if (wrmsr_safe(msr, a, b) < 0) { printk(KERN_ERR "MTRR: CPU %u: Writing MSR %x to %x:%x failed\n", smp_processor_id(), msr, a, b); + } } /** - * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have + * set_fixed_range - checks & updates a fixed-range MTRR if it + * differs from the value it should have * @msr: MSR address of the MTTR which should be checked and updated * @changed: pointer which indicates whether the MTRR needed to be changed * @msrwords: pointer to the MSR values which the MSR should have @@ -401,20 +409,23 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) * * Returns: The index of the region on success, else negative on error. */ -int generic_get_free_region(unsigned long base, unsigned long size, int replace_reg) +int +generic_get_free_region(unsigned long base, unsigned long size, int replace_reg) { - int i, max; - mtrr_type ltype; unsigned long lbase, lsize; + mtrr_type ltype; + int i, max; max = num_var_ranges; if (replace_reg >= 0 && replace_reg < max) return replace_reg; + for (i = 0; i < max; ++i) { mtrr_if->get(i, &lbase, &lsize, <ype); if (lsize == 0) return i; } + return -ENOSPC; } @@ -434,7 +445,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi); if ((mask_lo & 0x800) == 0) { - /* Invalid (i.e. free) range */ + /* Invalid (i.e. free) range */ *base = 0; *size = 0; *type = 0; @@ -471,27 +482,31 @@ out_put_cpu: } /** - * set_fixed_ranges - checks & updates the fixed-range MTRRs if they differ from the saved set + * set_fixed_ranges - checks & updates the fixed-range MTRRs if they + * differ from the saved set * @frs: pointer to fixed-range MTRR values, saved by get_fixed_ranges() */ -static int set_fixed_ranges(mtrr_type * frs) +static int set_fixed_ranges(mtrr_type *frs) { - unsigned long long *saved = (unsigned long long *) frs; + unsigned long long *saved = (unsigned long long *)frs; bool changed = false; - int block=-1, range; + int block = -1, range; k8_check_syscfg_dram_mod_en(); - while (fixed_range_blocks[++block].ranges) - for (range=0; range < fixed_range_blocks[block].ranges; range++) - set_fixed_range(fixed_range_blocks[block].base_msr + range, - &changed, (unsigned int *) saved++); + while (fixed_range_blocks[++block].ranges) { + for (range = 0; range < fixed_range_blocks[block].ranges; range++) + set_fixed_range(fixed_range_blocks[block].base_msr + range, + &changed, (unsigned int *)saved++); + } return changed; } -/* Set the MSR pair relating to a var range. Returns TRUE if - changes are made */ +/* + * Set the MSR pair relating to a var range. + * Returns true if changes are made. + */ static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr) { unsigned int lo, hi; @@ -501,6 +516,7 @@ static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr) if ((vr->base_lo & 0xfffff0ffUL) != (lo & 0xfffff0ffUL) || (vr->base_hi & (size_and_mask >> (32 - PAGE_SHIFT))) != (hi & (size_and_mask >> (32 - PAGE_SHIFT)))) { + mtrr_wrmsr(MTRRphysBase_MSR(index), vr->base_lo, vr->base_hi); changed = true; } @@ -526,21 +542,26 @@ static u32 deftype_lo, deftype_hi; */ static unsigned long set_mtrr_state(void) { - unsigned int i; unsigned long change_mask = 0; + unsigned int i; - for (i = 0; i < num_var_ranges; i++) + for (i = 0; i < num_var_ranges; i++) { if (set_mtrr_var_ranges(i, &mtrr_state.var_ranges[i])) change_mask |= MTRR_CHANGE_MASK_VARIABLE; + } if (mtrr_state.have_fixed && set_fixed_ranges(mtrr_state.fixed_ranges)) change_mask |= MTRR_CHANGE_MASK_FIXED; - /* Set_mtrr_restore restores the old value of MTRRdefType, - so to set it we fiddle with the saved value */ + /* + * Set_mtrr_restore restores the old value of MTRRdefType, + * so to set it we fiddle with the saved value: + */ if ((deftype_lo & 0xff) != mtrr_state.def_type || ((deftype_lo & 0xc00) >> 10) != mtrr_state.enabled) { - deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type | (mtrr_state.enabled << 10); + + deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type | + (mtrr_state.enabled << 10); change_mask |= MTRR_CHANGE_MASK_DEFTYPE; } @@ -548,33 +569,36 @@ static unsigned long set_mtrr_state(void) } -static unsigned long cr4 = 0; +static unsigned long cr4; static DEFINE_SPINLOCK(set_atomicity_lock); /* - * Since we are disabling the cache don't allow any interrupts - they - * would run extremely slow and would only increase the pain. The caller must - * ensure that local interrupts are disabled and are reenabled after post_set() - * has been called. + * Since we are disabling the cache don't allow any interrupts, + * they would run extremely slow and would only increase the pain. + * + * The caller must ensure that local interrupts are disabled and + * are reenabled after post_set() has been called. */ - static void prepare_set(void) __acquires(set_atomicity_lock) { unsigned long cr0; - /* Note that this is not ideal, since the cache is only flushed/disabled - for this CPU while the MTRRs are changed, but changing this requires - more invasive changes to the way the kernel boots */ + /* + * Note that this is not ideal + * since the cache is only flushed/disabled for this CPU while the + * MTRRs are changed, but changing this requires more invasive + * changes to the way the kernel boots + */ spin_lock(&set_atomicity_lock); - /* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */ + /* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */ cr0 = read_cr0() | X86_CR0_CD; write_cr0(cr0); wbinvd(); - /* Save value of CR4 and clear Page Global Enable (bit 7) */ - if ( cpu_has_pge ) { + /* Save value of CR4 and clear Page Global Enable (bit 7) */ + if (cpu_has_pge) { cr4 = read_cr4(); write_cr4(cr4 & ~X86_CR4_PGE); } @@ -582,26 +606,26 @@ static void prepare_set(void) __acquires(set_atomicity_lock) /* Flush all TLBs via a mov %cr3, %reg; mov %reg, %cr3 */ __flush_tlb(); - /* Save MTRR state */ + /* Save MTRR state */ rdmsr(MSR_MTRRdefType, deftype_lo, deftype_hi); - /* Disable MTRRs, and set the default type to uncached */ + /* Disable MTRRs, and set the default type to uncached */ mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi); } static void post_set(void) __releases(set_atomicity_lock) { - /* Flush TLBs (no need to flush caches - they are disabled) */ + /* Flush TLBs (no need to flush caches - they are disabled) */ __flush_tlb(); /* Intel (P6) standard MTRRs */ mtrr_wrmsr(MSR_MTRRdefType, deftype_lo, deftype_hi); - - /* Enable caches */ + + /* Enable caches */ write_cr0(read_cr0() & 0xbfffffff); - /* Restore value of CR4 */ - if ( cpu_has_pge ) + /* Restore value of CR4 */ + if (cpu_has_pge) write_cr4(cr4); spin_unlock(&set_atomicity_lock); } @@ -623,24 +647,27 @@ static void generic_set_all(void) post_set(); local_irq_restore(flags); - /* Use the atomic bitops to update the global mask */ + /* Use the atomic bitops to update the global mask */ for (count = 0; count < sizeof mask * 8; ++count) { if (mask & 0x01) set_bit(count, &smp_changes_mask); mask >>= 1; } - + } +/** + * generic_set_mtrr - set variable MTRR register on the local CPU. + * + * @reg: The register to set. + * @base: The base address of the region. + * @size: The size of the region. If this is 0 the region is disabled. + * @type: The type of the region. + * + * Returns nothing. + */ static void generic_set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type) -/* [SUMMARY] Set variable MTRR register on the local CPU. - <reg> The register to set. - <base> The base address of the region. - <size> The size of the region. If this is 0 the region is disabled. - <type> The type of the region. - [RETURNS] Nothing. -*/ { unsigned long flags; struct mtrr_var_range *vr; @@ -651,8 +678,10 @@ static void generic_set_mtrr(unsigned int reg, unsigned long base, prepare_set(); if (size == 0) { - /* The invalid bit is kept in the mask, so we simply clear the - relevant mask register to disable a range. */ + /* + * The invalid bit is kept in the mask, so we simply + * clear the relevant mask register to disable a range. + */ mtrr_wrmsr(MTRRphysMask_MSR(reg), 0, 0); memset(vr, 0, sizeof(struct mtrr_var_range)); } else { @@ -669,46 +698,50 @@ static void generic_set_mtrr(unsigned int reg, unsigned long base, local_irq_restore(flags); } -int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type) +int generic_validate_add_page(unsigned long base, unsigned long size, + unsigned int type) { unsigned long lbase, last; - /* For Intel PPro stepping <= 7, must be 4 MiB aligned - and not touch 0x70000000->0x7003FFFF */ + /* + * For Intel PPro stepping <= 7 + * must be 4 MiB aligned and not touch 0x70000000 -> 0x7003FFFF + */ if (is_cpu(INTEL) && boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 1 && boot_cpu_data.x86_mask <= 7) { if (base & ((1 << (22 - PAGE_SHIFT)) - 1)) { - printk(KERN_WARNING "mtrr: base(0x%lx000) is not 4 MiB aligned\n", base); + pr_warning("mtrr: base(0x%lx000) is not 4 MiB aligned\n", base); return -EINVAL; } if (!(base + size < 0x70000 || base > 0x7003F) && (type == MTRR_TYPE_WRCOMB || type == MTRR_TYPE_WRBACK)) { - printk(KERN_WARNING "mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n"); + pr_warning("mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n"); return -EINVAL; } } - /* Check upper bits of base and last are equal and lower bits are 0 - for base and 1 for last */ + /* + * Check upper bits of base and last are equal and lower bits are 0 + * for base and 1 for last + */ last = base + size - 1; for (lbase = base; !(lbase & 1) && (last & 1); - lbase = lbase >> 1, last = last >> 1) ; + lbase = lbase >> 1, last = last >> 1) + ; if (lbase != last) { - printk(KERN_WARNING "mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n", - base, size); + pr_warning("mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n", base, size); return -EINVAL; } return 0; } - static int generic_have_wrcomb(void) { unsigned long config, dummy; rdmsr(MSR_MTRRcap, config, dummy); - return (config & (1 << 10)); + return config & (1 << 10); } int positive_have_wrcomb(void) @@ -716,14 +749,15 @@ int positive_have_wrcomb(void) return 1; } -/* generic structure... +/* + * Generic structure... */ struct mtrr_ops generic_mtrr_ops = { - .use_intel_if = 1, - .set_all = generic_set_all, - .get = generic_get_mtrr, - .get_free_region = generic_get_free_region, - .set = generic_set_mtrr, - .validate_add_page = generic_validate_add_page, - .have_wrcomb = generic_have_wrcomb, + .use_intel_if = 1, + .set_all = generic_set_all, + .get = generic_get_mtrr, + .get_free_region = generic_get_free_region, + .set = generic_set_mtrr, + .validate_add_page = generic_validate_add_page, + .have_wrcomb = generic_have_wrcomb, }; diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index fb73a52913a4..08b6ea4c62b4 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -1,27 +1,28 @@ -#include <linux/init.h> -#include <linux/proc_fs.h> #include <linux/capability.h> -#include <linux/ctype.h> -#include <linux/module.h> #include <linux/seq_file.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> +#include <linux/proc_fs.h> +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/init.h> #define LINE_SIZE 80 #include <asm/mtrr.h> + #include "mtrr.h" #define FILE_FCOUNT(f) (((struct seq_file *)((f)->private_data))->private) static const char *const mtrr_strings[MTRR_NUM_TYPES] = { - "uncachable", /* 0 */ - "write-combining", /* 1 */ - "?", /* 2 */ - "?", /* 3 */ - "write-through", /* 4 */ - "write-protect", /* 5 */ - "write-back", /* 6 */ + "uncachable", /* 0 */ + "write-combining", /* 1 */ + "?", /* 2 */ + "?", /* 3 */ + "write-through", /* 4 */ + "write-protect", /* 5 */ + "write-back", /* 6 */ }; const char *mtrr_attrib_to_str(int x) @@ -35,8 +36,8 @@ static int mtrr_file_add(unsigned long base, unsigned long size, unsigned int type, bool increment, struct file *file, int page) { + unsigned int *fcount = FILE_FCOUNT(file); int reg, max; - unsigned int *fcount = FILE_FCOUNT(file); max = num_var_ranges; if (fcount == NULL) { @@ -61,8 +62,8 @@ static int mtrr_file_del(unsigned long base, unsigned long size, struct file *file, int page) { - int reg; unsigned int *fcount = FILE_FCOUNT(file); + int reg; if (!page) { if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) @@ -81,13 +82,14 @@ mtrr_file_del(unsigned long base, unsigned long size, return reg; } -/* RED-PEN: seq_file can seek now. this is ignored. */ +/* + * seq_file can seek but we ignore it. + * + * Format of control line: + * "base=%Lx size=%Lx type=%s" or "disable=%d" + */ static ssize_t mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) -/* Format of control line: - "base=%Lx size=%Lx type=%s" OR: - "disable=%d" -*/ { int i, err; unsigned long reg; @@ -100,15 +102,18 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) return -EPERM; if (!len) return -EINVAL; + memset(line, 0, LINE_SIZE); if (len > LINE_SIZE) len = LINE_SIZE; if (copy_from_user(line, buf, len - 1)) return -EFAULT; + linelen = strlen(line); ptr = line + linelen - 1; if (linelen && *ptr == '\n') *ptr = '\0'; + if (!strncmp(line, "disable=", 8)) { reg = simple_strtoul(line + 8, &ptr, 0); err = mtrr_del_page(reg, 0, 0); @@ -116,28 +121,35 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) return err; return len; } + if (strncmp(line, "base=", 5)) return -EINVAL; + base = simple_strtoull(line + 5, &ptr, 0); - for (; isspace(*ptr); ++ptr) ; + for (; isspace(*ptr); ++ptr) + ; + if (strncmp(ptr, "size=", 5)) return -EINVAL; + size = simple_strtoull(ptr + 5, &ptr, 0); if ((base & 0xfff) || (size & 0xfff)) return -EINVAL; - for (; isspace(*ptr); ++ptr) ; + for (; isspace(*ptr); ++ptr) + ; + if (strncmp(ptr, "type=", 5)) return -EINVAL; ptr += 5; - for (; isspace(*ptr); ++ptr) ; + for (; isspace(*ptr); ++ptr) + ; + for (i = 0; i < MTRR_NUM_TYPES; ++i) { if (strcmp(ptr, mtrr_strings[i])) continue; base >>= PAGE_SHIFT; size >>= PAGE_SHIFT; - err = - mtrr_add_page((unsigned long) base, (unsigned long) size, i, - true); + err = mtrr_add_page((unsigned long)base, (unsigned long)size, i, true); if (err < 0) return err; return len; @@ -181,7 +193,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) case MTRRIOC32_SET_PAGE_ENTRY: case MTRRIOC32_DEL_PAGE_ENTRY: case MTRRIOC32_KILL_PAGE_ENTRY: { - struct mtrr_sentry32 __user *s32 = (struct mtrr_sentry32 __user *)__arg; + struct mtrr_sentry32 __user *s32; + + s32 = (struct mtrr_sentry32 __user *)__arg; err = get_user(sentry.base, &s32->base); err |= get_user(sentry.size, &s32->size); err |= get_user(sentry.type, &s32->type); @@ -191,7 +205,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) } case MTRRIOC32_GET_ENTRY: case MTRRIOC32_GET_PAGE_ENTRY: { - struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg; + struct mtrr_gentry32 __user *g32; + + g32 = (struct mtrr_gentry32 __user *)__arg; err = get_user(gentry.regnum, &g32->regnum); err |= get_user(gentry.base, &g32->base); err |= get_user(gentry.size, &g32->size); @@ -314,7 +330,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) if (err) return err; - switch(cmd) { + switch (cmd) { case MTRRIOC_GET_ENTRY: case MTRRIOC_GET_PAGE_ENTRY: if (copy_to_user(arg, &gentry, sizeof gentry)) @@ -323,7 +339,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) #ifdef CONFIG_COMPAT case MTRRIOC32_GET_ENTRY: case MTRRIOC32_GET_PAGE_ENTRY: { - struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg; + struct mtrr_gentry32 __user *g32; + + g32 = (struct mtrr_gentry32 __user *)__arg; err = put_user(gentry.base, &g32->base); err |= put_user(gentry.size, &g32->size); err |= put_user(gentry.regnum, &g32->regnum); @@ -335,11 +353,10 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg) return err; } -static int -mtrr_close(struct inode *ino, struct file *file) +static int mtrr_close(struct inode *ino, struct file *file) { - int i, max; unsigned int *fcount = FILE_FCOUNT(file); + int i, max; if (fcount != NULL) { max = num_var_ranges; @@ -359,22 +376,22 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset); static int mtrr_open(struct inode *inode, struct file *file) { - if (!mtrr_if) + if (!mtrr_if) return -EIO; - if (!mtrr_if->get) - return -ENXIO; + if (!mtrr_if->get) + return -ENXIO; return single_open(file, mtrr_seq_show, NULL); } static const struct file_operations mtrr_fops = { - .owner = THIS_MODULE, - .open = mtrr_open, - .read = seq_read, - .llseek = seq_lseek, - .write = mtrr_write, - .unlocked_ioctl = mtrr_ioctl, - .compat_ioctl = mtrr_ioctl, - .release = mtrr_close, + .owner = THIS_MODULE, + .open = mtrr_open, + .read = seq_read, + .llseek = seq_lseek, + .write = mtrr_write, + .unlocked_ioctl = mtrr_ioctl, + .compat_ioctl = mtrr_ioctl, + .release = mtrr_close, }; static int mtrr_seq_show(struct seq_file *seq, void *offset) @@ -388,23 +405,24 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset) max = num_var_ranges; for (i = 0; i < max; i++) { mtrr_if->get(i, &base, &size, &type); - if (size == 0) + if (size == 0) { mtrr_usage_table[i] = 0; - else { - if (size < (0x100000 >> PAGE_SHIFT)) { - /* less than 1MB */ - factor = 'K'; - size <<= PAGE_SHIFT - 10; - } else { - factor = 'M'; - size >>= 20 - PAGE_SHIFT; - } - /* RED-PEN: base can be > 32bit */ - len += seq_printf(seq, - "reg%02i: base=0x%06lx000 (%5luMB), size=%5lu%cB, count=%d: %s\n", - i, base, base >> (20 - PAGE_SHIFT), size, factor, - mtrr_usage_table[i], mtrr_attrib_to_str(type)); + continue; } + if (size < (0x100000 >> PAGE_SHIFT)) { + /* less than 1MB */ + factor = 'K'; + size <<= PAGE_SHIFT - 10; + } else { + factor = 'M'; + size >>= 20 - PAGE_SHIFT; + } + /* Base can be > 32bit */ + len += seq_printf(seq, "reg%02i: base=0x%06lx000 " + "(%5luMB), size=%5lu%cB, count=%d: %s\n", + i, base, base >> (20 - PAGE_SHIFT), size, + factor, mtrr_usage_table[i], + mtrr_attrib_to_str(type)); } return 0; } @@ -422,6 +440,5 @@ static int __init mtrr_if_init(void) proc_create("mtrr", S_IWUSR | S_IRUGO, NULL, &mtrr_fops); return 0; } - arch_initcall(mtrr_if_init); #endif /* CONFIG_PROC_FS */ diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 8fc248b5aeaf..7af0f88a4163 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -25,43 +25,48 @@ Operating System Writer's Guide" (Intel document number 242692), section 11.11.7 - This was cleaned and made readable by Patrick Mochel <mochel@osdl.org> - on 6-7 March 2002. - Source: Intel Architecture Software Developers Manual, Volume 3: + This was cleaned and made readable by Patrick Mochel <mochel@osdl.org> + on 6-7 March 2002. + Source: Intel Architecture Software Developers Manual, Volume 3: System Programming Guide; Section 9.11. (1997 edition - PPro). */ +#define DEBUG + +#include <linux/types.h> /* FIXME: kvm_para.h needs this */ + +#include <linux/kvm_para.h> +#include <linux/uaccess.h> #include <linux/module.h> +#include <linux/mutex.h> #include <linux/init.h> +#include <linux/sort.h> +#include <linux/cpu.h> #include <linux/pci.h> #include <linux/smp.h> -#include <linux/cpu.h> -#include <linux/mutex.h> -#include <linux/sort.h> +#include <asm/processor.h> #include <asm/e820.h> #include <asm/mtrr.h> -#include <asm/uaccess.h> -#include <asm/processor.h> #include <asm/msr.h> -#include <asm/kvm_para.h> + #include "mtrr.h" -u32 num_var_ranges = 0; +u32 num_var_ranges; unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES]; static DEFINE_MUTEX(mtrr_mutex); u64 size_or_mask, size_and_mask; -static struct mtrr_ops * mtrr_ops[X86_VENDOR_NUM] = {}; +static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM]; -struct mtrr_ops * mtrr_if = NULL; +struct mtrr_ops *mtrr_if; static void set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type); -void set_mtrr_ops(struct mtrr_ops * ops) +void set_mtrr_ops(struct mtrr_ops *ops) { if (ops->vendor && ops->vendor < X86_VENDOR_NUM) mtrr_ops[ops->vendor] = ops; @@ -72,30 +77,36 @@ static int have_wrcomb(void) { struct pci_dev *dev; u8 rev; - - if ((dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL)) != NULL) { - /* ServerWorks LE chipsets < rev 6 have problems with write-combining - Don't allow it and leave room for other chipsets to be tagged */ + + dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL); + if (dev != NULL) { + /* + * ServerWorks LE chipsets < rev 6 have problems with + * write-combining. Don't allow it and leave room for other + * chipsets to be tagged + */ if (dev->vendor == PCI_VENDOR_ID_SERVERWORKS && dev->device == PCI_DEVICE_ID_SERVERWORKS_LE) { pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev); if (rev <= 5) { - printk(KERN_INFO "mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n"); + pr_info("mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n"); pci_dev_put(dev); return 0; } } - /* Intel 450NX errata # 23. Non ascending cacheline evictions to - write combining memory may resulting in data corruption */ + /* + * Intel 450NX errata # 23. Non ascending cacheline evictions to + * write combining memory may resulting in data corruption + */ if (dev->vendor == PCI_VENDOR_ID_INTEL && dev->device == PCI_DEVICE_ID_INTEL_82451NX) { - printk(KERN_INFO "mtrr: Intel 450NX MMC detected. Write-combining disabled.\n"); + pr_info("mtrr: Intel 450NX MMC detected. Write-combining disabled.\n"); pci_dev_put(dev); return 0; } pci_dev_put(dev); - } - return (mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0); + } + return mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0; } /* This function returns the number of variable MTRRs */ @@ -103,12 +114,13 @@ static void __init set_num_var_ranges(void) { unsigned long config = 0, dummy; - if (use_intel()) { + if (use_intel()) rdmsr(MSR_MTRRcap, config, dummy); - } else if (is_cpu(AMD)) + else if (is_cpu(AMD)) config = 2; else if (is_cpu(CYRIX) || is_cpu(CENTAUR)) config = 8; + num_var_ranges = config & 0xff; } @@ -130,10 +142,12 @@ struct set_mtrr_data { mtrr_type smp_type; }; +/** + * ipi_handler - Synchronisation handler. Executed by "other" CPUs. + * + * Returns nothing. + */ static void ipi_handler(void *info) -/* [SUMMARY] Synchronisation handler. Executed by "other" CPUs. - [RETURNS] Nothing. -*/ { #ifdef CONFIG_SMP struct set_mtrr_data *data = info; @@ -142,18 +156,19 @@ static void ipi_handler(void *info) local_irq_save(flags); atomic_dec(&data->count); - while(!atomic_read(&data->gate)) + while (!atomic_read(&data->gate)) cpu_relax(); /* The master has cleared me to execute */ - if (data->smp_reg != ~0U) - mtrr_if->set(data->smp_reg, data->smp_base, + if (data->smp_reg != ~0U) { + mtrr_if->set(data->smp_reg, data->smp_base, data->smp_size, data->smp_type); - else + } else { mtrr_if->set_all(); + } atomic_dec(&data->count); - while(atomic_read(&data->gate)) + while (atomic_read(&data->gate)) cpu_relax(); atomic_dec(&data->count); @@ -161,7 +176,8 @@ static void ipi_handler(void *info) #endif } -static inline int types_compatible(mtrr_type type1, mtrr_type type2) { +static inline int types_compatible(mtrr_type type1, mtrr_type type2) +{ return type1 == MTRR_TYPE_UNCACHABLE || type2 == MTRR_TYPE_UNCACHABLE || (type1 == MTRR_TYPE_WRTHROUGH && type2 == MTRR_TYPE_WRBACK) || @@ -176,10 +192,10 @@ static inline int types_compatible(mtrr_type type1, mtrr_type type2) { * @type: mtrr type * * This is kinda tricky, but fortunately, Intel spelled it out for us cleanly: - * + * * 1. Send IPI to do the following: * 2. Disable Interrupts - * 3. Wait for all procs to do so + * 3. Wait for all procs to do so * 4. Enter no-fill cache mode * 5. Flush caches * 6. Clear PGE bit @@ -189,26 +205,27 @@ static inline int types_compatible(mtrr_type type1, mtrr_type type2) { * 10. Enable all range registers * 11. Flush all TLBs and caches again * 12. Enter normal cache mode and reenable caching - * 13. Set PGE + * 13. Set PGE * 14. Wait for buddies to catch up * 15. Enable interrupts. - * + * * What does that mean for us? Well, first we set data.count to the number * of CPUs. As each CPU disables interrupts, it'll decrement it once. We wait * until it hits 0 and proceed. We set the data.gate flag and reset data.count. - * Meanwhile, they are waiting for that flag to be set. Once it's set, each - * CPU goes through the transition of updating MTRRs. The CPU vendors may each do it - * differently, so we call mtrr_if->set() callback and let them take care of it. - * When they're done, they again decrement data->count and wait for data.gate to - * be reset. - * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag. + * Meanwhile, they are waiting for that flag to be set. Once it's set, each + * CPU goes through the transition of updating MTRRs. + * The CPU vendors may each do it differently, + * so we call mtrr_if->set() callback and let them take care of it. + * When they're done, they again decrement data->count and wait for data.gate + * to be reset. + * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag * Everyone then enables interrupts and we all continue on. * * Note that the mechanism is the same for UP systems, too; all the SMP stuff * becomes nops. */ -static void set_mtrr(unsigned int reg, unsigned long base, - unsigned long size, mtrr_type type) +static void +set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type) { struct set_mtrr_data data; unsigned long flags; @@ -218,121 +235,122 @@ static void set_mtrr(unsigned int reg, unsigned long base, data.smp_size = size; data.smp_type = type; atomic_set(&data.count, num_booting_cpus() - 1); - /* make sure data.count is visible before unleashing other CPUs */ + + /* Make sure data.count is visible before unleashing other CPUs */ smp_wmb(); - atomic_set(&data.gate,0); + atomic_set(&data.gate, 0); - /* Start the ball rolling on other CPUs */ + /* Start the ball rolling on other CPUs */ if (smp_call_function(ipi_handler, &data, 0) != 0) panic("mtrr: timed out waiting for other CPUs\n"); local_irq_save(flags); - while(atomic_read(&data.count)) + while (atomic_read(&data.count)) cpu_relax(); - /* ok, reset count and toggle gate */ + /* Ok, reset count and toggle gate */ atomic_set(&data.count, num_booting_cpus() - 1); smp_wmb(); - atomic_set(&data.gate,1); + atomic_set(&data.gate, 1); - /* do our MTRR business */ + /* Do our MTRR business */ - /* HACK! + /* + * HACK! * We use this same function to initialize the mtrrs on boot. * The state of the boot cpu's mtrrs has been saved, and we want - * to replicate across all the APs. + * to replicate across all the APs. * If we're doing that @reg is set to something special... */ - if (reg != ~0U) - mtrr_if->set(reg,base,size,type); + if (reg != ~0U) + mtrr_if->set(reg, base, size, type); - /* wait for the others */ - while(atomic_read(&data.count)) + /* Wait for the others */ + while (atomic_read(&data.count)) cpu_relax(); atomic_set(&data.count, num_booting_cpus() - 1); smp_wmb(); - atomic_set(&data.gate,0); + atomic_set(&data.gate, 0); /* * Wait here for everyone to have seen the gate change * So we're the last ones to touch 'data' */ - while(atomic_read(&data.count)) + while (atomic_read(&data.count)) cpu_relax(); local_irq_restore(flags); } /** - * mtrr_add_page - Add a memory type region - * @base: Physical base address of region in pages (in units of 4 kB!) - * @size: Physical size of region in pages (4 kB) - * @type: Type of MTRR desired - * @increment: If this is true do usage counting on the region + * mtrr_add_page - Add a memory type region + * @base: Physical base address of region in pages (in units of 4 kB!) + * @size: Physical size of region in pages (4 kB) + * @type: Type of MTRR desired + * @increment: If this is true do usage counting on the region * - * Memory type region registers control the caching on newer Intel and - * non Intel processors. This function allows drivers to request an - * MTRR is added. The details and hardware specifics of each processor's - * implementation are hidden from the caller, but nevertheless the - * caller should expect to need to provide a power of two size on an - * equivalent power of two boundary. + * Memory type region registers control the caching on newer Intel and + * non Intel processors. This function allows drivers to request an + * MTRR is added. The details and hardware specifics of each processor's + * implementation are hidden from the caller, but nevertheless the + * caller should expect to need to provide a power of two size on an + * equivalent power of two boundary. * - * If the region cannot be added either because all regions are in use - * or the CPU cannot support it a negative value is returned. On success - * the register number for this entry is returned, but should be treated - * as a cookie only. + * If the region cannot be added either because all regions are in use + * or the CPU cannot support it a negative value is returned. On success + * the register number for this entry is returned, but should be treated + * as a cookie only. * - * On a multiprocessor machine the changes are made to all processors. - * This is required on x86 by the Intel processors. + * On a multiprocessor machine the changes are made to all processors. + * This is required on x86 by the Intel processors. * - * The available types are + * The available types are * - * %MTRR_TYPE_UNCACHABLE - No caching + * %MTRR_TYPE_UNCACHABLE - No caching * - * %MTRR_TYPE_WRBACK - Write data back in bursts whenever + * %MTRR_TYPE_WRBACK - Write data back in bursts whenever * - * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts + * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts * - * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes + * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes * - * BUGS: Needs a quiet flag for the cases where drivers do not mind - * failures and do not wish system log messages to be sent. + * BUGS: Needs a quiet flag for the cases where drivers do not mind + * failures and do not wish system log messages to be sent. */ - -int mtrr_add_page(unsigned long base, unsigned long size, +int mtrr_add_page(unsigned long base, unsigned long size, unsigned int type, bool increment) { + unsigned long lbase, lsize; int i, replace, error; mtrr_type ltype; - unsigned long lbase, lsize; if (!mtrr_if) return -ENXIO; - - if ((error = mtrr_if->validate_add_page(base,size,type))) + + error = mtrr_if->validate_add_page(base, size, type); + if (error) return error; if (type >= MTRR_NUM_TYPES) { - printk(KERN_WARNING "mtrr: type: %u invalid\n", type); + pr_warning("mtrr: type: %u invalid\n", type); return -EINVAL; } - /* If the type is WC, check that this processor supports it */ + /* If the type is WC, check that this processor supports it */ if ((type == MTRR_TYPE_WRCOMB) && !have_wrcomb()) { - printk(KERN_WARNING - "mtrr: your processor doesn't support write-combining\n"); + pr_warning("mtrr: your processor doesn't support write-combining\n"); return -ENOSYS; } if (!size) { - printk(KERN_WARNING "mtrr: zero sized request\n"); + pr_warning("mtrr: zero sized request\n"); return -EINVAL; } if (base & size_or_mask || size & size_or_mask) { - printk(KERN_WARNING "mtrr: base or size exceeds the MTRR width\n"); + pr_warning("mtrr: base or size exceeds the MTRR width\n"); return -EINVAL; } @@ -341,36 +359,40 @@ int mtrr_add_page(unsigned long base, unsigned long size, /* No CPU hotplug when we change MTRR entries */ get_online_cpus(); - /* Search for existing MTRR */ + + /* Search for existing MTRR */ mutex_lock(&mtrr_mutex); for (i = 0; i < num_var_ranges; ++i) { mtrr_if->get(i, &lbase, &lsize, <ype); - if (!lsize || base > lbase + lsize - 1 || base + size - 1 < lbase) + if (!lsize || base > lbase + lsize - 1 || + base + size - 1 < lbase) continue; - /* At this point we know there is some kind of overlap/enclosure */ + /* + * At this point we know there is some kind of + * overlap/enclosure + */ if (base < lbase || base + size - 1 > lbase + lsize - 1) { - if (base <= lbase && base + size - 1 >= lbase + lsize - 1) { + if (base <= lbase && + base + size - 1 >= lbase + lsize - 1) { /* New region encloses an existing region */ if (type == ltype) { replace = replace == -1 ? i : -2; continue; - } - else if (types_compatible(type, ltype)) + } else if (types_compatible(type, ltype)) continue; } - printk(KERN_WARNING - "mtrr: 0x%lx000,0x%lx000 overlaps existing" - " 0x%lx000,0x%lx000\n", base, size, lbase, - lsize); + pr_warning("mtrr: 0x%lx000,0x%lx000 overlaps existing" + " 0x%lx000,0x%lx000\n", base, size, lbase, + lsize); goto out; } - /* New region is enclosed by an existing region */ + /* New region is enclosed by an existing region */ if (ltype != type) { if (types_compatible(type, ltype)) continue; - printk (KERN_WARNING "mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n", - base, size, mtrr_attrib_to_str(ltype), - mtrr_attrib_to_str(type)); + pr_warning("mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n", + base, size, mtrr_attrib_to_str(ltype), + mtrr_attrib_to_str(type)); goto out; } if (increment) @@ -378,7 +400,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, error = i; goto out; } - /* Search for an empty MTRR */ + /* Search for an empty MTRR */ i = mtrr_if->get_free_region(base, size, replace); if (i >= 0) { set_mtrr(i, base, size, type); @@ -393,8 +415,9 @@ int mtrr_add_page(unsigned long base, unsigned long size, mtrr_usage_table[replace] = 0; } } - } else - printk(KERN_INFO "mtrr: no more MTRRs available\n"); + } else { + pr_info("mtrr: no more MTRRs available\n"); + } error = i; out: mutex_unlock(&mtrr_mutex); @@ -405,10 +428,8 @@ int mtrr_add_page(unsigned long base, unsigned long size, static int mtrr_check(unsigned long base, unsigned long size) { if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) { - printk(KERN_WARNING - "mtrr: size and base must be multiples of 4 kiB\n"); - printk(KERN_DEBUG - "mtrr: size: 0x%lx base: 0x%lx\n", size, base); + pr_warning("mtrr: size and base must be multiples of 4 kiB\n"); + pr_debug("mtrr: size: 0x%lx base: 0x%lx\n", size, base); dump_stack(); return -1; } @@ -416,66 +437,64 @@ static int mtrr_check(unsigned long base, unsigned long size) } /** - * mtrr_add - Add a memory type region - * @base: Physical base address of region - * @size: Physical size of region - * @type: Type of MTRR desired - * @increment: If this is true do usage counting on the region + * mtrr_add - Add a memory type region + * @base: Physical base address of region + * @size: Physical size of region + * @type: Type of MTRR desired + * @increment: If this is true do usage counting on the region * - * Memory type region registers control the caching on newer Intel and - * non Intel processors. This function allows drivers to request an - * MTRR is added. The details and hardware specifics of each processor's - * implementation are hidden from the caller, but nevertheless the - * caller should expect to need to provide a power of two size on an - * equivalent power of two boundary. + * Memory type region registers control the caching on newer Intel and + * non Intel processors. This function allows drivers to request an + * MTRR is added. The details and hardware specifics of each processor's + * implementation are hidden from the caller, but nevertheless the + * caller should expect to need to provide a power of two size on an + * equivalent power of two boundary. * - * If the region cannot be added either because all regions are in use - * or the CPU cannot support it a negative value is returned. On success - * the register number for this entry is returned, but should be treated - * as a cookie only. + * If the region cannot be added either because all regions are in use + * or the CPU cannot support it a negative value is returned. On success + * the register number for this entry is returned, but should be treated + * as a cookie only. * - * On a multiprocessor machine the changes are made to all processors. - * This is required on x86 by the Intel processors. + * On a multiprocessor machine the changes are made to all processors. + * This is required on x86 by the Intel processors. * - * The available types are + * The available types are * - * %MTRR_TYPE_UNCACHABLE - No caching + * %MTRR_TYPE_UNCACHABLE - No caching * - * %MTRR_TYPE_WRBACK - Write data back in bursts whenever + * %MTRR_TYPE_WRBACK - Write data back in bursts whenever * - * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts + * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts * - * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes + * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes * - * BUGS: Needs a quiet flag for the cases where drivers do not mind - * failures and do not wish system log messages to be sent. + * BUGS: Needs a quiet flag for the cases where drivers do not mind + * failures and do not wish system log messages to be sent. */ - -int -mtrr_add(unsigned long base, unsigned long size, unsigned int type, - bool increment) +int mtrr_add(unsigned long base, unsigned long size, unsigned int type, + bool increment) { if (mtrr_check(base, size)) return -EINVAL; return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type, increment); } +EXPORT_SYMBOL(mtrr_add); /** - * mtrr_del_page - delete a memory type region - * @reg: Register returned by mtrr_add - * @base: Physical base address - * @size: Size of region + * mtrr_del_page - delete a memory type region + * @reg: Register returned by mtrr_add + * @base: Physical base address + * @size: Size of region * - * If register is supplied then base and size are ignored. This is - * how drivers should call it. + * If register is supplied then base and size are ignored. This is + * how drivers should call it. * - * Releases an MTRR region. If the usage count drops to zero the - * register is freed and the region returns to default state. - * On success the register is returned, on failure a negative error - * code. + * Releases an MTRR region. If the usage count drops to zero the + * register is freed and the region returns to default state. + * On success the register is returned, on failure a negative error + * code. */ - int mtrr_del_page(int reg, unsigned long base, unsigned long size) { int i, max; @@ -500,22 +519,22 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) } } if (reg < 0) { - printk(KERN_DEBUG "mtrr: no MTRR for %lx000,%lx000 found\n", base, - size); + pr_debug("mtrr: no MTRR for %lx000,%lx000 found\n", + base, size); goto out; } } if (reg >= max) { - printk(KERN_WARNING "mtrr: register: %d too big\n", reg); + pr_warning("mtrr: register: %d too big\n", reg); goto out; } mtrr_if->get(reg, &lbase, &lsize, <ype); if (lsize < 1) { - printk(KERN_WARNING "mtrr: MTRR %d not used\n", reg); + pr_warning("mtrr: MTRR %d not used\n", reg); goto out; } if (mtrr_usage_table[reg] < 1) { - printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg); + pr_warning("mtrr: reg: %d has count=0\n", reg); goto out; } if (--mtrr_usage_table[reg] < 1) @@ -526,33 +545,31 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) put_online_cpus(); return error; } + /** - * mtrr_del - delete a memory type region - * @reg: Register returned by mtrr_add - * @base: Physical base address - * @size: Size of region + * mtrr_del - delete a memory type region + * @reg: Register returned by mtrr_add + * @base: Physical base address + * @size: Size of region * - * If register is supplied then base and size are ignored. This is - * how drivers should call it. + * If register is supplied then base and size are ignored. This is + * how drivers should call it. * - * Releases an MTRR region. If the usage count drops to zero the - * register is freed and the region returns to default state. - * On success the register is returned, on failure a negative error - * code. + * Releases an MTRR region. If the usage count drops to zero the + * register is freed and the region returns to default state. + * On success the register is returned, on failure a negative error + * code. */ - -int -mtrr_del(int reg, unsigned long base, unsigned long size) +int mtrr_del(int reg, unsigned long base, unsigned long size) { if (mtrr_check(base, size)) return -EINVAL; return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT); } - -EXPORT_SYMBOL(mtrr_add); EXPORT_SYMBOL(mtrr_del); -/* HACK ALERT! +/* + * HACK ALERT! * These should be called implicitly, but we can't yet until all the initcall * stuff is done... */ @@ -576,29 +593,28 @@ struct mtrr_value { static struct mtrr_value mtrr_value[MTRR_MAX_VAR_RANGES]; -static int mtrr_save(struct sys_device * sysdev, pm_message_t state) +static int mtrr_save(struct sys_device *sysdev, pm_message_t state) { int i; for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, - &mtrr_value[i].lbase, - &mtrr_value[i].lsize, - &mtrr_value[i].ltype); + mtrr_if->get(i, &mtrr_value[i].lbase, + &mtrr_value[i].lsize, + &mtrr_value[i].ltype); } return 0; } -static int mtrr_restore(struct sys_device * sysdev) +static int mtrr_restore(struct sys_device *sysdev) { int i; for (i = 0; i < num_var_ranges; i++) { - if (mtrr_value[i].lsize) - set_mtrr(i, - mtrr_value[i].lbase, - mtrr_value[i].lsize, - mtrr_value[i].ltype); + if (mtrr_value[i].lsize) { + set_mtrr(i, mtrr_value[i].lbase, + mtrr_value[i].lsize, + mtrr_value[i].ltype); + } } return 0; } @@ -615,26 +631,29 @@ int __initdata changed_by_mtrr_cleanup; /** * mtrr_bp_init - initialize mtrrs on the boot CPU * - * This needs to be called early; before any of the other CPUs are + * This needs to be called early; before any of the other CPUs are * initialized (i.e. before smp_init()). - * + * */ void __init mtrr_bp_init(void) { u32 phys_addr; + init_ifs(); phys_addr = 32; if (cpu_has_mtrr) { mtrr_if = &generic_mtrr_ops; - size_or_mask = 0xff000000; /* 36 bits */ + size_or_mask = 0xff000000; /* 36 bits */ size_and_mask = 0x00f00000; phys_addr = 36; - /* This is an AMD specific MSR, but we assume(hope?) that - Intel will implement it to when they extend the address - bus of the Xeon. */ + /* + * This is an AMD specific MSR, but we assume(hope?) that + * Intel will implement it to when they extend the address + * bus of the Xeon. + */ if (cpuid_eax(0x80000000) >= 0x80000008) { phys_addr = cpuid_eax(0x80000008) & 0xff; /* CPUID workaround for Intel 0F33/0F34 CPU */ @@ -649,9 +668,11 @@ void __init mtrr_bp_init(void) size_and_mask = ~size_or_mask & 0xfffff00000ULL; } else if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR && boot_cpu_data.x86 == 6) { - /* VIA C* family have Intel style MTRRs, but - don't support PAE */ - size_or_mask = 0xfff00000; /* 32 bits */ + /* + * VIA C* family have Intel style MTRRs, + * but don't support PAE + */ + size_or_mask = 0xfff00000; /* 32 bits */ size_and_mask = 0; phys_addr = 32; } @@ -694,7 +715,6 @@ void __init mtrr_bp_init(void) changed_by_mtrr_cleanup = 1; mtrr_if->set_all(); } - } } } @@ -706,12 +726,17 @@ void mtrr_ap_init(void) if (!mtrr_if || !use_intel()) return; /* - * Ideally we should hold mtrr_mutex here to avoid mtrr entries changed, - * but this routine will be called in cpu boot time, holding the lock - * breaks it. This routine is called in two cases: 1.very earily time - * of software resume, when there absolutely isn't mtrr entry changes; - * 2.cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug lock to - * prevent mtrr entry changes + * Ideally we should hold mtrr_mutex here to avoid mtrr entries + * changed, but this routine will be called in cpu boot time, + * holding the lock breaks it. + * + * This routine is called in two cases: + * + * 1. very earily time of software resume, when there absolutely + * isn't mtrr entry changes; + * + * 2. cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug + * lock to prevent mtrr entry changes */ local_irq_save(flags); @@ -732,19 +757,23 @@ static int __init mtrr_init_finialize(void) { if (!mtrr_if) return 0; + if (use_intel()) { if (!changed_by_mtrr_cleanup) mtrr_state_warn(); - } else { - /* The CPUs haven't MTRR and seem to not support SMP. They have - * specific drivers, we use a tricky method to support - * suspend/resume for them. - * TBD: is there any system with such CPU which supports - * suspend/resume? if no, we should remove the code. - */ - sysdev_driver_register(&cpu_sysdev_class, - &mtrr_sysdev_driver); + return 0; } + + /* + * The CPU has no MTRR and seems to not support SMP. They have + * specific drivers, we use a tricky method to support + * suspend/resume for them. + * + * TBD: is there any system with such CPU which supports + * suspend/resume? If no, we should remove the code. + */ + sysdev_driver_register(&cpu_sysdev_class, &mtrr_sysdev_driver); + return 0; } subsys_initcall(mtrr_init_finialize); diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h index 7538b767f206..a501dee9a87a 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -1,5 +1,5 @@ /* - * local mtrr defines. + * local MTRR defines. */ #include <linux/types.h> @@ -14,13 +14,12 @@ extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES]; struct mtrr_ops { u32 vendor; u32 use_intel_if; -// void (*init)(void); void (*set)(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type); void (*set_all)(void); void (*get)(unsigned int reg, unsigned long *base, - unsigned long *size, mtrr_type * type); + unsigned long *size, mtrr_type *type); int (*get_free_region)(unsigned long base, unsigned long size, int replace_reg); int (*validate_add_page)(unsigned long base, unsigned long size, @@ -39,11 +38,11 @@ extern int positive_have_wrcomb(void); /* library functions for processor-specific routines */ struct set_mtrr_context { - unsigned long flags; - unsigned long cr4val; - u32 deftype_lo; - u32 deftype_hi; - u32 ccr3; + unsigned long flags; + unsigned long cr4val; + u32 deftype_lo; + u32 deftype_hi; + u32 ccr3; }; void set_mtrr_done(struct set_mtrr_context *ctxt); @@ -54,10 +53,10 @@ void fill_mtrr_var_range(unsigned int index, u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); void get_mtrr_state(void); -extern void set_mtrr_ops(struct mtrr_ops * ops); +extern void set_mtrr_ops(struct mtrr_ops *ops); extern u64 size_or_mask, size_and_mask; -extern struct mtrr_ops * mtrr_if; +extern struct mtrr_ops *mtrr_if; #define is_cpu(vnd) (mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd) #define use_intel() (mtrr_if && mtrr_if->use_intel_if == 1) diff --git a/arch/x86/kernel/cpu/mtrr/state.c b/arch/x86/kernel/cpu/mtrr/state.c index 1f5fb1588d1f..dfc80b4e6b0d 100644 --- a/arch/x86/kernel/cpu/mtrr/state.c +++ b/arch/x86/kernel/cpu/mtrr/state.c @@ -1,24 +1,25 @@ -#include <linux/mm.h> #include <linux/init.h> -#include <asm/io.h> -#include <asm/mtrr.h> -#include <asm/msr.h> +#include <linux/io.h> +#include <linux/mm.h> + #include <asm/processor-cyrix.h> #include <asm/processor-flags.h> -#include "mtrr.h" +#include <asm/mtrr.h> +#include <asm/msr.h> +#include "mtrr.h" -/* Put the processor into a state where MTRRs can be safely set */ +/* Put the processor into a state where MTRRs can be safely set */ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt) { unsigned int cr0; - /* Disable interrupts locally */ + /* Disable interrupts locally */ local_irq_save(ctxt->flags); if (use_intel() || is_cpu(CYRIX)) { - /* Save value of CR4 and clear Page Global Enable (bit 7) */ + /* Save value of CR4 and clear Page Global Enable (bit 7) */ if (cpu_has_pge) { ctxt->cr4val = read_cr4(); write_cr4(ctxt->cr4val & ~X86_CR4_PGE); @@ -33,50 +34,61 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt) write_cr0(cr0); wbinvd(); - if (use_intel()) - /* Save MTRR state */ + if (use_intel()) { + /* Save MTRR state */ rdmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi); - else - /* Cyrix ARRs - everything else were excluded at the top */ + } else { + /* + * Cyrix ARRs - + * everything else were excluded at the top + */ ctxt->ccr3 = getCx86(CX86_CCR3); + } } } void set_mtrr_cache_disable(struct set_mtrr_context *ctxt) { - if (use_intel()) - /* Disable MTRRs, and set the default type to uncached */ + if (use_intel()) { + /* Disable MTRRs, and set the default type to uncached */ mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo & 0xf300UL, ctxt->deftype_hi); - else if (is_cpu(CYRIX)) - /* Cyrix ARRs - everything else were excluded at the top */ - setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10); + } else { + if (is_cpu(CYRIX)) { + /* Cyrix ARRs - everything else were excluded at the top */ + setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10); + } + } } -/* Restore the processor after a set_mtrr_prepare */ +/* Restore the processor after a set_mtrr_prepare */ void set_mtrr_done(struct set_mtrr_context *ctxt) { if (use_intel() || is_cpu(CYRIX)) { - /* Flush caches and TLBs */ + /* Flush caches and TLBs */ wbinvd(); - /* Restore MTRRdefType */ - if (use_intel()) + /* Restore MTRRdefType */ + if (use_intel()) { /* Intel (P6) standard MTRRs */ - mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi); - else - /* Cyrix ARRs - everything else was excluded at the top */ + mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, + ctxt->deftype_hi); + } else { + /* + * Cyrix ARRs - + * everything else was excluded at the top + */ setCx86(CX86_CCR3, ctxt->ccr3); + } - /* Enable caches */ + /* Enable caches */ write_cr0(read_cr0() & 0xbfffffff); - /* Restore value of CR4 */ + /* Restore value of CR4 */ if (cpu_has_pge) write_cr4(ctxt->cr4val); } - /* Re-enable interrupts locally (if enabled previously) */ + /* Re-enable interrupts locally (if enabled previously) */ local_irq_restore(ctxt->flags); } - diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c index a7aa8f900954..b237c181aa41 100644 --- a/arch/x86/kernel/cpu/perf_counter.c +++ b/arch/x86/kernel/cpu/perf_counter.c @@ -6,6 +6,7 @@ * Copyright (C) 2009 Jaswinder Singh Rajput * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> * * For licencing details see kernel-base/COPYING */ @@ -20,6 +21,7 @@ #include <linux/sched.h> #include <linux/uaccess.h> #include <linux/highmem.h> +#include <linux/cpu.h> #include <asm/apic.h> #include <asm/stacktrace.h> @@ -27,12 +29,52 @@ static u64 perf_counter_mask __read_mostly; +/* The maximal number of PEBS counters: */ +#define MAX_PEBS_COUNTERS 4 + +/* The size of a BTS record in bytes: */ +#define BTS_RECORD_SIZE 24 + +/* The size of a per-cpu BTS buffer in bytes: */ +#define BTS_BUFFER_SIZE (BTS_RECORD_SIZE * 1024) + +/* The BTS overflow threshold in bytes from the end of the buffer: */ +#define BTS_OVFL_TH (BTS_RECORD_SIZE * 64) + + +/* + * 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) + +/* + * 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_counter_reset[MAX_PEBS_COUNTERS]; +}; + struct cpu_hw_counters { struct perf_counter *counters[X86_PMC_IDX_MAX]; unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; unsigned long interrupts; int enabled; + struct debug_store *ds; }; /* @@ -57,6 +99,8 @@ struct x86_pmu { u64 counter_mask; u64 max_period; u64 intel_ctrl; + void (*enable_bts)(u64 config); + void (*disable_bts)(void); }; static struct x86_pmu x86_pmu __read_mostly; @@ -576,6 +620,9 @@ x86_perf_counter_update(struct perf_counter *counter, u64 prev_raw_count, new_raw_count; s64 delta; + if (idx == X86_PMC_IDX_FIXED_BTS) + return 0; + /* * Careful: an NMI might modify the previous counter value. * @@ -659,10 +706,109 @@ static void release_pmc_hardware(void) enable_lapic_nmi_watchdog(); } +static inline bool bts_available(void) +{ + return x86_pmu.enable_bts != NULL; +} + +static inline void init_debug_store_on_cpu(int cpu) +{ + struct debug_store *ds = per_cpu(cpu_hw_counters, cpu).ds; + + if (!ds) + return; + + wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, + (u32)((u64)(long)ds), (u32)((u64)(long)ds >> 32)); +} + +static inline void fini_debug_store_on_cpu(int cpu) +{ + if (!per_cpu(cpu_hw_counters, 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_counters, cpu).ds; + + if (!ds) + continue; + + per_cpu(cpu_hw_counters, cpu).ds = NULL; + + kfree((void *)(long)ds->bts_buffer_base); + kfree(ds); + } + + put_online_cpus(); +} + +static int reserve_bts_hardware(void) +{ + int cpu, err = 0; + + if (!bts_available()) + return -EOPNOTSUPP; + + 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)(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; + + per_cpu(cpu_hw_counters, cpu).ds = ds; + err = 0; + } + + if (err) + release_bts_hardware(); + else { + for_each_online_cpu(cpu) + init_debug_store_on_cpu(cpu); + } + + put_online_cpus(); + + return err; +} + static void hw_perf_counter_destroy(struct perf_counter *counter) { if (atomic_dec_and_mutex_lock(&active_counters, &pmc_reserve_mutex)) { release_pmc_hardware(); + release_bts_hardware(); mutex_unlock(&pmc_reserve_mutex); } } @@ -705,6 +851,42 @@ set_ext_hw_attr(struct hw_perf_counter *hwc, struct perf_counter_attr *attr) return 0; } +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_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + 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); +} + /* * Setup the hardware configuration for a given attr_type */ @@ -721,9 +903,13 @@ static int __hw_perf_counter_init(struct perf_counter *counter) err = 0; if (!atomic_inc_not_zero(&active_counters)) { mutex_lock(&pmc_reserve_mutex); - if (atomic_read(&active_counters) == 0 && !reserve_pmc_hardware()) - err = -EBUSY; - else + if (atomic_read(&active_counters) == 0) { + if (!reserve_pmc_hardware()) + err = -EBUSY; + else + reserve_bts_hardware(); + } + if (!err) atomic_inc(&active_counters); mutex_unlock(&pmc_reserve_mutex); } @@ -801,7 +987,18 @@ static void p6_pmu_disable_all(void) static void intel_pmu_disable_all(void) { + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + + if (!cpuc->enabled) + return; + + cpuc->enabled = 0; + barrier(); + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0); + + if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) + intel_pmu_disable_bts(); } static void amd_pmu_disable_all(void) @@ -859,7 +1056,25 @@ static void p6_pmu_enable_all(void) static void intel_pmu_enable_all(void) { + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + + if (cpuc->enabled) + return; + + cpuc->enabled = 1; + barrier(); + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); + + if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) { + struct perf_counter *counter = + cpuc->counters[X86_PMC_IDX_FIXED_BTS]; + + if (WARN_ON_ONCE(!counter)) + return; + + intel_pmu_enable_bts(counter->hw.config); + } } static void amd_pmu_enable_all(void) @@ -946,6 +1161,11 @@ p6_pmu_disable_counter(struct hw_perf_counter *hwc, int idx) static inline void intel_pmu_disable_counter(struct hw_perf_counter *hwc, int idx) { + if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) { + intel_pmu_disable_bts(); + return; + } + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { intel_pmu_disable_fixed(hwc, idx); return; @@ -974,6 +1194,9 @@ x86_perf_counter_set_period(struct perf_counter *counter, s64 period = hwc->sample_period; int err, ret = 0; + if (idx == X86_PMC_IDX_FIXED_BTS) + return 0; + /* * If we are way outside a reasoable range then just skip forward: */ @@ -1056,6 +1279,14 @@ static void p6_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) static void intel_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) { + if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) { + if (!__get_cpu_var(cpu_hw_counters).enabled) + return; + + intel_pmu_enable_bts(hwc->config); + return; + } + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { intel_pmu_enable_fixed(hwc, idx); return; @@ -1077,11 +1308,16 @@ fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc) { unsigned int event; + event = hwc->config & ARCH_PERFMON_EVENT_MASK; + + if (unlikely((event == + x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) && + (hwc->sample_period == 1))) + return X86_PMC_IDX_FIXED_BTS; + if (!x86_pmu.num_counters_fixed) return -1; - event = hwc->config & ARCH_PERFMON_EVENT_MASK; - if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS))) return X86_PMC_IDX_FIXED_INSTRUCTIONS; if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_CPU_CYCLES))) @@ -1102,7 +1338,25 @@ static int x86_pmu_enable(struct perf_counter *counter) int idx; idx = fixed_mode_idx(counter, hwc); - if (idx >= 0) { + if (idx == X86_PMC_IDX_FIXED_BTS) { + /* + * Try to use BTS for branch tracing. If that is not + * available, try to get a generic counter. + */ + if (unlikely(!cpuc->ds)) + goto try_generic; + + /* + * Try to get the fixed counter, if that is already taken + * then try to get a generic counter: + */ + if (test_and_set_bit(idx, cpuc->used_mask)) + goto try_generic; + + hwc->config_base = 0; + hwc->counter_base = 0; + hwc->idx = idx; + } else if (idx >= 0) { /* * Try to get the fixed counter, if that is already taken * then try to get a generic counter: @@ -1213,6 +1467,45 @@ void perf_counter_print_debug(void) local_irq_restore(flags); } +static void intel_pmu_drain_bts_buffer(struct cpu_hw_counters *cpuc, + struct perf_sample_data *data) +{ + struct debug_store *ds = cpuc->ds; + struct bts_record { + u64 from; + u64 to; + u64 flags; + }; + struct perf_counter *counter = cpuc->counters[X86_PMC_IDX_FIXED_BTS]; + unsigned long orig_ip = data->regs->ip; + u64 at; + + if (!counter) + return; + + if (!ds) + return; + + for (at = ds->bts_buffer_base; + at < ds->bts_index; + at += sizeof(struct bts_record)) { + struct bts_record *rec = (struct bts_record *)(long)at; + + data->regs->ip = rec->from; + data->addr = rec->to; + + perf_counter_output(counter, 1, data); + } + + ds->bts_index = ds->bts_buffer_base; + + data->regs->ip = orig_ip; + data->addr = 0; + + /* There's new data available. */ + counter->pending_kill = POLL_IN; +} + static void x86_pmu_disable(struct perf_counter *counter) { struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); @@ -1237,6 +1530,15 @@ static void x86_pmu_disable(struct perf_counter *counter) * that we are disabling: */ x86_perf_counter_update(counter, hwc, idx); + + /* Drain the remaining BTS records. */ + if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) { + struct perf_sample_data data; + struct pt_regs regs; + + data.regs = ®s; + intel_pmu_drain_bts_buffer(cpuc, &data); + } cpuc->counters[idx] = NULL; clear_bit(idx, cpuc->used_mask); @@ -1264,6 +1566,7 @@ static int intel_pmu_save_and_restart(struct perf_counter *counter) static void intel_pmu_reset(void) { + struct debug_store *ds = __get_cpu_var(cpu_hw_counters).ds; unsigned long flags; int idx; @@ -1281,6 +1584,8 @@ static void intel_pmu_reset(void) 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; local_irq_restore(flags); } @@ -1346,6 +1651,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_counters); perf_disable(); + intel_pmu_drain_bts_buffer(cpuc, &data); status = intel_pmu_get_status(); if (!status) { perf_enable(); @@ -1547,6 +1853,8 @@ static struct x86_pmu intel_pmu = { * the generic counter period: */ .max_period = (1ULL << 31) - 1, + .enable_bts = intel_pmu_enable_bts, + .disable_bts = intel_pmu_disable_bts, }; static struct x86_pmu amd_pmu = { @@ -1936,3 +2244,8 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) return entry; } + +void hw_perf_counter_setup_online(int cpu) +{ + init_debug_store_on_cpu(cpu); +} diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index e60ed740d2b3..392bea43b890 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -68,16 +68,16 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr) /* returns the bit offset of the performance counter register */ switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: - return (msr - MSR_K7_PERFCTR0); + return msr - MSR_K7_PERFCTR0; case X86_VENDOR_INTEL: if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) - return (msr - MSR_ARCH_PERFMON_PERFCTR0); + return msr - MSR_ARCH_PERFMON_PERFCTR0; switch (boot_cpu_data.x86) { case 6: - return (msr - MSR_P6_PERFCTR0); + return msr - MSR_P6_PERFCTR0; case 15: - return (msr - MSR_P4_BPU_PERFCTR0); + return msr - MSR_P4_BPU_PERFCTR0; } } return 0; @@ -92,16 +92,16 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr) /* returns the bit offset of the event selection register */ switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: - return (msr - MSR_K7_EVNTSEL0); + return msr - MSR_K7_EVNTSEL0; case X86_VENDOR_INTEL: if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) - return (msr - MSR_ARCH_PERFMON_EVENTSEL0); + return msr - MSR_ARCH_PERFMON_EVENTSEL0; switch (boot_cpu_data.x86) { case 6: - return (msr - MSR_P6_EVNTSEL0); + return msr - MSR_P6_EVNTSEL0; case 15: - return (msr - MSR_P4_BSU_ESCR0); + return msr - MSR_P4_BSU_ESCR0; } } return 0; @@ -113,7 +113,7 @@ int avail_to_resrv_perfctr_nmi_bit(unsigned int counter) { BUG_ON(counter > NMI_MAX_COUNTER_BITS); - return (!test_bit(counter, perfctr_nmi_owner)); + return !test_bit(counter, perfctr_nmi_owner); } /* checks the an msr for availability */ @@ -124,7 +124,7 @@ int avail_to_resrv_perfctr_nmi(unsigned int msr) counter = nmi_perfctr_msr_to_bit(msr); BUG_ON(counter > NMI_MAX_COUNTER_BITS); - return (!test_bit(counter, perfctr_nmi_owner)); + return !test_bit(counter, perfctr_nmi_owner); } EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit); @@ -237,7 +237,7 @@ static unsigned int adjust_for_32bit_ctr(unsigned int hz) */ counter_val = (u64)cpu_khz * 1000; do_div(counter_val, retval); - if (counter_val > 0x7fffffffULL) { + if (counter_val > 0x7fffffffULL) { u64 count = (u64)cpu_khz * 1000; do_div(count, 0x7fffffffUL); retval = count + 1; @@ -251,7 +251,7 @@ static void write_watchdog_counter(unsigned int perfctr_msr, u64 count = (u64)cpu_khz * 1000; do_div(count, nmi_hz); - if(descr) + if (descr) pr_debug("setting %s to -0x%08Lx\n", descr, count); wrmsrl(perfctr_msr, 0 - count); } @@ -262,7 +262,7 @@ static void write_watchdog_counter32(unsigned int perfctr_msr, u64 count = (u64)cpu_khz * 1000; do_div(count, nmi_hz); - if(descr) + if (descr) pr_debug("setting %s to -0x%08Lx\n", descr, count); wrmsr(perfctr_msr, (u32)(-count), 0); } @@ -296,7 +296,7 @@ static int setup_k7_watchdog(unsigned nmi_hz) /* setup the timer */ wrmsr(evntsel_msr, evntsel, 0); - write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz); + write_watchdog_counter(perfctr_msr, "K7_PERFCTR0", nmi_hz); /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; @@ -387,7 +387,7 @@ static int setup_p6_watchdog(unsigned nmi_hz) /* setup the timer */ wrmsr(evntsel_msr, evntsel, 0); nmi_hz = adjust_for_32bit_ctr(nmi_hz); - write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz); + write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0", nmi_hz); /* initialize the wd struct before enabling */ wd->perfctr_msr = perfctr_msr; @@ -415,7 +415,7 @@ static void __kprobes p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) apic_write(APIC_LVTPC, APIC_DM_NMI); /* P6/ARCH_PERFMON has 32 bit counter write */ - write_watchdog_counter32(wd->perfctr_msr, NULL,nmi_hz); + write_watchdog_counter32(wd->perfctr_msr, NULL, nmi_hz); } static const struct wd_ops p6_wd_ops = { @@ -490,9 +490,9 @@ static int setup_p4_watchdog(unsigned nmi_hz) if (smp_num_siblings == 2) { unsigned int ebx, apicid; - ebx = cpuid_ebx(1); - apicid = (ebx >> 24) & 0xff; - ht_num = apicid & 1; + ebx = cpuid_ebx(1); + apicid = (ebx >> 24) & 0xff; + ht_num = apicid & 1; } else #endif ht_num = 0; @@ -544,7 +544,7 @@ static int setup_p4_watchdog(unsigned nmi_hz) } evntsel = P4_ESCR_EVENT_SELECT(0x3F) - | P4_ESCR_OS + | P4_ESCR_OS | P4_ESCR_USR; cccr_val |= P4_CCCR_THRESHOLD(15) @@ -612,7 +612,7 @@ static void __kprobes p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) { unsigned dummy; /* - * P4 quirks: + * P4 quirks: * - An overflown perfctr will assert its interrupt * until the OVF flag in its CCCR is cleared. * - LVTPC is masked on interrupt and must be @@ -662,7 +662,8 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz) * NOTE: Corresponding bit = 0 in ebx indicates event present. */ cpuid(10, &(eax.full), &ebx, &unused, &unused); - if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) || + if ((eax.split.mask_length < + (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) || (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT)) return 0; diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index d5e30397246b..62ac8cb6ba27 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -116,11 +116,9 @@ static int show_cpuinfo(struct seq_file *m, void *v) seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize); #endif seq_printf(m, "clflush size\t: %u\n", c->x86_clflush_size); -#ifdef CONFIG_X86_64 seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment); seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n", c->x86_phys_bits, c->x86_virt_bits); -#endif seq_printf(m, "power management:"); for (i = 0; i < 32; i++) { @@ -128,7 +126,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) if (i < ARRAY_SIZE(x86_power_flags) && x86_power_flags[i]) seq_printf(m, "%s%s", - x86_power_flags[i][0]?" ":"", + x86_power_flags[i][0] ? " " : "", x86_power_flags[i]); else seq_printf(m, " [%d]", i); diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c index 284c399e3234..bc24f514ec93 100644 --- a/arch/x86/kernel/cpu/vmware.c +++ b/arch/x86/kernel/cpu/vmware.c @@ -49,17 +49,17 @@ static inline int __vmware_platform(void) static unsigned long __vmware_get_tsc_khz(void) { - uint64_t tsc_hz; - uint32_t eax, ebx, ecx, edx; + uint64_t tsc_hz; + uint32_t eax, ebx, ecx, edx; - VMWARE_PORT(GETHZ, eax, ebx, ecx, edx); + VMWARE_PORT(GETHZ, eax, ebx, ecx, edx); - if (ebx == UINT_MAX) - return 0; - tsc_hz = eax | (((uint64_t)ebx) << 32); - do_div(tsc_hz, 1000); - BUG_ON(tsc_hz >> 32); - return tsc_hz; + if (ebx == UINT_MAX) + return 0; + tsc_hz = eax | (((uint64_t)ebx) << 32); + do_div(tsc_hz, 1000); + BUG_ON(tsc_hz >> 32); + return tsc_hz; } /* diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c index b4f14c6c09d9..37250fe490b1 100644 --- a/arch/x86/kernel/doublefault_32.c +++ b/arch/x86/kernel/doublefault_32.c @@ -27,9 +27,7 @@ static void doublefault_fn(void) if (ptr_ok(gdt)) { gdt += GDT_ENTRY_TSS << 3; - tss = *(u16 *)(gdt+2); - tss += *(u8 *)(gdt+4) << 16; - tss += *(u8 *)(gdt+7) << 24; + tss = get_desc_base((struct desc_struct *)gdt); printk(KERN_EMERG "double fault, tss at %08lx\n", tss); if (ptr_ok(tss)) { diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 48bfe1386038..ef42a038f1a6 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -509,15 +509,15 @@ enum bts_field { bts_escape = ((unsigned long)-1 & ~bts_qual_mask) }; -static inline unsigned long bts_get(const char *base, enum bts_field field) +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, enum bts_field field, unsigned long val) +static inline void bts_set(char *base, unsigned long field, unsigned long val) { - base += (ds_cfg.sizeof_ptr_field * field);; + base += (ds_cfg.sizeof_ptr_field * field); (*(unsigned long *)base) = val; } diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index c8405718a4c3..2d8a371d4339 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -15,7 +15,6 @@ #include <linux/bug.h> #include <linux/nmi.h> #include <linux/sysfs.h> -#include <linux/ftrace.h> #include <asm/stacktrace.h> diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index d94e1ea3b9fe..8e9663413b7f 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -417,10 +417,6 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, unsigned long return_hooker = (unsigned long) &return_to_handler; - /* Nmi's are currently unsupported */ - if (unlikely(in_nmi())) - return; - if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index b0cdde6932f5..74656d1d4e30 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -104,7 +104,7 @@ static int show_other_interrupts(struct seq_file *p, int prec) seq_printf(p, " Threshold APIC interrupts\n"); # endif #endif -#ifdef CONFIG_X86_NEW_MCE +#ifdef CONFIG_X86_MCE seq_printf(p, "%*s: ", prec, "MCE"); for_each_online_cpu(j) seq_printf(p, "%10u ", per_cpu(mce_exception_count, j)); @@ -200,7 +200,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu) sum += irq_stats(cpu)->irq_threshold_count; # endif #endif -#ifdef CONFIG_X86_NEW_MCE +#ifdef CONFIG_X86_MCE sum += per_cpu(mce_exception_count, cpu); sum += per_cpu(mce_poll_count, cpu); #endif diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 92b7703d3d58..ccf8ab54f31a 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -190,7 +190,7 @@ static void __init apic_intr_init(void) #ifdef CONFIG_X86_MCE_THRESHOLD alloc_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt); #endif -#if defined(CONFIG_X86_NEW_MCE) && defined(CONFIG_X86_LOCAL_APIC) +#if defined(CONFIG_X86_MCE) && defined(CONFIG_X86_LOCAL_APIC) alloc_intr_gate(MCE_SELF_VECTOR, mce_self_interrupt); #endif diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index ae13e34f7248..2cf12fd32c83 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -147,7 +147,7 @@ again: return NULL; addr = page_to_phys(page); - if (!is_buffer_dma_capable(dma_mask, addr, size)) { + if (addr + size > dma_mask) { __free_pages(page, get_order(size)); if (dma_mask < DMA_BIT_MASK(32) && !(flag & GFP_DMA)) { diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index d2e56b8f48e7..98a827ee9ed7 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -190,14 +190,13 @@ static void iommu_full(struct device *dev, size_t size, int dir) static inline int need_iommu(struct device *dev, unsigned long addr, size_t size) { - return force_iommu || - !is_buffer_dma_capable(*dev->dma_mask, addr, size); + return force_iommu || !dma_capable(dev, addr, size); } static inline int nonforced_iommu(struct device *dev, unsigned long addr, size_t size) { - return !is_buffer_dma_capable(*dev->dma_mask, addr, size); + return !dma_capable(dev, addr, size); } /* Map a single continuous physical area into the IOMMU. diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index 71d412a09f30..c0a4222bf62b 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -14,7 +14,7 @@ static int check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size) { - if (hwdev && !is_buffer_dma_capable(*hwdev->dma_mask, bus, size)) { + if (hwdev && !dma_capable(hwdev, bus, size)) { if (*hwdev->dma_mask >= DMA_BIT_MASK(32)) printk(KERN_ERR "nommu_%s: overflow %Lx+%zu of device mask %Lx\n", diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index 1e66b18f45cb..aaa6b7839f1e 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -13,31 +13,6 @@ int swiotlb __read_mostly; -void * __init swiotlb_alloc_boot(size_t size, unsigned long nslabs) -{ - return alloc_bootmem_low_pages(size); -} - -void *swiotlb_alloc(unsigned order, unsigned long nslabs) -{ - return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order); -} - -dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr) -{ - return paddr; -} - -phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr) -{ - return baddr; -} - -int __weak swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size) -{ - return 0; -} - static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags) { diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 59f4524984af..4cf79567cdab 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -61,9 +61,6 @@ asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); -DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; -EXPORT_PER_CPU_SYMBOL(current_task); - /* * Return saved PC of a blocked thread. */ @@ -350,14 +347,21 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) *next = &next_p->thread; int cpu = smp_processor_id(); struct tss_struct *tss = &per_cpu(init_tss, cpu); + bool preload_fpu; /* never put a printk in __switch_to... printk() calls wake_up*() indirectly */ - __unlazy_fpu(prev_p); + /* + * If the task has used fpu the last 5 timeslices, just do a full + * restore of the math state immediately to avoid the trap; the + * chances of needing FPU soon are obviously high now + */ + preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5; + __unlazy_fpu(prev_p); /* we're going to use this soon, after a few expensive things */ - if (next_p->fpu_counter > 5) + if (preload_fpu) prefetch(next->xstate); /* @@ -398,6 +402,11 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT)) __switch_to_xtra(prev_p, next_p, tss); + /* If we're going to preload the fpu context, make sure clts + is run while we're batching the cpu state updates. */ + if (preload_fpu) + clts(); + /* * Leave lazy mode, flushing any hypercalls made here. * This must be done before restoring TLS segments so @@ -407,15 +416,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) */ arch_end_context_switch(next_p); - /* If the task has used fpu the last 5 timeslices, just do a full - * restore of the math state immediately to avoid the trap; the - * chances of needing FPU soon are obviously high now - * - * tsk_used_math() checks prevent calling math_state_restore(), - * which can sleep in the case of !tsk_used_math() - */ - if (tsk_used_math(next_p) && next_p->fpu_counter > 5) - math_state_restore(); + if (preload_fpu) + __math_state_restore(); /* * Restore %gs if needed (which is common) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index ebefb5407b9d..ad535b683170 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -55,9 +55,6 @@ asmlinkage extern void ret_from_fork(void); -DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; -EXPORT_PER_CPU_SYMBOL(current_task); - DEFINE_PER_CPU(unsigned long, old_rsp); static DEFINE_PER_CPU(unsigned char, is_idle); @@ -386,9 +383,17 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) int cpu = smp_processor_id(); struct tss_struct *tss = &per_cpu(init_tss, cpu); unsigned fsindex, gsindex; + bool preload_fpu; + + /* + * If the task has used fpu the last 5 timeslices, just do a full + * restore of the math state immediately to avoid the trap; the + * chances of needing FPU soon are obviously high now + */ + preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5; /* we're going to use this soon, after a few expensive things */ - if (next_p->fpu_counter > 5) + if (preload_fpu) prefetch(next->xstate); /* @@ -419,6 +424,13 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) load_TLS(next, cpu); + /* Must be after DS reload */ + unlazy_fpu(prev_p); + + /* Make sure cpu is ready for new context */ + if (preload_fpu) + clts(); + /* * Leave lazy mode, flushing any hypercalls made here. * This must be done before restoring TLS segments so @@ -459,9 +471,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) wrmsrl(MSR_KERNEL_GS_BASE, next->gs); prev->gsindex = gsindex; - /* Must be after DS reload */ - unlazy_fpu(prev_p); - /* * Switch the PDA and FPU contexts. */ @@ -480,15 +489,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV)) __switch_to_xtra(prev_p, next_p, tss); - /* If the task has used fpu the last 5 timeslices, just do a full - * restore of the math state immediately to avoid the trap; the - * chances of needing FPU soon are obviously high now - * - * tsk_used_math() checks prevent calling math_state_restore(), - * which can sleep in the case of !tsk_used_math() + /* + * Preload the FPU context, now that we've determined that the + * task is likely to be using it. */ - if (tsk_used_math(next_p) && next_p->fpu_counter > 5) - math_state_restore(); + if (preload_fpu) + __math_state_restore(); return prev_p; } diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 4c578751e94e..cc26ad4c3070 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -856,7 +856,7 @@ static void do_signal(struct pt_regs *regs) void do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) { -#ifdef CONFIG_X86_NEW_MCE +#ifdef CONFIG_X86_MCE /* notify userspace of pending MCEs */ if (thread_info_flags & _TIF_MCE_NOTIFY) mce_notify_process(); diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c index e8b9863ef8c4..3149032ff107 100644 --- a/arch/x86/kernel/step.c +++ b/arch/x86/kernel/step.c @@ -4,6 +4,7 @@ #include <linux/sched.h> #include <linux/mm.h> #include <linux/ptrace.h> +#include <asm/desc.h> unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs) { @@ -23,7 +24,7 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re * and APM bios ones we just ignore here. */ if ((seg & SEGMENT_TI_MASK) == SEGMENT_LDT) { - u32 *desc; + struct desc_struct *desc; unsigned long base; seg &= ~7UL; @@ -33,12 +34,10 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re addr = -1L; /* bogus selector, access would fault */ else { desc = child->mm->context.ldt + seg; - base = ((desc[0] >> 16) | - ((desc[1] & 0xff) << 16) | - (desc[1] & 0xff000000)); + base = get_desc_base(desc); /* 16-bit code segment? */ - if (!((desc[1] >> 22) & 1)) + if (!desc->d) addr &= 0xffff; addr += base; } diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 5204332f475d..83264922a878 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -76,7 +76,7 @@ char ignore_fpu_irq; * F0 0F bug workaround.. We have a special link segment * for this. */ -gate_desc idt_table[256] +gate_desc idt_table[NR_VECTORS] __attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, }; #endif @@ -786,33 +786,34 @@ do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) #endif } -#ifdef CONFIG_X86_32 -unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) +asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) { - struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id()); - unsigned long base = (kesp - uesp) & -THREAD_SIZE; - unsigned long new_kesp = kesp - base; - unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT; - __u64 desc = *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS]; - - /* Set up base for espfix segment */ - desc &= 0x00f0ff0000000000ULL; - desc |= ((((__u64)base) << 16) & 0x000000ffffff0000ULL) | - ((((__u64)base) << 32) & 0xff00000000000000ULL) | - ((((__u64)lim_pages) << 32) & 0x000f000000000000ULL) | - (lim_pages & 0xffff); - *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS] = desc; - - return new_kesp; } -#endif -asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) +asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void) { } -asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void) +/* + * __math_state_restore assumes that cr0.TS is already clear and the + * fpu state is all ready for use. Used during context switch. + */ +void __math_state_restore(void) { + struct thread_info *thread = current_thread_info(); + struct task_struct *tsk = thread->task; + + /* + * Paranoid restore. send a SIGSEGV if we fail to restore the state. + */ + if (unlikely(restore_fpu_checking(tsk))) { + stts(); + force_sig(SIGSEGV, tsk); + return; + } + + thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */ + tsk->fpu_counter++; } /* @@ -846,17 +847,8 @@ asmlinkage void math_state_restore(void) } clts(); /* Allow maths ops (or we recurse) */ - /* - * Paranoid restore. send a SIGSEGV if we fail to restore the state. - */ - if (unlikely(restore_fpu_checking(tsk))) { - stts(); - force_sig(SIGSEGV, tsk); - return; - } - thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */ - tsk->fpu_counter++; + __math_state_restore(); } EXPORT_SYMBOL_GPL(math_state_restore); diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index bfae139182ff..775a020990a5 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -285,26 +285,25 @@ check_v8086_mode(struct pt_regs *regs, unsigned long address, tsk->thread.screen_bitmap |= 1 << bit; } -static void dump_pagetable(unsigned long address) +static bool low_pfn(unsigned long pfn) { - __typeof__(pte_val(__pte(0))) page; + return pfn < max_low_pfn; +} - page = read_cr3(); - page = ((__typeof__(page) *) __va(page))[address >> PGDIR_SHIFT]; +static void dump_pagetable(unsigned long address) +{ + pgd_t *base = __va(read_cr3()); + pgd_t *pgd = &base[pgd_index(address)]; + pmd_t *pmd; + pte_t *pte; #ifdef CONFIG_X86_PAE - printk("*pdpt = %016Lx ", page); - if ((page >> PAGE_SHIFT) < max_low_pfn - && page & _PAGE_PRESENT) { - page &= PAGE_MASK; - page = ((__typeof__(page) *) __va(page))[(address >> PMD_SHIFT) - & (PTRS_PER_PMD - 1)]; - printk(KERN_CONT "*pde = %016Lx ", page); - page &= ~_PAGE_NX; - } -#else - printk("*pde = %08lx ", page); + printk("*pdpt = %016Lx ", pgd_val(*pgd)); + if (!low_pfn(pgd_val(*pgd) >> PAGE_SHIFT) || !pgd_present(*pgd)) + goto out; #endif + pmd = pmd_offset(pud_offset(pgd, address), address); + printk(KERN_CONT "*pde = %0*Lx ", sizeof(*pmd) * 2, (u64)pmd_val(*pmd)); /* * We must not directly access the pte in the highpte @@ -312,16 +311,12 @@ static void dump_pagetable(unsigned long address) * And let's rather not kmap-atomic the pte, just in case * it's allocated already: */ - if ((page >> PAGE_SHIFT) < max_low_pfn - && (page & _PAGE_PRESENT) - && !(page & _PAGE_PSE)) { - - page &= PAGE_MASK; - page = ((__typeof__(page) *) __va(page))[(address >> PAGE_SHIFT) - & (PTRS_PER_PTE - 1)]; - printk("*pte = %0*Lx ", sizeof(page)*2, (u64)page); - } + if (!low_pfn(pmd_pfn(*pmd)) || !pmd_present(*pmd) || pmd_large(*pmd)) + goto out; + pte = pte_offset_kernel(pmd, address); + printk("*pte = %0*Lx ", sizeof(*pte) * 2, (u64)pte_val(*pte)); +out: printk("\n"); } @@ -450,16 +445,12 @@ static int bad_address(void *p) static void dump_pagetable(unsigned long address) { - pgd_t *pgd; + pgd_t *base = __va(read_cr3() & PHYSICAL_PAGE_MASK); + pgd_t *pgd = base + pgd_index(address); pud_t *pud; pmd_t *pmd; pte_t *pte; - pgd = (pgd_t *)read_cr3(); - - pgd = __va((unsigned long)pgd & PHYSICAL_PAGE_MASK); - - pgd += pgd_index(address); if (bad_address(pgd)) goto bad; diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c index 2112ed55e7ea..1617958a3805 100644 --- a/arch/x86/mm/highmem_32.c +++ b/arch/x86/mm/highmem_32.c @@ -24,7 +24,7 @@ void kunmap(struct page *page) * no global lock is needed and because the kmap code must perform a global TLB * invalidation when the kmap pool wraps. * - * However when holding an atomic kmap is is not legal to sleep, so atomic + * However when holding an atomic kmap it is not legal to sleep, so atomic * kmaps are appropriate for short, tight code paths only. */ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot) diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 89b9a5cd63da..cb88b1a0bd5f 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -1,11 +1,14 @@ /** * @file nmi_int.c * - * @remark Copyright 2002-2008 OProfile authors + * @remark Copyright 2002-2009 OProfile authors * @remark Read the file COPYING * * @author John Levon <levon@movementarian.org> * @author Robert Richter <robert.richter@amd.com> + * @author Barry Kasindorf <barry.kasindorf@amd.com> + * @author Jason Yeh <jason.yeh@amd.com> + * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> */ #include <linux/init.h> @@ -24,13 +27,35 @@ #include "op_counter.h" #include "op_x86_model.h" -static struct op_x86_model_spec const *model; +static struct op_x86_model_spec *model; static DEFINE_PER_CPU(struct op_msrs, cpu_msrs); static DEFINE_PER_CPU(unsigned long, saved_lvtpc); /* 0 == registered but off, 1 == registered and on */ static int nmi_enabled = 0; +struct op_counter_config counter_config[OP_MAX_COUNTER]; + +/* common functions */ + +u64 op_x86_get_ctrl(struct op_x86_model_spec const *model, + struct op_counter_config *counter_config) +{ + u64 val = 0; + u16 event = (u16)counter_config->event; + + val |= ARCH_PERFMON_EVENTSEL_INT; + val |= counter_config->user ? ARCH_PERFMON_EVENTSEL_USR : 0; + val |= counter_config->kernel ? ARCH_PERFMON_EVENTSEL_OS : 0; + val |= (counter_config->unit_mask & 0xFF) << 8; + event &= model->event_mask ? model->event_mask : 0xFF; + val |= event & 0xFF; + val |= (event & 0x0F00) << 24; + + return val; +} + + static int profile_exceptions_notify(struct notifier_block *self, unsigned long val, void *data) { @@ -52,36 +77,214 @@ static int profile_exceptions_notify(struct notifier_block *self, static void nmi_cpu_save_registers(struct op_msrs *msrs) { - unsigned int const nr_ctrs = model->num_counters; - unsigned int const nr_ctrls = model->num_controls; struct op_msr *counters = msrs->counters; struct op_msr *controls = msrs->controls; unsigned int i; - for (i = 0; i < nr_ctrs; ++i) { - if (counters[i].addr) { - rdmsr(counters[i].addr, - counters[i].saved.low, - counters[i].saved.high); - } + for (i = 0; i < model->num_counters; ++i) { + if (counters[i].addr) + rdmsrl(counters[i].addr, counters[i].saved); + } + + for (i = 0; i < model->num_controls; ++i) { + if (controls[i].addr) + rdmsrl(controls[i].addr, controls[i].saved); + } +} + +static void nmi_cpu_start(void *dummy) +{ + struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs); + model->start(msrs); +} + +static int nmi_start(void) +{ + on_each_cpu(nmi_cpu_start, NULL, 1); + return 0; +} + +static void nmi_cpu_stop(void *dummy) +{ + struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs); + model->stop(msrs); +} + +static void nmi_stop(void) +{ + on_each_cpu(nmi_cpu_stop, NULL, 1); +} + +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + +static DEFINE_PER_CPU(int, switch_index); + +static inline int has_mux(void) +{ + return !!model->switch_ctrl; +} + +inline int op_x86_phys_to_virt(int phys) +{ + return __get_cpu_var(switch_index) + phys; +} + +inline int op_x86_virt_to_phys(int virt) +{ + return virt % model->num_counters; +} + +static void nmi_shutdown_mux(void) +{ + int i; + + if (!has_mux()) + return; + + for_each_possible_cpu(i) { + kfree(per_cpu(cpu_msrs, i).multiplex); + per_cpu(cpu_msrs, i).multiplex = NULL; + per_cpu(switch_index, i) = 0; } +} + +static int nmi_setup_mux(void) +{ + size_t multiplex_size = + sizeof(struct op_msr) * model->num_virt_counters; + int i; + + if (!has_mux()) + return 1; + + for_each_possible_cpu(i) { + per_cpu(cpu_msrs, i).multiplex = + kmalloc(multiplex_size, GFP_KERNEL); + if (!per_cpu(cpu_msrs, i).multiplex) + return 0; + } + + return 1; +} + +static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) +{ + int i; + struct op_msr *multiplex = msrs->multiplex; + + if (!has_mux()) + return; - for (i = 0; i < nr_ctrls; ++i) { - if (controls[i].addr) { - rdmsr(controls[i].addr, - controls[i].saved.low, - controls[i].saved.high); + for (i = 0; i < model->num_virt_counters; ++i) { + if (counter_config[i].enabled) { + multiplex[i].saved = -(u64)counter_config[i].count; + } else { + multiplex[i].addr = 0; + multiplex[i].saved = 0; } } + + per_cpu(switch_index, cpu) = 0; +} + +static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs) +{ + struct op_msr *multiplex = msrs->multiplex; + int i; + + for (i = 0; i < model->num_counters; ++i) { + int virt = op_x86_phys_to_virt(i); + if (multiplex[virt].addr) + rdmsrl(multiplex[virt].addr, multiplex[virt].saved); + } +} + +static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs) +{ + struct op_msr *multiplex = msrs->multiplex; + int i; + + for (i = 0; i < model->num_counters; ++i) { + int virt = op_x86_phys_to_virt(i); + if (multiplex[virt].addr) + wrmsrl(multiplex[virt].addr, multiplex[virt].saved); + } } -static void nmi_save_registers(void *dummy) +static void nmi_cpu_switch(void *dummy) { int cpu = smp_processor_id(); + int si = per_cpu(switch_index, cpu); struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu); - nmi_cpu_save_registers(msrs); + + nmi_cpu_stop(NULL); + nmi_cpu_save_mpx_registers(msrs); + + /* move to next set */ + si += model->num_counters; + if ((si > model->num_virt_counters) || (counter_config[si].count == 0)) + per_cpu(switch_index, cpu) = 0; + else + per_cpu(switch_index, cpu) = si; + + model->switch_ctrl(model, msrs); + nmi_cpu_restore_mpx_registers(msrs); + + nmi_cpu_start(NULL); +} + + +/* + * Quick check to see if multiplexing is necessary. + * The check should be sufficient since counters are used + * in ordre. + */ +static int nmi_multiplex_on(void) +{ + return counter_config[model->num_counters].count ? 0 : -EINVAL; +} + +static int nmi_switch_event(void) +{ + if (!has_mux()) + return -ENOSYS; /* not implemented */ + if (nmi_multiplex_on() < 0) + return -EINVAL; /* not necessary */ + + on_each_cpu(nmi_cpu_switch, NULL, 1); + + return 0; +} + +static inline void mux_init(struct oprofile_operations *ops) +{ + if (has_mux()) + ops->switch_events = nmi_switch_event; +} + +static void mux_clone(int cpu) +{ + if (!has_mux()) + return; + + memcpy(per_cpu(cpu_msrs, cpu).multiplex, + per_cpu(cpu_msrs, 0).multiplex, + sizeof(struct op_msr) * model->num_virt_counters); } +#else + +inline int op_x86_phys_to_virt(int phys) { return phys; } +inline int op_x86_virt_to_phys(int virt) { return virt; } +static inline void nmi_shutdown_mux(void) { } +static inline int nmi_setup_mux(void) { return 1; } +static inline void +nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { } +static inline void mux_init(struct oprofile_operations *ops) { } +static void mux_clone(int cpu) { } + +#endif + static void free_msrs(void) { int i; @@ -95,38 +298,32 @@ static void free_msrs(void) static int allocate_msrs(void) { - int success = 1; size_t controls_size = sizeof(struct op_msr) * model->num_controls; size_t counters_size = sizeof(struct op_msr) * model->num_counters; int i; for_each_possible_cpu(i) { per_cpu(cpu_msrs, i).counters = kmalloc(counters_size, - GFP_KERNEL); - if (!per_cpu(cpu_msrs, i).counters) { - success = 0; - break; - } + GFP_KERNEL); + if (!per_cpu(cpu_msrs, i).counters) + return 0; per_cpu(cpu_msrs, i).controls = kmalloc(controls_size, - GFP_KERNEL); - if (!per_cpu(cpu_msrs, i).controls) { - success = 0; - break; - } + GFP_KERNEL); + if (!per_cpu(cpu_msrs, i).controls) + return 0; } - if (!success) - free_msrs(); - - return success; + return 1; } static void nmi_cpu_setup(void *dummy) { int cpu = smp_processor_id(); struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu); + nmi_cpu_save_registers(msrs); spin_lock(&oprofilefs_lock); - model->setup_ctrs(msrs); + model->setup_ctrs(model, msrs); + nmi_cpu_setup_mux(cpu, msrs); spin_unlock(&oprofilefs_lock); per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC); apic_write(APIC_LVTPC, APIC_DM_NMI); @@ -144,11 +341,15 @@ static int nmi_setup(void) int cpu; if (!allocate_msrs()) - return -ENOMEM; + err = -ENOMEM; + else if (!nmi_setup_mux()) + err = -ENOMEM; + else + err = register_die_notifier(&profile_exceptions_nb); - err = register_die_notifier(&profile_exceptions_nb); if (err) { free_msrs(); + nmi_shutdown_mux(); return err; } @@ -159,45 +360,38 @@ static int nmi_setup(void) /* Assume saved/restored counters are the same on all CPUs */ model->fill_in_addresses(&per_cpu(cpu_msrs, 0)); for_each_possible_cpu(cpu) { - if (cpu != 0) { - memcpy(per_cpu(cpu_msrs, cpu).counters, - per_cpu(cpu_msrs, 0).counters, - sizeof(struct op_msr) * model->num_counters); - - memcpy(per_cpu(cpu_msrs, cpu).controls, - per_cpu(cpu_msrs, 0).controls, - sizeof(struct op_msr) * model->num_controls); - } + if (!cpu) + continue; + + memcpy(per_cpu(cpu_msrs, cpu).counters, + per_cpu(cpu_msrs, 0).counters, + sizeof(struct op_msr) * model->num_counters); + + memcpy(per_cpu(cpu_msrs, cpu).controls, + per_cpu(cpu_msrs, 0).controls, + sizeof(struct op_msr) * model->num_controls); + mux_clone(cpu); } - on_each_cpu(nmi_save_registers, NULL, 1); on_each_cpu(nmi_cpu_setup, NULL, 1); nmi_enabled = 1; return 0; } -static void nmi_restore_registers(struct op_msrs *msrs) +static void nmi_cpu_restore_registers(struct op_msrs *msrs) { - unsigned int const nr_ctrs = model->num_counters; - unsigned int const nr_ctrls = model->num_controls; struct op_msr *counters = msrs->counters; struct op_msr *controls = msrs->controls; unsigned int i; - for (i = 0; i < nr_ctrls; ++i) { - if (controls[i].addr) { - wrmsr(controls[i].addr, - controls[i].saved.low, - controls[i].saved.high); - } + for (i = 0; i < model->num_controls; ++i) { + if (controls[i].addr) + wrmsrl(controls[i].addr, controls[i].saved); } - for (i = 0; i < nr_ctrs; ++i) { - if (counters[i].addr) { - wrmsr(counters[i].addr, - counters[i].saved.low, - counters[i].saved.high); - } + for (i = 0; i < model->num_counters; ++i) { + if (counters[i].addr) + wrmsrl(counters[i].addr, counters[i].saved); } } @@ -205,7 +399,7 @@ static void nmi_cpu_shutdown(void *dummy) { unsigned int v; int cpu = smp_processor_id(); - struct op_msrs *msrs = &__get_cpu_var(cpu_msrs); + struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu); /* restoring APIC_LVTPC can trigger an apic error because the delivery * mode and vector nr combination can be illegal. That's by design: on @@ -216,7 +410,7 @@ static void nmi_cpu_shutdown(void *dummy) apic_write(APIC_LVTERR, v | APIC_LVT_MASKED); apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu)); apic_write(APIC_LVTERR, v); - nmi_restore_registers(msrs); + nmi_cpu_restore_registers(msrs); } static void nmi_shutdown(void) @@ -226,42 +420,18 @@ static void nmi_shutdown(void) nmi_enabled = 0; on_each_cpu(nmi_cpu_shutdown, NULL, 1); unregister_die_notifier(&profile_exceptions_nb); + nmi_shutdown_mux(); msrs = &get_cpu_var(cpu_msrs); model->shutdown(msrs); free_msrs(); put_cpu_var(cpu_msrs); } -static void nmi_cpu_start(void *dummy) -{ - struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs); - model->start(msrs); -} - -static int nmi_start(void) -{ - on_each_cpu(nmi_cpu_start, NULL, 1); - return 0; -} - -static void nmi_cpu_stop(void *dummy) -{ - struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs); - model->stop(msrs); -} - -static void nmi_stop(void) -{ - on_each_cpu(nmi_cpu_stop, NULL, 1); -} - -struct op_counter_config counter_config[OP_MAX_COUNTER]; - static int nmi_create_files(struct super_block *sb, struct dentry *root) { unsigned int i; - for (i = 0; i < model->num_counters; ++i) { + for (i = 0; i < model->num_virt_counters; ++i) { struct dentry *dir; char buf[4]; @@ -270,7 +440,7 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root) * NOTE: assumes 1:1 mapping here (that counters are organized * sequentially in their struct assignment). */ - if (unlikely(!avail_to_resrv_perfctr_nmi_bit(i))) + if (!avail_to_resrv_perfctr_nmi_bit(op_x86_virt_to_phys(i))) continue; snprintf(buf, sizeof(buf), "%d", i); @@ -402,6 +572,7 @@ module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0); static int __init ppro_init(char **cpu_type) { __u8 cpu_model = boot_cpu_data.x86_model; + struct op_x86_model_spec *spec = &op_ppro_spec; /* default */ if (force_arch_perfmon && cpu_has_arch_perfmon) return 0; @@ -428,7 +599,7 @@ static int __init ppro_init(char **cpu_type) *cpu_type = "i386/core_2"; break; case 26: - arch_perfmon_setup_counters(); + spec = &op_arch_perfmon_spec; *cpu_type = "i386/core_i7"; break; case 28: @@ -439,17 +610,7 @@ static int __init ppro_init(char **cpu_type) return 0; } - model = &op_ppro_spec; - return 1; -} - -static int __init arch_perfmon_init(char **cpu_type) -{ - if (!cpu_has_arch_perfmon) - return 0; - *cpu_type = "i386/arch_perfmon"; - model = &op_arch_perfmon_spec; - arch_perfmon_setup_counters(); + model = spec; return 1; } @@ -471,27 +632,26 @@ int __init op_nmi_init(struct oprofile_operations *ops) /* Needs to be at least an Athlon (or hammer in 32bit mode) */ switch (family) { - default: - return -ENODEV; case 6: - model = &op_amd_spec; cpu_type = "i386/athlon"; break; case 0xf: - model = &op_amd_spec; - /* Actually it could be i386/hammer too, but give - user space an consistent name. */ + /* + * Actually it could be i386/hammer too, but + * give user space an consistent name. + */ cpu_type = "x86-64/hammer"; break; case 0x10: - model = &op_amd_spec; cpu_type = "x86-64/family10"; break; case 0x11: - model = &op_amd_spec; cpu_type = "x86-64/family11h"; break; + default: + return -ENODEV; } + model = &op_amd_spec; break; case X86_VENDOR_INTEL: @@ -510,8 +670,15 @@ int __init op_nmi_init(struct oprofile_operations *ops) break; } - if (!cpu_type && !arch_perfmon_init(&cpu_type)) + if (cpu_type) + break; + + if (!cpu_has_arch_perfmon) return -ENODEV; + + /* use arch perfmon as fallback */ + cpu_type = "i386/arch_perfmon"; + model = &op_arch_perfmon_spec; break; default: @@ -522,18 +689,23 @@ int __init op_nmi_init(struct oprofile_operations *ops) register_cpu_notifier(&oprofile_cpu_nb); #endif /* default values, can be overwritten by model */ - ops->create_files = nmi_create_files; - ops->setup = nmi_setup; - ops->shutdown = nmi_shutdown; - ops->start = nmi_start; - ops->stop = nmi_stop; - ops->cpu_type = cpu_type; + ops->create_files = nmi_create_files; + ops->setup = nmi_setup; + ops->shutdown = nmi_shutdown; + ops->start = nmi_start; + ops->stop = nmi_stop; + ops->cpu_type = cpu_type; if (model->init) ret = model->init(ops); if (ret) return ret; + if (!model->num_virt_counters) + model->num_virt_counters = model->num_counters; + + mux_init(ops); + init_sysfs(); using_nmi = 1; printk(KERN_INFO "oprofile: using NMI interrupt.\n"); diff --git a/arch/x86/oprofile/op_counter.h b/arch/x86/oprofile/op_counter.h index 91b6a116165e..e28398df0df2 100644 --- a/arch/x86/oprofile/op_counter.h +++ b/arch/x86/oprofile/op_counter.h @@ -10,7 +10,7 @@ #ifndef OP_COUNTER_H #define OP_COUNTER_H -#define OP_MAX_COUNTER 8 +#define OP_MAX_COUNTER 32 /* Per-perfctr configuration as set via * oprofilefs. diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index 8fdf06e4edf9..39686c29f03a 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -9,12 +9,15 @@ * @author Philippe Elie * @author Graydon Hoare * @author Robert Richter <robert.richter@amd.com> - * @author Barry Kasindorf + * @author Barry Kasindorf <barry.kasindorf@amd.com> + * @author Jason Yeh <jason.yeh@amd.com> + * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> */ #include <linux/oprofile.h> #include <linux/device.h> #include <linux/pci.h> +#include <linux/percpu.h> #include <asm/ptrace.h> #include <asm/msr.h> @@ -25,43 +28,36 @@ #define NUM_COUNTERS 4 #define NUM_CONTROLS 4 +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX +#define NUM_VIRT_COUNTERS 32 +#define NUM_VIRT_CONTROLS 32 +#else +#define NUM_VIRT_COUNTERS NUM_COUNTERS +#define NUM_VIRT_CONTROLS NUM_CONTROLS +#endif + +#define OP_EVENT_MASK 0x0FFF +#define OP_CTR_OVERFLOW (1ULL<<31) -#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0) -#define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0) -#define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0) -#define CTR_OVERFLOWED(n) (!((n) & (1U<<31))) - -#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0) -#define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0) -#define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0) -#define CTRL_SET_ACTIVE(n) (n |= (1<<22)) -#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22)) -#define CTRL_CLEAR_LO(x) (x &= (1<<21)) -#define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0) -#define CTRL_SET_ENABLE(val) (val |= 1<<20) -#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16)) -#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17)) -#define CTRL_SET_UM(val, m) (val |= (m << 8)) -#define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff)) -#define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf)) -#define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9)) -#define CTRL_SET_GUEST_ONLY(val, h) (val |= ((h & 1) << 8)) - -static unsigned long reset_value[NUM_COUNTERS]; +#define MSR_AMD_EVENTSEL_RESERVED ((0xFFFFFCF0ULL<<32)|(1ULL<<21)) + +static unsigned long reset_value[NUM_VIRT_COUNTERS]; #ifdef CONFIG_OPROFILE_IBS /* IbsFetchCtl bits/masks */ -#define IBS_FETCH_HIGH_VALID_BIT (1UL << 17) /* bit 49 */ -#define IBS_FETCH_HIGH_ENABLE (1UL << 16) /* bit 48 */ -#define IBS_FETCH_LOW_MAX_CNT_MASK 0x0000FFFFUL /* MaxCnt mask */ +#define IBS_FETCH_RAND_EN (1ULL<<57) +#define IBS_FETCH_VAL (1ULL<<49) +#define IBS_FETCH_ENABLE (1ULL<<48) +#define IBS_FETCH_CNT_MASK 0xFFFF0000ULL /*IbsOpCtl bits */ -#define IBS_OP_LOW_VALID_BIT (1ULL<<18) /* bit 18 */ -#define IBS_OP_LOW_ENABLE (1ULL<<17) /* bit 17 */ +#define IBS_OP_CNT_CTL (1ULL<<19) +#define IBS_OP_VAL (1ULL<<18) +#define IBS_OP_ENABLE (1ULL<<17) -#define IBS_FETCH_SIZE 6 -#define IBS_OP_SIZE 12 +#define IBS_FETCH_SIZE 6 +#define IBS_OP_SIZE 12 static int has_ibs; /* AMD Family10h and later */ @@ -78,6 +74,45 @@ static struct op_ibs_config ibs_config; #endif +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + +static void op_mux_fill_in_addresses(struct op_msrs * const msrs) +{ + int i; + + for (i = 0; i < NUM_VIRT_COUNTERS; i++) { + int hw_counter = op_x86_virt_to_phys(i); + if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i)) + msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter; + else + msrs->multiplex[i].addr = 0; + } +} + +static void op_mux_switch_ctrl(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs) +{ + u64 val; + int i; + + /* enable active counters */ + for (i = 0; i < NUM_COUNTERS; ++i) { + int virt = op_x86_phys_to_virt(i); + if (!counter_config[virt].enabled) + continue; + rdmsrl(msrs->controls[i].addr, val); + val &= model->reserved; + val |= op_x86_get_ctrl(model, &counter_config[virt]); + wrmsrl(msrs->controls[i].addr, val); + } +} + +#else + +static inline void op_mux_fill_in_addresses(struct op_msrs * const msrs) { } + +#endif + /* functions for op_amd_spec */ static void op_amd_fill_in_addresses(struct op_msrs * const msrs) @@ -97,150 +132,174 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs) else msrs->controls[i].addr = 0; } -} + op_mux_fill_in_addresses(msrs); +} -static void op_amd_setup_ctrs(struct op_msrs const * const msrs) +static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; + /* setup reset_value */ + for (i = 0; i < NUM_VIRT_COUNTERS; ++i) { + if (counter_config[i].enabled) + reset_value[i] = counter_config[i].count; + else + reset_value[i] = 0; + } + /* clear all counters */ - for (i = 0 ; i < NUM_CONTROLS; ++i) { - if (unlikely(!CTRL_IS_RESERVED(msrs, i))) + for (i = 0; i < NUM_CONTROLS; ++i) { + if (unlikely(!msrs->controls[i].addr)) continue; - CTRL_READ(low, high, msrs, i); - CTRL_CLEAR_LO(low); - CTRL_CLEAR_HI(high); - CTRL_WRITE(low, high, msrs, i); + rdmsrl(msrs->controls[i].addr, val); + val &= model->reserved; + wrmsrl(msrs->controls[i].addr, val); } /* avoid a false detection of ctr overflows in NMI handler */ for (i = 0; i < NUM_COUNTERS; ++i) { - if (unlikely(!CTR_IS_RESERVED(msrs, i))) + if (unlikely(!msrs->counters[i].addr)) continue; - CTR_WRITE(1, msrs, i); + wrmsrl(msrs->counters[i].addr, -1LL); } /* enable active counters */ for (i = 0; i < NUM_COUNTERS; ++i) { - if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) { - reset_value[i] = counter_config[i].count; + int virt = op_x86_phys_to_virt(i); + if (!counter_config[virt].enabled) + continue; + if (!msrs->counters[i].addr) + continue; - CTR_WRITE(counter_config[i].count, msrs, i); - - CTRL_READ(low, high, msrs, i); - CTRL_CLEAR_LO(low); - CTRL_CLEAR_HI(high); - CTRL_SET_ENABLE(low); - CTRL_SET_USR(low, counter_config[i].user); - CTRL_SET_KERN(low, counter_config[i].kernel); - CTRL_SET_UM(low, counter_config[i].unit_mask); - CTRL_SET_EVENT_LOW(low, counter_config[i].event); - CTRL_SET_EVENT_HIGH(high, counter_config[i].event); - CTRL_SET_HOST_ONLY(high, 0); - CTRL_SET_GUEST_ONLY(high, 0); - - CTRL_WRITE(low, high, msrs, i); - } else { - reset_value[i] = 0; - } + /* setup counter registers */ + wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]); + + /* setup control registers */ + rdmsrl(msrs->controls[i].addr, val); + val &= model->reserved; + val |= op_x86_get_ctrl(model, &counter_config[virt]); + wrmsrl(msrs->controls[i].addr, val); } } #ifdef CONFIG_OPROFILE_IBS -static inline int +static inline void op_amd_handle_ibs(struct pt_regs * const regs, struct op_msrs const * const msrs) { - u32 low, high; - u64 msr; + u64 val, ctl; struct op_entry entry; if (!has_ibs) - return 1; + return; if (ibs_config.fetch_enabled) { - rdmsr(MSR_AMD64_IBSFETCHCTL, low, high); - if (high & IBS_FETCH_HIGH_VALID_BIT) { - rdmsrl(MSR_AMD64_IBSFETCHLINAD, msr); - oprofile_write_reserve(&entry, regs, msr, + rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl); + if (ctl & IBS_FETCH_VAL) { + rdmsrl(MSR_AMD64_IBSFETCHLINAD, val); + oprofile_write_reserve(&entry, regs, val, IBS_FETCH_CODE, IBS_FETCH_SIZE); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - oprofile_add_data(&entry, low); - oprofile_add_data(&entry, high); - rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); + oprofile_add_data64(&entry, val); + oprofile_add_data64(&entry, ctl); + rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, val); + oprofile_add_data64(&entry, val); oprofile_write_commit(&entry); /* reenable the IRQ */ - high &= ~IBS_FETCH_HIGH_VALID_BIT; - high |= IBS_FETCH_HIGH_ENABLE; - low &= IBS_FETCH_LOW_MAX_CNT_MASK; - wrmsr(MSR_AMD64_IBSFETCHCTL, low, high); + ctl &= ~(IBS_FETCH_VAL | IBS_FETCH_CNT_MASK); + ctl |= IBS_FETCH_ENABLE; + wrmsrl(MSR_AMD64_IBSFETCHCTL, ctl); } } if (ibs_config.op_enabled) { - rdmsr(MSR_AMD64_IBSOPCTL, low, high); - if (low & IBS_OP_LOW_VALID_BIT) { - rdmsrl(MSR_AMD64_IBSOPRIP, msr); - oprofile_write_reserve(&entry, regs, msr, + rdmsrl(MSR_AMD64_IBSOPCTL, ctl); + if (ctl & IBS_OP_VAL) { + rdmsrl(MSR_AMD64_IBSOPRIP, val); + oprofile_write_reserve(&entry, regs, val, IBS_OP_CODE, IBS_OP_SIZE); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - rdmsrl(MSR_AMD64_IBSOPDATA, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - rdmsrl(MSR_AMD64_IBSOPDATA2, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - rdmsrl(MSR_AMD64_IBSOPDATA3, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - rdmsrl(MSR_AMD64_IBSDCLINAD, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); - rdmsrl(MSR_AMD64_IBSDCPHYSAD, msr); - oprofile_add_data(&entry, (u32)msr); - oprofile_add_data(&entry, (u32)(msr >> 32)); + oprofile_add_data64(&entry, val); + rdmsrl(MSR_AMD64_IBSOPDATA, val); + oprofile_add_data64(&entry, val); + rdmsrl(MSR_AMD64_IBSOPDATA2, val); + oprofile_add_data64(&entry, val); + rdmsrl(MSR_AMD64_IBSOPDATA3, val); + oprofile_add_data64(&entry, val); + rdmsrl(MSR_AMD64_IBSDCLINAD, val); + oprofile_add_data64(&entry, val); + rdmsrl(MSR_AMD64_IBSDCPHYSAD, val); + oprofile_add_data64(&entry, val); oprofile_write_commit(&entry); /* reenable the IRQ */ - high = 0; - low &= ~IBS_OP_LOW_VALID_BIT; - low |= IBS_OP_LOW_ENABLE; - wrmsr(MSR_AMD64_IBSOPCTL, low, high); + ctl &= ~IBS_OP_VAL & 0xFFFFFFFF; + ctl |= IBS_OP_ENABLE; + wrmsrl(MSR_AMD64_IBSOPCTL, ctl); } } +} - return 1; +static inline void op_amd_start_ibs(void) +{ + u64 val; + if (has_ibs && ibs_config.fetch_enabled) { + val = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF; + val |= ibs_config.rand_en ? IBS_FETCH_RAND_EN : 0; + val |= IBS_FETCH_ENABLE; + wrmsrl(MSR_AMD64_IBSFETCHCTL, val); + } + + if (has_ibs && ibs_config.op_enabled) { + val = (ibs_config.max_cnt_op >> 4) & 0xFFFF; + val |= ibs_config.dispatched_ops ? IBS_OP_CNT_CTL : 0; + val |= IBS_OP_ENABLE; + wrmsrl(MSR_AMD64_IBSOPCTL, val); + } +} + +static void op_amd_stop_ibs(void) +{ + if (has_ibs && ibs_config.fetch_enabled) + /* clear max count and enable */ + wrmsrl(MSR_AMD64_IBSFETCHCTL, 0); + + if (has_ibs && ibs_config.op_enabled) + /* clear max count and enable */ + wrmsrl(MSR_AMD64_IBSOPCTL, 0); } +#else + +static inline void op_amd_handle_ibs(struct pt_regs * const regs, + struct op_msrs const * const msrs) { } +static inline void op_amd_start_ibs(void) { } +static inline void op_amd_stop_ibs(void) { } + #endif static int op_amd_check_ctrs(struct pt_regs * const regs, struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; - for (i = 0 ; i < NUM_COUNTERS; ++i) { - if (!reset_value[i]) + for (i = 0; i < NUM_COUNTERS; ++i) { + int virt = op_x86_phys_to_virt(i); + if (!reset_value[virt]) continue; - CTR_READ(low, high, msrs, i); - if (CTR_OVERFLOWED(low)) { - oprofile_add_sample(regs, i); - CTR_WRITE(reset_value[i], msrs, i); - } + rdmsrl(msrs->counters[i].addr, val); + /* bit is clear if overflowed: */ + if (val & OP_CTR_OVERFLOW) + continue; + oprofile_add_sample(regs, virt); + wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]); } -#ifdef CONFIG_OPROFILE_IBS op_amd_handle_ibs(regs, msrs); -#endif /* See op_model_ppro.c */ return 1; @@ -248,79 +307,50 @@ static int op_amd_check_ctrs(struct pt_regs * const regs, static void op_amd_start(struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; - for (i = 0 ; i < NUM_COUNTERS ; ++i) { - if (reset_value[i]) { - CTRL_READ(low, high, msrs, i); - CTRL_SET_ACTIVE(low); - CTRL_WRITE(low, high, msrs, i); - } - } -#ifdef CONFIG_OPROFILE_IBS - if (has_ibs && ibs_config.fetch_enabled) { - low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF; - high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */ - + IBS_FETCH_HIGH_ENABLE; - wrmsr(MSR_AMD64_IBSFETCHCTL, low, high); + for (i = 0; i < NUM_COUNTERS; ++i) { + if (!reset_value[op_x86_phys_to_virt(i)]) + continue; + rdmsrl(msrs->controls[i].addr, val); + val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(msrs->controls[i].addr, val); } - if (has_ibs && ibs_config.op_enabled) { - low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF) - + ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */ - + IBS_OP_LOW_ENABLE; - high = 0; - wrmsr(MSR_AMD64_IBSOPCTL, low, high); - } -#endif + op_amd_start_ibs(); } - static void op_amd_stop(struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; /* * Subtle: stop on all counters to avoid race with setting our * pm callback */ - for (i = 0 ; i < NUM_COUNTERS ; ++i) { - if (!reset_value[i]) + for (i = 0; i < NUM_COUNTERS; ++i) { + if (!reset_value[op_x86_phys_to_virt(i)]) continue; - CTRL_READ(low, high, msrs, i); - CTRL_SET_INACTIVE(low); - CTRL_WRITE(low, high, msrs, i); - } - -#ifdef CONFIG_OPROFILE_IBS - if (has_ibs && ibs_config.fetch_enabled) { - /* clear max count and enable */ - low = 0; - high = 0; - wrmsr(MSR_AMD64_IBSFETCHCTL, low, high); + rdmsrl(msrs->controls[i].addr, val); + val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(msrs->controls[i].addr, val); } - if (has_ibs && ibs_config.op_enabled) { - /* clear max count and enable */ - low = 0; - high = 0; - wrmsr(MSR_AMD64_IBSOPCTL, low, high); - } -#endif + op_amd_stop_ibs(); } static void op_amd_shutdown(struct op_msrs const * const msrs) { int i; - for (i = 0 ; i < NUM_COUNTERS ; ++i) { - if (CTR_IS_RESERVED(msrs, i)) + for (i = 0; i < NUM_COUNTERS; ++i) { + if (msrs->counters[i].addr) release_perfctr_nmi(MSR_K7_PERFCTR0 + i); } - for (i = 0 ; i < NUM_CONTROLS ; ++i) { - if (CTRL_IS_RESERVED(msrs, i)) + for (i = 0; i < NUM_CONTROLS; ++i) { + if (msrs->controls[i].addr) release_evntsel_nmi(MSR_K7_EVNTSEL0 + i); } } @@ -490,15 +520,21 @@ static void op_amd_exit(void) {} #endif /* CONFIG_OPROFILE_IBS */ -struct op_x86_model_spec const op_amd_spec = { - .init = op_amd_init, - .exit = op_amd_exit, +struct op_x86_model_spec op_amd_spec = { .num_counters = NUM_COUNTERS, .num_controls = NUM_CONTROLS, + .num_virt_counters = NUM_VIRT_COUNTERS, + .reserved = MSR_AMD_EVENTSEL_RESERVED, + .event_mask = OP_EVENT_MASK, + .init = op_amd_init, + .exit = op_amd_exit, .fill_in_addresses = &op_amd_fill_in_addresses, .setup_ctrs = &op_amd_setup_ctrs, .check_ctrs = &op_amd_check_ctrs, .start = &op_amd_start, .stop = &op_amd_stop, - .shutdown = &op_amd_shutdown + .shutdown = &op_amd_shutdown, +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + .switch_ctrl = &op_mux_switch_ctrl, +#endif }; diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c index 819b131fd752..ac6b354becdf 100644 --- a/arch/x86/oprofile/op_model_p4.c +++ b/arch/x86/oprofile/op_model_p4.c @@ -32,6 +32,8 @@ #define NUM_CCCRS_HT2 9 #define NUM_CONTROLS_HT2 (NUM_ESCRS_HT2 + NUM_CCCRS_HT2) +#define OP_CTR_OVERFLOW (1ULL<<31) + static unsigned int num_counters = NUM_COUNTERS_NON_HT; static unsigned int num_controls = NUM_CONTROLS_NON_HT; @@ -350,8 +352,6 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { #define ESCR_SET_OS_1(escr, os) ((escr) |= (((os) & 1) << 1)) #define ESCR_SET_EVENT_SELECT(escr, sel) ((escr) |= (((sel) & 0x3f) << 25)) #define ESCR_SET_EVENT_MASK(escr, mask) ((escr) |= (((mask) & 0xffff) << 9)) -#define ESCR_READ(escr, high, ev, i) do {rdmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0) -#define ESCR_WRITE(escr, high, ev, i) do {wrmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0) #define CCCR_RESERVED_BITS 0x38030FFF #define CCCR_CLEAR(cccr) ((cccr) &= CCCR_RESERVED_BITS) @@ -361,17 +361,9 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = { #define CCCR_SET_PMI_OVF_1(cccr) ((cccr) |= (1<<27)) #define CCCR_SET_ENABLE(cccr) ((cccr) |= (1<<12)) #define CCCR_SET_DISABLE(cccr) ((cccr) &= ~(1<<12)) -#define CCCR_READ(low, high, i) do {rdmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0) -#define CCCR_WRITE(low, high, i) do {wrmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0) #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31)) #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31))) -#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0) -#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0) -#define CTR_READ(l, h, i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h)); } while (0) -#define CTR_WRITE(l, i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1); } while (0) -#define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000)) - /* this assigns a "stagger" to the current CPU, which is used throughout the code in this module as an extra array offset, to select the "even" @@ -515,7 +507,7 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) if (ev->bindings[i].virt_counter & counter_bit) { /* modify ESCR */ - ESCR_READ(escr, high, ev, i); + rdmsr(ev->bindings[i].escr_address, escr, high); ESCR_CLEAR(escr); if (stag == 0) { ESCR_SET_USR_0(escr, counter_config[ctr].user); @@ -526,10 +518,11 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) } ESCR_SET_EVENT_SELECT(escr, ev->event_select); ESCR_SET_EVENT_MASK(escr, counter_config[ctr].unit_mask); - ESCR_WRITE(escr, high, ev, i); + wrmsr(ev->bindings[i].escr_address, escr, high); /* modify CCCR */ - CCCR_READ(cccr, high, VIRT_CTR(stag, ctr)); + rdmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address, + cccr, high); CCCR_CLEAR(cccr); CCCR_SET_REQUIRED_BITS(cccr); CCCR_SET_ESCR_SELECT(cccr, ev->escr_select); @@ -537,7 +530,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) CCCR_SET_PMI_OVF_0(cccr); else CCCR_SET_PMI_OVF_1(cccr); - CCCR_WRITE(cccr, high, VIRT_CTR(stag, ctr)); + wrmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address, + cccr, high); return; } } @@ -548,7 +542,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr) } -static void p4_setup_ctrs(struct op_msrs const * const msrs) +static void p4_setup_ctrs(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs) { unsigned int i; unsigned int low, high; @@ -563,8 +558,8 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs) } /* clear the cccrs we will use */ - for (i = 0 ; i < num_counters ; i++) { - if (unlikely(!CTRL_IS_RESERVED(msrs, i))) + for (i = 0; i < num_counters; i++) { + if (unlikely(!msrs->controls[i].addr)) continue; rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); CCCR_CLEAR(low); @@ -574,17 +569,18 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs) /* clear all escrs (including those outside our concern) */ for (i = num_counters; i < num_controls; i++) { - if (unlikely(!CTRL_IS_RESERVED(msrs, i))) + if (unlikely(!msrs->controls[i].addr)) continue; wrmsr(msrs->controls[i].addr, 0, 0); } /* setup all counters */ - for (i = 0 ; i < num_counters ; ++i) { - if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs, i))) { + for (i = 0; i < num_counters; ++i) { + if (counter_config[i].enabled && msrs->controls[i].addr) { reset_value[i] = counter_config[i].count; pmc_setup_one_p4_counter(i); - CTR_WRITE(counter_config[i].count, VIRT_CTR(stag, i)); + wrmsrl(p4_counters[VIRT_CTR(stag, i)].counter_address, + -(u64)counter_config[i].count); } else { reset_value[i] = 0; } @@ -624,14 +620,16 @@ static int p4_check_ctrs(struct pt_regs * const regs, real = VIRT_CTR(stag, i); - CCCR_READ(low, high, real); - CTR_READ(ctr, high, real); - if (CCCR_OVF_P(low) || CTR_OVERFLOW_P(ctr)) { + rdmsr(p4_counters[real].cccr_address, low, high); + rdmsr(p4_counters[real].counter_address, ctr, high); + if (CCCR_OVF_P(low) || !(ctr & OP_CTR_OVERFLOW)) { oprofile_add_sample(regs, i); - CTR_WRITE(reset_value[i], real); + wrmsrl(p4_counters[real].counter_address, + -(u64)reset_value[i]); CCCR_CLEAR_OVF(low); - CCCR_WRITE(low, high, real); - CTR_WRITE(reset_value[i], real); + wrmsr(p4_counters[real].cccr_address, low, high); + wrmsrl(p4_counters[real].counter_address, + -(u64)reset_value[i]); } } @@ -653,9 +651,9 @@ static void p4_start(struct op_msrs const * const msrs) for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; - CCCR_READ(low, high, VIRT_CTR(stag, i)); + rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); CCCR_SET_ENABLE(low); - CCCR_WRITE(low, high, VIRT_CTR(stag, i)); + wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); } } @@ -670,9 +668,9 @@ static void p4_stop(struct op_msrs const * const msrs) for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; - CCCR_READ(low, high, VIRT_CTR(stag, i)); + rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); CCCR_SET_DISABLE(low); - CCCR_WRITE(low, high, VIRT_CTR(stag, i)); + wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high); } } @@ -680,8 +678,8 @@ static void p4_shutdown(struct op_msrs const * const msrs) { int i; - for (i = 0 ; i < num_counters ; ++i) { - if (CTR_IS_RESERVED(msrs, i)) + for (i = 0; i < num_counters; ++i) { + if (msrs->counters[i].addr) release_perfctr_nmi(msrs->counters[i].addr); } /* @@ -689,15 +687,15 @@ static void p4_shutdown(struct op_msrs const * const msrs) * conjunction with the counter registers (hence the starting offset). * This saves a few bits. */ - for (i = num_counters ; i < num_controls ; ++i) { - if (CTRL_IS_RESERVED(msrs, i)) + for (i = num_counters; i < num_controls; ++i) { + if (msrs->controls[i].addr) release_evntsel_nmi(msrs->controls[i].addr); } } #ifdef CONFIG_SMP -struct op_x86_model_spec const op_p4_ht2_spec = { +struct op_x86_model_spec op_p4_ht2_spec = { .num_counters = NUM_COUNTERS_HT2, .num_controls = NUM_CONTROLS_HT2, .fill_in_addresses = &p4_fill_in_addresses, @@ -709,7 +707,7 @@ struct op_x86_model_spec const op_p4_ht2_spec = { }; #endif -struct op_x86_model_spec const op_p4_spec = { +struct op_x86_model_spec op_p4_spec = { .num_counters = NUM_COUNTERS_NON_HT, .num_controls = NUM_CONTROLS_NON_HT, .fill_in_addresses = &p4_fill_in_addresses, diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 4da7230b3d17..4899215999de 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -10,6 +10,7 @@ * @author Philippe Elie * @author Graydon Hoare * @author Andi Kleen + * @author Robert Richter <robert.richter@amd.com> */ #include <linux/oprofile.h> @@ -18,7 +19,6 @@ #include <asm/msr.h> #include <asm/apic.h> #include <asm/nmi.h> -#include <asm/perf_counter.h> #include "op_x86_model.h" #include "op_counter.h" @@ -26,20 +26,7 @@ static int num_counters = 2; static int counter_width = 32; -#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0) -#define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1)))) - -#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0) -#define CTRL_READ(l, h, msrs, c) do {rdmsr((msrs->controls[(c)].addr), (l), (h)); } while (0) -#define CTRL_WRITE(l, h, msrs, c) do {wrmsr((msrs->controls[(c)].addr), (l), (h)); } while (0) -#define CTRL_SET_ACTIVE(n) (n |= (1<<22)) -#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22)) -#define CTRL_CLEAR(x) (x &= (1<<21)) -#define CTRL_SET_ENABLE(val) (val |= 1<<20) -#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16)) -#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17)) -#define CTRL_SET_UM(val, m) (val |= (m << 8)) -#define CTRL_SET_EVENT(val, e) (val |= e) +#define MSR_PPRO_EVENTSEL_RESERVED ((0xFFFFFFFFULL<<32)|(1ULL<<21)) static u64 *reset_value; @@ -63,9 +50,10 @@ static void ppro_fill_in_addresses(struct op_msrs * const msrs) } -static void ppro_setup_ctrs(struct op_msrs const * const msrs) +static void ppro_setup_ctrs(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; if (!reset_value) { @@ -93,36 +81,30 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs) } /* clear all counters */ - for (i = 0 ; i < num_counters; ++i) { - if (unlikely(!CTRL_IS_RESERVED(msrs, i))) + for (i = 0; i < num_counters; ++i) { + if (unlikely(!msrs->controls[i].addr)) continue; - CTRL_READ(low, high, msrs, i); - CTRL_CLEAR(low); - CTRL_WRITE(low, high, msrs, i); + rdmsrl(msrs->controls[i].addr, val); + val &= model->reserved; + wrmsrl(msrs->controls[i].addr, val); } /* avoid a false detection of ctr overflows in NMI handler */ for (i = 0; i < num_counters; ++i) { - if (unlikely(!CTR_IS_RESERVED(msrs, i))) + if (unlikely(!msrs->counters[i].addr)) continue; wrmsrl(msrs->counters[i].addr, -1LL); } /* enable active counters */ for (i = 0; i < num_counters; ++i) { - if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) { + if (counter_config[i].enabled && msrs->counters[i].addr) { reset_value[i] = counter_config[i].count; - wrmsrl(msrs->counters[i].addr, -reset_value[i]); - - CTRL_READ(low, high, msrs, i); - CTRL_CLEAR(low); - CTRL_SET_ENABLE(low); - CTRL_SET_USR(low, counter_config[i].user); - CTRL_SET_KERN(low, counter_config[i].kernel); - CTRL_SET_UM(low, counter_config[i].unit_mask); - CTRL_SET_EVENT(low, counter_config[i].event); - CTRL_WRITE(low, high, msrs, i); + rdmsrl(msrs->controls[i].addr, val); + val &= model->reserved; + val |= op_x86_get_ctrl(model, &counter_config[i]); + wrmsrl(msrs->controls[i].addr, val); } else { reset_value[i] = 0; } @@ -143,14 +125,14 @@ static int ppro_check_ctrs(struct pt_regs * const regs, if (unlikely(!reset_value)) goto out; - for (i = 0 ; i < num_counters; ++i) { + for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; rdmsrl(msrs->counters[i].addr, val); - if (CTR_OVERFLOWED(val)) { - oprofile_add_sample(regs, i); - wrmsrl(msrs->counters[i].addr, -reset_value[i]); - } + if (val & (1ULL << (counter_width - 1))) + continue; + oprofile_add_sample(regs, i); + wrmsrl(msrs->counters[i].addr, -reset_value[i]); } out: @@ -171,16 +153,16 @@ out: static void ppro_start(struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; if (!reset_value) return; for (i = 0; i < num_counters; ++i) { if (reset_value[i]) { - CTRL_READ(low, high, msrs, i); - CTRL_SET_ACTIVE(low); - CTRL_WRITE(low, high, msrs, i); + rdmsrl(msrs->controls[i].addr, val); + val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(msrs->controls[i].addr, val); } } } @@ -188,7 +170,7 @@ static void ppro_start(struct op_msrs const * const msrs) static void ppro_stop(struct op_msrs const * const msrs) { - unsigned int low, high; + u64 val; int i; if (!reset_value) @@ -196,9 +178,9 @@ static void ppro_stop(struct op_msrs const * const msrs) for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; - CTRL_READ(low, high, msrs, i); - CTRL_SET_INACTIVE(low); - CTRL_WRITE(low, high, msrs, i); + rdmsrl(msrs->controls[i].addr, val); + val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(msrs->controls[i].addr, val); } } @@ -206,12 +188,12 @@ static void ppro_shutdown(struct op_msrs const * const msrs) { int i; - for (i = 0 ; i < num_counters ; ++i) { - if (CTR_IS_RESERVED(msrs, i)) + for (i = 0; i < num_counters; ++i) { + if (msrs->counters[i].addr) release_perfctr_nmi(MSR_P6_PERFCTR0 + i); } - for (i = 0 ; i < num_counters ; ++i) { - if (CTRL_IS_RESERVED(msrs, i)) + for (i = 0; i < num_counters; ++i) { + if (msrs->controls[i].addr) release_evntsel_nmi(MSR_P6_EVNTSEL0 + i); } if (reset_value) { @@ -222,8 +204,9 @@ static void ppro_shutdown(struct op_msrs const * const msrs) struct op_x86_model_spec op_ppro_spec = { - .num_counters = 2, /* can be overriden */ - .num_controls = 2, /* dito */ + .num_counters = 2, + .num_controls = 2, + .reserved = MSR_PPRO_EVENTSEL_RESERVED, .fill_in_addresses = &ppro_fill_in_addresses, .setup_ctrs = &ppro_setup_ctrs, .check_ctrs = &ppro_check_ctrs, @@ -241,7 +224,7 @@ struct op_x86_model_spec op_ppro_spec = { * the specific CPU. */ -void arch_perfmon_setup_counters(void) +static void arch_perfmon_setup_counters(void) { union cpuid10_eax eax; @@ -259,11 +242,17 @@ void arch_perfmon_setup_counters(void) op_arch_perfmon_spec.num_counters = num_counters; op_arch_perfmon_spec.num_controls = num_counters; - op_ppro_spec.num_counters = num_counters; - op_ppro_spec.num_controls = num_counters; +} + +static int arch_perfmon_init(struct oprofile_operations *ignore) +{ + arch_perfmon_setup_counters(); + return 0; } struct op_x86_model_spec op_arch_perfmon_spec = { + .reserved = MSR_PPRO_EVENTSEL_RESERVED, + .init = &arch_perfmon_init, /* num_counters/num_controls filled in at runtime */ .fill_in_addresses = &ppro_fill_in_addresses, /* user space does the cpuid check for available events */ diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h index 825e79064d64..b83776180c7f 100644 --- a/arch/x86/oprofile/op_x86_model.h +++ b/arch/x86/oprofile/op_x86_model.h @@ -6,51 +6,66 @@ * @remark Read the file COPYING * * @author Graydon Hoare + * @author Robert Richter <robert.richter@amd.com> */ #ifndef OP_X86_MODEL_H #define OP_X86_MODEL_H -struct op_saved_msr { - unsigned int high; - unsigned int low; -}; +#include <asm/types.h> +#include <asm/perf_counter.h> struct op_msr { - unsigned long addr; - struct op_saved_msr saved; + unsigned long addr; + u64 saved; }; struct op_msrs { struct op_msr *counters; struct op_msr *controls; + struct op_msr *multiplex; }; struct pt_regs; +struct oprofile_operations; + /* The model vtable abstracts the differences between * various x86 CPU models' perfctr support. */ struct op_x86_model_spec { - int (*init)(struct oprofile_operations *ops); - void (*exit)(void); - unsigned int num_counters; - unsigned int num_controls; - void (*fill_in_addresses)(struct op_msrs * const msrs); - void (*setup_ctrs)(struct op_msrs const * const msrs); - int (*check_ctrs)(struct pt_regs * const regs, - struct op_msrs const * const msrs); - void (*start)(struct op_msrs const * const msrs); - void (*stop)(struct op_msrs const * const msrs); - void (*shutdown)(struct op_msrs const * const msrs); + unsigned int num_counters; + unsigned int num_controls; + unsigned int num_virt_counters; + u64 reserved; + u16 event_mask; + int (*init)(struct oprofile_operations *ops); + void (*exit)(void); + void (*fill_in_addresses)(struct op_msrs * const msrs); + void (*setup_ctrs)(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs); + int (*check_ctrs)(struct pt_regs * const regs, + struct op_msrs const * const msrs); + void (*start)(struct op_msrs const * const msrs); + void (*stop)(struct op_msrs const * const msrs); + void (*shutdown)(struct op_msrs const * const msrs); +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + void (*switch_ctrl)(struct op_x86_model_spec const *model, + struct op_msrs const * const msrs); +#endif }; +struct op_counter_config; + +extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model, + struct op_counter_config *counter_config); +extern int op_x86_phys_to_virt(int phys); +extern int op_x86_virt_to_phys(int virt); + extern struct op_x86_model_spec op_ppro_spec; -extern struct op_x86_model_spec const op_p4_spec; -extern struct op_x86_model_spec const op_p4_ht2_spec; -extern struct op_x86_model_spec const op_amd_spec; +extern struct op_x86_model_spec op_p4_spec; +extern struct op_x86_model_spec op_p4_ht2_spec; +extern struct op_x86_model_spec op_amd_spec; extern struct op_x86_model_spec op_arch_perfmon_spec; -extern void arch_perfmon_setup_counters(void); - #endif /* OP_X86_MODEL_H */ diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index 5d7a02f63e1c..50eecfe1d724 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -24,6 +24,7 @@ #include <linux/sysrq.h> #include <linux/kbd_kern.h> #include <linux/proc_fs.h> +#include <linux/nmi.h> #include <linux/quotaops.h> #include <linux/perf_counter.h> #include <linux/kernel.h> @@ -222,12 +223,20 @@ static DECLARE_WORK(sysrq_showallcpus, sysrq_showregs_othercpus); static void sysrq_handle_showallcpus(int key, struct tty_struct *tty) { - struct pt_regs *regs = get_irq_regs(); - if (regs) { - printk(KERN_INFO "CPU%d:\n", smp_processor_id()); - show_regs(regs); + /* + * Fall back to the workqueue based printing if the + * backtrace printing did not succeed or the + * architecture has no support for it: + */ + if (!trigger_all_cpu_backtrace()) { + struct pt_regs *regs = get_irq_regs(); + + if (regs) { + printk(KERN_INFO "CPU%d:\n", smp_processor_id()); + show_regs(regs); + } + schedule_work(&sysrq_showallcpus); } - schedule_work(&sysrq_showallcpus); } static struct sysrq_key_op sysrq_showallcpus_op = { diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index 242257b19441..a7aae24f2889 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -21,7 +21,6 @@ #include <linux/sched.h> #include <linux/oprofile.h> -#include <linux/vmalloc.h> #include <linux/errno.h> #include "event_buffer.h" @@ -407,6 +406,21 @@ int oprofile_add_data(struct op_entry *entry, unsigned long val) return op_cpu_buffer_add_data(entry, val); } +int oprofile_add_data64(struct op_entry *entry, u64 val) +{ + if (!entry->event) + return 0; + if (op_cpu_buffer_get_size(entry) < 2) + /* + * the function returns 0 to indicate a too small + * buffer, even if there is some space left + */ + return 0; + if (!op_cpu_buffer_add_data(entry, (u32)val)) + return 0; + return op_cpu_buffer_add_data(entry, (u32)(val >> 32)); +} + int oprofile_write_commit(struct op_entry *entry) { if (!entry->event) diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index 3cffce90f82a..dc8a0428260d 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -12,6 +12,8 @@ #include <linux/init.h> #include <linux/oprofile.h> #include <linux/moduleparam.h> +#include <linux/workqueue.h> +#include <linux/time.h> #include <asm/mutex.h> #include "oprof.h" @@ -87,6 +89,69 @@ out: return err; } +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + +static void switch_worker(struct work_struct *work); +static DECLARE_DELAYED_WORK(switch_work, switch_worker); + +static void start_switch_worker(void) +{ + if (oprofile_ops.switch_events) + schedule_delayed_work(&switch_work, oprofile_time_slice); +} + +static void stop_switch_worker(void) +{ + cancel_delayed_work_sync(&switch_work); +} + +static void switch_worker(struct work_struct *work) +{ + if (oprofile_ops.switch_events()) + return; + + atomic_inc(&oprofile_stats.multiplex_counter); + start_switch_worker(); +} + +/* User inputs in ms, converts to jiffies */ +int oprofile_set_timeout(unsigned long val_msec) +{ + int err = 0; + unsigned long time_slice; + + mutex_lock(&start_mutex); + + if (oprofile_started) { + err = -EBUSY; + goto out; + } + + if (!oprofile_ops.switch_events) { + err = -EINVAL; + goto out; + } + + time_slice = msecs_to_jiffies(val_msec); + if (time_slice == MAX_JIFFY_OFFSET) { + err = -EINVAL; + goto out; + } + + oprofile_time_slice = time_slice; + +out: + mutex_unlock(&start_mutex); + return err; + +} + +#else + +static inline void start_switch_worker(void) { } +static inline void stop_switch_worker(void) { } + +#endif /* Actually start profiling (echo 1>/dev/oprofile/enable) */ int oprofile_start(void) @@ -108,6 +173,8 @@ int oprofile_start(void) if ((err = oprofile_ops.start())) goto out; + start_switch_worker(); + oprofile_started = 1; out: mutex_unlock(&start_mutex); @@ -123,6 +190,9 @@ void oprofile_stop(void) goto out; oprofile_ops.stop(); oprofile_started = 0; + + stop_switch_worker(); + /* wake up the daemon to read what remains */ wake_up_buffer_waiter(); out: @@ -155,7 +225,6 @@ post_sync: mutex_unlock(&start_mutex); } - int oprofile_set_backtrace(unsigned long val) { int err = 0; diff --git a/drivers/oprofile/oprof.h b/drivers/oprofile/oprof.h index c288d3c24b50..cb92f5c98c1a 100644 --- a/drivers/oprofile/oprof.h +++ b/drivers/oprofile/oprof.h @@ -24,6 +24,8 @@ struct oprofile_operations; extern unsigned long oprofile_buffer_size; extern unsigned long oprofile_cpu_buffer_size; extern unsigned long oprofile_buffer_watershed; +extern unsigned long oprofile_time_slice; + extern struct oprofile_operations oprofile_ops; extern unsigned long oprofile_started; extern unsigned long oprofile_backtrace_depth; @@ -35,5 +37,6 @@ void oprofile_create_files(struct super_block *sb, struct dentry *root); void oprofile_timer_init(struct oprofile_operations *ops); int oprofile_set_backtrace(unsigned long depth); +int oprofile_set_timeout(unsigned long time); #endif /* OPROF_H */ diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c index 5d36ffc30dd5..bbd7516e0869 100644 --- a/drivers/oprofile/oprofile_files.c +++ b/drivers/oprofile/oprofile_files.c @@ -9,6 +9,7 @@ #include <linux/fs.h> #include <linux/oprofile.h> +#include <linux/jiffies.h> #include "event_buffer.h" #include "oprofile_stats.h" @@ -17,10 +18,51 @@ #define BUFFER_SIZE_DEFAULT 131072 #define CPU_BUFFER_SIZE_DEFAULT 8192 #define BUFFER_WATERSHED_DEFAULT 32768 /* FIXME: tune */ +#define TIME_SLICE_DEFAULT 1 unsigned long oprofile_buffer_size; unsigned long oprofile_cpu_buffer_size; unsigned long oprofile_buffer_watershed; +unsigned long oprofile_time_slice; + +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + +static ssize_t timeout_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(jiffies_to_msecs(oprofile_time_slice), + buf, count, offset); +} + + +static ssize_t timeout_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval) + return retval; + + retval = oprofile_set_timeout(val); + + if (retval) + return retval; + return count; +} + + +static const struct file_operations timeout_fops = { + .read = timeout_read, + .write = timeout_write, +}; + +#endif + static ssize_t depth_read(struct file *file, char __user *buf, size_t count, loff_t *offset) { @@ -129,6 +171,7 @@ void oprofile_create_files(struct super_block *sb, struct dentry *root) oprofile_buffer_size = BUFFER_SIZE_DEFAULT; oprofile_cpu_buffer_size = CPU_BUFFER_SIZE_DEFAULT; oprofile_buffer_watershed = BUFFER_WATERSHED_DEFAULT; + oprofile_time_slice = msecs_to_jiffies(TIME_SLICE_DEFAULT); oprofilefs_create_file(sb, root, "enable", &enable_fops); oprofilefs_create_file_perm(sb, root, "dump", &dump_fops, 0666); @@ -139,6 +182,9 @@ void oprofile_create_files(struct super_block *sb, struct dentry *root) oprofilefs_create_file(sb, root, "cpu_type", &cpu_type_fops); oprofilefs_create_file(sb, root, "backtrace_depth", &depth_fops); oprofilefs_create_file(sb, root, "pointer_size", &pointer_size_fops); +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + oprofilefs_create_file(sb, root, "time_slice", &timeout_fops); +#endif oprofile_create_stats_files(sb, root); if (oprofile_ops.create_files) oprofile_ops.create_files(sb, root); diff --git a/drivers/oprofile/oprofile_stats.c b/drivers/oprofile/oprofile_stats.c index 3c2270a8300c..61689e814d46 100644 --- a/drivers/oprofile/oprofile_stats.c +++ b/drivers/oprofile/oprofile_stats.c @@ -34,6 +34,7 @@ void oprofile_reset_stats(void) atomic_set(&oprofile_stats.sample_lost_no_mapping, 0); atomic_set(&oprofile_stats.event_lost_overflow, 0); atomic_set(&oprofile_stats.bt_lost_no_mapping, 0); + atomic_set(&oprofile_stats.multiplex_counter, 0); } @@ -76,4 +77,8 @@ void oprofile_create_stats_files(struct super_block *sb, struct dentry *root) &oprofile_stats.event_lost_overflow); oprofilefs_create_ro_atomic(sb, dir, "bt_lost_no_mapping", &oprofile_stats.bt_lost_no_mapping); +#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX + oprofilefs_create_ro_atomic(sb, dir, "multiplex_counter", + &oprofile_stats.multiplex_counter); +#endif } diff --git a/drivers/oprofile/oprofile_stats.h b/drivers/oprofile/oprofile_stats.h index 3da0d08dc1f9..0b54e46c3c14 100644 --- a/drivers/oprofile/oprofile_stats.h +++ b/drivers/oprofile/oprofile_stats.h @@ -17,6 +17,7 @@ struct oprofile_stat_struct { atomic_t sample_lost_no_mapping; atomic_t bt_lost_no_mapping; atomic_t event_lost_overflow; + atomic_t multiplex_counter; }; extern struct oprofile_stat_struct oprofile_stats; diff --git a/drivers/pnp/pnpbios/bioscalls.c b/drivers/pnp/pnpbios/bioscalls.c index 7e6b5a3b3281..bd035e3d3550 100644 --- a/drivers/pnp/pnpbios/bioscalls.c +++ b/drivers/pnp/pnpbios/bioscalls.c @@ -55,12 +55,12 @@ __asm__(".text \n" #define Q2_SET_SEL(cpu, selname, address, size) \ do { \ -struct desc_struct *gdt = get_cpu_gdt_table((cpu)); \ -set_base(gdt[(selname) >> 3], (u32)(address)); \ -set_limit(gdt[(selname) >> 3], size); \ + struct desc_struct *gdt = get_cpu_gdt_table((cpu)); \ + set_desc_base(&gdt[(selname) >> 3], (u32)(address)); \ + set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \ } while(0) -static struct desc_struct bad_bios_desc; +static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0); /* * At some point we want to use this stack frame pointer to unwind @@ -476,19 +476,17 @@ void pnpbios_calls_init(union pnp_bios_install_struct *header) pnp_bios_callpoint.offset = header->fields.pm16offset; pnp_bios_callpoint.segment = PNP_CS16; - bad_bios_desc.a = 0; - bad_bios_desc.b = 0x00409200; - - set_base(bad_bios_desc, __va((unsigned long)0x40 << 4)); - _set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4)); + set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4)); + set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4)); for_each_possible_cpu(i) { struct desc_struct *gdt = get_cpu_gdt_table(i); if (!gdt) continue; - set_base(gdt[GDT_ENTRY_PNPBIOS_CS32], &pnp_bios_callfunc); - set_base(gdt[GDT_ENTRY_PNPBIOS_CS16], - __va(header->fields.pm16cseg)); - set_base(gdt[GDT_ENTRY_PNPBIOS_DS], - __va(header->fields.pm16dseg)); + set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_CS32], + (unsigned long)&pnp_bios_callfunc); + set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_CS16], + (unsigned long)__va(header->fields.pm16cseg)); + set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_DS], + (unsigned long)__va(header->fields.pm16dseg)); } } diff --git a/fs/dcache.c b/fs/dcache.c index 9e5cd3c3a6ba..a100fa35a48f 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -32,6 +32,7 @@ #include <linux/swap.h> #include <linux/bootmem.h> #include <linux/fs_struct.h> +#include <linux/hardirq.h> #include "internal.h" int sysctl_vfs_cache_pressure __read_mostly = 100; diff --git a/fs/locks.c b/fs/locks.c index 52366e877d76..19ee18a6829b 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -768,7 +768,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) * give it the opportunity to lock the file. */ if (found) - cond_resched_bkl(); + cond_resched(); find_conflict: for_each_lock(inode, before) { diff --git a/include/asm-generic/cputime.h b/include/asm-generic/cputime.h index 1c1fa422d18a..ca0f239f0e13 100644 --- a/include/asm-generic/cputime.h +++ b/include/asm-generic/cputime.h @@ -7,6 +7,7 @@ typedef unsigned long cputime_t; #define cputime_zero (0UL) +#define cputime_one_jiffy jiffies_to_cputime(1) #define cputime_max ((~0UL >> 1) - 1) #define cputime_add(__a, __b) ((__a) + (__b)) #define cputime_sub(__a, __b) ((__a) - (__b)) diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 07dfd460d286..c0f6c3cd788c 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -98,11 +98,6 @@ static inline int is_device_dma_capable(struct device *dev) return dev->dma_mask != NULL && *dev->dma_mask != DMA_MASK_NONE; } -static inline int is_buffer_dma_capable(u64 mask, dma_addr_t addr, size_t size) -{ - return addr + size <= mask; -} - #ifdef CONFIG_HAS_DMA #include <asm/dma-mapping.h> #else diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index a81170de7f6b..ac8c6f8cf242 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -103,6 +103,8 @@ void trace_current_buffer_discard_commit(struct ring_buffer_event *event); void tracing_record_cmdline(struct task_struct *tsk); +struct event_filter; + struct ftrace_event_call { struct list_head list; char *name; @@ -118,7 +120,7 @@ struct ftrace_event_call { int (*define_fields)(void); struct list_head fields; int filter_active; - void *filter; + struct event_filter *filter; void *mod; atomic_t profile_count; diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 8246c697863d..0d885fd75111 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -64,6 +64,12 @@ #define HARDIRQ_OFFSET (1UL << HARDIRQ_SHIFT) #define NMI_OFFSET (1UL << NMI_SHIFT) +#ifndef PREEMPT_ACTIVE +#define PREEMPT_ACTIVE_BITS 1 +#define PREEMPT_ACTIVE_SHIFT (NMI_SHIFT + NMI_BITS) +#define PREEMPT_ACTIVE (__IRQ_MASK(PREEMPT_ACTIVE_BITS) << PREEMPT_ACTIVE_SHIFT) +#endif + #if PREEMPT_ACTIVE < (1 << (NMI_SHIFT + NMI_BITS)) #error PREEMPT_ACTIVE is too low! #endif diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 4759917adc71..ff037f0b1b4e 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -91,7 +91,6 @@ enum hrtimer_restart { * @function: timer expiry callback function * @base: pointer to the timer base (per cpu and per clock) * @state: state information (See bit values above) - * @cb_entry: list head to enqueue an expired timer into the callback list * @start_site: timer statistics field to store the site where the timer * was started * @start_comm: timer statistics field to store the name of the process which @@ -108,7 +107,6 @@ struct hrtimer { enum hrtimer_restart (*function)(struct hrtimer *); struct hrtimer_clock_base *base; unsigned long state; - struct list_head cb_entry; #ifdef CONFIG_TIMER_STATS int start_pid; void *start_site; diff --git a/include/linux/irq.h b/include/linux/irq.h index cb2e77a3f7f7..6956df9961ab 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -220,13 +220,6 @@ static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node); /* - * Migration helpers for obsolete names, they will go away: - */ -#define hw_interrupt_type irq_chip -#define no_irq_type no_irq_chip -typedef struct irq_desc irq_desc_t; - -/* * Pick up the arch-dependent methods: */ #include <asm/hw_irq.h> diff --git a/include/linux/kernel.h b/include/linux/kernel.h index d6320a3e8def..2b5b1e0899a8 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -125,7 +125,7 @@ extern int _cond_resched(void); #endif #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP - void __might_sleep(char *file, int line); + void __might_sleep(char *file, int line, int preempt_offset); /** * might_sleep - annotation for functions that can sleep * @@ -137,8 +137,9 @@ extern int _cond_resched(void); * supposed to. */ # define might_sleep() \ - do { __might_sleep(__FILE__, __LINE__); might_resched(); } while (0) + do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0) #else + static inline void __might_sleep(char *file, int line, int preempt_offset) { } # define might_sleep() do { might_resched(); } while (0) #endif diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index b25d1b53df0d..9ccf0e286b2a 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -149,6 +149,12 @@ struct lock_list { struct lock_class *class; struct stack_trace trace; int distance; + + /* + * The parent field is used to implement breadth-first search, and the + * bit 0 is reused to indicate if the lock has been accessed in BFS. + */ + struct lock_list *parent; }; /* @@ -208,10 +214,12 @@ struct held_lock { * interrupt context: */ unsigned int irq_context:2; /* bit 0 - soft, bit 1 - hard */ - unsigned int trylock:1; + unsigned int trylock:1; /* 16 bits */ + unsigned int read:2; /* see lock_acquire() comment */ unsigned int check:2; /* see lock_acquire() comment */ unsigned int hardirqs_off:1; + unsigned int references:11; /* 32 bits */ }; /* @@ -291,6 +299,10 @@ extern void lock_acquire(struct lockdep_map *lock, unsigned int subclass, extern void lock_release(struct lockdep_map *lock, int nested, unsigned long ip); +#define lockdep_is_held(lock) lock_is_held(&(lock)->dep_map) + +extern int lock_is_held(struct lockdep_map *lock); + extern void lock_set_class(struct lockdep_map *lock, const char *name, struct lock_class_key *key, unsigned int subclass, unsigned long ip); @@ -309,6 +321,8 @@ extern void lockdep_trace_alloc(gfp_t mask); #define lockdep_depth(tsk) (debug_locks ? (tsk)->lockdep_depth : 0) +#define lockdep_assert_held(l) WARN_ON(debug_locks && !lockdep_is_held(l)) + #else /* !LOCKDEP */ static inline void lockdep_off(void) @@ -353,6 +367,8 @@ struct lock_class_key { }; #define lockdep_depth(tsk) (0) +#define lockdep_assert_held(l) do { } while (0) + #endif /* !LOCKDEP */ #ifdef CONFIG_LOCK_STAT diff --git a/include/linux/nmi.h b/include/linux/nmi.h index 29af2d5df097..b752e807adde 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -28,8 +28,23 @@ static inline void acpi_nmi_disable(void) { } static inline void acpi_nmi_enable(void) { } #endif -#ifndef trigger_all_cpu_backtrace -#define trigger_all_cpu_backtrace() do { } while (0) +/* + * Create trigger_all_cpu_backtrace() out of the arch-provided + * base function. Return whether such support was available, + * to allow calling code to fall back to some other mechanism: + */ +#ifdef arch_trigger_all_cpu_backtrace +static inline bool trigger_all_cpu_backtrace(void) +{ + arch_trigger_all_cpu_backtrace(); + + return true; +} +#else +static inline bool trigger_all_cpu_backtrace(void) +{ + return false; +} #endif #endif diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h index 1d9518bc4c58..5171639ecf0f 100644 --- a/include/linux/oprofile.h +++ b/include/linux/oprofile.h @@ -67,6 +67,9 @@ struct oprofile_operations { /* Initiate a stack backtrace. Optional. */ void (*backtrace)(struct pt_regs * const regs, unsigned int depth); + + /* Multiplex between different events. Optional. */ + int (*switch_events)(void); /* CPU identification string. */ char * cpu_type; }; @@ -171,7 +174,6 @@ struct op_sample; struct op_entry { struct ring_buffer_event *event; struct op_sample *sample; - unsigned long irq_flags; unsigned long size; unsigned long *data; }; @@ -180,6 +182,7 @@ void oprofile_write_reserve(struct op_entry *entry, struct pt_regs * const regs, unsigned long pc, int code, int size); int oprofile_add_data(struct op_entry *entry, unsigned long val); +int oprofile_add_data64(struct op_entry *entry, u64 val); int oprofile_write_commit(struct op_entry *entry); #endif /* OPROFILE_H */ diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h index 68438e18fff4..afd5f8b7061f 100644 --- a/include/linux/percpu-defs.h +++ b/include/linux/percpu-defs.h @@ -69,11 +69,13 @@ /* * Declaration/definition used for per-CPU variables that must be page aligned. */ -#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name) \ - DECLARE_PER_CPU_SECTION(type, name, ".page_aligned") +#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, ".page_aligned") \ + __aligned(PAGE_SIZE) #define DEFINE_PER_CPU_PAGE_ALIGNED(type, name) \ - DEFINE_PER_CPU_SECTION(type, name, ".page_aligned") + DEFINE_PER_CPU_SECTION(type, name, ".page_aligned") \ + __aligned(PAGE_SIZE) /* * Intermodule exports for per-CPU variables. diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h index a67dd5c5b6d3..0a6f3209c9de 100644 --- a/include/linux/perf_counter.h +++ b/include/linux/perf_counter.h @@ -121,7 +121,7 @@ enum perf_counter_sample_format { PERF_SAMPLE_CPU = 1U << 7, PERF_SAMPLE_PERIOD = 1U << 8, PERF_SAMPLE_STREAM_ID = 1U << 9, - PERF_SAMPLE_TP_RECORD = 1U << 10, + PERF_SAMPLE_RAW = 1U << 10, PERF_SAMPLE_MAX = 1U << 11, /* non-ABI */ }; @@ -414,9 +414,9 @@ struct perf_callchain_entry { __u64 ip[PERF_MAX_STACK_DEPTH]; }; -struct perf_tracepoint_record { - int size; - char *record; +struct perf_raw_record { + u32 size; + void *data; }; struct task_struct; @@ -687,11 +687,13 @@ struct perf_sample_data { struct pt_regs *regs; u64 addr; u64 period; - void *private; + struct perf_raw_record *raw; }; extern int perf_counter_overflow(struct perf_counter *counter, int nmi, struct perf_sample_data *data); +extern void perf_counter_output(struct perf_counter *counter, int nmi, + struct perf_sample_data *data); /* * Return 1 for a software counter, 0 for a hardware counter diff --git a/include/linux/poison.h b/include/linux/poison.h index 6729f7dcd60e..229322707f5f 100644 --- a/include/linux/poison.h +++ b/include/linux/poison.h @@ -2,13 +2,25 @@ #define _LINUX_POISON_H /********** include/linux/list.h **********/ + +/* + * Architectures might want to move the poison pointer offset + * into some well-recognized area such as 0xdead000000000000, + * that is also not mappable by user-space exploits: + */ +#ifdef CONFIG_ILLEGAL_POINTER_VALUE +# define POISON_POINTER_DELTA _AC(CONFIG_ILLEGAL_POINTER_VALUE, UL) +#else +# define POISON_POINTER_DELTA 0 +#endif + /* * These are non-NULL pointers that will result in page faults * under normal circumstances, used to verify that nobody uses * non-initialized list entries. */ -#define LIST_POISON1 ((void *) 0x00100100) -#define LIST_POISON2 ((void *) 0x00200200) +#define LIST_POISON1 ((void *) 0x00100100 + POISON_POINTER_DELTA) +#define LIST_POISON2 ((void *) 0x00200200 + POISON_POINTER_DELTA) /********** include/linux/timer.h **********/ /* diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h deleted file mode 100644 index bfd92e1e5d2c..000000000000 --- a/include/linux/rcuclassic.h +++ /dev/null @@ -1,178 +0,0 @@ -/* - * Read-Copy Update mechanism for mutual exclusion (classic version) - * - * 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 IBM Corporation, 2001 - * - * Author: Dipankar Sarma <dipankar@in.ibm.com> - * - * Based on the original work by Paul McKenney <paulmck@us.ibm.com> - * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. - * Papers: - * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf - * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001) - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU - * - */ - -#ifndef __LINUX_RCUCLASSIC_H -#define __LINUX_RCUCLASSIC_H - -#include <linux/cache.h> -#include <linux/spinlock.h> -#include <linux/threads.h> -#include <linux/cpumask.h> -#include <linux/seqlock.h> - -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR -#define RCU_SECONDS_TILL_STALL_CHECK (10 * HZ) /* for rcp->jiffies_stall */ -#define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ) /* for rcp->jiffies_stall */ -#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - -/* Global control variables for rcupdate callback mechanism. */ -struct rcu_ctrlblk { - long cur; /* Current batch number. */ - long completed; /* Number of the last completed batch */ - long pending; /* Number of the last pending batch */ -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR - unsigned long gp_start; /* Time at which GP started in jiffies. */ - unsigned long jiffies_stall; - /* Time at which to check for CPU stalls. */ -#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - - int signaled; - - spinlock_t lock ____cacheline_internodealigned_in_smp; - DECLARE_BITMAP(cpumask, NR_CPUS); /* CPUs that need to switch for */ - /* current batch to proceed. */ -} ____cacheline_internodealigned_in_smp; - -/* Is batch a before batch b ? */ -static inline int rcu_batch_before(long a, long b) -{ - return (a - b) < 0; -} - -/* Is batch a after batch b ? */ -static inline int rcu_batch_after(long a, long b) -{ - return (a - b) > 0; -} - -/* Per-CPU data for Read-Copy UPdate. */ -struct rcu_data { - /* 1) quiescent state handling : */ - long quiescbatch; /* Batch # for grace period */ - int passed_quiesc; /* User-mode/idle loop etc. */ - int qs_pending; /* core waits for quiesc state */ - - /* 2) batch handling */ - /* - * if nxtlist is not NULL, then: - * batch: - * The batch # for the last entry of nxtlist - * [*nxttail[1], NULL = *nxttail[2]): - * Entries that batch # <= batch - * [*nxttail[0], *nxttail[1]): - * Entries that batch # <= batch - 1 - * [nxtlist, *nxttail[0]): - * Entries that batch # <= batch - 2 - * The grace period for these entries has completed, and - * the other grace-period-completed entries may be moved - * here temporarily in rcu_process_callbacks(). - */ - long batch; - struct rcu_head *nxtlist; - struct rcu_head **nxttail[3]; - long qlen; /* # of queued callbacks */ - struct rcu_head *donelist; - struct rcu_head **donetail; - long blimit; /* Upper limit on a processed batch */ - int cpu; - struct rcu_head barrier; -}; - -/* - * Increment the quiescent state counter. - * The counter is a bit degenerated: We do not need to know - * how many quiescent states passed, just if there was at least - * one since the start of the grace period. Thus just a flag. - */ -extern void rcu_qsctr_inc(int cpu); -extern void rcu_bh_qsctr_inc(int cpu); - -extern int rcu_pending(int cpu); -extern int rcu_needs_cpu(int cpu); - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -extern struct lockdep_map rcu_lock_map; -# define rcu_read_acquire() \ - lock_acquire(&rcu_lock_map, 0, 0, 2, 1, NULL, _THIS_IP_) -# define rcu_read_release() lock_release(&rcu_lock_map, 1, _THIS_IP_) -#else -# define rcu_read_acquire() do { } while (0) -# define rcu_read_release() do { } while (0) -#endif - -#define __rcu_read_lock() \ - do { \ - preempt_disable(); \ - __acquire(RCU); \ - rcu_read_acquire(); \ - } while (0) -#define __rcu_read_unlock() \ - do { \ - rcu_read_release(); \ - __release(RCU); \ - preempt_enable(); \ - } while (0) -#define __rcu_read_lock_bh() \ - do { \ - local_bh_disable(); \ - __acquire(RCU_BH); \ - rcu_read_acquire(); \ - } while (0) -#define __rcu_read_unlock_bh() \ - do { \ - rcu_read_release(); \ - __release(RCU_BH); \ - local_bh_enable(); \ - } while (0) - -#define __synchronize_sched() synchronize_rcu() - -#define call_rcu_sched(head, func) call_rcu(head, func) - -extern void __rcu_init(void); -#define rcu_init_sched() do { } while (0) -extern void rcu_check_callbacks(int cpu, int user); -extern void rcu_restart_cpu(int cpu); - -extern long rcu_batches_completed(void); -extern long rcu_batches_completed_bh(void); - -#define rcu_enter_nohz() do { } while (0) -#define rcu_exit_nohz() do { } while (0) - -/* A context switch is a grace period for rcuclassic. */ -static inline int rcu_blocking_is_gp(void) -{ - return num_online_cpus() == 1; -} - -#endif /* __LINUX_RCUCLASSIC_H */ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 19b82b96affa..f8cad2b20778 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -42,12 +42,22 @@ #include <linux/lockdep.h> #include <linux/completion.h> -/* Internal to kernel, but needed by rcupreempt.h. */ +/* Exported common interfaces */ +extern void synchronize_rcu(void); +extern void synchronize_rcu_bh(void); +extern void rcu_barrier(void); +extern void rcu_barrier_bh(void); +extern void rcu_barrier_sched(void); +extern void synchronize_sched_expedited(void); +extern int sched_expedited_torture_stats(char *page); + +/* Internal to kernel */ +extern void rcu_init(void); +extern void rcu_scheduler_starting(void); +extern int rcu_needs_cpu(int cpu); extern int rcu_scheduler_active; -#if defined(CONFIG_CLASSIC_RCU) -#include <linux/rcuclassic.h> -#elif defined(CONFIG_TREE_RCU) +#if defined(CONFIG_TREE_RCU) #include <linux/rcutree.h> #elif defined(CONFIG_PREEMPT_RCU) #include <linux/rcupreempt.h> @@ -250,15 +260,4 @@ extern void call_rcu(struct rcu_head *head, extern void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *head)); -/* Exported common interfaces */ -extern void synchronize_rcu(void); -extern void rcu_barrier(void); -extern void rcu_barrier_bh(void); -extern void rcu_barrier_sched(void); - -/* Internal to kernel */ -extern void rcu_init(void); -extern void rcu_scheduler_starting(void); -extern int rcu_needs_cpu(int cpu); - #endif /* __LINUX_RCUPDATE_H */ diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h index fce522782ffa..f164ac9b7807 100644 --- a/include/linux/rcupreempt.h +++ b/include/linux/rcupreempt.h @@ -74,6 +74,16 @@ extern int rcu_needs_cpu(int cpu); extern void __synchronize_sched(void); +static inline void synchronize_rcu_expedited(void) +{ + synchronize_rcu(); /* Placeholder for new rcupreempt implementation. */ +} + +static inline void synchronize_rcu_bh_expedited(void) +{ + synchronize_rcu_bh(); /* Placeholder for new rcupreempt impl. */ +} + extern void __rcu_init(void); extern void rcu_init_sched(void); extern void rcu_check_callbacks(int cpu, int user); diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index 5a5153806c42..d4dfd2489633 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -286,8 +286,14 @@ static inline void __rcu_read_unlock_bh(void) #define call_rcu_sched(head, func) call_rcu(head, func) -static inline void rcu_init_sched(void) +static inline void synchronize_rcu_expedited(void) +{ + synchronize_sched_expedited(); +} + +static inline void synchronize_rcu_bh_expedited(void) { + synchronize_sched_expedited(); } extern void __rcu_init(void); @@ -297,6 +303,10 @@ extern void rcu_restart_cpu(int cpu); extern long rcu_batches_completed(void); extern long rcu_batches_completed_bh(void); +static inline void rcu_init_sched(void) +{ +} + #ifdef CONFIG_NO_HZ void rcu_enter_nohz(void); void rcu_exit_nohz(void); diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 29f8599e6bea..7fca71693ae7 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -170,7 +170,6 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer); unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu); unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu); unsigned long ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_nmi_dropped_cpu(struct ring_buffer *buffer, int cpu); u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu); void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, diff --git a/include/linux/sched.h b/include/linux/sched.h index 5c733563cf95..2bc71800558e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -38,6 +38,8 @@ #define SCHED_BATCH 3 /* SCHED_ISO: reserved but not implemented yet */ #define SCHED_IDLE 5 +/* Can be ORed in to make sure the process is reverted back to SCHED_NORMAL on fork */ +#define SCHED_RESET_ON_FORK 0x40000000 #ifdef __KERNEL__ @@ -474,6 +476,13 @@ struct pacct_struct { unsigned long ac_minflt, ac_majflt; }; +struct cpu_itimer { + cputime_t expires; + cputime_t incr; + u32 error; + u32 incr_error; +}; + /** * struct task_cputime - collected CPU time counts * @utime: time spent in user mode, in &cputime_t units @@ -568,9 +577,12 @@ struct signal_struct { struct pid *leader_pid; ktime_t it_real_incr; - /* ITIMER_PROF and ITIMER_VIRTUAL timers for the process */ - cputime_t it_prof_expires, it_virt_expires; - cputime_t it_prof_incr, it_virt_incr; + /* + * ITIMER_PROF and ITIMER_VIRTUAL timers for the process, we use + * CPUCLOCK_PROF and CPUCLOCK_VIRT for indexing array as these + * values are defined to 0 and 1 respectively + */ + struct cpu_itimer it[2]; /* * Thread group totals for process CPU timers. @@ -1049,7 +1061,6 @@ struct sched_class { struct rq *busiest, struct sched_domain *sd, enum cpu_idle_type idle); void (*pre_schedule) (struct rq *this_rq, struct task_struct *task); - int (*needs_post_schedule) (struct rq *this_rq); void (*post_schedule) (struct rq *this_rq); void (*task_wake_up) (struct rq *this_rq, struct task_struct *task); @@ -1233,6 +1244,10 @@ struct task_struct { unsigned did_exec:1; unsigned in_execve:1; /* Tell the LSMs that the process is doing an * execve */ + + /* Revert to default priority/policy when forking */ + unsigned sched_reset_on_fork:1; + pid_t pid; pid_t tgid; @@ -2284,23 +2299,31 @@ static inline int need_resched(void) * cond_resched_softirq() will enable bhs before scheduling. */ extern int _cond_resched(void); -#ifdef CONFIG_PREEMPT_BKL -static inline int cond_resched(void) -{ - return 0; -} + +#define cond_resched() ({ \ + __might_sleep(__FILE__, __LINE__, 0); \ + _cond_resched(); \ +}) + +extern int __cond_resched_lock(spinlock_t *lock); + +#ifdef CONFIG_PREEMPT +#define PREEMPT_LOCK_OFFSET PREEMPT_OFFSET #else -static inline int cond_resched(void) -{ - return _cond_resched(); -} +#define PREEMPT_LOCK_OFFSET 0 #endif -extern int cond_resched_lock(spinlock_t * lock); -extern int cond_resched_softirq(void); -static inline int cond_resched_bkl(void) -{ - return _cond_resched(); -} + +#define cond_resched_lock(lock) ({ \ + __might_sleep(__FILE__, __LINE__, PREEMPT_LOCK_OFFSET); \ + __cond_resched_lock(lock); \ +}) + +extern int __cond_resched_softirq(void); + +#define cond_resched_softirq() ({ \ + __might_sleep(__FILE__, __LINE__, SOFTIRQ_OFFSET); \ + __cond_resched_softirq(); \ +}) /* * Does a critical section need to be broken due to another diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h index cb1a6631b8f4..73b1f1cec423 100644 --- a/include/linux/swiotlb.h +++ b/include/linux/swiotlb.h @@ -14,7 +14,6 @@ struct scatterlist; */ #define IO_TLB_SEGSIZE 128 - /* * log of the size of each IO TLB slab. The number of slabs is command line * controllable. @@ -24,16 +23,6 @@ struct scatterlist; extern void swiotlb_init(void); -extern void *swiotlb_alloc_boot(size_t bytes, unsigned long nslabs); -extern void *swiotlb_alloc(unsigned order, unsigned long nslabs); - -extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, - phys_addr_t address); -extern phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, - dma_addr_t address); - -extern int swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size); - extern void *swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags); diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 7fb16d90e7b1..7b0834c6d3bc 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -25,7 +25,7 @@ #define __array(type, item, len) type item[len]; #undef __dynamic_array -#define __dynamic_array(type, item, len) unsigned short __data_loc_##item; +#define __dynamic_array(type, item, len) u32 __data_loc_##item; #undef __string #define __string(item, src) __dynamic_array(char, item, -1) @@ -51,13 +51,14 @@ * Include the following: * * struct ftrace_data_offsets_<call> { - * int <item1>; - * int <item2>; + * u32 <item1>; + * u32 <item2>; * [...] * }; * - * The __dynamic_array() macro will create each int <item>, this is + * The __dynamic_array() macro will create each u32 <item>, this is * to keep the offset of each array from the beginning of the event. + * The size of an array is also encoded, in the higher 16 bits of <item>. */ #undef __field @@ -67,7 +68,7 @@ #define __array(type, item, len) #undef __dynamic_array -#define __dynamic_array(type, item, len) int item; +#define __dynamic_array(type, item, len) u32 item; #undef __string #define __string(item, src) __dynamic_array(char, item, -1) @@ -120,7 +121,7 @@ #undef __dynamic_array #define __dynamic_array(type, item, len) \ - ret = trace_seq_printf(s, "\tfield:__data_loc " #item ";\t" \ + ret = trace_seq_printf(s, "\tfield:__data_loc " #type "[] " #item ";\t"\ "offset:%u;\tsize:%u;\n", \ (unsigned int)offsetof(typeof(field), \ __data_loc_##item), \ @@ -210,7 +211,7 @@ ftrace_format_##call(struct trace_seq *s) \ #undef __get_dynamic_array #define __get_dynamic_array(field) \ - ((void *)__entry + __entry->__data_loc_##field) + ((void *)__entry + (__entry->__data_loc_##field & 0xffff)) #undef __get_str #define __get_str(field) (char *)__get_dynamic_array(field) @@ -282,7 +283,7 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags) \ #undef __dynamic_array #define __dynamic_array(type, item, len) \ - ret = trace_define_field(event_call, "__data_loc" "[" #type "]", #item,\ + ret = trace_define_field(event_call, "__data_loc " #type "[]", #item, \ offsetof(typeof(field), __data_loc_##item), \ sizeof(field.__data_loc_##item), 0); @@ -328,6 +329,7 @@ ftrace_define_fields_##call(void) \ #define __dynamic_array(type, item, len) \ __data_offsets->item = __data_size + \ offsetof(typeof(*entry), __data); \ + __data_offsets->item |= (len * sizeof(type)) << 16; \ __data_size += (len) * sizeof(type); #undef __string diff --git a/init/Kconfig b/init/Kconfig index 7659bc5bdeeb..58ee16398537 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -320,21 +320,13 @@ choice prompt "RCU Implementation" default TREE_RCU -config CLASSIC_RCU - bool "Classic RCU" - help - This option selects the classic RCU implementation that is - designed for best read-side performance on non-realtime - systems. - - Select this option if you are unsure. - config TREE_RCU bool "Tree-based hierarchical RCU" help This option selects the RCU implementation that is designed for very large SMP system with hundreds or - thousands of CPUs. + thousands of CPUs. It also scales down nicely to + smaller systems. config PREEMPT_RCU bool "Preemptible RCU" diff --git a/kernel/Makefile b/kernel/Makefile index 55a6d6aa2395..a00eb1c3a8af 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -81,7 +81,6 @@ obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o -obj-$(CONFIG_CLASSIC_RCU) += rcuclassic.o obj-$(CONFIG_TREE_RCU) += rcutree.o obj-$(CONFIG_PREEMPT_RCU) += rcupreempt.o obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o diff --git a/kernel/fork.c b/kernel/fork.c index 2d7d99b7ef45..1008c1c61abd 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -63,6 +63,7 @@ #include <linux/magic.h> #include <linux/perf_counter.h> #include <linux/kmemleak.h> +#include <linux/posix-timers.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -795,10 +796,10 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) thread_group_cputime_init(sig); /* Expiration times and increments. */ - sig->it_virt_expires = cputime_zero; - sig->it_virt_incr = cputime_zero; - sig->it_prof_expires = cputime_zero; - sig->it_prof_incr = cputime_zero; + sig->it[CPUCLOCK_PROF].expires = cputime_zero; + sig->it[CPUCLOCK_PROF].incr = cputime_zero; + sig->it[CPUCLOCK_VIRT].expires = cputime_zero; + sig->it[CPUCLOCK_VIRT].incr = cputime_zero; /* Cached expiration times. */ sig->cputime_expires.prof_exp = cputime_zero; diff --git a/kernel/futex.c b/kernel/futex.c index 0672ff88f159..956e58f469a3 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1010,19 +1010,24 @@ void requeue_futex(struct futex_q *q, struct futex_hash_bucket *hb1, * requeue_pi_wake_futex() - Wake a task that acquired the lock during requeue * q: the futex_q * key: the key of the requeue target futex + * hb: the hash_bucket of the requeue target futex * * During futex_requeue, with requeue_pi=1, it is possible to acquire the * target futex if it is uncontended or via a lock steal. Set the futex_q key * to the requeue target futex so the waiter can detect the wakeup on the right * futex, but remove it from the hb and NULL the rt_waiter so it can detect - * atomic lock acquisition. Must be called with the q->lock_ptr held. + * atomic lock acquisition. Set the q->lock_ptr to the requeue target hb->lock + * to protect access to the pi_state to fixup the owner later. Must be called + * with the q->lock_ptr held. */ static inline -void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key) +void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, + struct futex_hash_bucket *hb) { drop_futex_key_refs(&q->key); get_futex_key_refs(key); q->key = *key; + q->lock_ptr = &hb->lock; WARN_ON(plist_node_empty(&q->list)); plist_del(&q->list, &q->list.plist); @@ -1088,7 +1093,7 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, ret = futex_lock_pi_atomic(pifutex, hb2, key2, ps, top_waiter->task, set_waiters); if (ret == 1) - requeue_pi_wake_futex(top_waiter, key2); + requeue_pi_wake_futex(top_waiter, key2, hb2); return ret; } @@ -1273,7 +1278,7 @@ retry_private: this->task, 1); if (ret == 1) { /* We got the lock. */ - requeue_pi_wake_futex(this, &key2); + requeue_pi_wake_futex(this, &key2, hb2); continue; } else if (ret) { /* -EDEADLK */ @@ -2102,11 +2107,11 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * We call schedule in futex_wait_queue_me() when we enqueue and return there * via the following: * 1) wakeup on uaddr2 after an atomic lock acquisition by futex_requeue() - * 2) wakeup on uaddr2 after a requeue and subsequent unlock - * 3) signal (before or after requeue) - * 4) timeout (before or after requeue) + * 2) wakeup on uaddr2 after a requeue + * 3) signal + * 4) timeout * - * If 3, we setup a restart_block with futex_wait_requeue_pi() as the function. + * If 3, cleanup and return -ERESTARTNOINTR. * * If 2, we may then block on trying to take the rt_mutex and return via: * 5) successful lock @@ -2114,7 +2119,7 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * 7) timeout * 8) other lock acquisition failure * - * If 6, we setup a restart_block with futex_lock_pi() as the function. + * If 6, return -EWOULDBLOCK (restarting the syscall would do the same). * * If 4 or 7, we cleanup and return with -ETIMEDOUT. * @@ -2232,14 +2237,11 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, rt_mutex_unlock(pi_mutex); } else if (ret == -EINTR) { /* - * We've already been requeued, but we have no way to - * restart by calling futex_lock_pi() directly. We - * could restart the syscall, but that will look at - * the user space value and return right away. So we - * drop back with EWOULDBLOCK to tell user space that - * "val" has been changed. That's the same what the - * restart of the syscall would do in - * futex_wait_setup(). + * We've already been requeued, but cannot restart by calling + * futex_lock_pi() directly. We could restart this syscall, but + * it would detect that the user space "val" changed and return + * -EWOULDBLOCK. Save the overhead of the restart and return + * -EWOULDBLOCK directly. */ ret = -EWOULDBLOCK; } diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 49da79ab8486..e2f91ecc01a8 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -48,37 +48,6 @@ #include <asm/uaccess.h> -/** - * ktime_get - get the monotonic time in ktime_t format - * - * returns the time in ktime_t format - */ -ktime_t ktime_get(void) -{ - struct timespec now; - - ktime_get_ts(&now); - - return timespec_to_ktime(now); -} -EXPORT_SYMBOL_GPL(ktime_get); - -/** - * ktime_get_real - get the real (wall-) time in ktime_t format - * - * returns the time in ktime_t format - */ -ktime_t ktime_get_real(void) -{ - struct timespec now; - - getnstimeofday(&now); - - return timespec_to_ktime(now); -} - -EXPORT_SYMBOL_GPL(ktime_get_real); - /* * The timer bases: * @@ -106,31 +75,6 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = } }; -/** - * ktime_get_ts - get the monotonic clock in timespec format - * @ts: pointer to timespec variable - * - * The function calculates the monotonic clock from the realtime - * clock and the wall_to_monotonic offset and stores the result - * in normalized timespec format in the variable pointed to by @ts. - */ -void ktime_get_ts(struct timespec *ts) -{ - struct timespec tomono; - unsigned long seq; - - do { - seq = read_seqbegin(&xtime_lock); - getnstimeofday(ts); - tomono = wall_to_monotonic; - - } while (read_seqretry(&xtime_lock, seq)); - - set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, - ts->tv_nsec + tomono.tv_nsec); -} -EXPORT_SYMBOL_GPL(ktime_get_ts); - /* * Get the coarse grained time at the softirq based on xtime and * wall_to_monotonic. @@ -1154,7 +1098,6 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, clock_id = CLOCK_MONOTONIC; timer->base = &cpu_base->clock_base[clock_id]; - INIT_LIST_HEAD(&timer->cb_entry); hrtimer_init_timer_hres(timer); #ifdef CONFIG_TIMER_STATS diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 61c679db4687..d1538bd4a5f7 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -978,7 +978,7 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, kfree(action); #ifdef CONFIG_DEBUG_SHIRQ - if (irqflags & IRQF_SHARED) { + if (!retval && (irqflags & IRQF_SHARED)) { /* * It's a shared IRQ -- the driver ought to be prepared for it * to happen immediately, so let's make sure.... diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 2f69bee57bf2..3fd30197da2e 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -107,8 +107,8 @@ out_unlock: struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) { - /* those all static, do move them */ - if (desc->irq < NR_IRQS_LEGACY) + /* those static or target node is -1, do not move them */ + if (desc->irq < NR_IRQS_LEGACY || node == -1) return desc; if (desc->node != node) diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c index 89c7117acf2b..090c3763f3a2 100644 --- a/kernel/irq/resend.c +++ b/kernel/irq/resend.c @@ -70,8 +70,7 @@ void check_irq_resend(struct irq_desc *desc, unsigned int irq) if ((status & (IRQ_LEVEL | IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) { desc->status = (status & ~IRQ_PENDING) | IRQ_REPLAY; - if (!desc->chip || !desc->chip->retrigger || - !desc->chip->retrigger(irq)) { + if (!desc->chip->retrigger || !desc->chip->retrigger(irq)) { #ifdef CONFIG_HARDIRQS_SW_RESEND /* Set it pending and activate the softirq: */ set_bit(irq, irqs_resend); diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 4d568294de3e..114e704760fe 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -297,7 +297,6 @@ static int __init irqfixup_setup(char *str) __setup("irqfixup", irqfixup_setup); module_param(irqfixup, int, 0644); -MODULE_PARM_DESC("irqfixup", "0: No fixup, 1: irqfixup mode, 2: irqpoll mode"); static int __init irqpoll_setup(char *str) { diff --git a/kernel/itimer.c b/kernel/itimer.c index 58762f7077ec..8078a32d3b10 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -41,10 +41,43 @@ static struct timeval itimer_get_remtime(struct hrtimer *timer) return ktime_to_timeval(rem); } +static void get_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, + struct itimerval *const value) +{ + cputime_t cval, cinterval; + struct cpu_itimer *it = &tsk->signal->it[clock_id]; + + spin_lock_irq(&tsk->sighand->siglock); + + cval = it->expires; + cinterval = it->incr; + if (!cputime_eq(cval, cputime_zero)) { + struct task_cputime cputime; + cputime_t t; + + thread_group_cputimer(tsk, &cputime); + if (clock_id == CPUCLOCK_PROF) + t = cputime_add(cputime.utime, cputime.stime); + else + /* CPUCLOCK_VIRT */ + t = cputime.utime; + + if (cputime_le(cval, t)) + /* about to fire */ + cval = cputime_one_jiffy; + else + cval = cputime_sub(cval, t); + } + + spin_unlock_irq(&tsk->sighand->siglock); + + cputime_to_timeval(cval, &value->it_value); + cputime_to_timeval(cinterval, &value->it_interval); +} + int do_getitimer(int which, struct itimerval *value) { struct task_struct *tsk = current; - cputime_t cinterval, cval; switch (which) { case ITIMER_REAL: @@ -55,44 +88,10 @@ int do_getitimer(int which, struct itimerval *value) spin_unlock_irq(&tsk->sighand->siglock); break; case ITIMER_VIRTUAL: - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_virt_expires; - cinterval = tsk->signal->it_virt_incr; - if (!cputime_eq(cval, cputime_zero)) { - struct task_cputime cputime; - cputime_t utime; - - thread_group_cputimer(tsk, &cputime); - utime = cputime.utime; - if (cputime_le(cval, utime)) { /* about to fire */ - cval = jiffies_to_cputime(1); - } else { - cval = cputime_sub(cval, utime); - } - } - spin_unlock_irq(&tsk->sighand->siglock); - cputime_to_timeval(cval, &value->it_value); - cputime_to_timeval(cinterval, &value->it_interval); + get_cpu_itimer(tsk, CPUCLOCK_VIRT, value); break; case ITIMER_PROF: - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_prof_expires; - cinterval = tsk->signal->it_prof_incr; - if (!cputime_eq(cval, cputime_zero)) { - struct task_cputime times; - cputime_t ptime; - - thread_group_cputimer(tsk, ×); - ptime = cputime_add(times.utime, times.stime); - if (cputime_le(cval, ptime)) { /* about to fire */ - cval = jiffies_to_cputime(1); - } else { - cval = cputime_sub(cval, ptime); - } - } - spin_unlock_irq(&tsk->sighand->siglock); - cputime_to_timeval(cval, &value->it_value); - cputime_to_timeval(cinterval, &value->it_interval); + get_cpu_itimer(tsk, CPUCLOCK_PROF, value); break; default: return(-EINVAL); @@ -128,6 +127,54 @@ enum hrtimer_restart it_real_fn(struct hrtimer *timer) return HRTIMER_NORESTART; } +static inline u32 cputime_sub_ns(cputime_t ct, s64 real_ns) +{ + struct timespec ts; + s64 cpu_ns; + + cputime_to_timespec(ct, &ts); + cpu_ns = timespec_to_ns(&ts); + + return (cpu_ns <= real_ns) ? 0 : cpu_ns - real_ns; +} + +static void set_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, + const struct itimerval *const value, + struct itimerval *const ovalue) +{ + cputime_t cval, nval, cinterval, ninterval; + s64 ns_ninterval, ns_nval; + struct cpu_itimer *it = &tsk->signal->it[clock_id]; + + nval = timeval_to_cputime(&value->it_value); + ns_nval = timeval_to_ns(&value->it_value); + ninterval = timeval_to_cputime(&value->it_interval); + ns_ninterval = timeval_to_ns(&value->it_interval); + + it->incr_error = cputime_sub_ns(ninterval, ns_ninterval); + it->error = cputime_sub_ns(nval, ns_nval); + + spin_lock_irq(&tsk->sighand->siglock); + + cval = it->expires; + cinterval = it->incr; + if (!cputime_eq(cval, cputime_zero) || + !cputime_eq(nval, cputime_zero)) { + if (cputime_gt(nval, cputime_zero)) + nval = cputime_add(nval, cputime_one_jiffy); + set_process_cpu_timer(tsk, clock_id, &nval, &cval); + } + it->expires = nval; + it->incr = ninterval; + + spin_unlock_irq(&tsk->sighand->siglock); + + if (ovalue) { + cputime_to_timeval(cval, &ovalue->it_value); + cputime_to_timeval(cinterval, &ovalue->it_interval); + } +} + /* * Returns true if the timeval is in canonical form */ @@ -139,7 +186,6 @@ int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue) struct task_struct *tsk = current; struct hrtimer *timer; ktime_t expires; - cputime_t cval, cinterval, nval, ninterval; /* * Validate the timevals in value. @@ -174,48 +220,10 @@ again: spin_unlock_irq(&tsk->sighand->siglock); break; case ITIMER_VIRTUAL: - nval = timeval_to_cputime(&value->it_value); - ninterval = timeval_to_cputime(&value->it_interval); - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_virt_expires; - cinterval = tsk->signal->it_virt_incr; - if (!cputime_eq(cval, cputime_zero) || - !cputime_eq(nval, cputime_zero)) { - if (cputime_gt(nval, cputime_zero)) - nval = cputime_add(nval, - jiffies_to_cputime(1)); - set_process_cpu_timer(tsk, CPUCLOCK_VIRT, - &nval, &cval); - } - tsk->signal->it_virt_expires = nval; - tsk->signal->it_virt_incr = ninterval; - spin_unlock_irq(&tsk->sighand->siglock); - if (ovalue) { - cputime_to_timeval(cval, &ovalue->it_value); - cputime_to_timeval(cinterval, &ovalue->it_interval); - } + set_cpu_itimer(tsk, CPUCLOCK_VIRT, value, ovalue); break; case ITIMER_PROF: - nval = timeval_to_cputime(&value->it_value); - ninterval = timeval_to_cputime(&value->it_interval); - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_prof_expires; - cinterval = tsk->signal->it_prof_incr; - if (!cputime_eq(cval, cputime_zero) || - !cputime_eq(nval, cputime_zero)) { - if (cputime_gt(nval, cputime_zero)) - nval = cputime_add(nval, - jiffies_to_cputime(1)); - set_process_cpu_timer(tsk, CPUCLOCK_PROF, - &nval, &cval); - } - tsk->signal->it_prof_expires = nval; - tsk->signal->it_prof_incr = ninterval; - spin_unlock_irq(&tsk->sighand->siglock); - if (ovalue) { - cputime_to_timeval(cval, &ovalue->it_value); - cputime_to_timeval(cinterval, &ovalue->it_interval); - } + set_cpu_itimer(tsk, CPUCLOCK_PROF, value, ovalue); break; default: return -EINVAL; diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 0540948e29ab..ef177d653b2c 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -103,7 +103,7 @@ static struct kprobe_blackpoint kprobe_blacklist[] = { #define INSNS_PER_PAGE (PAGE_SIZE/(MAX_INSN_SIZE * sizeof(kprobe_opcode_t))) struct kprobe_insn_page { - struct hlist_node hlist; + struct list_head list; kprobe_opcode_t *insns; /* Page of instruction slots */ char slot_used[INSNS_PER_PAGE]; int nused; @@ -117,7 +117,7 @@ enum kprobe_slot_state { }; static DEFINE_MUTEX(kprobe_insn_mutex); /* Protects kprobe_insn_pages */ -static struct hlist_head kprobe_insn_pages; +static LIST_HEAD(kprobe_insn_pages); static int kprobe_garbage_slots; static int collect_garbage_slots(void); @@ -152,10 +152,9 @@ loop_end: static kprobe_opcode_t __kprobes *__get_insn_slot(void) { struct kprobe_insn_page *kip; - struct hlist_node *pos; retry: - hlist_for_each_entry(kip, pos, &kprobe_insn_pages, hlist) { + list_for_each_entry(kip, &kprobe_insn_pages, list) { if (kip->nused < INSNS_PER_PAGE) { int i; for (i = 0; i < INSNS_PER_PAGE; i++) { @@ -189,8 +188,8 @@ static kprobe_opcode_t __kprobes *__get_insn_slot(void) kfree(kip); return NULL; } - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, &kprobe_insn_pages); + INIT_LIST_HEAD(&kip->list); + list_add(&kip->list, &kprobe_insn_pages); memset(kip->slot_used, SLOT_CLEAN, INSNS_PER_PAGE); kip->slot_used[0] = SLOT_USED; kip->nused = 1; @@ -219,12 +218,8 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) * so as not to have to set it up again the * next time somebody inserts a probe. */ - hlist_del(&kip->hlist); - if (hlist_empty(&kprobe_insn_pages)) { - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, - &kprobe_insn_pages); - } else { + if (!list_is_singular(&kprobe_insn_pages)) { + list_del(&kip->list); module_free(NULL, kip->insns); kfree(kip); } @@ -235,14 +230,13 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) static int __kprobes collect_garbage_slots(void) { - struct kprobe_insn_page *kip; - struct hlist_node *pos, *next; + struct kprobe_insn_page *kip, *next; /* Ensure no-one is preepmted on the garbages */ if (check_safety()) return -EAGAIN; - hlist_for_each_entry_safe(kip, pos, next, &kprobe_insn_pages, hlist) { + list_for_each_entry_safe(kip, next, &kprobe_insn_pages, list) { int i; if (kip->ngarbage == 0) continue; @@ -260,19 +254,17 @@ static int __kprobes collect_garbage_slots(void) void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) { struct kprobe_insn_page *kip; - struct hlist_node *pos; mutex_lock(&kprobe_insn_mutex); - hlist_for_each_entry(kip, pos, &kprobe_insn_pages, hlist) { + list_for_each_entry(kip, &kprobe_insn_pages, list) { if (kip->insns <= slot && slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; if (dirty) { kip->slot_used[i] = SLOT_DIRTY; kip->ngarbage++; - } else { + } else collect_one_slot(kip, i); - } break; } } diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 8bbeef996c76..0843584aed5e 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -42,6 +42,7 @@ #include <linux/hash.h> #include <linux/ftrace.h> #include <linux/stringify.h> +#include <linux/bitops.h> #include <asm/sections.h> @@ -366,11 +367,21 @@ static int save_trace(struct stack_trace *trace) save_stack_trace(trace); + /* + * Some daft arches put -1 at the end to indicate its a full trace. + * + * <rant> this is buggy anyway, since it takes a whole extra entry so a + * complete trace that maxes out the entries provided will be reported + * as incomplete, friggin useless </rant> + */ + if (trace->entries[trace->nr_entries-1] == ULONG_MAX) + trace->nr_entries--; + trace->max_entries = trace->nr_entries; nr_stack_trace_entries += trace->nr_entries; - if (nr_stack_trace_entries == MAX_STACK_TRACE_ENTRIES) { + if (nr_stack_trace_entries >= MAX_STACK_TRACE_ENTRIES-1) { if (!debug_locks_off_graph_unlock()) return 0; @@ -390,19 +401,6 @@ unsigned int nr_process_chains; unsigned int max_lockdep_depth; unsigned int max_recursion_depth; -static unsigned int lockdep_dependency_gen_id; - -static bool lockdep_dependency_visit(struct lock_class *source, - unsigned int depth) -{ - if (!depth) - lockdep_dependency_gen_id++; - if (source->dep_gen_id == lockdep_dependency_gen_id) - return true; - source->dep_gen_id = lockdep_dependency_gen_id; - return false; -} - #ifdef CONFIG_DEBUG_LOCKDEP /* * We cannot printk in early bootup code. Not even early_printk() @@ -551,58 +549,6 @@ static void lockdep_print_held_locks(struct task_struct *curr) } } -static void print_lock_class_header(struct lock_class *class, int depth) -{ - int bit; - - printk("%*s->", depth, ""); - print_lock_name(class); - printk(" ops: %lu", class->ops); - printk(" {\n"); - - for (bit = 0; bit < LOCK_USAGE_STATES; bit++) { - if (class->usage_mask & (1 << bit)) { - int len = depth; - - len += printk("%*s %s", depth, "", usage_str[bit]); - len += printk(" at:\n"); - print_stack_trace(class->usage_traces + bit, len); - } - } - printk("%*s }\n", depth, ""); - - printk("%*s ... key at: ",depth,""); - print_ip_sym((unsigned long)class->key); -} - -/* - * printk all lock dependencies starting at <entry>: - */ -static void __used -print_lock_dependencies(struct lock_class *class, int depth) -{ - struct lock_list *entry; - - if (lockdep_dependency_visit(class, depth)) - return; - - if (DEBUG_LOCKS_WARN_ON(depth >= 20)) - return; - - print_lock_class_header(class, depth); - - list_for_each_entry(entry, &class->locks_after, entry) { - if (DEBUG_LOCKS_WARN_ON(!entry->class)) - return; - - print_lock_dependencies(entry->class, depth + 1); - - printk("%*s ... acquired at:\n",depth,""); - print_stack_trace(&entry->trace, 2); - printk("\n"); - } -} - static void print_kernel_version(void) { printk("%s %.*s\n", init_utsname()->release, @@ -898,22 +844,203 @@ static int add_lock_to_list(struct lock_class *class, struct lock_class *this, } /* + * For good efficiency of modular, we use power of 2 + */ +#define MAX_CIRCULAR_QUEUE_SIZE 4096UL +#define CQ_MASK (MAX_CIRCULAR_QUEUE_SIZE-1) + +/* + * The circular_queue and helpers is used to implement the + * breadth-first search(BFS)algorithem, by which we can build + * the shortest path from the next lock to be acquired to the + * previous held lock if there is a circular between them. + */ +struct circular_queue { + unsigned long element[MAX_CIRCULAR_QUEUE_SIZE]; + unsigned int front, rear; +}; + +static struct circular_queue lock_cq; + +unsigned int max_bfs_queue_depth; + +static unsigned int lockdep_dependency_gen_id; + +static inline void __cq_init(struct circular_queue *cq) +{ + cq->front = cq->rear = 0; + lockdep_dependency_gen_id++; +} + +static inline int __cq_empty(struct circular_queue *cq) +{ + return (cq->front == cq->rear); +} + +static inline int __cq_full(struct circular_queue *cq) +{ + return ((cq->rear + 1) & CQ_MASK) == cq->front; +} + +static inline int __cq_enqueue(struct circular_queue *cq, unsigned long elem) +{ + if (__cq_full(cq)) + return -1; + + cq->element[cq->rear] = elem; + cq->rear = (cq->rear + 1) & CQ_MASK; + return 0; +} + +static inline int __cq_dequeue(struct circular_queue *cq, unsigned long *elem) +{ + if (__cq_empty(cq)) + return -1; + + *elem = cq->element[cq->front]; + cq->front = (cq->front + 1) & CQ_MASK; + return 0; +} + +static inline unsigned int __cq_get_elem_count(struct circular_queue *cq) +{ + return (cq->rear - cq->front) & CQ_MASK; +} + +static inline void mark_lock_accessed(struct lock_list *lock, + struct lock_list *parent) +{ + unsigned long nr; + + nr = lock - list_entries; + WARN_ON(nr >= nr_list_entries); + lock->parent = parent; + lock->class->dep_gen_id = lockdep_dependency_gen_id; +} + +static inline unsigned long lock_accessed(struct lock_list *lock) +{ + unsigned long nr; + + nr = lock - list_entries; + WARN_ON(nr >= nr_list_entries); + return lock->class->dep_gen_id == lockdep_dependency_gen_id; +} + +static inline struct lock_list *get_lock_parent(struct lock_list *child) +{ + return child->parent; +} + +static inline int get_lock_depth(struct lock_list *child) +{ + int depth = 0; + struct lock_list *parent; + + while ((parent = get_lock_parent(child))) { + child = parent; + depth++; + } + return depth; +} + +static int __bfs(struct lock_list *source_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry, + int forward) +{ + struct lock_list *entry; + struct list_head *head; + struct circular_queue *cq = &lock_cq; + int ret = 1; + + if (match(source_entry, data)) { + *target_entry = source_entry; + ret = 0; + goto exit; + } + + if (forward) + head = &source_entry->class->locks_after; + else + head = &source_entry->class->locks_before; + + if (list_empty(head)) + goto exit; + + __cq_init(cq); + __cq_enqueue(cq, (unsigned long)source_entry); + + while (!__cq_empty(cq)) { + struct lock_list *lock; + + __cq_dequeue(cq, (unsigned long *)&lock); + + if (!lock->class) { + ret = -2; + goto exit; + } + + if (forward) + head = &lock->class->locks_after; + else + head = &lock->class->locks_before; + + list_for_each_entry(entry, head, entry) { + if (!lock_accessed(entry)) { + unsigned int cq_depth; + mark_lock_accessed(entry, lock); + if (match(entry, data)) { + *target_entry = entry; + ret = 0; + goto exit; + } + + if (__cq_enqueue(cq, (unsigned long)entry)) { + ret = -1; + goto exit; + } + cq_depth = __cq_get_elem_count(cq); + if (max_bfs_queue_depth < cq_depth) + max_bfs_queue_depth = cq_depth; + } + } + } +exit: + return ret; +} + +static inline int __bfs_forwards(struct lock_list *src_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry) +{ + return __bfs(src_entry, data, match, target_entry, 1); + +} + +static inline int __bfs_backwards(struct lock_list *src_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry) +{ + return __bfs(src_entry, data, match, target_entry, 0); + +} + +/* * Recursive, forwards-direction lock-dependency checking, used for * both noncyclic checking and for hardirq-unsafe/softirq-unsafe * checking. - * - * (to keep the stackframe of the recursive functions small we - * use these global variables, and we also mark various helper - * functions as noinline.) */ -static struct held_lock *check_source, *check_target; /* * Print a dependency chain entry (this is only done when a deadlock * has been detected): */ static noinline int -print_circular_bug_entry(struct lock_list *target, unsigned int depth) +print_circular_bug_entry(struct lock_list *target, int depth) { if (debug_locks_silent) return 0; @@ -930,11 +1057,13 @@ print_circular_bug_entry(struct lock_list *target, unsigned int depth) * header first: */ static noinline int -print_circular_bug_header(struct lock_list *entry, unsigned int depth) +print_circular_bug_header(struct lock_list *entry, unsigned int depth, + struct held_lock *check_src, + struct held_lock *check_tgt) { struct task_struct *curr = current; - if (!debug_locks_off_graph_unlock() || debug_locks_silent) + if (debug_locks_silent) return 0; printk("\n=======================================================\n"); @@ -943,9 +1072,9 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth) printk( "-------------------------------------------------------\n"); printk("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); - print_lock(check_source); + print_lock(check_src); printk("\nbut task is already holding lock:\n"); - print_lock(check_target); + print_lock(check_tgt); printk("\nwhich lock already depends on the new lock.\n\n"); printk("\nthe existing dependency chain (in reverse order) is:\n"); @@ -954,19 +1083,36 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth) return 0; } -static noinline int print_circular_bug_tail(void) +static inline int class_equal(struct lock_list *entry, void *data) +{ + return entry->class == data; +} + +static noinline int print_circular_bug(struct lock_list *this, + struct lock_list *target, + struct held_lock *check_src, + struct held_lock *check_tgt) { struct task_struct *curr = current; - struct lock_list this; + struct lock_list *parent; + int depth; - if (debug_locks_silent) + if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - this.class = hlock_class(check_source); - if (!save_trace(&this.trace)) + if (!save_trace(&this->trace)) return 0; - print_circular_bug_entry(&this, 0); + depth = get_lock_depth(target); + + print_circular_bug_header(target, depth, check_src, check_tgt); + + parent = get_lock_parent(target); + + while (parent) { + print_circular_bug_entry(parent, --depth); + parent = get_lock_parent(parent); + } printk("\nother info that might help us debug this:\n\n"); lockdep_print_held_locks(curr); @@ -977,73 +1123,69 @@ static noinline int print_circular_bug_tail(void) return 0; } -#define RECURSION_LIMIT 40 - -static int noinline print_infinite_recursion_bug(void) +static noinline int print_bfs_bug(int ret) { if (!debug_locks_off_graph_unlock()) return 0; - WARN_ON(1); + WARN(1, "lockdep bfs error:%d\n", ret); return 0; } -unsigned long __lockdep_count_forward_deps(struct lock_class *class, - unsigned int depth) +static int noop_count(struct lock_list *entry, void *data) { - struct lock_list *entry; - unsigned long ret = 1; + (*(unsigned long *)data)++; + return 0; +} - if (lockdep_dependency_visit(class, depth)) - return 0; +unsigned long __lockdep_count_forward_deps(struct lock_list *this) +{ + unsigned long count = 0; + struct lock_list *uninitialized_var(target_entry); - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_after, entry) - ret += __lockdep_count_forward_deps(entry->class, depth + 1); + __bfs_forwards(this, (void *)&count, noop_count, &target_entry); - return ret; + return count; } - unsigned long lockdep_count_forward_deps(struct lock_class *class) { unsigned long ret, flags; + struct lock_list this; + + this.parent = NULL; + this.class = class; local_irq_save(flags); __raw_spin_lock(&lockdep_lock); - ret = __lockdep_count_forward_deps(class, 0); + ret = __lockdep_count_forward_deps(&this); __raw_spin_unlock(&lockdep_lock); local_irq_restore(flags); return ret; } -unsigned long __lockdep_count_backward_deps(struct lock_class *class, - unsigned int depth) +unsigned long __lockdep_count_backward_deps(struct lock_list *this) { - struct lock_list *entry; - unsigned long ret = 1; + unsigned long count = 0; + struct lock_list *uninitialized_var(target_entry); - if (lockdep_dependency_visit(class, depth)) - return 0; - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_before, entry) - ret += __lockdep_count_backward_deps(entry->class, depth + 1); + __bfs_backwards(this, (void *)&count, noop_count, &target_entry); - return ret; + return count; } unsigned long lockdep_count_backward_deps(struct lock_class *class) { unsigned long ret, flags; + struct lock_list this; + + this.parent = NULL; + this.class = class; local_irq_save(flags); __raw_spin_lock(&lockdep_lock); - ret = __lockdep_count_backward_deps(class, 0); + ret = __lockdep_count_backward_deps(&this); __raw_spin_unlock(&lockdep_lock); local_irq_restore(flags); @@ -1055,29 +1197,16 @@ unsigned long lockdep_count_backward_deps(struct lock_class *class) * lead to <target>. Print an error and return 0 if it does. */ static noinline int -check_noncircular(struct lock_class *source, unsigned int depth) +check_noncircular(struct lock_list *root, struct lock_class *target, + struct lock_list **target_entry) { - struct lock_list *entry; + int result; - if (lockdep_dependency_visit(source, depth)) - return 1; + debug_atomic_inc(&nr_cyclic_checks); - debug_atomic_inc(&nr_cyclic_check_recursions); - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_after, entry) { - if (entry->class == hlock_class(check_target)) - return print_circular_bug_header(entry, depth+1); - debug_atomic_inc(&nr_cyclic_checks); - if (!check_noncircular(entry->class, depth+1)) - return print_circular_bug_entry(entry, depth+1); - } - return 1; + result = __bfs_forwards(root, target, class_equal, target_entry); + + return result; } #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) @@ -1086,103 +1215,121 @@ check_noncircular(struct lock_class *source, unsigned int depth) * proving that two subgraphs can be connected by a new dependency * without creating any illegal irq-safe -> irq-unsafe lock dependency. */ -static enum lock_usage_bit find_usage_bit; -static struct lock_class *forwards_match, *backwards_match; + +static inline int usage_match(struct lock_list *entry, void *bit) +{ + return entry->class->usage_mask & (1 << (enum lock_usage_bit)bit); +} + + /* * Find a node in the forwards-direction dependency sub-graph starting - * at <source> that matches <find_usage_bit>. + * at @root->class that matches @bit. * - * Return 2 if such a node exists in the subgraph, and put that node - * into <forwards_match>. + * Return 0 if such a node exists in the subgraph, and put that node + * into *@target_entry. * - * Return 1 otherwise and keep <forwards_match> unchanged. - * Return 0 on error. + * Return 1 otherwise and keep *@target_entry unchanged. + * Return <0 on error. */ -static noinline int -find_usage_forwards(struct lock_class *source, unsigned int depth) +static int +find_usage_forwards(struct lock_list *root, enum lock_usage_bit bit, + struct lock_list **target_entry) { - struct lock_list *entry; - int ret; - - if (lockdep_dependency_visit(source, depth)) - return 1; - - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); + int result; debug_atomic_inc(&nr_find_usage_forwards_checks); - if (source->usage_mask & (1 << find_usage_bit)) { - forwards_match = source; - return 2; - } - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_after, entry) { - debug_atomic_inc(&nr_find_usage_forwards_recursions); - ret = find_usage_forwards(entry->class, depth+1); - if (ret == 2 || ret == 0) - return ret; - } - return 1; + result = __bfs_forwards(root, (void *)bit, usage_match, target_entry); + + return result; } /* * Find a node in the backwards-direction dependency sub-graph starting - * at <source> that matches <find_usage_bit>. + * at @root->class that matches @bit. * - * Return 2 if such a node exists in the subgraph, and put that node - * into <backwards_match>. + * Return 0 if such a node exists in the subgraph, and put that node + * into *@target_entry. * - * Return 1 otherwise and keep <backwards_match> unchanged. - * Return 0 on error. + * Return 1 otherwise and keep *@target_entry unchanged. + * Return <0 on error. */ -static noinline int -find_usage_backwards(struct lock_class *source, unsigned int depth) +static int +find_usage_backwards(struct lock_list *root, enum lock_usage_bit bit, + struct lock_list **target_entry) { - struct lock_list *entry; - int ret; + int result; - if (lockdep_dependency_visit(source, depth)) - return 1; + debug_atomic_inc(&nr_find_usage_backwards_checks); - if (!__raw_spin_is_locked(&lockdep_lock)) - return DEBUG_LOCKS_WARN_ON(1); + result = __bfs_backwards(root, (void *)bit, usage_match, target_entry); - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); + return result; +} - debug_atomic_inc(&nr_find_usage_backwards_checks); - if (source->usage_mask & (1 << find_usage_bit)) { - backwards_match = source; - return 2; - } +static void print_lock_class_header(struct lock_class *class, int depth) +{ + int bit; - if (!source && debug_locks_off_graph_unlock()) { - WARN_ON(1); - return 0; - } + printk("%*s->", depth, ""); + print_lock_name(class); + printk(" ops: %lu", class->ops); + printk(" {\n"); - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_before, entry) { - debug_atomic_inc(&nr_find_usage_backwards_recursions); - ret = find_usage_backwards(entry->class, depth+1); - if (ret == 2 || ret == 0) - return ret; + for (bit = 0; bit < LOCK_USAGE_STATES; bit++) { + if (class->usage_mask & (1 << bit)) { + int len = depth; + + len += printk("%*s %s", depth, "", usage_str[bit]); + len += printk(" at:\n"); + print_stack_trace(class->usage_traces + bit, len); + } } - return 1; + printk("%*s }\n", depth, ""); + + printk("%*s ... key at: ",depth,""); + print_ip_sym((unsigned long)class->key); +} + +/* + * printk the shortest lock dependencies from @start to @end in reverse order: + */ +static void __used +print_shortest_lock_dependencies(struct lock_list *leaf, + struct lock_list *root) +{ + struct lock_list *entry = leaf; + int depth; + + /*compute depth from generated tree by BFS*/ + depth = get_lock_depth(leaf); + + do { + print_lock_class_header(entry->class, depth); + printk("%*s ... acquired at:\n", depth, ""); + print_stack_trace(&entry->trace, 2); + printk("\n"); + + if (depth == 0 && (entry != root)) { + printk("lockdep:%s bad BFS generated tree\n", __func__); + break; + } + + entry = get_lock_parent(entry); + depth--; + } while (entry && (depth >= 0)); + + return; } static int print_bad_irq_dependency(struct task_struct *curr, + struct lock_list *prev_root, + struct lock_list *next_root, + struct lock_list *backwards_entry, + struct lock_list *forwards_entry, struct held_lock *prev, struct held_lock *next, enum lock_usage_bit bit1, @@ -1215,26 +1362,32 @@ print_bad_irq_dependency(struct task_struct *curr, printk("\nbut this new dependency connects a %s-irq-safe lock:\n", irqclass); - print_lock_name(backwards_match); + print_lock_name(backwards_entry->class); printk("\n... which became %s-irq-safe at:\n", irqclass); - print_stack_trace(backwards_match->usage_traces + bit1, 1); + print_stack_trace(backwards_entry->class->usage_traces + bit1, 1); printk("\nto a %s-irq-unsafe lock:\n", irqclass); - print_lock_name(forwards_match); + print_lock_name(forwards_entry->class); printk("\n... which became %s-irq-unsafe at:\n", irqclass); printk("..."); - print_stack_trace(forwards_match->usage_traces + bit2, 1); + print_stack_trace(forwards_entry->class->usage_traces + bit2, 1); printk("\nother info that might help us debug this:\n\n"); lockdep_print_held_locks(curr); - printk("\nthe %s-irq-safe lock's dependencies:\n", irqclass); - print_lock_dependencies(backwards_match, 0); + printk("\nthe dependencies between %s-irq-safe lock", irqclass); + printk(" and the holding lock:\n"); + if (!save_trace(&prev_root->trace)) + return 0; + print_shortest_lock_dependencies(backwards_entry, prev_root); - printk("\nthe %s-irq-unsafe lock's dependencies:\n", irqclass); - print_lock_dependencies(forwards_match, 0); + printk("\nthe dependencies between the lock to be acquired"); + printk(" and %s-irq-unsafe lock:\n", irqclass); + if (!save_trace(&next_root->trace)) + return 0; + print_shortest_lock_dependencies(forwards_entry, next_root); printk("\nstack backtrace:\n"); dump_stack(); @@ -1248,19 +1401,30 @@ check_usage(struct task_struct *curr, struct held_lock *prev, enum lock_usage_bit bit_forwards, const char *irqclass) { int ret; + struct lock_list this, that; + struct lock_list *uninitialized_var(target_entry); + struct lock_list *uninitialized_var(target_entry1); - find_usage_bit = bit_backwards; - /* fills in <backwards_match> */ - ret = find_usage_backwards(hlock_class(prev), 0); - if (!ret || ret == 1) + this.parent = NULL; + + this.class = hlock_class(prev); + ret = find_usage_backwards(&this, bit_backwards, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - find_usage_bit = bit_forwards; - ret = find_usage_forwards(hlock_class(next), 0); - if (!ret || ret == 1) + that.parent = NULL; + that.class = hlock_class(next); + ret = find_usage_forwards(&that, bit_forwards, &target_entry1); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - /* ret == 2 */ - return print_bad_irq_dependency(curr, prev, next, + + return print_bad_irq_dependency(curr, &this, &that, + target_entry, target_entry1, + prev, next, bit_backwards, bit_forwards, irqclass); } @@ -1472,6 +1636,8 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, { struct lock_list *entry; int ret; + struct lock_list this; + struct lock_list *uninitialized_var(target_entry); /* * Prove that the new <prev> -> <next> dependency would not @@ -1482,10 +1648,13 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, * We are using global variables to control the recursion, to * keep the stackframe size of the recursive functions low: */ - check_source = next; - check_target = prev; - if (!(check_noncircular(hlock_class(next), 0))) - return print_circular_bug_tail(); + this.class = hlock_class(next); + this.parent = NULL; + ret = check_noncircular(&this, hlock_class(prev), &target_entry); + if (unlikely(!ret)) + return print_circular_bug(&this, target_entry, next, prev); + else if (unlikely(ret < 0)) + return print_bfs_bug(ret); if (!check_prev_add_irq(curr, prev, next)) return 0; @@ -1884,7 +2053,8 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, * print irq inversion bug: */ static int -print_irq_inversion_bug(struct task_struct *curr, struct lock_class *other, +print_irq_inversion_bug(struct task_struct *curr, + struct lock_list *root, struct lock_list *other, struct held_lock *this, int forwards, const char *irqclass) { @@ -1902,17 +2072,16 @@ print_irq_inversion_bug(struct task_struct *curr, struct lock_class *other, printk("but this lock took another, %s-unsafe lock in the past:\n", irqclass); else printk("but this lock was taken by another, %s-safe lock in the past:\n", irqclass); - print_lock_name(other); + print_lock_name(other->class); printk("\n\nand interrupts could create inverse lock ordering between them.\n\n"); printk("\nother info that might help us debug this:\n"); lockdep_print_held_locks(curr); - printk("\nthe first lock's dependencies:\n"); - print_lock_dependencies(hlock_class(this), 0); - - printk("\nthe second lock's dependencies:\n"); - print_lock_dependencies(other, 0); + printk("\nthe shortest dependencies between 2nd lock and 1st lock:\n"); + if (!save_trace(&root->trace)) + return 0; + print_shortest_lock_dependencies(other, root); printk("\nstack backtrace:\n"); dump_stack(); @@ -1929,14 +2098,19 @@ check_usage_forwards(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit bit, const char *irqclass) { int ret; - - find_usage_bit = bit; - /* fills in <forwards_match> */ - ret = find_usage_forwards(hlock_class(this), 0); - if (!ret || ret == 1) + struct lock_list root; + struct lock_list *uninitialized_var(target_entry); + + root.parent = NULL; + root.class = hlock_class(this); + ret = find_usage_forwards(&root, bit, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - return print_irq_inversion_bug(curr, forwards_match, this, 1, irqclass); + return print_irq_inversion_bug(curr, &root, target_entry, + this, 1, irqclass); } /* @@ -1948,14 +2122,19 @@ check_usage_backwards(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit bit, const char *irqclass) { int ret; - - find_usage_bit = bit; - /* fills in <backwards_match> */ - ret = find_usage_backwards(hlock_class(this), 0); - if (!ret || ret == 1) + struct lock_list root; + struct lock_list *uninitialized_var(target_entry); + + root.parent = NULL; + root.class = hlock_class(this); + ret = find_usage_backwards(&root, bit, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - return print_irq_inversion_bug(curr, backwards_match, this, 0, irqclass); + return print_irq_inversion_bug(curr, &root, target_entry, + this, 1, irqclass); } void print_irqtrace_events(struct task_struct *curr) @@ -2530,13 +2709,15 @@ EXPORT_SYMBOL_GPL(lockdep_init_map); */ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int trylock, int read, int check, int hardirqs_off, - struct lockdep_map *nest_lock, unsigned long ip) + struct lockdep_map *nest_lock, unsigned long ip, + int references) { struct task_struct *curr = current; struct lock_class *class = NULL; struct held_lock *hlock; unsigned int depth, id; int chain_head = 0; + int class_idx; u64 chain_key; if (!prove_locking) @@ -2584,10 +2765,24 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (DEBUG_LOCKS_WARN_ON(depth >= MAX_LOCK_DEPTH)) return 0; + class_idx = class - lock_classes + 1; + + if (depth) { + hlock = curr->held_locks + depth - 1; + if (hlock->class_idx == class_idx && nest_lock) { + if (hlock->references) + hlock->references++; + else + hlock->references = 2; + + return 1; + } + } + hlock = curr->held_locks + depth; if (DEBUG_LOCKS_WARN_ON(!class)) return 0; - hlock->class_idx = class - lock_classes + 1; + hlock->class_idx = class_idx; hlock->acquire_ip = ip; hlock->instance = lock; hlock->nest_lock = nest_lock; @@ -2595,6 +2790,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->read = read; hlock->check = check; hlock->hardirqs_off = !!hardirqs_off; + hlock->references = references; #ifdef CONFIG_LOCK_STAT hlock->waittime_stamp = 0; hlock->holdtime_stamp = sched_clock(); @@ -2703,6 +2899,30 @@ static int check_unlock(struct task_struct *curr, struct lockdep_map *lock, return 1; } +static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) +{ + if (hlock->instance == lock) + return 1; + + if (hlock->references) { + struct lock_class *class = lock->class_cache; + + if (!class) + class = look_up_lock_class(lock, 0); + + if (DEBUG_LOCKS_WARN_ON(!class)) + return 0; + + if (DEBUG_LOCKS_WARN_ON(!hlock->nest_lock)) + return 0; + + if (hlock->class_idx == class - lock_classes + 1) + return 1; + } + + return 0; +} + static int __lock_set_class(struct lockdep_map *lock, const char *name, struct lock_class_key *key, unsigned int subclass, @@ -2726,7 +2946,7 @@ __lock_set_class(struct lockdep_map *lock, const char *name, */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -2745,7 +2965,8 @@ found_it: if (!__lock_acquire(hlock->instance, hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip)) + hlock->nest_lock, hlock->acquire_ip, + hlock->references)) return 0; } @@ -2784,20 +3005,34 @@ lock_release_non_nested(struct task_struct *curr, */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } return print_unlock_inbalance_bug(curr, lock, ip); found_it: - lock_release_holdtime(hlock); + if (hlock->instance == lock) + lock_release_holdtime(hlock); + + if (hlock->references) { + hlock->references--; + if (hlock->references) { + /* + * We had, and after removing one, still have + * references, the current lock stack is still + * valid. We're done! + */ + return 1; + } + } /* * We have the right lock to unlock, 'hlock' points to it. * Now we remove it from the stack, and add back the other * entries (if any), recalculating the hash along the way: */ + curr->lockdep_depth = i; curr->curr_chain_key = hlock->prev_chain_key; @@ -2806,7 +3041,8 @@ found_it: if (!__lock_acquire(hlock->instance, hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip)) + hlock->nest_lock, hlock->acquire_ip, + hlock->references)) return 0; } @@ -2836,7 +3072,7 @@ static int lock_release_nested(struct task_struct *curr, /* * Is the unlock non-nested: */ - if (hlock->instance != lock) + if (hlock->instance != lock || hlock->references) return lock_release_non_nested(curr, lock, ip); curr->lockdep_depth--; @@ -2881,6 +3117,21 @@ __lock_release(struct lockdep_map *lock, int nested, unsigned long ip) check_chain_key(curr); } +static int __lock_is_held(struct lockdep_map *lock) +{ + struct task_struct *curr = current; + int i; + + for (i = 0; i < curr->lockdep_depth; i++) { + struct held_lock *hlock = curr->held_locks + i; + + if (match_held_lock(hlock, lock)) + return 1; + } + + return 0; +} + /* * Check whether we follow the irq-flags state precisely: */ @@ -2957,7 +3208,7 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, current->lockdep_recursion = 1; __lock_acquire(lock, subclass, trylock, read, check, - irqs_disabled_flags(flags), nest_lock, ip); + irqs_disabled_flags(flags), nest_lock, ip, 0); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } @@ -2982,6 +3233,26 @@ void lock_release(struct lockdep_map *lock, int nested, } EXPORT_SYMBOL_GPL(lock_release); +int lock_is_held(struct lockdep_map *lock) +{ + unsigned long flags; + int ret = 0; + + if (unlikely(current->lockdep_recursion)) + return ret; + + raw_local_irq_save(flags); + check_flags(flags); + + current->lockdep_recursion = 1; + ret = __lock_is_held(lock); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); + + return ret; +} +EXPORT_SYMBOL_GPL(lock_is_held); + void lockdep_set_current_reclaim_state(gfp_t gfp_mask) { current->lockdep_reclaim_gfp = gfp_mask; @@ -3041,7 +3312,7 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -3049,6 +3320,9 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) return; found_it: + if (hlock->instance != lock) + return; + hlock->waittime_stamp = sched_clock(); contention_point = lock_point(hlock_class(hlock)->contention_point, ip); @@ -3088,7 +3362,7 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -3096,6 +3370,9 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) return; found_it: + if (hlock->instance != lock) + return; + cpu = smp_processor_id(); if (hlock->waittime_stamp) { now = sched_clock(); @@ -3326,7 +3603,12 @@ void __init lockdep_info(void) sizeof(struct list_head) * CLASSHASH_SIZE + sizeof(struct lock_list) * MAX_LOCKDEP_ENTRIES + sizeof(struct lock_chain) * MAX_LOCKDEP_CHAINS + - sizeof(struct list_head) * CHAINHASH_SIZE) / 1024); + sizeof(struct list_head) * CHAINHASH_SIZE +#ifdef CONFIG_PROVE_LOCKING + + sizeof(struct circular_queue) +#endif + ) / 1024 + ); printk(" per task-struct memory footprint: %lu bytes\n", sizeof(struct held_lock) * MAX_LOCK_DEPTH); diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index 699a2ac3a0d7..a2ee95ad1313 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -91,6 +91,8 @@ extern unsigned int nr_process_chains; extern unsigned int max_lockdep_depth; extern unsigned int max_recursion_depth; +extern unsigned int max_bfs_queue_depth; + #ifdef CONFIG_PROVE_LOCKING extern unsigned long lockdep_count_forward_deps(struct lock_class *); extern unsigned long lockdep_count_backward_deps(struct lock_class *); diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index e94caa666dba..98cc79654750 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -411,6 +411,10 @@ static int lockdep_stats_show(struct seq_file *m, void *v) max_lockdep_depth); seq_printf(m, " max recursion depth: %11u\n", max_recursion_depth); +#ifdef CONFIG_PROVE_LOCKING + seq_printf(m, " max bfs queue depth: %11u\n", + max_bfs_queue_depth); +#endif lockdep_stats_debug_show(m); seq_printf(m, " debug_locks: %11u\n", debug_locks); diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c index 868102172aa4..bf8110b35c51 100644 --- a/kernel/perf_counter.c +++ b/kernel/perf_counter.c @@ -88,6 +88,7 @@ void __weak hw_perf_disable(void) { barrier(); } void __weak hw_perf_enable(void) { barrier(); } void __weak hw_perf_counter_setup(int cpu) { barrier(); } +void __weak hw_perf_counter_setup_online(int cpu) { barrier(); } int __weak hw_perf_group_sched_in(struct perf_counter *group_leader, @@ -2630,7 +2631,7 @@ static u32 perf_counter_tid(struct perf_counter *counter, struct task_struct *p) return task_pid_nr_ns(p, counter->ns); } -static void perf_counter_output(struct perf_counter *counter, int nmi, +void perf_counter_output(struct perf_counter *counter, int nmi, struct perf_sample_data *data) { int ret; @@ -2646,7 +2647,7 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, u64 counter; } group_entry; struct perf_callchain_entry *callchain = NULL; - struct perf_tracepoint_record *tp; + struct perf_raw_record *raw = NULL; int callchain_size = 0; u64 time; struct { @@ -2715,9 +2716,10 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, header.size += sizeof(u64); } - if (sample_type & PERF_SAMPLE_TP_RECORD) { - tp = data->private; - header.size += tp->size; + if (sample_type & PERF_SAMPLE_RAW) { + raw = data->raw; + if (raw) + header.size += raw->size; } ret = perf_output_begin(&handle, counter, header.size, nmi, 1); @@ -2783,8 +2785,8 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, } } - if (sample_type & PERF_SAMPLE_TP_RECORD) - perf_output_copy(&handle, tp->record, tp->size); + if ((sample_type & PERF_SAMPLE_RAW) && raw) + perf_output_copy(&handle, raw->data, raw->size); perf_output_end(&handle); } @@ -2849,7 +2851,8 @@ perf_counter_read_event(struct perf_counter *counter, */ struct perf_task_event { - struct task_struct *task; + struct task_struct *task; + struct perf_counter_context *task_ctx; struct { struct perf_event_header header; @@ -2909,24 +2912,23 @@ static void perf_counter_task_ctx(struct perf_counter_context *ctx, static void perf_counter_task_event(struct perf_task_event *task_event) { struct perf_cpu_context *cpuctx; - struct perf_counter_context *ctx; + struct perf_counter_context *ctx = task_event->task_ctx; cpuctx = &get_cpu_var(perf_cpu_context); perf_counter_task_ctx(&cpuctx->ctx, task_event); put_cpu_var(perf_cpu_context); rcu_read_lock(); - /* - * doesn't really matter which of the child contexts the - * events ends up in. - */ - ctx = rcu_dereference(current->perf_counter_ctxp); + if (!ctx) + ctx = rcu_dereference(task_event->task->perf_counter_ctxp); if (ctx) perf_counter_task_ctx(ctx, task_event); rcu_read_unlock(); } -static void perf_counter_task(struct task_struct *task, int new) +static void perf_counter_task(struct task_struct *task, + struct perf_counter_context *task_ctx, + int new) { struct perf_task_event task_event; @@ -2936,8 +2938,9 @@ static void perf_counter_task(struct task_struct *task, int new) return; task_event = (struct perf_task_event){ - .task = task, - .event = { + .task = task, + .task_ctx = task_ctx, + .event = { .header = { .type = new ? PERF_EVENT_FORK : PERF_EVENT_EXIT, .misc = 0, @@ -2955,7 +2958,7 @@ static void perf_counter_task(struct task_struct *task, int new) void perf_counter_fork(struct task_struct *task) { - perf_counter_task(task, 1); + perf_counter_task(task, NULL, 1); } /* @@ -3344,87 +3347,81 @@ int perf_counter_overflow(struct perf_counter *counter, int nmi, * Generic software counter infrastructure */ -static void perf_swcounter_update(struct perf_counter *counter) +/* + * We directly increment counter->count and keep a second value in + * counter->hw.period_left to count intervals. This period counter + * is kept in the range [-sample_period, 0] so that we can use the + * sign as trigger. + */ + +static u64 perf_swcounter_set_period(struct perf_counter *counter) { struct hw_perf_counter *hwc = &counter->hw; - u64 prev, now; - s64 delta; + u64 period = hwc->last_period; + u64 nr, offset; + s64 old, val; + + hwc->last_period = hwc->sample_period; again: - prev = atomic64_read(&hwc->prev_count); - now = atomic64_read(&hwc->count); - if (atomic64_cmpxchg(&hwc->prev_count, prev, now) != prev) - goto again; + old = val = atomic64_read(&hwc->period_left); + if (val < 0) + return 0; - delta = now - prev; + nr = div64_u64(period + val, period); + offset = nr * period; + val -= offset; + if (atomic64_cmpxchg(&hwc->period_left, old, val) != old) + goto again; - atomic64_add(delta, &counter->count); - atomic64_sub(delta, &hwc->period_left); + return nr; } -static void perf_swcounter_set_period(struct perf_counter *counter) +static void perf_swcounter_overflow(struct perf_counter *counter, + int nmi, struct perf_sample_data *data) { struct hw_perf_counter *hwc = &counter->hw; - s64 left = atomic64_read(&hwc->period_left); - s64 period = hwc->sample_period; + u64 overflow; - if (unlikely(left <= -period)) { - left = period; - atomic64_set(&hwc->period_left, left); - hwc->last_period = period; - } + data->period = counter->hw.last_period; + overflow = perf_swcounter_set_period(counter); - if (unlikely(left <= 0)) { - left += period; - atomic64_add(period, &hwc->period_left); - hwc->last_period = period; - } + if (hwc->interrupts == MAX_INTERRUPTS) + return; - atomic64_set(&hwc->prev_count, -left); - atomic64_set(&hwc->count, -left); + for (; overflow; overflow--) { + if (perf_counter_overflow(counter, nmi, data)) { + /* + * We inhibit the overflow from happening when + * hwc->interrupts == MAX_INTERRUPTS. + */ + break; + } + } } -static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer) +static void perf_swcounter_unthrottle(struct perf_counter *counter) { - enum hrtimer_restart ret = HRTIMER_RESTART; - struct perf_sample_data data; - struct perf_counter *counter; - u64 period; - - counter = container_of(hrtimer, struct perf_counter, hw.hrtimer); - counter->pmu->read(counter); - - data.addr = 0; - data.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. + * Nothing to do, we already reset hwc->interrupts. */ - if ((counter->attr.exclude_kernel || !data.regs) && - !counter->attr.exclude_user) - data.regs = task_pt_regs(current); +} - if (data.regs) { - if (perf_counter_overflow(counter, 0, &data)) - ret = HRTIMER_NORESTART; - } +static void perf_swcounter_add(struct perf_counter *counter, u64 nr, + int nmi, struct perf_sample_data *data) +{ + struct hw_perf_counter *hwc = &counter->hw; - period = max_t(u64, 10000, counter->hw.sample_period); - hrtimer_forward_now(hrtimer, ns_to_ktime(period)); + atomic64_add(nr, &counter->count); - return ret; -} + if (!hwc->sample_period) + return; -static void perf_swcounter_overflow(struct perf_counter *counter, - int nmi, struct perf_sample_data *data) -{ - data->period = counter->hw.last_period; + if (!data->regs) + return; - perf_swcounter_update(counter); - perf_swcounter_set_period(counter); - if (perf_counter_overflow(counter, nmi, data)) - /* soft-disable the counter */ - ; + if (!atomic64_add_negative(nr, &hwc->period_left)) + perf_swcounter_overflow(counter, nmi, data); } static int perf_swcounter_is_counting(struct perf_counter *counter) @@ -3488,15 +3485,6 @@ static int perf_swcounter_match(struct perf_counter *counter, return 1; } -static void perf_swcounter_add(struct perf_counter *counter, u64 nr, - int nmi, struct perf_sample_data *data) -{ - int neg = atomic64_add_negative(nr, &counter->hw.count); - - if (counter->hw.sample_period && !neg && data->regs) - perf_swcounter_overflow(counter, nmi, data); -} - static void perf_swcounter_ctx_event(struct perf_counter_context *ctx, enum perf_type_id type, u32 event, u64 nr, int nmi, @@ -3575,27 +3563,66 @@ void __perf_swcounter_event(u32 event, u64 nr, int nmi, static void perf_swcounter_read(struct perf_counter *counter) { - perf_swcounter_update(counter); } static int perf_swcounter_enable(struct perf_counter *counter) { - perf_swcounter_set_period(counter); + struct hw_perf_counter *hwc = &counter->hw; + + if (hwc->sample_period) { + hwc->last_period = hwc->sample_period; + perf_swcounter_set_period(counter); + } return 0; } static void perf_swcounter_disable(struct perf_counter *counter) { - perf_swcounter_update(counter); } static const struct pmu perf_ops_generic = { .enable = perf_swcounter_enable, .disable = perf_swcounter_disable, .read = perf_swcounter_read, + .unthrottle = perf_swcounter_unthrottle, }; /* + * hrtimer based swcounter callback + */ + +static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer) +{ + enum hrtimer_restart ret = HRTIMER_RESTART; + struct perf_sample_data data; + struct perf_counter *counter; + u64 period; + + counter = container_of(hrtimer, struct perf_counter, hw.hrtimer); + counter->pmu->read(counter); + + data.addr = 0; + data.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 ((counter->attr.exclude_kernel || !data.regs) && + !counter->attr.exclude_user) + data.regs = task_pt_regs(current); + + if (data.regs) { + if (perf_counter_overflow(counter, 0, &data)) + ret = HRTIMER_NORESTART; + } + + period = max_t(u64, 10000, counter->hw.sample_period); + hrtimer_forward_now(hrtimer, ns_to_ktime(period)); + + return ret; +} + +/* * Software counter: cpu wall time clock */ @@ -3715,15 +3742,15 @@ static const struct pmu perf_ops_task_clock = { void perf_tpcounter_event(int event_id, u64 addr, u64 count, void *record, int entry_size) { - struct perf_tracepoint_record tp = { + struct perf_raw_record raw = { .size = entry_size, - .record = record, + .data = record, }; struct perf_sample_data data = { .regs = get_irq_regs(), .addr = addr, - .private = &tp, + .raw = &raw, }; if (!data.regs) @@ -4285,7 +4312,7 @@ void perf_counter_exit_task(struct task_struct *child) unsigned long flags; if (likely(!child->perf_counter_ctxp)) { - perf_counter_task(child, 0); + perf_counter_task(child, NULL, 0); return; } @@ -4305,6 +4332,7 @@ void perf_counter_exit_task(struct task_struct *child) * incremented the context's refcount before we do put_ctx below. */ spin_lock(&child_ctx->lock); + child->perf_counter_ctxp = NULL; /* * If this context is a clone; unclone it so it can't get * swapped to another process while we're removing all @@ -4318,9 +4346,7 @@ void perf_counter_exit_task(struct task_struct *child) * won't get any samples after PERF_EVENT_EXIT. We can however still * get a few PERF_EVENT_READ events. */ - perf_counter_task(child, 0); - - child->perf_counter_ctxp = NULL; + perf_counter_task(child, child_ctx, 0); /* * We can recurse on the same lock type through: @@ -4541,6 +4567,11 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) perf_counter_init_cpu(cpu); break; + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + hw_perf_counter_setup_online(cpu); + break; + case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: perf_counter_exit_cpu(cpu); @@ -4565,6 +4596,8 @@ void __init perf_counter_init(void) { perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE, (void *)(long)smp_processor_id()); + perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE, + (void *)(long)smp_processor_id()); register_cpu_notifier(&perf_cpu_nb); } diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index e33a21cb9407..12161f74744e 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -14,11 +14,11 @@ */ void update_rlimit_cpu(unsigned long rlim_new) { - cputime_t cputime; + cputime_t cputime = secs_to_cputime(rlim_new); + struct signal_struct *const sig = current->signal; - cputime = secs_to_cputime(rlim_new); - if (cputime_eq(current->signal->it_prof_expires, cputime_zero) || - cputime_gt(current->signal->it_prof_expires, cputime)) { + 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); @@ -542,6 +542,17 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } +static inline int expires_gt(cputime_t expires, cputime_t new_exp) +{ + return cputime_eq(expires, cputime_zero) || + 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 @@ -586,34 +597,32 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) */ 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 (cputime_eq(p->cputime_expires.prof_exp, - cputime_zero) || - cputime_gt(p->cputime_expires.prof_exp, - nt->expires.cpu)) - p->cputime_expires.prof_exp = - nt->expires.cpu; + if (expires_gt(p->cputime_expires.prof_exp, + exp->cpu)) + p->cputime_expires.prof_exp = exp->cpu; break; case CPUCLOCK_VIRT: - if (cputime_eq(p->cputime_expires.virt_exp, - cputime_zero) || - cputime_gt(p->cputime_expires.virt_exp, - nt->expires.cpu)) - p->cputime_expires.virt_exp = - nt->expires.cpu; + 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 > - nt->expires.sched) + p->cputime_expires.sched_exp > exp->sched) p->cputime_expires.sched_exp = - nt->expires.sched; + 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. */ @@ -621,30 +630,23 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) default: BUG(); case CPUCLOCK_VIRT: - if (!cputime_eq(p->signal->it_virt_expires, - cputime_zero) && - cputime_lt(p->signal->it_virt_expires, - timer->it.cpu.expires.cpu)) + if (expires_le(sig->it[CPUCLOCK_VIRT].expires, + exp->cpu)) break; - p->signal->cputime_expires.virt_exp = - timer->it.cpu.expires.cpu; + sig->cputime_expires.virt_exp = exp->cpu; break; case CPUCLOCK_PROF: - if (!cputime_eq(p->signal->it_prof_expires, - cputime_zero) && - cputime_lt(p->signal->it_prof_expires, - timer->it.cpu.expires.cpu)) + if (expires_le(sig->it[CPUCLOCK_PROF].expires, + exp->cpu)) break; - i = p->signal->rlim[RLIMIT_CPU].rlim_cur; + i = sig->rlim[RLIMIT_CPU].rlim_cur; if (i != RLIM_INFINITY && - i <= cputime_to_secs(timer->it.cpu.expires.cpu)) + i <= cputime_to_secs(exp->cpu)) break; - p->signal->cputime_expires.prof_exp = - timer->it.cpu.expires.cpu; + sig->cputime_expires.prof_exp = exp->cpu; break; case CPUCLOCK_SCHED: - p->signal->cputime_expires.sched_exp = - timer->it.cpu.expires.sched; + sig->cputime_expires.sched_exp = exp->sched; break; } } @@ -1071,6 +1073,36 @@ static void stop_process_timers(struct task_struct *tsk) spin_unlock_irqrestore(&cputimer->lock, flags); } +static u32 onecputick; + +static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it, + cputime_t *expires, cputime_t cur_time, int signo) +{ + if (cputime_eq(it->expires, cputime_zero)) + return; + + if (cputime_ge(cur_time, it->expires)) { + if (!cputime_eq(it->incr, cputime_zero)) { + it->expires = cputime_add(it->expires, it->incr); + it->error += it->incr_error; + if (it->error >= onecputick) { + it->expires = cputime_sub(it->expires, + cputime_one_jiffy); + it->error -= onecputick; + } + } else + it->expires = cputime_zero; + + __group_send_sig_info(signo, SEND_SIG_PRIV, tsk); + } + + if (!cputime_eq(it->expires, cputime_zero) && + (cputime_eq(*expires, cputime_zero) || + cputime_lt(it->expires, *expires))) { + *expires = it->expires; + } +} + /* * Check for any per-thread CPU timers that have fired and move them * off the tsk->*_timers list onto the firing list. Per-thread timers @@ -1090,10 +1122,10 @@ static void check_process_timers(struct task_struct *tsk, * Don't sample the current process CPU clocks if there are no timers. */ if (list_empty(&timers[CPUCLOCK_PROF]) && - cputime_eq(sig->it_prof_expires, cputime_zero) && + cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) && sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY && list_empty(&timers[CPUCLOCK_VIRT]) && - cputime_eq(sig->it_virt_expires, cputime_zero) && + cputime_eq(sig->it[CPUCLOCK_VIRT].expires, cputime_zero) && list_empty(&timers[CPUCLOCK_SCHED])) { stop_process_timers(tsk); return; @@ -1153,38 +1185,11 @@ static void check_process_timers(struct task_struct *tsk, /* * Check for the special case process timers. */ - if (!cputime_eq(sig->it_prof_expires, cputime_zero)) { - if (cputime_ge(ptime, sig->it_prof_expires)) { - /* ITIMER_PROF fires and reloads. */ - sig->it_prof_expires = sig->it_prof_incr; - if (!cputime_eq(sig->it_prof_expires, cputime_zero)) { - sig->it_prof_expires = cputime_add( - sig->it_prof_expires, ptime); - } - __group_send_sig_info(SIGPROF, SEND_SIG_PRIV, tsk); - } - if (!cputime_eq(sig->it_prof_expires, cputime_zero) && - (cputime_eq(prof_expires, cputime_zero) || - cputime_lt(sig->it_prof_expires, prof_expires))) { - prof_expires = sig->it_prof_expires; - } - } - if (!cputime_eq(sig->it_virt_expires, cputime_zero)) { - if (cputime_ge(utime, sig->it_virt_expires)) { - /* ITIMER_VIRTUAL fires and reloads. */ - sig->it_virt_expires = sig->it_virt_incr; - if (!cputime_eq(sig->it_virt_expires, cputime_zero)) { - sig->it_virt_expires = cputime_add( - sig->it_virt_expires, utime); - } - __group_send_sig_info(SIGVTALRM, SEND_SIG_PRIV, tsk); - } - if (!cputime_eq(sig->it_virt_expires, cputime_zero) && - (cputime_eq(virt_expires, cputime_zero) || - cputime_lt(sig->it_virt_expires, virt_expires))) { - virt_expires = sig->it_virt_expires; - } - } + check_cpu_itimer(tsk, &sig->it[CPUCLOCK_PROF], &prof_expires, ptime, + SIGPROF); + check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime, + SIGVTALRM); + if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { unsigned long psecs = cputime_to_secs(ptime); cputime_t x; @@ -1457,7 +1462,7 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, if (!cputime_eq(*oldval, cputime_zero)) { if (cputime_le(*oldval, now.cpu)) { /* Just about to fire. */ - *oldval = jiffies_to_cputime(1); + *oldval = cputime_one_jiffy; } else { *oldval = cputime_sub(*oldval, now.cpu); } @@ -1703,10 +1708,15 @@ static __init int init_posix_cpu_timers(void) .nsleep = thread_cpu_nsleep, .nsleep_restart = thread_cpu_nsleep_restart, }; + struct timespec ts; register_posix_clock(CLOCK_PROCESS_CPUTIME_ID, &process); register_posix_clock(CLOCK_THREAD_CPUTIME_ID, &thread); + cputime_to_timespec(cputime_one_jiffy, &ts); + onecputick = ts.tv_nsec; + WARN_ON(ts.tv_sec != 0); + return 0; } __initcall(init_posix_cpu_timers); diff --git a/kernel/printk.c b/kernel/printk.c index b4d97b54c1ec..e10d193a833a 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -37,6 +37,12 @@ #include <asm/uaccess.h> /* + * for_each_console() allows you to iterate on each console + */ +#define for_each_console(con) \ + for (con = console_drivers; con != NULL; con = con->next) + +/* * Architectures can override it: */ void asmlinkage __attribute__((weak)) early_printk(const char *fmt, ...) @@ -61,6 +67,8 @@ int console_printk[4] = { DEFAULT_CONSOLE_LOGLEVEL, /* default_console_loglevel */ }; +static int saved_console_loglevel = -1; + /* * Low level drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it. @@ -372,10 +380,15 @@ int do_syslog(int type, char __user *buf, int len) logged_chars = 0; break; case 6: /* Disable logging to console */ + if (saved_console_loglevel == -1) + saved_console_loglevel = console_loglevel; console_loglevel = minimum_console_loglevel; break; case 7: /* Enable logging to console */ - console_loglevel = default_console_loglevel; + if (saved_console_loglevel != -1) { + console_loglevel = saved_console_loglevel; + saved_console_loglevel = -1; + } break; case 8: /* Set level of messages printed to console */ error = -EINVAL; @@ -384,6 +397,8 @@ int do_syslog(int type, char __user *buf, int len) if (len < minimum_console_loglevel) len = minimum_console_loglevel; console_loglevel = len; + /* Implicitly re-enable logging to console */ + saved_console_loglevel = -1; error = 0; break; case 9: /* Number of chars in the log buffer */ @@ -412,7 +427,7 @@ static void __call_console_drivers(unsigned start, unsigned end) { struct console *con; - for (con = console_drivers; con; con = con->next) { + for_each_console(con) { if ((con->flags & CON_ENABLED) && con->write && (cpu_online(smp_processor_id()) || (con->flags & CON_ANYTIME))) @@ -544,7 +559,7 @@ static int have_callable_console(void) { struct console *con; - for (con = console_drivers; con; con = con->next) + for_each_console(con) if (con->flags & CON_ANYTIME) return 1; @@ -1082,7 +1097,7 @@ void console_unblank(void) console_locked = 1; console_may_schedule = 0; - for (c = console_drivers; c != NULL; c = c->next) + for_each_console(c) if ((c->flags & CON_ENABLED) && c->unblank) c->unblank(); release_console_sem(); @@ -1097,7 +1112,7 @@ struct tty_driver *console_device(int *index) struct tty_driver *driver = NULL; acquire_console_sem(); - for (c = console_drivers; c != NULL; c = c->next) { + for_each_console(c) { if (!c->device) continue; driver = c->device(c, index); @@ -1134,25 +1149,49 @@ EXPORT_SYMBOL(console_start); * to register the console printing procedure with printk() and to * print any messages that were printed by the kernel before the * console driver was initialized. + * + * This can happen pretty early during the boot process (because of + * early_printk) - sometimes before setup_arch() completes - be careful + * of what kernel features are used - they may not be initialised yet. + * + * There are two types of consoles - bootconsoles (early_printk) and + * "real" consoles (everything which is not a bootconsole) which are + * handled differently. + * - Any number of bootconsoles can be registered at any time. + * - As soon as a "real" console is registered, all bootconsoles + * will be unregistered automatically. + * - Once a "real" console is registered, any attempt to register a + * bootconsoles will be rejected */ -void register_console(struct console *console) +void register_console(struct console *newcon) { int i; unsigned long flags; - struct console *bootconsole = NULL; + struct console *bcon = NULL; - if (console_drivers) { - if (console->flags & CON_BOOT) - return; - if (console_drivers->flags & CON_BOOT) - bootconsole = console_drivers; + /* + * before we register a new CON_BOOT console, make sure we don't + * already have a valid console + */ + if (console_drivers && newcon->flags & CON_BOOT) { + /* find the last or real console */ + for_each_console(bcon) { + if (!(bcon->flags & CON_BOOT)) { + printk(KERN_INFO "Too late to register bootconsole %s%d\n", + newcon->name, newcon->index); + return; + } + } } - if (preferred_console < 0 || bootconsole || !console_drivers) + if (console_drivers && console_drivers->flags & CON_BOOT) + bcon = console_drivers; + + if (preferred_console < 0 || bcon || !console_drivers) preferred_console = selected_console; - if (console->early_setup) - console->early_setup(); + if (newcon->early_setup) + newcon->early_setup(); /* * See if we want to use this console driver. If we @@ -1160,13 +1199,13 @@ void register_console(struct console *console) * that registers here. */ if (preferred_console < 0) { - if (console->index < 0) - console->index = 0; - if (console->setup == NULL || - console->setup(console, NULL) == 0) { - console->flags |= CON_ENABLED; - if (console->device) { - console->flags |= CON_CONSDEV; + if (newcon->index < 0) + newcon->index = 0; + if (newcon->setup == NULL || + newcon->setup(newcon, NULL) == 0) { + newcon->flags |= CON_ENABLED; + if (newcon->device) { + newcon->flags |= CON_CONSDEV; preferred_console = 0; } } @@ -1178,64 +1217,62 @@ void register_console(struct console *console) */ for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++) { - if (strcmp(console_cmdline[i].name, console->name) != 0) + if (strcmp(console_cmdline[i].name, newcon->name) != 0) continue; - if (console->index >= 0 && - console->index != console_cmdline[i].index) + if (newcon->index >= 0 && + newcon->index != console_cmdline[i].index) continue; - if (console->index < 0) - console->index = console_cmdline[i].index; + if (newcon->index < 0) + newcon->index = console_cmdline[i].index; #ifdef CONFIG_A11Y_BRAILLE_CONSOLE if (console_cmdline[i].brl_options) { - console->flags |= CON_BRL; - braille_register_console(console, + newcon->flags |= CON_BRL; + braille_register_console(newcon, console_cmdline[i].index, console_cmdline[i].options, console_cmdline[i].brl_options); return; } #endif - if (console->setup && - console->setup(console, console_cmdline[i].options) != 0) + if (newcon->setup && + newcon->setup(newcon, console_cmdline[i].options) != 0) break; - console->flags |= CON_ENABLED; - console->index = console_cmdline[i].index; + newcon->flags |= CON_ENABLED; + newcon->index = console_cmdline[i].index; if (i == selected_console) { - console->flags |= CON_CONSDEV; + newcon->flags |= CON_CONSDEV; preferred_console = selected_console; } break; } - if (!(console->flags & CON_ENABLED)) + if (!(newcon->flags & CON_ENABLED)) return; - if (bootconsole && (console->flags & CON_CONSDEV)) { - printk(KERN_INFO "console handover: boot [%s%d] -> real [%s%d]\n", - bootconsole->name, bootconsole->index, - console->name, console->index); - unregister_console(bootconsole); - console->flags &= ~CON_PRINTBUFFER; - } else { - printk(KERN_INFO "console [%s%d] enabled\n", - console->name, console->index); - } + /* + * If we have a bootconsole, and are switching to a real console, + * don't print everything out again, since when the boot console, and + * the real console are the same physical device, it's annoying to + * see the beginning boot messages twice + */ + if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) + newcon->flags &= ~CON_PRINTBUFFER; /* * Put this console in the list - keep the * preferred driver at the head of the list. */ acquire_console_sem(); - if ((console->flags & CON_CONSDEV) || console_drivers == NULL) { - console->next = console_drivers; - console_drivers = console; - if (console->next) - console->next->flags &= ~CON_CONSDEV; + if ((newcon->flags & CON_CONSDEV) || console_drivers == NULL) { + newcon->next = console_drivers; + console_drivers = newcon; + if (newcon->next) + newcon->next->flags &= ~CON_CONSDEV; } else { - console->next = console_drivers->next; - console_drivers->next = console; + newcon->next = console_drivers->next; + console_drivers->next = newcon; } - if (console->flags & CON_PRINTBUFFER) { + if (newcon->flags & CON_PRINTBUFFER) { /* * release_console_sem() will print out the buffered messages * for us. @@ -1245,6 +1282,28 @@ void register_console(struct console *console) spin_unlock_irqrestore(&logbuf_lock, flags); } release_console_sem(); + + /* + * By unregistering the bootconsoles after we enable the real console + * we get the "console xxx enabled" message on all the consoles - + * boot consoles, real consoles, etc - this is to ensure that end + * users know there might be something in the kernel's log buffer that + * went to the bootconsole (that they do not see on the real console) + */ + if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) { + /* we need to iterate through twice, to make sure we print + * everything out, before we unregister the console(s) + */ + printk(KERN_INFO "console [%s%d] enabled, bootconsole disabled\n", + newcon->name, newcon->index); + for_each_console(bcon) + if (bcon->flags & CON_BOOT) + unregister_console(bcon); + } else { + printk(KERN_INFO "%sconsole [%s%d] enabled\n", + (newcon->flags & CON_BOOT) ? "boot" : "" , + newcon->name, newcon->index); + } } EXPORT_SYMBOL(register_console); @@ -1287,11 +1346,13 @@ EXPORT_SYMBOL(unregister_console); static int __init disable_boot_consoles(void) { - if (console_drivers != NULL) { - if (console_drivers->flags & CON_BOOT) { + struct console *con; + + for_each_console(con) { + if (con->flags & CON_BOOT) { printk(KERN_INFO "turn off boot console %s%d\n", - console_drivers->name, console_drivers->index); - return unregister_console(console_drivers); + con->name, con->index); + unregister_console(con); } } return 0; diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c deleted file mode 100644 index 0f2b0b311304..000000000000 --- a/kernel/rcuclassic.c +++ /dev/null @@ -1,807 +0,0 @@ -/* - * Read-Copy Update mechanism for mutual exclusion - * - * 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 IBM Corporation, 2001 - * - * Authors: Dipankar Sarma <dipankar@in.ibm.com> - * Manfred Spraul <manfred@colorfullife.com> - * - * Based on the original work by Paul McKenney <paulmck@us.ibm.com> - * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. - * Papers: - * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf - * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001) - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/rcupdate.h> -#include <linux/interrupt.h> -#include <linux/sched.h> -#include <asm/atomic.h> -#include <linux/bitops.h> -#include <linux/module.h> -#include <linux/completion.h> -#include <linux/moduleparam.h> -#include <linux/percpu.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/mutex.h> -#include <linux/time.h> - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -static struct lock_class_key rcu_lock_key; -struct lockdep_map rcu_lock_map = - STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); -EXPORT_SYMBOL_GPL(rcu_lock_map); -#endif - - -/* Definition for rcupdate control block. */ -static struct rcu_ctrlblk rcu_ctrlblk = { - .cur = -300, - .completed = -300, - .pending = -300, - .lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock), - .cpumask = CPU_BITS_NONE, -}; - -static struct rcu_ctrlblk rcu_bh_ctrlblk = { - .cur = -300, - .completed = -300, - .pending = -300, - .lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock), - .cpumask = CPU_BITS_NONE, -}; - -static DEFINE_PER_CPU(struct rcu_data, rcu_data); -static DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); - -/* - * Increment the quiescent state counter. - * The counter is a bit degenerated: We do not need to know - * how many quiescent states passed, just if there was at least - * one since the start of the grace period. Thus just a flag. - */ -void rcu_qsctr_inc(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_data, cpu); - rdp->passed_quiesc = 1; -} - -void rcu_bh_qsctr_inc(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu); - rdp->passed_quiesc = 1; -} - -static int blimit = 10; -static int qhimark = 10000; -static int qlowmark = 100; - -#ifdef CONFIG_SMP -static void force_quiescent_state(struct rcu_data *rdp, - struct rcu_ctrlblk *rcp) -{ - int cpu; - unsigned long flags; - - set_need_resched(); - spin_lock_irqsave(&rcp->lock, flags); - if (unlikely(!rcp->signaled)) { - rcp->signaled = 1; - /* - * Don't send IPI to itself. With irqs disabled, - * rdp->cpu is the current cpu. - * - * cpu_online_mask is updated by the _cpu_down() - * using __stop_machine(). Since we're in irqs disabled - * section, __stop_machine() is not exectuting, hence - * the cpu_online_mask is stable. - * - * However, a cpu might have been offlined _just_ before - * we disabled irqs while entering here. - * And rcu subsystem might not yet have handled the CPU_DEAD - * notification, leading to the offlined cpu's bit - * being set in the rcp->cpumask. - * - * Hence cpumask = (rcp->cpumask & cpu_online_mask) to prevent - * sending smp_reschedule() to an offlined CPU. - */ - for_each_cpu_and(cpu, - to_cpumask(rcp->cpumask), cpu_online_mask) { - if (cpu != rdp->cpu) - smp_send_reschedule(cpu); - } - } - spin_unlock_irqrestore(&rcp->lock, flags); -} -#else -static inline void force_quiescent_state(struct rcu_data *rdp, - struct rcu_ctrlblk *rcp) -{ - set_need_resched(); -} -#endif - -static void __call_rcu(struct rcu_head *head, struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - long batch; - - head->next = NULL; - smp_mb(); /* Read of rcu->cur must happen after any change by caller. */ - - /* - * Determine the batch number of this callback. - * - * Using ACCESS_ONCE to avoid the following error when gcc eliminates - * local variable "batch" and emits codes like this: - * 1) rdp->batch = rcp->cur + 1 # gets old value - * ...... - * 2)rcu_batch_after(rcp->cur + 1, rdp->batch) # gets new value - * then [*nxttail[0], *nxttail[1]) may contain callbacks - * that batch# = rdp->batch, see the comment of struct rcu_data. - */ - batch = ACCESS_ONCE(rcp->cur) + 1; - - if (rdp->nxtlist && rcu_batch_after(batch, rdp->batch)) { - /* process callbacks */ - rdp->nxttail[0] = rdp->nxttail[1]; - rdp->nxttail[1] = rdp->nxttail[2]; - if (rcu_batch_after(batch - 1, rdp->batch)) - rdp->nxttail[0] = rdp->nxttail[2]; - } - - rdp->batch = batch; - *rdp->nxttail[2] = head; - rdp->nxttail[2] = &head->next; - - if (unlikely(++rdp->qlen > qhimark)) { - rdp->blimit = INT_MAX; - force_quiescent_state(rdp, &rcu_ctrlblk); - } -} - -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR - -static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) -{ - rcp->gp_start = jiffies; - rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_CHECK; -} - -static void print_other_cpu_stall(struct rcu_ctrlblk *rcp) -{ - int cpu; - long delta; - unsigned long flags; - - /* Only let one CPU complain about others per time interval. */ - - spin_lock_irqsave(&rcp->lock, flags); - delta = jiffies - rcp->jiffies_stall; - if (delta < 2 || rcp->cur != rcp->completed) { - spin_unlock_irqrestore(&rcp->lock, flags); - return; - } - rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; - spin_unlock_irqrestore(&rcp->lock, flags); - - /* OK, time to rat on our buddy... */ - - printk(KERN_ERR "INFO: RCU detected CPU stalls:"); - for_each_possible_cpu(cpu) { - if (cpumask_test_cpu(cpu, to_cpumask(rcp->cpumask))) - printk(" %d", cpu); - } - printk(" (detected by %d, t=%ld jiffies)\n", - smp_processor_id(), (long)(jiffies - rcp->gp_start)); -} - -static void print_cpu_stall(struct rcu_ctrlblk *rcp) -{ - unsigned long flags; - - printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu/%lu jiffies)\n", - smp_processor_id(), jiffies, - jiffies - rcp->gp_start); - dump_stack(); - spin_lock_irqsave(&rcp->lock, flags); - if ((long)(jiffies - rcp->jiffies_stall) >= 0) - rcp->jiffies_stall = - jiffies + RCU_SECONDS_TILL_STALL_RECHECK; - spin_unlock_irqrestore(&rcp->lock, flags); - set_need_resched(); /* kick ourselves to get things going. */ -} - -static void check_cpu_stall(struct rcu_ctrlblk *rcp) -{ - long delta; - - delta = jiffies - rcp->jiffies_stall; - if (cpumask_test_cpu(smp_processor_id(), to_cpumask(rcp->cpumask)) && - delta >= 0) { - - /* We haven't checked in, so go dump stack. */ - print_cpu_stall(rcp); - - } else if (rcp->cur != rcp->completed && delta >= 2) { - - /* They had two seconds to dump stack, so complain. */ - print_other_cpu_stall(rcp); - } -} - -#else /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - -static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) -{ -} - -static inline void check_cpu_stall(struct rcu_ctrlblk *rcp) -{ -} - -#endif /* #else #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - -/** - * call_rcu - Queue an RCU callback for invocation after a grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual update function to be invoked after the grace period - * - * The update function will be invoked some time after a full grace - * period elapses, in other words after all currently executing RCU - * read-side critical sections have completed. RCU read-side critical - * sections are delimited by rcu_read_lock() and rcu_read_unlock(), - * and may be nested. - */ -void call_rcu(struct rcu_head *head, - void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - - head->func = func; - local_irq_save(flags); - __call_rcu(head, &rcu_ctrlblk, &__get_cpu_var(rcu_data)); - local_irq_restore(flags); -} -EXPORT_SYMBOL_GPL(call_rcu); - -/** - * call_rcu_bh - Queue an RCU for invocation after a quicker grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual update function to be invoked after the grace period - * - * The update function will be invoked some time after a full grace - * period elapses, in other words after all currently executing RCU - * read-side critical sections have completed. call_rcu_bh() assumes - * that the read-side critical sections end on completion of a softirq - * handler. This means that read-side critical sections in process - * context must not be interrupted by softirqs. This interface is to be - * used when most of the read-side critical sections are in softirq context. - * RCU read-side critical sections are delimited by rcu_read_lock() and - * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh() - * and rcu_read_unlock_bh(), if in process context. These may be nested. - */ -void call_rcu_bh(struct rcu_head *head, - void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - - head->func = func; - local_irq_save(flags); - __call_rcu(head, &rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); - local_irq_restore(flags); -} -EXPORT_SYMBOL_GPL(call_rcu_bh); - -/* - * Return the number of RCU batches processed thus far. Useful - * for debug and statistics. - */ -long rcu_batches_completed(void) -{ - return rcu_ctrlblk.completed; -} -EXPORT_SYMBOL_GPL(rcu_batches_completed); - -/* - * Return the number of RCU batches processed thus far. Useful - * for debug and statistics. - */ -long rcu_batches_completed_bh(void) -{ - return rcu_bh_ctrlblk.completed; -} -EXPORT_SYMBOL_GPL(rcu_batches_completed_bh); - -/* Raises the softirq for processing rcu_callbacks. */ -static inline void raise_rcu_softirq(void) -{ - raise_softirq(RCU_SOFTIRQ); -} - -/* - * Invoke the completed RCU callbacks. They are expected to be in - * a per-cpu list. - */ -static void rcu_do_batch(struct rcu_data *rdp) -{ - unsigned long flags; - struct rcu_head *next, *list; - int count = 0; - - list = rdp->donelist; - while (list) { - next = list->next; - prefetch(next); - list->func(list); - list = next; - if (++count >= rdp->blimit) - break; - } - rdp->donelist = list; - - local_irq_save(flags); - rdp->qlen -= count; - local_irq_restore(flags); - if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark) - rdp->blimit = blimit; - - if (!rdp->donelist) - rdp->donetail = &rdp->donelist; - else - raise_rcu_softirq(); -} - -/* - * Grace period handling: - * The grace period handling consists out of two steps: - * - A new grace period is started. - * This is done by rcu_start_batch. The start is not broadcasted to - * all cpus, they must pick this up by comparing rcp->cur with - * rdp->quiescbatch. All cpus are recorded in the - * rcu_ctrlblk.cpumask bitmap. - * - All cpus must go through a quiescent state. - * Since the start of the grace period is not broadcasted, at least two - * calls to rcu_check_quiescent_state are required: - * The first call just notices that a new grace period is running. The - * following calls check if there was a quiescent state since the beginning - * of the grace period. If so, it updates rcu_ctrlblk.cpumask. If - * the bitmap is empty, then the grace period is completed. - * rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace - * period (if necessary). - */ - -/* - * Register a new batch of callbacks, and start it up if there is currently no - * active batch and the batch to be registered has not already occurred. - * Caller must hold rcu_ctrlblk.lock. - */ -static void rcu_start_batch(struct rcu_ctrlblk *rcp) -{ - if (rcp->cur != rcp->pending && - rcp->completed == rcp->cur) { - rcp->cur++; - record_gp_stall_check_time(rcp); - - /* - * Accessing nohz_cpu_mask before incrementing rcp->cur needs a - * Barrier Otherwise it can cause tickless idle CPUs to be - * included in rcp->cpumask, which will extend graceperiods - * unnecessarily. - */ - smp_mb(); - cpumask_andnot(to_cpumask(rcp->cpumask), - cpu_online_mask, nohz_cpu_mask); - - rcp->signaled = 0; - } -} - -/* - * cpu went through a quiescent state since the beginning of the grace period. - * Clear it from the cpu mask and complete the grace period if it was the last - * cpu. Start another grace period if someone has further entries pending - */ -static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp) -{ - cpumask_clear_cpu(cpu, to_cpumask(rcp->cpumask)); - if (cpumask_empty(to_cpumask(rcp->cpumask))) { - /* batch completed ! */ - rcp->completed = rcp->cur; - rcu_start_batch(rcp); - } -} - -/* - * Check if the cpu has gone through a quiescent state (say context - * switch). If so and if it already hasn't done so in this RCU - * quiescent cycle, then indicate that it has done so. - */ -static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - - if (rdp->quiescbatch != rcp->cur) { - /* start new grace period: */ - rdp->qs_pending = 1; - rdp->passed_quiesc = 0; - rdp->quiescbatch = rcp->cur; - return; - } - - /* Grace period already completed for this cpu? - * qs_pending is checked instead of the actual bitmap to avoid - * cacheline trashing. - */ - if (!rdp->qs_pending) - return; - - /* - * Was there a quiescent state since the beginning of the grace - * period? If no, then exit and wait for the next call. - */ - if (!rdp->passed_quiesc) - return; - rdp->qs_pending = 0; - - spin_lock_irqsave(&rcp->lock, flags); - /* - * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync - * during cpu startup. Ignore the quiescent state. - */ - if (likely(rdp->quiescbatch == rcp->cur)) - cpu_quiet(rdp->cpu, rcp); - - spin_unlock_irqrestore(&rcp->lock, flags); -} - - -#ifdef CONFIG_HOTPLUG_CPU - -/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing - * locking requirements, the list it's pulling from has to belong to a cpu - * which is dead and hence not processing interrupts. - */ -static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list, - struct rcu_head **tail, long batch) -{ - unsigned long flags; - - if (list) { - local_irq_save(flags); - this_rdp->batch = batch; - *this_rdp->nxttail[2] = list; - this_rdp->nxttail[2] = tail; - local_irq_restore(flags); - } -} - -static void __rcu_offline_cpu(struct rcu_data *this_rdp, - struct rcu_ctrlblk *rcp, struct rcu_data *rdp) -{ - unsigned long flags; - - /* - * if the cpu going offline owns the grace period - * we can block indefinitely waiting for it, so flush - * it here - */ - spin_lock_irqsave(&rcp->lock, flags); - if (rcp->cur != rcp->completed) - cpu_quiet(rdp->cpu, rcp); - rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail, rcp->cur + 1); - rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail[2], rcp->cur + 1); - spin_unlock(&rcp->lock); - - this_rdp->qlen += rdp->qlen; - local_irq_restore(flags); -} - -static void rcu_offline_cpu(int cpu) -{ - struct rcu_data *this_rdp = &get_cpu_var(rcu_data); - struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data); - - __rcu_offline_cpu(this_rdp, &rcu_ctrlblk, - &per_cpu(rcu_data, cpu)); - __rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk, - &per_cpu(rcu_bh_data, cpu)); - put_cpu_var(rcu_data); - put_cpu_var(rcu_bh_data); -} - -#else - -static void rcu_offline_cpu(int cpu) -{ -} - -#endif - -/* - * This does the RCU processing work from softirq context. - */ -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - long completed_snap; - - if (rdp->nxtlist) { - local_irq_save(flags); - completed_snap = ACCESS_ONCE(rcp->completed); - - /* - * move the other grace-period-completed entries to - * [rdp->nxtlist, *rdp->nxttail[0]) temporarily - */ - if (!rcu_batch_before(completed_snap, rdp->batch)) - rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2]; - else if (!rcu_batch_before(completed_snap, rdp->batch - 1)) - rdp->nxttail[0] = rdp->nxttail[1]; - - /* - * the grace period for entries in - * [rdp->nxtlist, *rdp->nxttail[0]) has completed and - * move these entries to donelist - */ - if (rdp->nxttail[0] != &rdp->nxtlist) { - *rdp->donetail = rdp->nxtlist; - rdp->donetail = rdp->nxttail[0]; - rdp->nxtlist = *rdp->nxttail[0]; - *rdp->donetail = NULL; - - if (rdp->nxttail[1] == rdp->nxttail[0]) - rdp->nxttail[1] = &rdp->nxtlist; - if (rdp->nxttail[2] == rdp->nxttail[0]) - rdp->nxttail[2] = &rdp->nxtlist; - rdp->nxttail[0] = &rdp->nxtlist; - } - - local_irq_restore(flags); - - if (rcu_batch_after(rdp->batch, rcp->pending)) { - unsigned long flags2; - - /* and start it/schedule start if it's a new batch */ - spin_lock_irqsave(&rcp->lock, flags2); - if (rcu_batch_after(rdp->batch, rcp->pending)) { - rcp->pending = rdp->batch; - rcu_start_batch(rcp); - } - spin_unlock_irqrestore(&rcp->lock, flags2); - } - } - - rcu_check_quiescent_state(rcp, rdp); - if (rdp->donelist) - rcu_do_batch(rdp); -} - -static void rcu_process_callbacks(struct softirq_action *unused) -{ - /* - * Memory references from any prior RCU read-side critical sections - * executed by the interrupted code must be see before any RCU - * grace-period manupulations below. - */ - - smp_mb(); /* See above block comment. */ - - __rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data)); - __rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); - - /* - * Memory references from any later RCU read-side critical sections - * executed by the interrupted code must be see after any RCU - * grace-period manupulations above. - */ - - smp_mb(); /* See above block comment. */ -} - -static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp) -{ - /* Check for CPU stalls, if enabled. */ - check_cpu_stall(rcp); - - if (rdp->nxtlist) { - long completed_snap = ACCESS_ONCE(rcp->completed); - - /* - * This cpu has pending rcu entries and the grace period - * for them has completed. - */ - if (!rcu_batch_before(completed_snap, rdp->batch)) - return 1; - if (!rcu_batch_before(completed_snap, rdp->batch - 1) && - rdp->nxttail[0] != rdp->nxttail[1]) - return 1; - if (rdp->nxttail[0] != &rdp->nxtlist) - return 1; - - /* - * This cpu has pending rcu entries and the new batch - * for then hasn't been started nor scheduled start - */ - if (rcu_batch_after(rdp->batch, rcp->pending)) - return 1; - } - - /* This cpu has finished callbacks to invoke */ - if (rdp->donelist) - return 1; - - /* The rcu core waits for a quiescent state from the cpu */ - if (rdp->quiescbatch != rcp->cur || rdp->qs_pending) - return 1; - - /* nothing to do */ - return 0; -} - -/* - * Check to see if there is any immediate RCU-related work to be done - * by the current CPU, returning 1 if so. This function is part of the - * RCU implementation; it is -not- an exported member of the RCU API. - */ -int rcu_pending(int cpu) -{ - return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) || - __rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu)); -} - -/* - * Check to see if any future RCU-related work will need to be done - * by the current CPU, even if none need be done immediately, returning - * 1 if so. This function is part of the RCU implementation; it is -not- - * an exported member of the RCU API. - */ -int rcu_needs_cpu(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_data, cpu); - struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu); - - return !!rdp->nxtlist || !!rdp_bh->nxtlist || rcu_pending(cpu); -} - -/* - * Top-level function driving RCU grace-period detection, normally - * invoked from the scheduler-clock interrupt. This function simply - * increments counters that are read only from softirq by this same - * CPU, so there are no memory barriers required. - */ -void rcu_check_callbacks(int cpu, int user) -{ - if (user || - (idle_cpu(cpu) && rcu_scheduler_active && - !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { - - /* - * Get here if this CPU took its interrupt from user - * mode or from the idle loop, and if this is not a - * nested interrupt. In this case, the CPU is in - * a quiescent state, so count it. - * - * Also do a memory barrier. This is needed to handle - * the case where writes from a preempt-disable section - * of code get reordered into schedule() by this CPU's - * write buffer. The memory barrier makes sure that - * the rcu_qsctr_inc() and rcu_bh_qsctr_inc() are see - * by other CPUs to happen after any such write. - */ - - smp_mb(); /* See above block comment. */ - rcu_qsctr_inc(cpu); - rcu_bh_qsctr_inc(cpu); - - } else if (!in_softirq()) { - - /* - * Get here if this CPU did not take its interrupt from - * softirq, in other words, if it is not interrupting - * a rcu_bh read-side critical section. This is an _bh - * critical section, so count it. The memory barrier - * is needed for the same reason as is the above one. - */ - - smp_mb(); /* See above block comment. */ - rcu_bh_qsctr_inc(cpu); - } - raise_rcu_softirq(); -} - -static void __cpuinit rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - - spin_lock_irqsave(&rcp->lock, flags); - memset(rdp, 0, sizeof(*rdp)); - rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2] = &rdp->nxtlist; - rdp->donetail = &rdp->donelist; - rdp->quiescbatch = rcp->completed; - rdp->qs_pending = 0; - rdp->cpu = cpu; - rdp->blimit = blimit; - spin_unlock_irqrestore(&rcp->lock, flags); -} - -static void __cpuinit rcu_online_cpu(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_data, cpu); - struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu); - - rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp); - rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp); - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); -} - -static int __cpuinit rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) -{ - long cpu = (long)hcpu; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - rcu_online_cpu(cpu); - break; - case CPU_DEAD: - case CPU_DEAD_FROZEN: - rcu_offline_cpu(cpu); - break; - default: - break; - } - return NOTIFY_OK; -} - -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - -/* - * Initializes rcu mechanism. Assumed to be called early. - * That is before local timer(SMP) or jiffie timer (uniproc) is setup. - * Note that rcu_qsctr and friends are implicitly - * initialized due to the choice of ``0'' for RCU_CTR_INVALID. - */ -void __init __rcu_init(void) -{ -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR - printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); -#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, - (void *)(long)smp_processor_id()); - /* Register notifier for non-boot CPUs */ - register_cpu_notifier(&rcu_nb); -} - -module_param(blimit, int, 0); -module_param(qhimark, int, 0); -module_param(qlowmark, int, 0); diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index a967c9feb90a..9f0584e84bb6 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -98,6 +98,30 @@ void synchronize_rcu(void) } EXPORT_SYMBOL_GPL(synchronize_rcu); +/** + * synchronize_rcu_bh - wait until an rcu_bh grace period has elapsed. + * + * Control will return to the caller some time after a full rcu_bh grace + * period has elapsed, in other words after all currently executing rcu_bh + * read-side critical sections have completed. RCU read-side critical + * sections are delimited by rcu_read_lock_bh() and rcu_read_unlock_bh(), + * and may be nested. + */ +void synchronize_rcu_bh(void) +{ + struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_bh(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_bh); + static void rcu_barrier_callback(struct rcu_head *notused) { if (atomic_dec_and_test(&rcu_barrier_cpu_count)) @@ -129,6 +153,7 @@ static void rcu_barrier_func(void *type) static inline void wait_migrated_callbacks(void) { wait_event(rcu_migrate_wq, !atomic_read(&rcu_migrate_type_count)); + smp_mb(); /* In case we didn't sleep. */ } /* @@ -192,9 +217,13 @@ static void rcu_migrate_callback(struct rcu_head *notused) wake_up(&rcu_migrate_wq); } +extern int rcu_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu); + static int __cpuinit rcu_barrier_cpu_hotplug(struct notifier_block *self, unsigned long action, void *hcpu) { + rcu_cpu_notify(self, action, hcpu); if (action == CPU_DYING) { /* * preempt_disable() in on_each_cpu() prevents stop_machine(), @@ -209,7 +238,7 @@ static int __cpuinit rcu_barrier_cpu_hotplug(struct notifier_block *self, call_rcu_bh(rcu_migrate_head, rcu_migrate_callback); call_rcu_sched(rcu_migrate_head + 1, rcu_migrate_callback); call_rcu(rcu_migrate_head + 2, rcu_migrate_callback); - } else if (action == CPU_POST_DEAD) { + } else if (action == CPU_DEAD) { /* rcu_migrate_head is protected by cpu_add_remove_lock */ wait_migrated_callbacks(); } @@ -219,8 +248,18 @@ static int __cpuinit rcu_barrier_cpu_hotplug(struct notifier_block *self, void __init rcu_init(void) { + int i; + __rcu_init(); hotcpu_notifier(rcu_barrier_cpu_hotplug, 0); + + /* + * We don't need protection against CPU-hotplug here because + * this is called early in boot, before either interrupts + * or the scheduler are operational. + */ + for_each_online_cpu(i) + rcu_barrier_cpu_hotplug(NULL, CPU_UP_PREPARE, (void *)(long)i); } void rcu_scheduler_starting(void) diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index beb0e659adcc..594835363550 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -1417,7 +1417,7 @@ int rcu_pending(int cpu) return 0; } -static int __cpuinit rcu_cpu_notify(struct notifier_block *self, +int __cpuinit rcu_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -1439,10 +1439,6 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - void __init __rcu_init(void) { int cpu; @@ -1471,23 +1467,6 @@ void __init __rcu_init(void) rdp->waitschedtail = &rdp->waitschedlist; rdp->rcu_sched_sleeping = 0; } - register_cpu_notifier(&rcu_nb); - - /* - * We don't need protection against CPU-Hotplug here - * since - * a) If a CPU comes online while we are iterating over the - * cpu_online_mask below, we would only end up making a - * duplicate call to rcu_online_cpu() which sets the corresponding - * CPU's mask in the rcu_cpu_online_map. - * - * b) A CPU cannot go offline at this point in time since the user - * does not have access to the sysfs interface, nor do we - * suspend the system. - */ - for_each_online_cpu(cpu) - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long) cpu); - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); } diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 9b4a975a4b4a..b33db539a8ad 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -257,14 +257,14 @@ struct rcu_torture_ops { void (*init)(void); void (*cleanup)(void); int (*readlock)(void); - void (*readdelay)(struct rcu_random_state *rrsp); + void (*read_delay)(struct rcu_random_state *rrsp); void (*readunlock)(int idx); int (*completed)(void); - void (*deferredfree)(struct rcu_torture *p); + void (*deferred_free)(struct rcu_torture *p); void (*sync)(void); void (*cb_barrier)(void); int (*stats)(char *page); - int irqcapable; + int irq_capable; char *name; }; static struct rcu_torture_ops *cur_ops = NULL; @@ -320,7 +320,7 @@ rcu_torture_cb(struct rcu_head *p) rp->rtort_mbtest = 0; rcu_torture_free(rp); } else - cur_ops->deferredfree(rp); + cur_ops->deferred_free(rp); } static void rcu_torture_deferred_free(struct rcu_torture *p) @@ -329,18 +329,18 @@ static void rcu_torture_deferred_free(struct rcu_torture *p) } static struct rcu_torture_ops rcu_ops = { - .init = NULL, - .cleanup = NULL, - .readlock = rcu_torture_read_lock, - .readdelay = rcu_read_delay, - .readunlock = rcu_torture_read_unlock, - .completed = rcu_torture_completed, - .deferredfree = rcu_torture_deferred_free, - .sync = synchronize_rcu, - .cb_barrier = rcu_barrier, - .stats = NULL, - .irqcapable = 1, - .name = "rcu" + .init = NULL, + .cleanup = NULL, + .readlock = rcu_torture_read_lock, + .read_delay = rcu_read_delay, + .readunlock = rcu_torture_read_unlock, + .completed = rcu_torture_completed, + .deferred_free = rcu_torture_deferred_free, + .sync = synchronize_rcu, + .cb_barrier = rcu_barrier, + .stats = NULL, + .irq_capable = 1, + .name = "rcu" }; static void rcu_sync_torture_deferred_free(struct rcu_torture *p) @@ -370,18 +370,18 @@ static void rcu_sync_torture_init(void) } static struct rcu_torture_ops rcu_sync_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = rcu_torture_read_lock, - .readdelay = rcu_read_delay, - .readunlock = rcu_torture_read_unlock, - .completed = rcu_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = synchronize_rcu, - .cb_barrier = NULL, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_sync" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_torture_read_lock, + .read_delay = rcu_read_delay, + .readunlock = rcu_torture_read_unlock, + .completed = rcu_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_rcu, + .cb_barrier = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_sync" }; /* @@ -432,33 +432,33 @@ static void rcu_bh_torture_synchronize(void) } static struct rcu_torture_ops rcu_bh_ops = { - .init = NULL, - .cleanup = NULL, - .readlock = rcu_bh_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferredfree = rcu_bh_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, - .cb_barrier = rcu_barrier_bh, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_bh" + .init = NULL, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferred_free = rcu_bh_torture_deferred_free, + .sync = rcu_bh_torture_synchronize, + .cb_barrier = rcu_barrier_bh, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_bh" }; static struct rcu_torture_ops rcu_bh_sync_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = rcu_bh_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, - .cb_barrier = NULL, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_bh_sync" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = rcu_bh_torture_synchronize, + .cb_barrier = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_bh_sync" }; /* @@ -530,17 +530,17 @@ static int srcu_torture_stats(char *page) } static struct rcu_torture_ops srcu_ops = { - .init = srcu_torture_init, - .cleanup = srcu_torture_cleanup, - .readlock = srcu_torture_read_lock, - .readdelay = srcu_read_delay, - .readunlock = srcu_torture_read_unlock, - .completed = srcu_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = srcu_torture_synchronize, - .cb_barrier = NULL, - .stats = srcu_torture_stats, - .name = "srcu" + .init = srcu_torture_init, + .cleanup = srcu_torture_cleanup, + .readlock = srcu_torture_read_lock, + .read_delay = srcu_read_delay, + .readunlock = srcu_torture_read_unlock, + .completed = srcu_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = srcu_torture_synchronize, + .cb_barrier = NULL, + .stats = srcu_torture_stats, + .name = "srcu" }; /* @@ -574,32 +574,49 @@ static void sched_torture_synchronize(void) } static struct rcu_torture_ops sched_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = sched_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = sched_torture_completed, - .deferredfree = rcu_sched_torture_deferred_free, - .sync = sched_torture_synchronize, - .cb_barrier = rcu_barrier_sched, - .stats = NULL, - .irqcapable = 1, - .name = "sched" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = sched_torture_completed, + .deferred_free = rcu_sched_torture_deferred_free, + .sync = sched_torture_synchronize, + .cb_barrier = rcu_barrier_sched, + .stats = NULL, + .irq_capable = 1, + .name = "sched" }; static struct rcu_torture_ops sched_ops_sync = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = sched_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = sched_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = sched_torture_synchronize, - .cb_barrier = NULL, - .stats = NULL, - .name = "sched_sync" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = sched_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = sched_torture_synchronize, + .cb_barrier = NULL, + .stats = NULL, + .name = "sched_sync" +}; + +extern int rcu_expedited_torture_stats(char *page); + +static struct rcu_torture_ops sched_expedited_ops = { + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = sched_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_sched_expedited, + .cb_barrier = NULL, + .stats = rcu_expedited_torture_stats, + .irq_capable = 1, + .name = "sched_expedited" }; /* @@ -635,7 +652,7 @@ rcu_torture_writer(void *arg) i = RCU_TORTURE_PIPE_LEN; atomic_inc(&rcu_torture_wcount[i]); old_rp->rtort_pipe_count++; - cur_ops->deferredfree(old_rp); + cur_ops->deferred_free(old_rp); } rcu_torture_current_version++; oldbatch = cur_ops->completed(); @@ -700,7 +717,7 @@ static void rcu_torture_timer(unsigned long unused) if (p->rtort_mbtest == 0) atomic_inc(&n_rcu_torture_mberror); spin_lock(&rand_lock); - cur_ops->readdelay(&rand); + cur_ops->read_delay(&rand); n_rcu_torture_timers++; spin_unlock(&rand_lock); preempt_disable(); @@ -738,11 +755,11 @@ rcu_torture_reader(void *arg) VERBOSE_PRINTK_STRING("rcu_torture_reader task started"); set_user_nice(current, 19); - if (irqreader && cur_ops->irqcapable) + if (irqreader && cur_ops->irq_capable) setup_timer_on_stack(&t, rcu_torture_timer, 0); do { - if (irqreader && cur_ops->irqcapable) { + if (irqreader && cur_ops->irq_capable) { if (!timer_pending(&t)) mod_timer(&t, 1); } @@ -757,7 +774,7 @@ rcu_torture_reader(void *arg) } if (p->rtort_mbtest == 0) atomic_inc(&n_rcu_torture_mberror); - cur_ops->readdelay(&rand); + cur_ops->read_delay(&rand); preempt_disable(); pipe_count = p->rtort_pipe_count; if (pipe_count > RCU_TORTURE_PIPE_LEN) { @@ -778,7 +795,7 @@ rcu_torture_reader(void *arg) } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); VERBOSE_PRINTK_STRING("rcu_torture_reader task stopping"); rcutorture_shutdown_absorb("rcu_torture_reader"); - if (irqreader && cur_ops->irqcapable) + if (irqreader && cur_ops->irq_capable) del_timer_sync(&t); while (!kthread_should_stop()) schedule_timeout_uninterruptible(1); @@ -1078,6 +1095,7 @@ rcu_torture_init(void) int firsterr = 0; static struct rcu_torture_ops *torture_ops[] = { &rcu_ops, &rcu_sync_ops, &rcu_bh_ops, &rcu_bh_sync_ops, + &sched_expedited_ops, &srcu_ops, &sched_ops, &sched_ops_sync, }; mutex_lock(&fullstop_mutex); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 7717b95c2027..6c5b37d4ce7b 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -35,6 +35,7 @@ #include <linux/rcupdate.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <linux/nmi.h> #include <asm/atomic.h> #include <linux/bitops.h> #include <linux/module.h> @@ -469,6 +470,8 @@ static void print_other_cpu_stall(struct rcu_state *rsp) } printk(" (detected by %d, t=%ld jiffies)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start)); + trigger_all_cpu_backtrace(); + force_quiescent_state(rsp, 0); /* Kick them all. */ } @@ -479,12 +482,14 @@ static void print_cpu_stall(struct rcu_state *rsp) printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu jiffies)\n", smp_processor_id(), jiffies - rsp->gp_start); - dump_stack(); + trigger_all_cpu_backtrace(); + spin_lock_irqsave(&rnp->lock, flags); if ((long)(jiffies - rsp->jiffies_stall) >= 0) rsp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; spin_unlock_irqrestore(&rnp->lock, flags); + set_need_resched(); /* kick ourselves to get things going. */ } @@ -1132,6 +1137,8 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; + WARN_ON_ONCE(rdp->beenonline == 0); + /* * If an RCU GP has gone long enough, go check for dyntick * idle CPUs and, if needed, send resched IPIs. @@ -1325,22 +1332,40 @@ int rcu_needs_cpu(int cpu) } /* - * Initialize a CPU's per-CPU RCU data. We take this "scorched earth" - * approach so that we don't have to worry about how long the CPU has - * been gone, or whether it ever was online previously. We do trust the - * ->mynode field, as it is constant for a given struct rcu_data and - * initialized during early boot. - * - * Note that only one online or offline event can be happening at a given - * time. Note also that we can accept some slop in the rsp->completed - * access due to the fact that this CPU cannot possibly have any RCU - * callbacks in flight yet. + * Do boot-time initialization of a CPU's per-CPU RCU data. + */ +static void __init +rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) +{ + unsigned long flags; + int i; + struct rcu_data *rdp = rsp->rda[cpu]; + struct rcu_node *rnp = rcu_get_root(rsp); + + /* Set up local state, ensuring consistent view of global state. */ + spin_lock_irqsave(&rnp->lock, flags); + rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); + rdp->nxtlist = NULL; + for (i = 0; i < RCU_NEXT_SIZE; i++) + rdp->nxttail[i] = &rdp->nxtlist; + rdp->qlen = 0; +#ifdef CONFIG_NO_HZ + rdp->dynticks = &per_cpu(rcu_dynticks, cpu); +#endif /* #ifdef CONFIG_NO_HZ */ + rdp->cpu = cpu; + spin_unlock_irqrestore(&rnp->lock, flags); +} + +/* + * Initialize a CPU's per-CPU RCU data. Note that only one online or + * offline event can be happening at a given time. Note also that we + * can accept some slop in the rsp->completed access due to the fact + * that this CPU cannot possibly have any RCU callbacks in flight yet. */ static void __cpuinit rcu_init_percpu_data(int cpu, struct rcu_state *rsp) { unsigned long flags; - int i; long lastcomp; unsigned long mask; struct rcu_data *rdp = rsp->rda[cpu]; @@ -1355,16 +1380,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->qs_pending = 1; /* so set up to respond to current GP. */ rdp->beenonline = 1; /* We have now been online. */ rdp->passed_quiesc_completed = lastcomp - 1; - rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; - rdp->qlen = 0; rdp->blimit = blimit; -#ifdef CONFIG_NO_HZ - rdp->dynticks = &per_cpu(rcu_dynticks, cpu); -#endif /* #ifdef CONFIG_NO_HZ */ - rdp->cpu = cpu; spin_unlock(&rnp->lock); /* irqs remain disabled. */ /* @@ -1407,13 +1423,12 @@ static void __cpuinit rcu_online_cpu(int cpu) { rcu_init_percpu_data(cpu, &rcu_state); rcu_init_percpu_data(cpu, &rcu_bh_state); - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); } /* * Handle CPU online/offline notifcation events. */ -static int __cpuinit rcu_cpu_notify(struct notifier_block *self, +int __cpuinit rcu_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -1523,10 +1538,6 @@ do { \ } \ } while (0) -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - void __init __rcu_init(void) { int i; /* All used by RCU_DATA_PTR_INIT(). */ @@ -1539,13 +1550,13 @@ void __init __rcu_init(void) #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ rcu_init_one(&rcu_state); RCU_DATA_PTR_INIT(&rcu_state, rcu_data); + for_each_possible_cpu(i) + rcu_boot_init_percpu_data(i, &rcu_state); rcu_init_one(&rcu_bh_state); RCU_DATA_PTR_INIT(&rcu_bh_state, rcu_bh_data); - - for_each_online_cpu(i) - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long)i); - /* Register notifier for non-boot CPUs */ - register_cpu_notifier(&rcu_nb); + for_each_possible_cpu(i) + rcu_boot_init_percpu_data(i, &rcu_bh_state); + open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); } module_param(blimit, int, 0); diff --git a/kernel/sched.c b/kernel/sched.c index 1b59e265273b..5184580613b7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -616,6 +616,7 @@ struct rq { unsigned char idle_at_tick; /* For active balancing */ + int post_schedule; int active_balance; int push_cpu; /* cpu of this runqueue: */ @@ -693,6 +694,7 @@ static inline int cpu_of(struct rq *rq) #define this_rq() (&__get_cpu_var(runqueues)) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) +#define raw_rq() (&__raw_get_cpu_var(runqueues)) inline void update_rq_clock(struct rq *rq) { @@ -1522,13 +1524,18 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long sd_shares, unsigned long sd_rq_weight) { - unsigned long shares; unsigned long rq_weight; + unsigned long shares; + int boost = 0; if (!tg->se[cpu]) return; rq_weight = tg->cfs_rq[cpu]->rq_weight; + if (!rq_weight) { + boost = 1; + rq_weight = NICE_0_LOAD; + } /* * \Sum shares * rq_weight @@ -1545,8 +1552,7 @@ update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long flags; spin_lock_irqsave(&rq->lock, flags); - tg->cfs_rq[cpu]->shares = shares; - + tg->cfs_rq[cpu]->shares = boost ? 0 : shares; __set_se_shares(tg->se[cpu], shares); spin_unlock_irqrestore(&rq->lock, flags); } @@ -1559,7 +1565,7 @@ update_group_shares_cpu(struct task_group *tg, int cpu, */ static int tg_shares_up(struct task_group *tg, void *data) { - unsigned long weight, rq_weight = 0; + unsigned long weight, rq_weight = 0, eff_weight = 0; unsigned long shares = 0; struct sched_domain *sd = data; int i; @@ -1571,11 +1577,13 @@ static int tg_shares_up(struct task_group *tg, void *data) * run here it will not get delayed by group starvation. */ weight = tg->cfs_rq[i]->load.weight; + tg->cfs_rq[i]->rq_weight = weight; + rq_weight += weight; + if (!weight) weight = NICE_0_LOAD; - tg->cfs_rq[i]->rq_weight = weight; - rq_weight += weight; + eff_weight += weight; shares += tg->cfs_rq[i]->shares; } @@ -1585,8 +1593,14 @@ static int tg_shares_up(struct task_group *tg, void *data) if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE)) shares = tg->shares; - for_each_cpu(i, sched_domain_span(sd)) - update_group_shares_cpu(tg, i, shares, rq_weight); + for_each_cpu(i, sched_domain_span(sd)) { + unsigned long sd_rq_weight = rq_weight; + + if (!tg->cfs_rq[i]->rq_weight) + sd_rq_weight = eff_weight; + + update_group_shares_cpu(tg, i, shares, sd_rq_weight); + } return 0; } @@ -1616,8 +1630,14 @@ static int tg_load_down(struct task_group *tg, void *data) static void update_shares(struct sched_domain *sd) { - u64 now = cpu_clock(raw_smp_processor_id()); - s64 elapsed = now - sd->last_update; + s64 elapsed; + u64 now; + + if (root_task_group_empty()) + return; + + now = cpu_clock(raw_smp_processor_id()); + elapsed = now - sd->last_update; if (elapsed >= (s64)(u64)sysctl_sched_shares_ratelimit) { sd->last_update = now; @@ -1627,6 +1647,9 @@ static void update_shares(struct sched_domain *sd) static void update_shares_locked(struct rq *rq, struct sched_domain *sd) { + if (root_task_group_empty()) + return; + spin_unlock(&rq->lock); update_shares(sd); spin_lock(&rq->lock); @@ -1634,6 +1657,9 @@ static void update_shares_locked(struct rq *rq, struct sched_domain *sd) static void update_h_load(long cpu) { + if (root_task_group_empty()) + return; + walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); } @@ -2637,9 +2663,32 @@ void sched_fork(struct task_struct *p, int clone_flags) set_task_cpu(p, cpu); /* - * Make sure we do not leak PI boosting priority to the child: + * Make sure we do not leak PI boosting priority to the child. */ p->prio = current->normal_prio; + + /* + * Revert to default priority/policy on fork if requested. + */ + if (unlikely(p->sched_reset_on_fork)) { + if (p->policy == SCHED_FIFO || p->policy == SCHED_RR) + p->policy = SCHED_NORMAL; + + if (p->normal_prio < DEFAULT_PRIO) + p->prio = DEFAULT_PRIO; + + if (PRIO_TO_NICE(p->static_prio) < 0) { + p->static_prio = NICE_TO_PRIO(0); + set_load_weight(p); + } + + /* + * We don't need the reset flag anymore after the fork. It has + * fulfilled its duty: + */ + p->sched_reset_on_fork = 0; + } + if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; @@ -2796,12 +2845,6 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) { struct mm_struct *mm = rq->prev_mm; long prev_state; -#ifdef CONFIG_SMP - int post_schedule = 0; - - if (current->sched_class->needs_post_schedule) - post_schedule = current->sched_class->needs_post_schedule(rq); -#endif rq->prev_mm = NULL; @@ -2820,10 +2863,6 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) finish_arch_switch(prev); perf_counter_task_sched_in(current, cpu_of(rq)); finish_lock_switch(rq, prev); -#ifdef CONFIG_SMP - if (post_schedule) - current->sched_class->post_schedule(rq); -#endif fire_sched_in_preempt_notifiers(current); if (mm) @@ -2838,6 +2877,42 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) } } +#ifdef CONFIG_SMP + +/* assumes rq->lock is held */ +static inline void pre_schedule(struct rq *rq, struct task_struct *prev) +{ + if (prev->sched_class->pre_schedule) + prev->sched_class->pre_schedule(rq, prev); +} + +/* rq->lock is NOT held, but preemption is disabled */ +static inline void post_schedule(struct rq *rq) +{ + if (rq->post_schedule) { + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + if (rq->curr->sched_class->post_schedule) + rq->curr->sched_class->post_schedule(rq); + spin_unlock_irqrestore(&rq->lock, flags); + + rq->post_schedule = 0; + } +} + +#else + +static inline void pre_schedule(struct rq *rq, struct task_struct *p) +{ +} + +static inline void post_schedule(struct rq *rq) +{ +} + +#endif + /** * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. @@ -2848,6 +2923,13 @@ asmlinkage void schedule_tail(struct task_struct *prev) struct rq *rq = this_rq(); finish_task_switch(rq, prev); + + /* + * FIXME: do we need to worry about rq being invalidated by the + * task_switch? + */ + post_schedule(rq); + #ifdef __ARCH_WANT_UNLOCKED_CTXSW /* In this case, finish_task_switch does not reenable preemption */ preempt_enable(); @@ -5031,17 +5113,16 @@ void account_idle_time(cputime_t cputime) */ void account_process_tick(struct task_struct *p, int user_tick) { - cputime_t one_jiffy = jiffies_to_cputime(1); - cputime_t one_jiffy_scaled = cputime_to_scaled(one_jiffy); + cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); struct rq *rq = this_rq(); if (user_tick) - account_user_time(p, one_jiffy, one_jiffy_scaled); + account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET)) - account_system_time(p, HARDIRQ_OFFSET, one_jiffy, + account_system_time(p, HARDIRQ_OFFSET, cputime_one_jiffy, one_jiffy_scaled); else - account_idle_time(one_jiffy); + account_idle_time(cputime_one_jiffy); } /* @@ -5349,10 +5430,7 @@ need_resched_nonpreemptible: switch_count = &prev->nvcsw; } -#ifdef CONFIG_SMP - if (prev->sched_class->pre_schedule) - prev->sched_class->pre_schedule(rq, prev); -#endif + pre_schedule(rq, prev); if (unlikely(!rq->nr_running)) idle_balance(cpu, rq); @@ -5378,6 +5456,8 @@ need_resched_nonpreemptible: } else spin_unlock_irq(&rq->lock); + post_schedule(rq); + if (unlikely(reacquire_kernel_lock(current) < 0)) goto need_resched_nonpreemptible; @@ -6123,17 +6203,25 @@ static int __sched_setscheduler(struct task_struct *p, int policy, unsigned long flags; const struct sched_class *prev_class = p->sched_class; struct rq *rq; + int reset_on_fork; /* may grab non-irq protected spin_locks */ BUG_ON(in_interrupt()); recheck: /* double check policy once rq lock held */ - if (policy < 0) + if (policy < 0) { + reset_on_fork = p->sched_reset_on_fork; policy = oldpolicy = p->policy; - else if (policy != SCHED_FIFO && policy != SCHED_RR && - policy != SCHED_NORMAL && policy != SCHED_BATCH && - policy != SCHED_IDLE) - return -EINVAL; + } else { + reset_on_fork = !!(policy & SCHED_RESET_ON_FORK); + policy &= ~SCHED_RESET_ON_FORK; + + if (policy != SCHED_FIFO && policy != SCHED_RR && + policy != SCHED_NORMAL && policy != SCHED_BATCH && + policy != SCHED_IDLE) + return -EINVAL; + } + /* * Valid priorities for SCHED_FIFO and SCHED_RR are * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL, @@ -6177,6 +6265,10 @@ recheck: /* can't change other user's priorities */ if (!check_same_owner(p)) return -EPERM; + + /* Normal users shall not reset the sched_reset_on_fork flag */ + if (p->sched_reset_on_fork && !reset_on_fork) + return -EPERM; } if (user) { @@ -6220,6 +6312,8 @@ recheck: if (running) p->sched_class->put_prev_task(rq, p); + p->sched_reset_on_fork = reset_on_fork; + oldprio = p->prio; __setscheduler(rq, p, policy, param->sched_priority); @@ -6336,14 +6430,15 @@ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) if (p) { retval = security_task_getscheduler(p); if (!retval) - retval = p->policy; + retval = p->policy + | (p->sched_reset_on_fork ? SCHED_RESET_ON_FORK : 0); } read_unlock(&tasklist_lock); return retval; } /** - * sys_sched_getscheduler - get the RT priority of a thread + * sys_sched_getparam - get the RT priority of a thread * @pid: the pid in question. * @param: structure containing the RT priority. */ @@ -6571,19 +6666,9 @@ static inline int should_resched(void) static void __cond_resched(void) { -#ifdef CONFIG_DEBUG_SPINLOCK_SLEEP - __might_sleep(__FILE__, __LINE__); -#endif - /* - * The BKS might be reacquired before we have dropped - * PREEMPT_ACTIVE, which could trigger a second - * cond_resched() call. - */ - do { - add_preempt_count(PREEMPT_ACTIVE); - schedule(); - sub_preempt_count(PREEMPT_ACTIVE); - } while (need_resched()); + add_preempt_count(PREEMPT_ACTIVE); + schedule(); + sub_preempt_count(PREEMPT_ACTIVE); } int __sched _cond_resched(void) @@ -6597,18 +6682,20 @@ int __sched _cond_resched(void) EXPORT_SYMBOL(_cond_resched); /* - * cond_resched_lock() - if a reschedule is pending, drop the given lock, + * __cond_resched_lock() - if a reschedule is pending, drop the given lock, * call schedule, and on return reacquire the lock. * * This works OK both with and without CONFIG_PREEMPT. We do strange low-level * operations here to prevent schedule() from being called twice (once via * spin_unlock(), once by hand). */ -int cond_resched_lock(spinlock_t *lock) +int __cond_resched_lock(spinlock_t *lock) { int resched = should_resched(); int ret = 0; + lockdep_assert_held(lock); + if (spin_needbreak(lock) || resched) { spin_unlock(lock); if (resched) @@ -6620,9 +6707,9 @@ int cond_resched_lock(spinlock_t *lock) } return ret; } -EXPORT_SYMBOL(cond_resched_lock); +EXPORT_SYMBOL(__cond_resched_lock); -int __sched cond_resched_softirq(void) +int __sched __cond_resched_softirq(void) { BUG_ON(!in_softirq()); @@ -6634,7 +6721,7 @@ int __sched cond_resched_softirq(void) } return 0; } -EXPORT_SYMBOL(cond_resched_softirq); +EXPORT_SYMBOL(__cond_resched_softirq); /** * yield - yield the current processor to other threads. @@ -6658,7 +6745,7 @@ EXPORT_SYMBOL(yield); */ void __sched io_schedule(void) { - struct rq *rq = &__raw_get_cpu_var(runqueues); + struct rq *rq = raw_rq(); delayacct_blkio_start(); atomic_inc(&rq->nr_iowait); @@ -6670,7 +6757,7 @@ EXPORT_SYMBOL(io_schedule); long __sched io_schedule_timeout(long timeout) { - struct rq *rq = &__raw_get_cpu_var(runqueues); + struct rq *rq = raw_rq(); long ret; delayacct_blkio_start(); @@ -6992,8 +7079,12 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) { /* Need help from migration thread: drop lock and wait. */ + struct task_struct *mt = rq->migration_thread; + + get_task_struct(mt); task_rq_unlock(rq, &flags); wake_up_process(rq->migration_thread); + put_task_struct(mt); wait_for_completion(&req.done); tlb_migrate_finish(p->mm); return 0; @@ -7051,6 +7142,11 @@ fail: return ret; } +#define RCU_MIGRATION_IDLE 0 +#define RCU_MIGRATION_NEED_QS 1 +#define RCU_MIGRATION_GOT_QS 2 +#define RCU_MIGRATION_MUST_SYNC 3 + /* * migration_thread - this is a highprio system thread that performs * thread migration by bumping thread off CPU then 'pushing' onto @@ -7058,6 +7154,7 @@ fail: */ static int migration_thread(void *data) { + int badcpu; int cpu = (long)data; struct rq *rq; @@ -7092,8 +7189,17 @@ static int migration_thread(void *data) req = list_entry(head->next, struct migration_req, list); list_del_init(head->next); - spin_unlock(&rq->lock); - __migrate_task(req->task, cpu, req->dest_cpu); + if (req->task != NULL) { + spin_unlock(&rq->lock); + __migrate_task(req->task, cpu, req->dest_cpu); + } else if (likely(cpu == (badcpu = smp_processor_id()))) { + req->dest_cpu = RCU_MIGRATION_GOT_QS; + spin_unlock(&rq->lock); + } else { + req->dest_cpu = RCU_MIGRATION_MUST_SYNC; + spin_unlock(&rq->lock); + WARN_ONCE(1, "migration_thread() on CPU %d, expected %d\n", badcpu, cpu); + } local_irq_enable(); complete(&req->done); @@ -7625,7 +7731,7 @@ static int __init migration_init(void) migration_call(&migration_notifier, CPU_ONLINE, cpu); register_cpu_notifier(&migration_notifier); - return err; + return 0; } early_initcall(migration_init); #endif @@ -7841,7 +7947,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) rq->rd = rd; cpumask_set_cpu(rq->cpu, rd->span); - if (cpumask_test_cpu(rq->cpu, cpu_online_mask)) + if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); spin_unlock_irqrestore(&rq->lock, flags); @@ -9334,6 +9440,7 @@ void __init sched_init(void) #ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; + rq->post_schedule = 0; rq->active_balance = 0; rq->next_balance = jiffies; rq->push_cpu = 0; @@ -9398,13 +9505,20 @@ void __init sched_init(void) } #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP -void __might_sleep(char *file, int line) +static inline int preempt_count_equals(int preempt_offset) +{ + int nested = preempt_count() & ~PREEMPT_ACTIVE; + + return (nested == PREEMPT_INATOMIC_BASE + preempt_offset); +} + +void __might_sleep(char *file, int line, int preempt_offset) { #ifdef in_atomic static unsigned long prev_jiffy; /* ratelimiting */ - if ((!in_atomic() && !irqs_disabled()) || - system_state != SYSTEM_RUNNING || oops_in_progress) + if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || + system_state != SYSTEM_RUNNING || oops_in_progress) return; if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) return; @@ -10581,3 +10695,113 @@ struct cgroup_subsys cpuacct_subsys = { .subsys_id = cpuacct_subsys_id, }; #endif /* CONFIG_CGROUP_CPUACCT */ + +#ifndef CONFIG_SMP + +int rcu_expedited_torture_stats(char *page) +{ + return 0; +} +EXPORT_SYMBOL_GPL(rcu_expedited_torture_stats); + +void synchronize_sched_expedited(void) +{ +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#else /* #ifndef CONFIG_SMP */ + +static DEFINE_PER_CPU(struct migration_req, rcu_migration_req); +static DEFINE_MUTEX(rcu_sched_expedited_mutex); + +#define RCU_EXPEDITED_STATE_POST -2 +#define RCU_EXPEDITED_STATE_IDLE -1 + +static int rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; + +int rcu_expedited_torture_stats(char *page) +{ + int cnt = 0; + int cpu; + + cnt += sprintf(&page[cnt], "state: %d /", rcu_expedited_state); + for_each_online_cpu(cpu) { + cnt += sprintf(&page[cnt], " %d:%d", + cpu, per_cpu(rcu_migration_req, cpu).dest_cpu); + } + cnt += sprintf(&page[cnt], "\n"); + return cnt; +} +EXPORT_SYMBOL_GPL(rcu_expedited_torture_stats); + +static long synchronize_sched_expedited_count; + +/* + * Wait for an rcu-sched grace period to elapse, but use "big hammer" + * approach to force grace period to end quickly. This consumes + * significant time on all CPUs, and is thus not recommended for + * any sort of common-case code. + * + * Note that it is illegal to call this function while holding any + * lock that is acquired by a CPU-hotplug notifier. Failing to + * observe this restriction will result in deadlock. + */ +void synchronize_sched_expedited(void) +{ + int cpu; + unsigned long flags; + bool need_full_sync = 0; + struct rq *rq; + struct migration_req *req; + long snap; + int trycount = 0; + + smp_mb(); /* ensure prior mod happens before capturing snap. */ + snap = ACCESS_ONCE(synchronize_sched_expedited_count) + 1; + get_online_cpus(); + while (!mutex_trylock(&rcu_sched_expedited_mutex)) { + put_online_cpus(); + if (trycount++ < 10) + udelay(trycount * num_online_cpus()); + else { + synchronize_sched(); + return; + } + if (ACCESS_ONCE(synchronize_sched_expedited_count) - snap > 0) { + smp_mb(); /* ensure test happens before caller kfree */ + return; + } + get_online_cpus(); + } + rcu_expedited_state = RCU_EXPEDITED_STATE_POST; + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + req = &per_cpu(rcu_migration_req, cpu); + init_completion(&req->done); + req->task = NULL; + req->dest_cpu = RCU_MIGRATION_NEED_QS; + spin_lock_irqsave(&rq->lock, flags); + list_add(&req->list, &rq->migration_queue); + spin_unlock_irqrestore(&rq->lock, flags); + wake_up_process(rq->migration_thread); + } + for_each_online_cpu(cpu) { + rcu_expedited_state = cpu; + req = &per_cpu(rcu_migration_req, cpu); + rq = cpu_rq(cpu); + wait_for_completion(&req->done); + spin_lock_irqsave(&rq->lock, flags); + if (unlikely(req->dest_cpu == RCU_MIGRATION_MUST_SYNC)) + need_full_sync = 1; + req->dest_cpu = RCU_MIGRATION_IDLE; + spin_unlock_irqrestore(&rq->lock, flags); + } + rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; + mutex_unlock(&rcu_sched_expedited_mutex); + put_online_cpus(); + if (need_full_sync) + synchronize_sched(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#endif /* #else #ifndef CONFIG_SMP */ diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index d014efbf947a..0f052fc674d5 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -127,21 +127,11 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) /* * If the cpu was currently mapped to a different value, we - * first need to unmap the old value + * need to map it to the new value then remove the old value. + * Note, we must add the new value first, otherwise we risk the + * cpu being cleared from pri_active, and this cpu could be + * missed for a push or pull. */ - if (likely(oldpri != CPUPRI_INVALID)) { - struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; - - spin_lock_irqsave(&vec->lock, flags); - - vec->count--; - if (!vec->count) - clear_bit(oldpri, cp->pri_active); - cpumask_clear_cpu(cpu, vec->mask); - - spin_unlock_irqrestore(&vec->lock, flags); - } - if (likely(newpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[newpri]; @@ -154,6 +144,18 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) spin_unlock_irqrestore(&vec->lock, flags); } + if (likely(oldpri != CPUPRI_INVALID)) { + struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; + + spin_lock_irqsave(&vec->lock, flags); + + vec->count--; + if (!vec->count) + clear_bit(oldpri, cp->pri_active); + cpumask_clear_cpu(cpu, vec->mask); + + spin_unlock_irqrestore(&vec->lock, flags); + } *currpri = newpri; } diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 652e8bdef9aa..342000b31ad6 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -79,11 +79,6 @@ static const struct sched_class fair_sched_class; * CFS operations on generic schedulable entities: */ -static inline struct task_struct *task_of(struct sched_entity *se) -{ - return container_of(se, struct task_struct, se); -} - #ifdef CONFIG_FAIR_GROUP_SCHED /* cpu runqueue to which this cfs_rq is attached */ @@ -95,6 +90,14 @@ static inline struct rq *rq_of(struct cfs_rq *cfs_rq) /* An entity is a task if it doesn't "own" a runqueue */ #define entity_is_task(se) (!se->my_q) +static inline struct task_struct *task_of(struct sched_entity *se) +{ +#ifdef CONFIG_SCHED_DEBUG + WARN_ON_ONCE(!entity_is_task(se)); +#endif + return container_of(se, struct task_struct, se); +} + /* Walk up scheduling entities hierarchy */ #define for_each_sched_entity(se) \ for (; se; se = se->parent) @@ -186,7 +189,12 @@ find_matching_se(struct sched_entity **se, struct sched_entity **pse) } } -#else /* CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED */ + +static inline struct task_struct *task_of(struct sched_entity *se) +{ + return container_of(se, struct task_struct, se); +} static inline struct rq *rq_of(struct cfs_rq *cfs_rq) { @@ -1046,17 +1054,21 @@ static void yield_task_fair(struct rq *rq) * search starts with cpus closest then further out as needed, * so we always favor a closer, idle cpu. * Domains may include CPUs that are not usable for migration, - * hence we need to mask them out (cpu_active_mask) + * hence we need to mask them out (rq->rd->online) * * Returns the CPU we should wake onto. */ #if defined(ARCH_HAS_SCHED_WAKE_IDLE) + +#define cpu_rd_active(cpu, rq) cpumask_test_cpu(cpu, rq->rd->online) + static int wake_idle(int cpu, struct task_struct *p) { struct sched_domain *sd; int i; unsigned int chosen_wakeup_cpu; int this_cpu; + struct rq *task_rq = task_rq(p); /* * At POWERSAVINGS_BALANCE_WAKEUP level, if both this_cpu and prev_cpu @@ -1089,10 +1101,10 @@ static int wake_idle(int cpu, struct task_struct *p) for_each_domain(cpu, sd) { if ((sd->flags & SD_WAKE_IDLE) || ((sd->flags & SD_WAKE_IDLE_FAR) - && !task_hot(p, task_rq(p)->clock, sd))) { + && !task_hot(p, task_rq->clock, sd))) { for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) { - if (cpu_active(i) && idle_cpu(i)) { + if (cpu_rd_active(i, task_rq) && idle_cpu(i)) { if (i != task_cpu(p)) { schedstat_inc(p, se.nr_wakeups_idle); diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 3918e01994e0..3d4020a9ba1b 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -3,15 +3,18 @@ * policies) */ +#ifdef CONFIG_RT_GROUP_SCHED + +#define rt_entity_is_task(rt_se) (!(rt_se)->my_q) + static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) { +#ifdef CONFIG_SCHED_DEBUG + WARN_ON_ONCE(!rt_entity_is_task(rt_se)); +#endif return container_of(rt_se, struct task_struct, rt); } -#ifdef CONFIG_RT_GROUP_SCHED - -#define rt_entity_is_task(rt_se) (!(rt_se)->my_q) - static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq) { return rt_rq->rq; @@ -26,6 +29,11 @@ static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se) #define rt_entity_is_task(rt_se) (1) +static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) +{ + return container_of(rt_se, struct task_struct, rt); +} + static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq) { return container_of(rt_rq, struct rq, rt); @@ -128,6 +136,11 @@ static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); } +static inline int has_pushable_tasks(struct rq *rq) +{ + return !plist_head_empty(&rq->rt.pushable_tasks); +} + #else static inline void enqueue_pushable_task(struct rq *rq, struct task_struct *p) @@ -1064,6 +1077,14 @@ static struct task_struct *pick_next_task_rt(struct rq *rq) if (p) dequeue_pushable_task(rq, p); +#ifdef CONFIG_SMP + /* + * We detect this state here so that we can avoid taking the RQ + * lock again later if there is no need to push + */ + rq->post_schedule = has_pushable_tasks(rq); +#endif + return p; } @@ -1162,13 +1183,6 @@ static int find_lowest_rq(struct task_struct *task) return -1; /* No targets found */ /* - * Only consider CPUs that are usable for migration. - * I guess we might want to change cpupri_find() to ignore those - * in the first place. - */ - cpumask_and(lowest_mask, lowest_mask, cpu_active_mask); - - /* * At this point we have built a mask of cpus representing the * lowest priority tasks in the system. Now we want to elect * the best one based on our affinity and topology. @@ -1262,11 +1276,6 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) return lowest_rq; } -static inline int has_pushable_tasks(struct rq *rq) -{ - return !plist_head_empty(&rq->rt.pushable_tasks); -} - static struct task_struct *pick_next_pushable_task(struct rq *rq) { struct task_struct *p; @@ -1466,23 +1475,9 @@ static void pre_schedule_rt(struct rq *rq, struct task_struct *prev) pull_rt_task(rq); } -/* - * assumes rq->lock is held - */ -static int needs_post_schedule_rt(struct rq *rq) -{ - return has_pushable_tasks(rq); -} - static void post_schedule_rt(struct rq *rq) { - /* - * This is only called if needs_post_schedule_rt() indicates that - * we need to push tasks away - */ - spin_lock_irq(&rq->lock); push_rt_tasks(rq); - spin_unlock_irq(&rq->lock); } /* @@ -1758,7 +1753,6 @@ static const struct sched_class rt_sched_class = { .rq_online = rq_online_rt, .rq_offline = rq_offline_rt, .pre_schedule = pre_schedule_rt, - .needs_post_schedule = needs_post_schedule_rt, .post_schedule = post_schedule_rt, .task_wake_up = task_wake_up_rt, .switched_from = switched_from_rt, diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index e8c77d9c633a..02c0b2c9c674 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -125,6 +125,75 @@ void getnstimeofday(struct timespec *ts) EXPORT_SYMBOL(getnstimeofday); +ktime_t ktime_get(void) +{ + cycle_t cycle_now, cycle_delta; + unsigned int seq; + s64 secs, nsecs; + + WARN_ON(timekeeping_suspended); + + do { + seq = read_seqbegin(&xtime_lock); + secs = xtime.tv_sec + wall_to_monotonic.tv_sec; + nsecs = xtime.tv_nsec + wall_to_monotonic.tv_nsec; + + /* read clocksource: */ + cycle_now = clocksource_read(clock); + + /* calculate the delta since the last update_wall_time: */ + cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; + + /* convert to nanoseconds: */ + nsecs += cyc2ns(clock, cycle_delta); + + } while (read_seqretry(&xtime_lock, seq)); + /* + * Use ktime_set/ktime_add_ns to create a proper ktime on + * 32-bit architectures without CONFIG_KTIME_SCALAR. + */ + return ktime_add_ns(ktime_set(secs, 0), nsecs); +} +EXPORT_SYMBOL_GPL(ktime_get); + +/** + * ktime_get_ts - get the monotonic clock in timespec format + * @ts: pointer to timespec variable + * + * The function calculates the monotonic clock from the realtime + * clock and the wall_to_monotonic offset and stores the result + * in normalized timespec format in the variable pointed to by @ts. + */ +void ktime_get_ts(struct timespec *ts) +{ + cycle_t cycle_now, cycle_delta; + struct timespec tomono; + unsigned int seq; + s64 nsecs; + + WARN_ON(timekeeping_suspended); + + do { + seq = read_seqbegin(&xtime_lock); + *ts = xtime; + tomono = wall_to_monotonic; + + /* read clocksource: */ + cycle_now = clocksource_read(clock); + + /* calculate the delta since the last update_wall_time: */ + cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; + + /* convert to nanoseconds: */ + nsecs = cyc2ns(clock, cycle_delta); + + } while (read_seqretry(&xtime_lock, seq)); + + set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, + ts->tv_nsec + tomono.tv_nsec + nsecs); +} +EXPORT_SYMBOL_GPL(ktime_get_ts); + /** * do_gettimeofday - Returns the time of day in a timeval * @tv: pointer to the timeval to be set @@ -221,10 +290,65 @@ static void change_clocksource(void) clock->name); */ } -#else +#else /* GENERIC_TIME */ static inline void clocksource_forward_now(void) { } static inline void change_clocksource(void) { } -#endif + +/** + * ktime_get - get the monotonic time in ktime_t format + * + * returns the time in ktime_t format + */ +ktime_t ktime_get(void) +{ + struct timespec now; + + ktime_get_ts(&now); + + return timespec_to_ktime(now); +} +EXPORT_SYMBOL_GPL(ktime_get); + +/** + * ktime_get_ts - get the monotonic clock in timespec format + * @ts: pointer to timespec variable + * + * The function calculates the monotonic clock from the realtime + * clock and the wall_to_monotonic offset and stores the result + * in normalized timespec format in the variable pointed to by @ts. + */ +void ktime_get_ts(struct timespec *ts) +{ + struct timespec tomono; + unsigned long seq; + + do { + seq = read_seqbegin(&xtime_lock); + getnstimeofday(ts); + tomono = wall_to_monotonic; + + } while (read_seqretry(&xtime_lock, seq)); + + set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, + ts->tv_nsec + tomono.tv_nsec); +} +EXPORT_SYMBOL_GPL(ktime_get_ts); +#endif /* !GENERIC_TIME */ + +/** + * ktime_get_real - get the real (wall-) time in ktime_t format + * + * returns the time in ktime_t format + */ +ktime_t ktime_get_real(void) +{ + struct timespec now; + + getnstimeofday(&now); + + return timespec_to_ktime(now); +} +EXPORT_SYMBOL_GPL(ktime_get_real); /** * getrawmonotonic - Returns the raw monotonic time in a timespec diff --git a/kernel/timer.c b/kernel/timer.c index a7f07d5a6241..33fc9d175f40 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -72,6 +72,7 @@ struct tvec_base { spinlock_t lock; struct timer_list *running_timer; unsigned long timer_jiffies; + unsigned long next_timer; struct tvec_root tv1; struct tvec tv2; struct tvec tv3; @@ -622,6 +623,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, if (timer_pending(timer)) { detach_timer(timer, 0); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } else { if (pending_only) @@ -663,6 +667,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, } timer->expires = expires; + if (time_before(timer->expires, base->next_timer) && + !tbase_get_deferrable(timer->base)) + base->next_timer = timer->expires; internal_add_timer(base, timer); out_unlock: @@ -781,6 +788,9 @@ void add_timer_on(struct timer_list *timer, int cpu) spin_lock_irqsave(&base->lock, flags); timer_set_base(timer, base); debug_timer_activate(timer); + if (time_before(timer->expires, base->next_timer) && + !tbase_get_deferrable(timer->base)) + base->next_timer = timer->expires; internal_add_timer(base, timer); /* * Check whether the other CPU is idle and needs to be @@ -817,6 +827,9 @@ int del_timer(struct timer_list *timer) base = lock_timer_base(timer, &flags); if (timer_pending(timer)) { detach_timer(timer, 1); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } spin_unlock_irqrestore(&base->lock, flags); @@ -850,6 +863,9 @@ int try_to_del_timer_sync(struct timer_list *timer) ret = 0; if (timer_pending(timer)) { detach_timer(timer, 1); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } out: @@ -1134,7 +1150,9 @@ unsigned long get_next_timer_interrupt(unsigned long now) unsigned long expires; spin_lock(&base->lock); - expires = __next_timer_interrupt(base); + if (time_before_eq(base->next_timer, base->timer_jiffies)) + base->next_timer = __next_timer_interrupt(base); + expires = base->next_timer; spin_unlock(&base->lock); if (time_before_eq(expires, now)) @@ -1523,6 +1541,7 @@ static int __cpuinit init_timers_cpu(int cpu) INIT_LIST_HEAD(base->tv1.vec + j); base->timer_jiffies = jiffies; + base->next_timer = base->timer_jiffies; return 0; } @@ -1535,6 +1554,9 @@ static void migrate_timer_list(struct tvec_base *new_base, struct list_head *hea timer = list_first_entry(head, struct timer_list, entry); detach_timer(timer, 0); timer_set_base(timer, new_base); + if (time_before(timer->expires, new_base->next_timer) && + !tbase_get_deferrable(timer->base)) + new_base->next_timer = timer->expires; internal_add_timer(new_base, timer); } } diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 1e1d23c26308..094863416b2e 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1016,71 +1016,35 @@ static int __ftrace_replace_code(struct dyn_ftrace *rec, int enable) { unsigned long ftrace_addr; - unsigned long ip, fl; + unsigned long flag = 0UL; ftrace_addr = (unsigned long)FTRACE_ADDR; - ip = rec->ip; - /* - * If this record is not to be traced and - * it is not enabled then do nothing. + * If this record is not to be traced or we want to disable it, + * then disable it. * - * If this record is not to be traced and - * it is enabled then disable it. + * If we want to enable it and filtering is off, then enable it. * + * If we want to enable it and filtering is on, enable it only if + * it's filtered */ - if (rec->flags & FTRACE_FL_NOTRACE) { - if (rec->flags & FTRACE_FL_ENABLED) - rec->flags &= ~FTRACE_FL_ENABLED; - else - return 0; - - } else if (ftrace_filtered && enable) { - /* - * Filtering is on: - */ - - fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED); - - /* Record is filtered and enabled, do nothing */ - if (fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) - return 0; - - /* Record is not filtered or enabled, do nothing */ - if (!fl) - return 0; - - /* Record is not filtered but enabled, disable it */ - if (fl == FTRACE_FL_ENABLED) - rec->flags &= ~FTRACE_FL_ENABLED; - else - /* Otherwise record is filtered but not enabled, enable it */ - rec->flags |= FTRACE_FL_ENABLED; - } else { - /* Disable or not filtered */ - - if (enable) { - /* if record is enabled, do nothing */ - if (rec->flags & FTRACE_FL_ENABLED) - return 0; - - rec->flags |= FTRACE_FL_ENABLED; - - } else { + if (enable && !(rec->flags & FTRACE_FL_NOTRACE)) { + if (!ftrace_filtered || (rec->flags & FTRACE_FL_FILTER)) + flag = FTRACE_FL_ENABLED; + } - /* if record is not enabled, do nothing */ - if (!(rec->flags & FTRACE_FL_ENABLED)) - return 0; + /* If the state of this record hasn't changed, then do nothing */ + if ((rec->flags & FTRACE_FL_ENABLED) == flag) + return 0; - rec->flags &= ~FTRACE_FL_ENABLED; - } + if (flag) { + rec->flags |= FTRACE_FL_ENABLED; + return ftrace_make_call(rec, ftrace_addr); } - if (rec->flags & FTRACE_FL_ENABLED) - return ftrace_make_call(rec, ftrace_addr); - else - return ftrace_make_nop(NULL, rec, ftrace_addr); + rec->flags &= ~FTRACE_FL_ENABLED; + return ftrace_make_nop(NULL, rec, ftrace_addr); } static void ftrace_replace_code(int enable) @@ -1375,7 +1339,6 @@ struct ftrace_iterator { unsigned flags; unsigned char buffer[FTRACE_BUFF_MAX+1]; unsigned buffer_idx; - unsigned filtered; }; static void * @@ -1438,18 +1401,13 @@ static int t_hash_show(struct seq_file *m, void *v) { struct ftrace_func_probe *rec; struct hlist_node *hnd = v; - char str[KSYM_SYMBOL_LEN]; rec = hlist_entry(hnd, struct ftrace_func_probe, node); if (rec->ops->print) return rec->ops->print(m, rec->ip, rec->ops, rec->data); - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); - seq_printf(m, "%s:", str); - - kallsyms_lookup((unsigned long)rec->ops->func, NULL, NULL, NULL, str); - seq_printf(m, "%s", str); + seq_printf(m, "%pf:%pf", (void *)rec->ip, (void *)rec->ops->func); if (rec->data) seq_printf(m, ":%p", rec->data); @@ -1547,7 +1505,6 @@ static int t_show(struct seq_file *m, void *v) { struct ftrace_iterator *iter = m->private; struct dyn_ftrace *rec = v; - char str[KSYM_SYMBOL_LEN]; if (iter->flags & FTRACE_ITER_HASH) return t_hash_show(m, v); @@ -1560,9 +1517,7 @@ static int t_show(struct seq_file *m, void *v) if (!rec) return 0; - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); - - seq_printf(m, "%s\n", str); + seq_printf(m, "%pf\n", (void *)rec->ip); return 0; } @@ -2312,7 +2267,6 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, } if (isspace(ch)) { - iter->filtered++; iter->buffer[iter->buffer_idx] = 0; ret = ftrace_process_regex(iter->buffer, iter->buffer_idx, enable); @@ -2443,7 +2397,6 @@ ftrace_regex_release(struct inode *inode, struct file *file, int enable) iter = file->private_data; if (iter->buffer_idx) { - iter->filtered++; iter->buffer[iter->buffer_idx] = 0; ftrace_match_records(iter->buffer, iter->buffer_idx, enable); } @@ -2543,7 +2496,6 @@ static void g_stop(struct seq_file *m, void *p) static int g_show(struct seq_file *m, void *v) { unsigned long *ptr = v; - char str[KSYM_SYMBOL_LEN]; if (!ptr) return 0; @@ -2553,9 +2505,7 @@ static int g_show(struct seq_file *m, void *v) return 0; } - kallsyms_lookup(*ptr, NULL, NULL, NULL, str); - - seq_printf(m, "%s\n", str); + seq_printf(m, "%pf\n", v); return 0; } diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index 1edaa9516e81..dda53ccf749b 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -239,12 +239,52 @@ struct kmemtrace_user_event_alloc { }; static enum print_line_t -kmemtrace_print_alloc_user(struct trace_iterator *iter, - struct kmemtrace_alloc_entry *entry) +kmemtrace_print_alloc(struct trace_iterator *iter, int flags) { - struct kmemtrace_user_event_alloc *ev_alloc; struct trace_seq *s = &iter->seq; + struct kmemtrace_alloc_entry *entry; + int ret; + + trace_assign_type(entry, iter->ent); + + ret = trace_seq_printf(s, "type_id %d call_site %pF ptr %lu " + "bytes_req %lu bytes_alloc %lu gfp_flags %lu node %d\n", + entry->type_id, (void *)entry->call_site, (unsigned long)entry->ptr, + (unsigned long)entry->bytes_req, (unsigned long)entry->bytes_alloc, + (unsigned long)entry->gfp_flags, entry->node); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_free(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_free_entry *entry; + int ret; + + trace_assign_type(entry, iter->ent); + + ret = trace_seq_printf(s, "type_id %d call_site %pF ptr %lu\n", + entry->type_id, (void *)entry->call_site, + (unsigned long)entry->ptr); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_alloc_user(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_alloc_entry *entry; struct kmemtrace_user_event *ev; + struct kmemtrace_user_event_alloc *ev_alloc; + + trace_assign_type(entry, iter->ent); ev = trace_seq_reserve(s, sizeof(*ev)); if (!ev) @@ -271,12 +311,14 @@ kmemtrace_print_alloc_user(struct trace_iterator *iter, } static enum print_line_t -kmemtrace_print_free_user(struct trace_iterator *iter, - struct kmemtrace_free_entry *entry) +kmemtrace_print_free_user(struct trace_iterator *iter, int flags) { struct trace_seq *s = &iter->seq; + struct kmemtrace_free_entry *entry; struct kmemtrace_user_event *ev; + trace_assign_type(entry, iter->ent); + ev = trace_seq_reserve(s, sizeof(*ev)); if (!ev) return TRACE_TYPE_PARTIAL_LINE; @@ -294,12 +336,14 @@ kmemtrace_print_free_user(struct trace_iterator *iter, /* The two other following provide a more minimalistic output */ static enum print_line_t -kmemtrace_print_alloc_compress(struct trace_iterator *iter, - struct kmemtrace_alloc_entry *entry) +kmemtrace_print_alloc_compress(struct trace_iterator *iter) { + struct kmemtrace_alloc_entry *entry; struct trace_seq *s = &iter->seq; int ret; + trace_assign_type(entry, iter->ent); + /* Alloc entry */ ret = trace_seq_printf(s, " + "); if (!ret) @@ -345,29 +389,24 @@ kmemtrace_print_alloc_compress(struct trace_iterator *iter, if (!ret) return TRACE_TYPE_PARTIAL_LINE; - /* Node */ - ret = trace_seq_printf(s, "%4d ", entry->node); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - /* Call site */ - ret = seq_print_ip_sym(s, entry->call_site, 0); + /* Node and call site*/ + ret = trace_seq_printf(s, "%4d %pf\n", entry->node, + (void *)entry->call_site); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - if (!trace_seq_printf(s, "\n")) - return TRACE_TYPE_PARTIAL_LINE; - return TRACE_TYPE_HANDLED; } static enum print_line_t -kmemtrace_print_free_compress(struct trace_iterator *iter, - struct kmemtrace_free_entry *entry) +kmemtrace_print_free_compress(struct trace_iterator *iter) { + struct kmemtrace_free_entry *entry; struct trace_seq *s = &iter->seq; int ret; + trace_assign_type(entry, iter->ent); + /* Free entry */ ret = trace_seq_printf(s, " - "); if (!ret) @@ -401,19 +440,11 @@ kmemtrace_print_free_compress(struct trace_iterator *iter, if (!ret) return TRACE_TYPE_PARTIAL_LINE; - /* Skip node */ - ret = trace_seq_printf(s, " "); + /* Skip node and print call site*/ + ret = trace_seq_printf(s, " %pf\n", (void *)entry->call_site); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - /* Call site */ - ret = seq_print_ip_sym(s, entry->call_site, 0); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - if (!trace_seq_printf(s, "\n")) - return TRACE_TYPE_PARTIAL_LINE; - return TRACE_TYPE_HANDLED; } @@ -421,32 +452,31 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) { struct trace_entry *entry = iter->ent; - switch (entry->type) { - case TRACE_KMEM_ALLOC: { - struct kmemtrace_alloc_entry *field; - - trace_assign_type(field, entry); - if (kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL) - return kmemtrace_print_alloc_compress(iter, field); - else - return kmemtrace_print_alloc_user(iter, field); - } - - case TRACE_KMEM_FREE: { - struct kmemtrace_free_entry *field; - - trace_assign_type(field, entry); - if (kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL) - return kmemtrace_print_free_compress(iter, field); - else - return kmemtrace_print_free_user(iter, field); - } + if (!(kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL)) + return TRACE_TYPE_UNHANDLED; + switch (entry->type) { + case TRACE_KMEM_ALLOC: + return kmemtrace_print_alloc_compress(iter); + case TRACE_KMEM_FREE: + return kmemtrace_print_free_compress(iter); default: return TRACE_TYPE_UNHANDLED; } } +static struct trace_event kmem_trace_alloc = { + .type = TRACE_KMEM_ALLOC, + .trace = kmemtrace_print_alloc, + .binary = kmemtrace_print_alloc_user, +}; + +static struct trace_event kmem_trace_free = { + .type = TRACE_KMEM_FREE, + .trace = kmemtrace_print_free, + .binary = kmemtrace_print_free_user, +}; + static struct tracer kmem_tracer __read_mostly = { .name = "kmemtrace", .init = kmem_trace_init, @@ -463,6 +493,21 @@ void kmemtrace_init(void) static int __init init_kmem_tracer(void) { - return register_tracer(&kmem_tracer); + if (!register_ftrace_event(&kmem_trace_alloc)) { + pr_warning("Warning: could not register kmem events\n"); + return 1; + } + + if (!register_ftrace_event(&kmem_trace_free)) { + pr_warning("Warning: could not register kmem events\n"); + return 1; + } + + if (!register_tracer(&kmem_tracer)) { + pr_warning("Warning: could not register the kmem tracer\n"); + return 1; + } + + return 0; } device_initcall(init_kmem_tracer); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index a330513d96ce..da2c59d8f486 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -322,6 +322,14 @@ struct buffer_data_page { unsigned char data[]; /* data of buffer page */ }; +/* + * Note, the buffer_page list must be first. The buffer pages + * are allocated in cache lines, which means that each buffer + * page will be at the beginning of a cache line, and thus + * the least significant bits will be zero. We use this to + * add flags in the list struct pointers, to make the ring buffer + * lockless. + */ struct buffer_page { struct list_head list; /* list of buffer pages */ local_t write; /* index for next write */ @@ -330,6 +338,21 @@ struct buffer_page { struct buffer_data_page *page; /* Actual data page */ }; +/* + * The buffer page counters, write and entries, must be reset + * atomically when crossing page boundaries. To synchronize this + * update, two counters are inserted into the number. One is + * the actual counter for the write position or count on the page. + * + * The other is a counter of updaters. Before an update happens + * the update partition of the counter is incremented. This will + * allow the updater to update the counter atomically. + * + * The counter is 20 bits, and the state data is 12. + */ +#define RB_WRITE_MASK 0xfffff +#define RB_WRITE_INTCNT (1 << 20) + static void rb_init_page(struct buffer_data_page *bpage) { local_set(&bpage->commit, 0); @@ -403,21 +426,20 @@ int ring_buffer_print_page_header(struct trace_seq *s) struct ring_buffer_per_cpu { int cpu; struct ring_buffer *buffer; - spinlock_t reader_lock; /* serialize readers */ + spinlock_t reader_lock; /* serialize readers */ raw_spinlock_t lock; struct lock_class_key lock_key; - struct list_head pages; + struct list_head *pages; struct buffer_page *head_page; /* read from head */ struct buffer_page *tail_page; /* write to tail */ struct buffer_page *commit_page; /* committed pages */ struct buffer_page *reader_page; - unsigned long nmi_dropped; - unsigned long commit_overrun; - unsigned long overrun; - unsigned long read; + local_t commit_overrun; + local_t overrun; local_t entries; local_t committing; local_t commits; + unsigned long read; u64 write_stamp; u64 read_stamp; atomic_t record_disabled; @@ -489,6 +511,390 @@ void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, } EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp); +/* + * Making the ring buffer lockless makes things tricky. + * Although writes only happen on the CPU that they are on, + * and they only need to worry about interrupts. Reads can + * happen on any CPU. + * + * The reader page is always off the ring buffer, but when the + * reader finishes with a page, it needs to swap its page with + * a new one from the buffer. The reader needs to take from + * the head (writes go to the tail). But if a writer is in overwrite + * mode and wraps, it must push the head page forward. + * + * Here lies the problem. + * + * The reader must be careful to replace only the head page, and + * not another one. As described at the top of the file in the + * ASCII art, the reader sets its old page to point to the next + * page after head. It then sets the page after head to point to + * the old reader page. But if the writer moves the head page + * during this operation, the reader could end up with the tail. + * + * We use cmpxchg to help prevent this race. We also do something + * special with the page before head. We set the LSB to 1. + * + * When the writer must push the page forward, it will clear the + * bit that points to the head page, move the head, and then set + * the bit that points to the new head page. + * + * We also don't want an interrupt coming in and moving the head + * page on another writer. Thus we use the second LSB to catch + * that too. Thus: + * + * head->list->prev->next bit 1 bit 0 + * ------- ------- + * Normal page 0 0 + * Points to head page 0 1 + * New head page 1 0 + * + * Note we can not trust the prev pointer of the head page, because: + * + * +----+ +-----+ +-----+ + * | |------>| T |---X--->| N | + * | |<------| | | | + * +----+ +-----+ +-----+ + * ^ ^ | + * | +-----+ | | + * +----------| R |----------+ | + * | |<-----------+ + * +-----+ + * + * Key: ---X--> HEAD flag set in pointer + * T Tail page + * R Reader page + * N Next page + * + * (see __rb_reserve_next() to see where this happens) + * + * What the above shows is that the reader just swapped out + * the reader page with a page in the buffer, but before it + * could make the new header point back to the new page added + * it was preempted by a writer. The writer moved forward onto + * the new page added by the reader and is about to move forward + * again. + * + * You can see, it is legitimate for the previous pointer of + * the head (or any page) not to point back to itself. But only + * temporarially. + */ + +#define RB_PAGE_NORMAL 0UL +#define RB_PAGE_HEAD 1UL +#define RB_PAGE_UPDATE 2UL + + +#define RB_FLAG_MASK 3UL + +/* PAGE_MOVED is not part of the mask */ +#define RB_PAGE_MOVED 4UL + +/* + * rb_list_head - remove any bit + */ +static struct list_head *rb_list_head(struct list_head *list) +{ + unsigned long val = (unsigned long)list; + + return (struct list_head *)(val & ~RB_FLAG_MASK); +} + +/* + * rb_is_head_page - test if the give page is the head page + * + * Because the reader may move the head_page pointer, we can + * not trust what the head page is (it may be pointing to + * the reader page). But if the next page is a header page, + * its flags will be non zero. + */ +static int inline +rb_is_head_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *page, struct list_head *list) +{ + unsigned long val; + + val = (unsigned long)list->next; + + if ((val & ~RB_FLAG_MASK) != (unsigned long)&page->list) + return RB_PAGE_MOVED; + + return val & RB_FLAG_MASK; +} + +/* + * rb_is_reader_page + * + * The unique thing about the reader page, is that, if the + * writer is ever on it, the previous pointer never points + * back to the reader page. + */ +static int rb_is_reader_page(struct buffer_page *page) +{ + struct list_head *list = page->list.prev; + + return rb_list_head(list->next) != &page->list; +} + +/* + * rb_set_list_to_head - set a list_head to be pointing to head. + */ +static void rb_set_list_to_head(struct ring_buffer_per_cpu *cpu_buffer, + struct list_head *list) +{ + unsigned long *ptr; + + ptr = (unsigned long *)&list->next; + *ptr |= RB_PAGE_HEAD; + *ptr &= ~RB_PAGE_UPDATE; +} + +/* + * rb_head_page_activate - sets up head page + */ +static void rb_head_page_activate(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct buffer_page *head; + + head = cpu_buffer->head_page; + if (!head) + return; + + /* + * Set the previous list pointer to have the HEAD flag. + */ + rb_set_list_to_head(cpu_buffer, head->list.prev); +} + +static void rb_list_head_clear(struct list_head *list) +{ + unsigned long *ptr = (unsigned long *)&list->next; + + *ptr &= ~RB_FLAG_MASK; +} + +/* + * rb_head_page_dactivate - clears head page ptr (for free list) + */ +static void +rb_head_page_deactivate(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct list_head *hd; + + /* Go through the whole list and clear any pointers found. */ + rb_list_head_clear(cpu_buffer->pages); + + list_for_each(hd, cpu_buffer->pages) + rb_list_head_clear(hd); +} + +static int rb_head_page_set(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag, int new_flag) +{ + struct list_head *list; + unsigned long val = (unsigned long)&head->list; + unsigned long ret; + + list = &prev->list; + + val &= ~RB_FLAG_MASK; + + ret = (unsigned long)cmpxchg(&list->next, + val | old_flag, val | new_flag); + + /* check if the reader took the page */ + if ((ret & ~RB_FLAG_MASK) != val) + return RB_PAGE_MOVED; + + return ret & RB_FLAG_MASK; +} + +static int rb_head_page_set_update(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_UPDATE); +} + +static int rb_head_page_set_head(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_HEAD); +} + +static int rb_head_page_set_normal(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_NORMAL); +} + +static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page **bpage) +{ + struct list_head *p = rb_list_head((*bpage)->list.next); + + *bpage = list_entry(p, struct buffer_page, list); +} + +static struct buffer_page * +rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct buffer_page *head; + struct buffer_page *page; + struct list_head *list; + int i; + + if (RB_WARN_ON(cpu_buffer, !cpu_buffer->head_page)) + return NULL; + + /* sanity check */ + list = cpu_buffer->pages; + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev->next) != list)) + return NULL; + + page = head = cpu_buffer->head_page; + /* + * It is possible that the writer moves the header behind + * where we started, and we miss in one loop. + * A second loop should grab the header, but we'll do + * three loops just because I'm paranoid. + */ + for (i = 0; i < 3; i++) { + do { + if (rb_is_head_page(cpu_buffer, page, page->list.prev)) { + cpu_buffer->head_page = page; + return page; + } + rb_inc_page(cpu_buffer, &page); + } while (page != head); + } + + RB_WARN_ON(cpu_buffer, 1); + + return NULL; +} + +static int rb_head_page_replace(struct buffer_page *old, + struct buffer_page *new) +{ + unsigned long *ptr = (unsigned long *)&old->list.prev->next; + unsigned long val; + unsigned long ret; + + val = *ptr & ~RB_FLAG_MASK; + val |= RB_PAGE_HEAD; + + ret = cmpxchg(ptr, val, &new->list); + + return ret == val; +} + +/* + * rb_tail_page_update - move the tail page forward + * + * Returns 1 if moved tail page, 0 if someone else did. + */ +static int rb_tail_page_update(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *tail_page, + struct buffer_page *next_page) +{ + struct buffer_page *old_tail; + unsigned long old_entries; + unsigned long old_write; + int ret = 0; + + /* + * The tail page now needs to be moved forward. + * + * We need to reset the tail page, but without messing + * with possible erasing of data brought in by interrupts + * that have moved the tail page and are currently on it. + * + * We add a counter to the write field to denote this. + */ + old_write = local_add_return(RB_WRITE_INTCNT, &next_page->write); + old_entries = local_add_return(RB_WRITE_INTCNT, &next_page->entries); + + /* + * Just make sure we have seen our old_write and synchronize + * with any interrupts that come in. + */ + barrier(); + + /* + * If the tail page is still the same as what we think + * it is, then it is up to us to update the tail + * pointer. + */ + if (tail_page == cpu_buffer->tail_page) { + /* Zero the write counter */ + unsigned long val = old_write & ~RB_WRITE_MASK; + unsigned long eval = old_entries & ~RB_WRITE_MASK; + + /* + * This will only succeed if an interrupt did + * not come in and change it. In which case, we + * do not want to modify it. + * + * We add (void) to let the compiler know that we do not care + * about the return value of these functions. We use the + * cmpxchg to only update if an interrupt did not already + * do it for us. If the cmpxchg fails, we don't care. + */ + (void)local_cmpxchg(&next_page->write, old_write, val); + (void)local_cmpxchg(&next_page->entries, old_entries, eval); + + /* + * No need to worry about races with clearing out the commit. + * it only can increment when a commit takes place. But that + * only happens in the outer most nested commit. + */ + local_set(&next_page->page->commit, 0); + + old_tail = cmpxchg(&cpu_buffer->tail_page, + tail_page, next_page); + + if (old_tail == tail_page) + ret = 1; + } + + return ret; +} + +static int rb_check_bpage(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *bpage) +{ + unsigned long val = (unsigned long)bpage; + + if (RB_WARN_ON(cpu_buffer, val & RB_FLAG_MASK)) + return 1; + + return 0; +} + +/** + * rb_check_list - make sure a pointer to a list has the last bits zero + */ +static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer, + struct list_head *list) +{ + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev) != list->prev)) + return 1; + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->next) != list->next)) + return 1; + return 0; +} + /** * check_pages - integrity check of buffer pages * @cpu_buffer: CPU buffer with pages to test @@ -498,14 +904,19 @@ EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp); */ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) { - struct list_head *head = &cpu_buffer->pages; + struct list_head *head = cpu_buffer->pages; struct buffer_page *bpage, *tmp; + rb_head_page_deactivate(cpu_buffer); + if (RB_WARN_ON(cpu_buffer, head->next->prev != head)) return -1; if (RB_WARN_ON(cpu_buffer, head->prev->next != head)) return -1; + if (rb_check_list(cpu_buffer, head)) + return -1; + list_for_each_entry_safe(bpage, tmp, head, list) { if (RB_WARN_ON(cpu_buffer, bpage->list.next->prev != &bpage->list)) @@ -513,25 +924,33 @@ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) if (RB_WARN_ON(cpu_buffer, bpage->list.prev->next != &bpage->list)) return -1; + if (rb_check_list(cpu_buffer, &bpage->list)) + return -1; } + rb_head_page_activate(cpu_buffer); + return 0; } static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) { - struct list_head *head = &cpu_buffer->pages; struct buffer_page *bpage, *tmp; unsigned long addr; LIST_HEAD(pages); unsigned i; + WARN_ON(!nr_pages); + for (i = 0; i < nr_pages; i++) { bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu_buffer->cpu)); if (!bpage) goto free_pages; + + rb_check_bpage(cpu_buffer, bpage); + list_add(&bpage->list, &pages); addr = __get_free_page(GFP_KERNEL); @@ -541,7 +960,13 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, rb_init_page(bpage->page); } - list_splice(&pages, head); + /* + * The ring buffer page list is a circular list that does not + * start and end with a list head. All page list items point to + * other pages. + */ + cpu_buffer->pages = pages.next; + list_del(&pages); rb_check_pages(cpu_buffer); @@ -573,13 +998,14 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) spin_lock_init(&cpu_buffer->reader_lock); lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key); cpu_buffer->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; - INIT_LIST_HEAD(&cpu_buffer->pages); bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); if (!bpage) goto fail_free_buffer; + rb_check_bpage(cpu_buffer, bpage); + cpu_buffer->reader_page = bpage; addr = __get_free_page(GFP_KERNEL); if (!addr) @@ -594,9 +1020,11 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) goto fail_free_reader; cpu_buffer->head_page - = list_entry(cpu_buffer->pages.next, struct buffer_page, list); + = list_entry(cpu_buffer->pages, struct buffer_page, list); cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page; + rb_head_page_activate(cpu_buffer); + return cpu_buffer; fail_free_reader: @@ -609,15 +1037,22 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) { - struct list_head *head = &cpu_buffer->pages; + struct list_head *head = cpu_buffer->pages; struct buffer_page *bpage, *tmp; free_buffer_page(cpu_buffer->reader_page); - list_for_each_entry_safe(bpage, tmp, head, list) { - list_del_init(&bpage->list); + rb_head_page_deactivate(cpu_buffer); + + if (head) { + list_for_each_entry_safe(bpage, tmp, head, list) { + list_del_init(&bpage->list); + free_buffer_page(bpage); + } + bpage = list_entry(head, struct buffer_page, list); free_buffer_page(bpage); } + kfree(cpu_buffer); } @@ -760,15 +1195,17 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); + rb_head_page_deactivate(cpu_buffer); + for (i = 0; i < nr_pages; i++) { - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages))) return; - p = cpu_buffer->pages.next; + p = cpu_buffer->pages->next; bpage = list_entry(p, struct buffer_page, list); list_del_init(&bpage->list); free_buffer_page(bpage); } - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages))) return; rb_reset_cpu(cpu_buffer); @@ -790,15 +1227,19 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); + spin_lock_irq(&cpu_buffer->reader_lock); + rb_head_page_deactivate(cpu_buffer); + for (i = 0; i < nr_pages; i++) { if (RB_WARN_ON(cpu_buffer, list_empty(pages))) return; p = pages->next; bpage = list_entry(p, struct buffer_page, list); list_del_init(&bpage->list); - list_add_tail(&bpage->list, &cpu_buffer->pages); + list_add_tail(&bpage->list, cpu_buffer->pages); } rb_reset_cpu(cpu_buffer); + spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); @@ -949,21 +1390,14 @@ rb_reader_event(struct ring_buffer_per_cpu *cpu_buffer) } static inline struct ring_buffer_event * -rb_head_event(struct ring_buffer_per_cpu *cpu_buffer) -{ - return __rb_page_index(cpu_buffer->head_page, - cpu_buffer->head_page->read); -} - -static inline struct ring_buffer_event * rb_iter_head_event(struct ring_buffer_iter *iter) { return __rb_page_index(iter->head_page, iter->head); } -static inline unsigned rb_page_write(struct buffer_page *bpage) +static inline unsigned long rb_page_write(struct buffer_page *bpage) { - return local_read(&bpage->write); + return local_read(&bpage->write) & RB_WRITE_MASK; } static inline unsigned rb_page_commit(struct buffer_page *bpage) @@ -971,6 +1405,11 @@ static inline unsigned rb_page_commit(struct buffer_page *bpage) return local_read(&bpage->page->commit); } +static inline unsigned long rb_page_entries(struct buffer_page *bpage) +{ + return local_read(&bpage->entries) & RB_WRITE_MASK; +} + /* Size is determined by what has been commited */ static inline unsigned rb_page_size(struct buffer_page *bpage) { @@ -983,22 +1422,6 @@ rb_commit_index(struct ring_buffer_per_cpu *cpu_buffer) return rb_page_commit(cpu_buffer->commit_page); } -static inline unsigned rb_head_size(struct ring_buffer_per_cpu *cpu_buffer) -{ - return rb_page_commit(cpu_buffer->head_page); -} - -static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer, - struct buffer_page **bpage) -{ - struct list_head *p = (*bpage)->list.next; - - if (p == &cpu_buffer->pages) - p = p->next; - - *bpage = list_entry(p, struct buffer_page, list); -} - static inline unsigned rb_event_index(struct ring_buffer_event *event) { @@ -1024,6 +1447,8 @@ rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer, static void rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) { + unsigned long max_count; + /* * We only race with interrupts and NMIs on this CPU. * If we own the commit event, then we can commit @@ -1033,9 +1458,16 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) * assign the commit to the tail. */ again: + max_count = cpu_buffer->buffer->pages * 100; + while (cpu_buffer->commit_page != cpu_buffer->tail_page) { - cpu_buffer->commit_page->page->commit = - cpu_buffer->commit_page->write; + if (RB_WARN_ON(cpu_buffer, !(--max_count))) + return; + if (RB_WARN_ON(cpu_buffer, + rb_is_reader_page(cpu_buffer->tail_page))) + return; + local_set(&cpu_buffer->commit_page->page->commit, + rb_page_write(cpu_buffer->commit_page)); rb_inc_page(cpu_buffer, &cpu_buffer->commit_page); cpu_buffer->write_stamp = cpu_buffer->commit_page->page->time_stamp; @@ -1044,8 +1476,12 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) } while (rb_commit_index(cpu_buffer) != rb_page_write(cpu_buffer->commit_page)) { - cpu_buffer->commit_page->page->commit = - cpu_buffer->commit_page->write; + + local_set(&cpu_buffer->commit_page->page->commit, + rb_page_write(cpu_buffer->commit_page)); + RB_WARN_ON(cpu_buffer, + local_read(&cpu_buffer->commit_page->page->commit) & + ~RB_WRITE_MASK); barrier(); } @@ -1078,7 +1514,7 @@ static void rb_inc_iter(struct ring_buffer_iter *iter) * to the head page instead of next. */ if (iter->head_page == cpu_buffer->reader_page) - iter->head_page = cpu_buffer->head_page; + iter->head_page = rb_set_head_page(cpu_buffer); else rb_inc_page(cpu_buffer, &iter->head_page); @@ -1122,6 +1558,163 @@ rb_update_event(struct ring_buffer_event *event, } } +/* + * rb_handle_head_page - writer hit the head page + * + * Returns: +1 to retry page + * 0 to continue + * -1 on error + */ +static int +rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *tail_page, + struct buffer_page *next_page) +{ + struct buffer_page *new_head; + int entries; + int type; + int ret; + + entries = rb_page_entries(next_page); + + /* + * The hard part is here. We need to move the head + * forward, and protect against both readers on + * other CPUs and writers coming in via interrupts. + */ + type = rb_head_page_set_update(cpu_buffer, next_page, tail_page, + RB_PAGE_HEAD); + + /* + * type can be one of four: + * NORMAL - an interrupt already moved it for us + * HEAD - we are the first to get here. + * UPDATE - we are the interrupt interrupting + * a current move. + * MOVED - a reader on another CPU moved the next + * pointer to its reader page. Give up + * and try again. + */ + + switch (type) { + case RB_PAGE_HEAD: + /* + * We changed the head to UPDATE, thus + * it is our responsibility to update + * the counters. + */ + local_add(entries, &cpu_buffer->overrun); + + /* + * The entries will be zeroed out when we move the + * tail page. + */ + + /* still more to do */ + break; + + case RB_PAGE_UPDATE: + /* + * This is an interrupt that interrupt the + * previous update. Still more to do. + */ + break; + case RB_PAGE_NORMAL: + /* + * An interrupt came in before the update + * and processed this for us. + * Nothing left to do. + */ + return 1; + case RB_PAGE_MOVED: + /* + * The reader is on another CPU and just did + * a swap with our next_page. + * Try again. + */ + return 1; + default: + RB_WARN_ON(cpu_buffer, 1); /* WTF??? */ + return -1; + } + + /* + * Now that we are here, the old head pointer is + * set to UPDATE. This will keep the reader from + * swapping the head page with the reader page. + * The reader (on another CPU) will spin till + * we are finished. + * + * We just need to protect against interrupts + * doing the job. We will set the next pointer + * to HEAD. After that, we set the old pointer + * to NORMAL, but only if it was HEAD before. + * otherwise we are an interrupt, and only + * want the outer most commit to reset it. + */ + new_head = next_page; + rb_inc_page(cpu_buffer, &new_head); + + ret = rb_head_page_set_head(cpu_buffer, new_head, next_page, + RB_PAGE_NORMAL); + + /* + * Valid returns are: + * HEAD - an interrupt came in and already set it. + * NORMAL - One of two things: + * 1) We really set it. + * 2) A bunch of interrupts came in and moved + * the page forward again. + */ + switch (ret) { + case RB_PAGE_HEAD: + case RB_PAGE_NORMAL: + /* OK */ + break; + default: + RB_WARN_ON(cpu_buffer, 1); + return -1; + } + + /* + * It is possible that an interrupt came in, + * set the head up, then more interrupts came in + * and moved it again. When we get back here, + * the page would have been set to NORMAL but we + * just set it back to HEAD. + * + * How do you detect this? Well, if that happened + * the tail page would have moved. + */ + if (ret == RB_PAGE_NORMAL) { + /* + * If the tail had moved passed next, then we need + * to reset the pointer. + */ + if (cpu_buffer->tail_page != tail_page && + cpu_buffer->tail_page != next_page) + rb_head_page_set_normal(cpu_buffer, new_head, + next_page, + RB_PAGE_HEAD); + } + + /* + * If this was the outer most commit (the one that + * changed the original pointer from HEAD to UPDATE), + * then it is up to us to reset it to NORMAL. + */ + if (type == RB_PAGE_HEAD) { + ret = rb_head_page_set_normal(cpu_buffer, next_page, + tail_page, + RB_PAGE_UPDATE); + if (RB_WARN_ON(cpu_buffer, + ret != RB_PAGE_UPDATE)) + return -1; + } + + return 0; +} + static unsigned rb_calculate_event_length(unsigned length) { struct ring_buffer_event event; /* Used only for sizeof array */ @@ -1200,96 +1793,93 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, struct buffer_page *commit_page, struct buffer_page *tail_page, u64 *ts) { - struct buffer_page *next_page, *head_page, *reader_page; struct ring_buffer *buffer = cpu_buffer->buffer; - bool lock_taken = false; - unsigned long flags; + struct buffer_page *next_page; + int ret; next_page = tail_page; - local_irq_save(flags); - /* - * Since the write to the buffer is still not - * fully lockless, we must be careful with NMIs. - * The locks in the writers are taken when a write - * crosses to a new page. The locks protect against - * races with the readers (this will soon be fixed - * with a lockless solution). - * - * Because we can not protect against NMIs, and we - * want to keep traces reentrant, we need to manage - * what happens when we are in an NMI. - * - * NMIs can happen after we take the lock. - * If we are in an NMI, only take the lock - * if it is not already taken. Otherwise - * simply fail. - */ - if (unlikely(in_nmi())) { - if (!__raw_spin_trylock(&cpu_buffer->lock)) { - cpu_buffer->nmi_dropped++; - goto out_reset; - } - } else - __raw_spin_lock(&cpu_buffer->lock); - - lock_taken = true; - rb_inc_page(cpu_buffer, &next_page); - head_page = cpu_buffer->head_page; - reader_page = cpu_buffer->reader_page; - - /* we grabbed the lock before incrementing */ - if (RB_WARN_ON(cpu_buffer, next_page == reader_page)) - goto out_reset; - /* * If for some reason, we had an interrupt storm that made * it all the way around the buffer, bail, and warn * about it. */ if (unlikely(next_page == commit_page)) { - cpu_buffer->commit_overrun++; + local_inc(&cpu_buffer->commit_overrun); goto out_reset; } - if (next_page == head_page) { - if (!(buffer->flags & RB_FL_OVERWRITE)) - goto out_reset; - - /* tail_page has not moved yet? */ - if (tail_page == cpu_buffer->tail_page) { - /* count overflows */ - cpu_buffer->overrun += - local_read(&head_page->entries); + /* + * This is where the fun begins! + * + * We are fighting against races between a reader that + * could be on another CPU trying to swap its reader + * page with the buffer head. + * + * We are also fighting against interrupts coming in and + * moving the head or tail on us as well. + * + * If the next page is the head page then we have filled + * the buffer, unless the commit page is still on the + * reader page. + */ + if (rb_is_head_page(cpu_buffer, next_page, &tail_page->list)) { - rb_inc_page(cpu_buffer, &head_page); - cpu_buffer->head_page = head_page; - cpu_buffer->head_page->read = 0; + /* + * If the commit is not on the reader page, then + * move the header page. + */ + if (!rb_is_reader_page(cpu_buffer->commit_page)) { + /* + * If we are not in overwrite mode, + * this is easy, just stop here. + */ + if (!(buffer->flags & RB_FL_OVERWRITE)) + goto out_reset; + + ret = rb_handle_head_page(cpu_buffer, + tail_page, + next_page); + if (ret < 0) + goto out_reset; + if (ret) + goto out_again; + } else { + /* + * We need to be careful here too. The + * commit page could still be on the reader + * page. We could have a small buffer, and + * have filled up the buffer with events + * from interrupts and such, and wrapped. + * + * Note, if the tail page is also the on the + * reader_page, we let it move out. + */ + if (unlikely((cpu_buffer->commit_page != + cpu_buffer->tail_page) && + (cpu_buffer->commit_page == + cpu_buffer->reader_page))) { + local_inc(&cpu_buffer->commit_overrun); + goto out_reset; + } } } - /* - * If the tail page is still the same as what we think - * it is, then it is up to us to update the tail - * pointer. - */ - if (tail_page == cpu_buffer->tail_page) { - local_set(&next_page->write, 0); - local_set(&next_page->entries, 0); - local_set(&next_page->page->commit, 0); - cpu_buffer->tail_page = next_page; - - /* reread the time stamp */ + ret = rb_tail_page_update(cpu_buffer, tail_page, next_page); + if (ret) { + /* + * Nested commits always have zero deltas, so + * just reread the time stamp + */ *ts = rb_time_stamp(buffer, cpu_buffer->cpu); - cpu_buffer->tail_page->page->time_stamp = *ts; + next_page->page->time_stamp = *ts; } - rb_reset_tail(cpu_buffer, tail_page, tail, length); + out_again: - __raw_spin_unlock(&cpu_buffer->lock); - local_irq_restore(flags); + rb_reset_tail(cpu_buffer, tail_page, tail, length); /* fail and let the caller try again */ return ERR_PTR(-EAGAIN); @@ -1298,9 +1888,6 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, /* reset write */ rb_reset_tail(cpu_buffer, tail_page, tail, length); - if (likely(lock_taken)) - __raw_spin_unlock(&cpu_buffer->lock); - local_irq_restore(flags); return NULL; } @@ -1317,6 +1904,9 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, barrier(); tail_page = cpu_buffer->tail_page; write = local_add_return(length, &tail_page->write); + + /* set write to only the index of the write */ + write &= RB_WRITE_MASK; tail = write - length; /* See if we shot pass the end of this buffer page */ @@ -1361,12 +1951,16 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, bpage = cpu_buffer->tail_page; if (bpage->page == (void *)addr && rb_page_write(bpage) == old_index) { + unsigned long write_mask = + local_read(&bpage->write) & ~RB_WRITE_MASK; /* * This is on the tail page. It is possible that * a write could come in and move the tail page * and write to the next page. That is fine * because we just shorten what is on this page. */ + old_index += write_mask; + new_index += write_mask; index = local_cmpxchg(&bpage->write, old_index, new_index); if (index == old_index) return 1; @@ -1875,9 +2469,13 @@ EXPORT_SYMBOL_GPL(ring_buffer_write); static int rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) { struct buffer_page *reader = cpu_buffer->reader_page; - struct buffer_page *head = cpu_buffer->head_page; + struct buffer_page *head = rb_set_head_page(cpu_buffer); struct buffer_page *commit = cpu_buffer->commit_page; + /* In case of error, head will be NULL */ + if (unlikely(!head)) + return 1; + return reader->read == rb_page_commit(reader) && (commit == reader || (commit == head && @@ -1968,7 +2566,7 @@ unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu) return 0; cpu_buffer = buffer->buffers[cpu]; - ret = (local_read(&cpu_buffer->entries) - cpu_buffer->overrun) + ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun)) - cpu_buffer->read; return ret; @@ -1989,33 +2587,13 @@ unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu) return 0; cpu_buffer = buffer->buffers[cpu]; - ret = cpu_buffer->overrun; + ret = local_read(&cpu_buffer->overrun); return ret; } EXPORT_SYMBOL_GPL(ring_buffer_overrun_cpu); /** - * ring_buffer_nmi_dropped_cpu - get the number of nmis that were dropped - * @buffer: The ring buffer - * @cpu: The per CPU buffer to get the number of overruns from - */ -unsigned long ring_buffer_nmi_dropped_cpu(struct ring_buffer *buffer, int cpu) -{ - struct ring_buffer_per_cpu *cpu_buffer; - unsigned long ret; - - if (!cpumask_test_cpu(cpu, buffer->cpumask)) - return 0; - - cpu_buffer = buffer->buffers[cpu]; - ret = cpu_buffer->nmi_dropped; - - return ret; -} -EXPORT_SYMBOL_GPL(ring_buffer_nmi_dropped_cpu); - -/** * ring_buffer_commit_overrun_cpu - get the number of overruns caused by commits * @buffer: The ring buffer * @cpu: The per CPU buffer to get the number of overruns from @@ -2030,7 +2608,7 @@ ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu) return 0; cpu_buffer = buffer->buffers[cpu]; - ret = cpu_buffer->commit_overrun; + ret = local_read(&cpu_buffer->commit_overrun); return ret; } @@ -2053,7 +2631,7 @@ unsigned long ring_buffer_entries(struct ring_buffer *buffer) for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; entries += (local_read(&cpu_buffer->entries) - - cpu_buffer->overrun) - cpu_buffer->read; + local_read(&cpu_buffer->overrun)) - cpu_buffer->read; } return entries; @@ -2076,7 +2654,7 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer) /* if you care about this being correct, lock the buffer */ for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; - overruns += cpu_buffer->overrun; + overruns += local_read(&cpu_buffer->overrun); } return overruns; @@ -2089,8 +2667,10 @@ static void rb_iter_reset(struct ring_buffer_iter *iter) /* Iterator usage is expected to have record disabled */ if (list_empty(&cpu_buffer->reader_page->list)) { - iter->head_page = cpu_buffer->head_page; - iter->head = cpu_buffer->head_page->read; + iter->head_page = rb_set_head_page(cpu_buffer); + if (unlikely(!iter->head_page)) + return; + iter->head = iter->head_page->read; } else { iter->head_page = cpu_buffer->reader_page; iter->head = cpu_buffer->reader_page->read; @@ -2207,6 +2787,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) struct buffer_page *reader = NULL; unsigned long flags; int nr_loops = 0; + int ret; local_irq_save(flags); __raw_spin_lock(&cpu_buffer->lock); @@ -2240,30 +2821,56 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) goto out; /* - * Splice the empty reader page into the list around the head. * Reset the reader page to size zero. */ + 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); - reader = cpu_buffer->head_page; + spin: + /* + * Splice the empty reader page into the list around the head. + */ + reader = rb_set_head_page(cpu_buffer); cpu_buffer->reader_page->list.next = reader->list.next; cpu_buffer->reader_page->list.prev = reader->list.prev; - 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->pages just needs to point to the buffer, it + * has no specific buffer page to point to. Lets move it out + * of our way so we don't accidently swap it. + */ + cpu_buffer->pages = reader->list.prev; - /* Make the reader page now replace the head */ - reader->list.prev->next = &cpu_buffer->reader_page->list; - reader->list.next->prev = &cpu_buffer->reader_page->list; + /* The reader page will be pointing to the new head */ + rb_set_list_to_head(cpu_buffer, &cpu_buffer->reader_page->list); + + /* + * Here's the tricky part. + * + * We need to move the pointer past the header page. + * But we can only do that if a writer is not currently + * moving it. The page before the header page has the + * flag bit '1' set if it is pointing to the page we want. + * but if the writer is in the process of moving it + * than it will be '2' or already moved '0'. + */ + + ret = rb_head_page_replace(reader, cpu_buffer->reader_page); /* - * If the tail is on the reader, then we must set the head - * to the inserted page, otherwise we set it one before. + * If we did not convert it, then we must try again. */ - cpu_buffer->head_page = cpu_buffer->reader_page; + if (!ret) + goto spin; - if (cpu_buffer->commit_page != reader) - rb_inc_page(cpu_buffer, &cpu_buffer->head_page); + /* + * Yeah! We succeeded in replacing the page. + * + * Now make the new head point back to the reader page. + */ + reader->list.next->prev = &cpu_buffer->reader_page->list; + rb_inc_page(cpu_buffer, &cpu_buffer->head_page); /* Finally update the reader page to the new head */ cpu_buffer->reader_page = reader; @@ -2717,8 +3324,10 @@ EXPORT_SYMBOL_GPL(ring_buffer_size); static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) { + rb_head_page_deactivate(cpu_buffer); + cpu_buffer->head_page - = list_entry(cpu_buffer->pages.next, struct buffer_page, list); + = list_entry(cpu_buffer->pages, struct buffer_page, list); local_set(&cpu_buffer->head_page->write, 0); local_set(&cpu_buffer->head_page->entries, 0); local_set(&cpu_buffer->head_page->page->commit, 0); @@ -2734,16 +3343,17 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) local_set(&cpu_buffer->reader_page->page->commit, 0); cpu_buffer->reader_page->read = 0; - cpu_buffer->nmi_dropped = 0; - cpu_buffer->commit_overrun = 0; - cpu_buffer->overrun = 0; - cpu_buffer->read = 0; + local_set(&cpu_buffer->commit_overrun, 0); + local_set(&cpu_buffer->overrun, 0); local_set(&cpu_buffer->entries, 0); local_set(&cpu_buffer->committing, 0); local_set(&cpu_buffer->commits, 0); + cpu_buffer->read = 0; cpu_buffer->write_stamp = 0; cpu_buffer->read_stamp = 0; + + rb_head_page_activate(cpu_buffer); } /** @@ -3091,7 +3701,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer, read = 0; } else { /* update the entry counter */ - cpu_buffer->read += local_read(&reader->entries); + cpu_buffer->read += rb_page_entries(reader); /* swap the pages */ rb_init_page(bpage); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 6904e381f51c..e2ab900e6fec 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -50,7 +50,7 @@ unsigned long __read_mostly tracing_thresh; * On boot up, the ring buffer is set to the minimum size, so that * we do not waste memory on systems that are not using tracing. */ -static int ring_buffer_expanded; +int ring_buffer_expanded; /* * We need to change this state when a selftest is running. @@ -64,7 +64,7 @@ static bool __read_mostly tracing_selftest_running; /* * If a tracer is running, we do not want to run SELFTEST. */ -static bool __read_mostly tracing_selftest_disabled; +bool __read_mostly tracing_selftest_disabled; /* For tracers that don't implement custom flags */ static struct tracer_opt dummy_tracer_opt[] = { @@ -89,7 +89,7 @@ static int dummy_set_flag(u32 old_flags, u32 bit, int set) */ static int tracing_disabled = 1; -static DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); +DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); static inline void ftrace_disable_cpu(void) { @@ -867,10 +867,6 @@ struct ring_buffer_event *trace_buffer_lock_reserve(struct trace_array *tr, return event; } -static void ftrace_trace_stack(struct trace_array *tr, - unsigned long flags, int skip, int pc); -static void ftrace_trace_userstack(struct trace_array *tr, - unsigned long flags, int pc); static inline void __trace_buffer_unlock_commit(struct trace_array *tr, struct ring_buffer_event *event, @@ -947,54 +943,6 @@ trace_function(struct trace_array *tr, ring_buffer_unlock_commit(tr->buffer, event); } -#ifdef CONFIG_FUNCTION_GRAPH_TRACER -static int __trace_graph_entry(struct trace_array *tr, - struct ftrace_graph_ent *trace, - unsigned long flags, - int pc) -{ - struct ftrace_event_call *call = &event_funcgraph_entry; - struct ring_buffer_event *event; - struct ftrace_graph_ent_entry *entry; - - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) - return 0; - - event = trace_buffer_lock_reserve(&global_trace, TRACE_GRAPH_ENT, - sizeof(*entry), flags, pc); - if (!event) - return 0; - entry = ring_buffer_event_data(event); - entry->graph_ent = *trace; - if (!filter_current_check_discard(call, entry, event)) - ring_buffer_unlock_commit(global_trace.buffer, event); - - return 1; -} - -static void __trace_graph_return(struct trace_array *tr, - struct ftrace_graph_ret *trace, - unsigned long flags, - int pc) -{ - struct ftrace_event_call *call = &event_funcgraph_exit; - struct ring_buffer_event *event; - struct ftrace_graph_ret_entry *entry; - - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) - return; - - event = trace_buffer_lock_reserve(&global_trace, TRACE_GRAPH_RET, - sizeof(*entry), flags, pc); - if (!event) - return; - entry = ring_buffer_event_data(event); - entry->ret = *trace; - if (!filter_current_check_discard(call, entry, event)) - ring_buffer_unlock_commit(global_trace.buffer, event); -} -#endif - void ftrace(struct trace_array *tr, struct trace_array_cpu *data, unsigned long ip, unsigned long parent_ip, unsigned long flags, @@ -1004,11 +952,11 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data, trace_function(tr, ip, parent_ip, flags, pc); } +#ifdef CONFIG_STACKTRACE static void __ftrace_trace_stack(struct trace_array *tr, unsigned long flags, int skip, int pc) { -#ifdef CONFIG_STACKTRACE struct ftrace_event_call *call = &event_kernel_stack; struct ring_buffer_event *event; struct stack_entry *entry; @@ -1029,12 +977,10 @@ static void __ftrace_trace_stack(struct trace_array *tr, save_stack_trace(&trace); if (!filter_check_discard(call, entry, tr->buffer, event)) ring_buffer_unlock_commit(tr->buffer, event); -#endif } -static void ftrace_trace_stack(struct trace_array *tr, - unsigned long flags, - int skip, int pc) +void ftrace_trace_stack(struct trace_array *tr, unsigned long flags, int skip, + int pc) { if (!(trace_flags & TRACE_ITER_STACKTRACE)) return; @@ -1042,17 +988,14 @@ static void ftrace_trace_stack(struct trace_array *tr, __ftrace_trace_stack(tr, flags, skip, pc); } -void __trace_stack(struct trace_array *tr, - unsigned long flags, - int skip, int pc) +void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, + int pc) { __ftrace_trace_stack(tr, flags, skip, pc); } -static void ftrace_trace_userstack(struct trace_array *tr, - unsigned long flags, int pc) +void ftrace_trace_userstack(struct trace_array *tr, unsigned long flags, int pc) { -#ifdef CONFIG_STACKTRACE struct ftrace_event_call *call = &event_user_stack; struct ring_buffer_event *event; struct userstack_entry *entry; @@ -1077,7 +1020,6 @@ static void ftrace_trace_userstack(struct trace_array *tr, save_stack_trace_user(&trace); if (!filter_check_discard(call, entry, tr->buffer, event)) ring_buffer_unlock_commit(tr->buffer, event); -#endif } #ifdef UNUSED @@ -1087,6 +1029,8 @@ static void __trace_userstack(struct trace_array *tr, unsigned long flags) } #endif /* UNUSED */ +#endif /* CONFIG_STACKTRACE */ + static void ftrace_trace_special(void *__tr, unsigned long arg1, unsigned long arg2, unsigned long arg3, @@ -1115,62 +1059,6 @@ __trace_special(void *__tr, void *__data, } void -tracing_sched_switch_trace(struct trace_array *tr, - struct task_struct *prev, - struct task_struct *next, - unsigned long flags, int pc) -{ - struct ftrace_event_call *call = &event_context_switch; - struct ring_buffer_event *event; - struct ctx_switch_entry *entry; - - event = trace_buffer_lock_reserve(tr, TRACE_CTX, - sizeof(*entry), flags, pc); - if (!event) - return; - entry = ring_buffer_event_data(event); - entry->prev_pid = prev->pid; - entry->prev_prio = prev->prio; - entry->prev_state = prev->state; - entry->next_pid = next->pid; - entry->next_prio = next->prio; - entry->next_state = next->state; - entry->next_cpu = task_cpu(next); - - if (!filter_check_discard(call, entry, tr->buffer, event)) - trace_buffer_unlock_commit(tr, event, flags, pc); -} - -void -tracing_sched_wakeup_trace(struct trace_array *tr, - struct task_struct *wakee, - struct task_struct *curr, - unsigned long flags, int pc) -{ - struct ftrace_event_call *call = &event_wakeup; - struct ring_buffer_event *event; - struct ctx_switch_entry *entry; - - event = trace_buffer_lock_reserve(tr, TRACE_WAKE, - sizeof(*entry), flags, pc); - if (!event) - return; - entry = ring_buffer_event_data(event); - entry->prev_pid = curr->pid; - entry->prev_prio = curr->prio; - entry->prev_state = curr->state; - entry->next_pid = wakee->pid; - entry->next_prio = wakee->prio; - entry->next_state = wakee->state; - entry->next_cpu = task_cpu(wakee); - - if (!filter_check_discard(call, entry, tr->buffer, event)) - ring_buffer_unlock_commit(tr->buffer, event); - ftrace_trace_stack(tr, flags, 6, pc); - ftrace_trace_userstack(tr, flags, pc); -} - -void ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { struct trace_array *tr = &global_trace; @@ -1194,68 +1082,6 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) local_irq_restore(flags); } -#ifdef CONFIG_FUNCTION_GRAPH_TRACER -int trace_graph_entry(struct ftrace_graph_ent *trace) -{ - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int ret; - int cpu; - int pc; - - if (!ftrace_trace_task(current)) - return 0; - - if (!ftrace_graph_addr(trace->func)) - return 0; - - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - pc = preempt_count(); - ret = __trace_graph_entry(tr, trace, flags, pc); - } else { - ret = 0; - } - /* Only do the atomic if it is not already set */ - if (!test_tsk_trace_graph(current)) - set_tsk_trace_graph(current); - - atomic_dec(&data->disabled); - local_irq_restore(flags); - - return ret; -} - -void trace_graph_return(struct ftrace_graph_ret *trace) -{ - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int cpu; - int pc; - - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - pc = preempt_count(); - __trace_graph_return(tr, trace, flags, pc); - } - if (!trace->depth) - clear_tsk_trace_graph(current); - atomic_dec(&data->disabled); - local_irq_restore(flags); -} -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ - - /** * trace_vbprintk - write binary msg to tracing buffer * @@ -2255,8 +2081,8 @@ tracing_trace_options_read(struct file *filp, char __user *ubuf, len += 3; /* "no" and newline */ } - /* +2 for \n and \0 */ - buf = kmalloc(len + 2, GFP_KERNEL); + /* +1 for \0 */ + buf = kmalloc(len + 1, GFP_KERNEL); if (!buf) { mutex_unlock(&trace_types_lock); return -ENOMEM; @@ -2279,7 +2105,7 @@ tracing_trace_options_read(struct file *filp, char __user *ubuf, } mutex_unlock(&trace_types_lock); - WARN_ON(r >= len + 2); + WARN_ON(r >= len + 1); r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); @@ -3631,9 +3457,6 @@ tracing_stats_read(struct file *filp, char __user *ubuf, cnt = ring_buffer_commit_overrun_cpu(tr->buffer, cpu); trace_seq_printf(s, "commit overrun: %ld\n", cnt); - cnt = ring_buffer_nmi_dropped_cpu(tr->buffer, cpu); - trace_seq_printf(s, "nmi dropped: %ld\n", cnt); - count = simple_read_from_buffer(ubuf, count, ppos, s->buffer, s->len); kfree(s); @@ -4271,7 +4094,6 @@ void ftrace_dump(void) __init static int tracer_alloc_buffers(void) { - struct trace_array_cpu *data; int ring_buf_size; int i; int ret = -ENOMEM; @@ -4320,7 +4142,7 @@ __init static int tracer_alloc_buffers(void) /* Allocate the first page for all buffers */ for_each_tracing_cpu(i) { - data = global_trace.data[i] = &per_cpu(global_trace_cpu, i); + global_trace.data[i] = &per_cpu(global_trace_cpu, i); max_tr.data[i] = &per_cpu(max_data, i); } diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 8b9f4f6e9559..d682357e4b1f 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -467,6 +467,7 @@ void trace_function(struct trace_array *tr, void trace_graph_return(struct ftrace_graph_ret *trace); int trace_graph_entry(struct ftrace_graph_ent *trace); +void set_graph_array(struct trace_array *tr); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); @@ -485,9 +486,31 @@ void update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu); void update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu); -void __trace_stack(struct trace_array *tr, - unsigned long flags, - int skip, int pc); +#ifdef CONFIG_STACKTRACE +void ftrace_trace_stack(struct trace_array *tr, unsigned long flags, + int skip, int pc); + +void ftrace_trace_userstack(struct trace_array *tr, unsigned long flags, + int pc); + +void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, + int pc); +#else +static inline void ftrace_trace_stack(struct trace_array *tr, + unsigned long flags, int skip, int pc) +{ +} + +static inline void ftrace_trace_userstack(struct trace_array *tr, + unsigned long flags, int pc) +{ +} + +static inline void __trace_stack(struct trace_array *tr, unsigned long flags, + int skip, int pc) +{ +} +#endif /* CONFIG_STACKTRACE */ extern cycle_t ftrace_now(int cpu); @@ -513,6 +536,10 @@ extern unsigned long ftrace_update_tot_cnt; extern int DYN_FTRACE_TEST_NAME(void); #endif +extern int ring_buffer_expanded; +extern bool tracing_selftest_disabled; +DECLARE_PER_CPU(local_t, ftrace_cpu_disabled); + #ifdef CONFIG_FTRACE_STARTUP_TEST extern int trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr); @@ -743,13 +770,15 @@ struct event_filter { int n_preds; struct filter_pred **preds; char *filter_string; + bool no_reset; }; struct event_subsystem { struct list_head list; const char *name; struct dentry *entry; - void *filter; + struct event_filter *filter; + int nr_events; }; struct filter_pred; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index e75276a49cf5..e0cbede96783 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -17,6 +17,8 @@ #include <linux/ctype.h> #include <linux/delay.h> +#include <asm/setup.h> + #include "trace_output.h" #define TRACE_SYSTEM "TRACE_SYSTEM" @@ -849,8 +851,10 @@ event_subsystem_dir(const char *name, struct dentry *d_events) /* First see if we did not already create this dir */ list_for_each_entry(system, &event_subsystems, list) { - if (strcmp(system->name, name) == 0) + if (strcmp(system->name, name) == 0) { + system->nr_events++; return system->entry; + } } /* need to create new entry */ @@ -869,6 +873,7 @@ event_subsystem_dir(const char *name, struct dentry *d_events) return d_events; } + system->nr_events = 1; system->name = kstrdup(name, GFP_KERNEL); if (!system->name) { debugfs_remove(system->entry); @@ -987,6 +992,32 @@ struct ftrace_module_file_ops { struct file_operations filter; }; +static void remove_subsystem_dir(const char *name) +{ + struct event_subsystem *system; + + if (strcmp(name, TRACE_SYSTEM) == 0) + return; + + list_for_each_entry(system, &event_subsystems, list) { + if (strcmp(system->name, name) == 0) { + if (!--system->nr_events) { + struct event_filter *filter = system->filter; + + debugfs_remove_recursive(system->entry); + list_del(&system->list); + if (filter) { + kfree(filter->filter_string); + kfree(filter); + } + kfree(system->name); + kfree(system); + } + break; + } + } +} + static struct ftrace_module_file_ops * trace_create_file_ops(struct module *mod) { @@ -1077,6 +1108,7 @@ static void trace_module_remove_events(struct module *mod) list_del(&call->list); trace_destroy_fields(call); destroy_preds(call); + remove_subsystem_dir(call->system); } } @@ -1133,6 +1165,18 @@ struct notifier_block trace_module_nb = { extern struct ftrace_event_call __start_ftrace_events[]; extern struct ftrace_event_call __stop_ftrace_events[]; +static char bootup_event_buf[COMMAND_LINE_SIZE] __initdata; + +static __init int setup_trace_event(char *str) +{ + strlcpy(bootup_event_buf, str, COMMAND_LINE_SIZE); + ring_buffer_expanded = 1; + tracing_selftest_disabled = 1; + + return 1; +} +__setup("trace_event=", setup_trace_event); + static __init int event_trace_init(void) { struct ftrace_event_call *call; @@ -1140,6 +1184,8 @@ static __init int event_trace_init(void) struct dentry *entry; struct dentry *d_events; int ret; + char *buf = bootup_event_buf; + char *token; d_tracer = tracing_init_dentry(); if (!d_tracer) @@ -1185,6 +1231,19 @@ static __init int event_trace_init(void) &ftrace_event_format_fops); } + while (true) { + token = strsep(&buf, ","); + + if (!token) + break; + if (!*token) + continue; + + ret = ftrace_set_clr_event(token, 1); + if (ret) + pr_warning("Failed to enable trace event: %s\n", token); + } + ret = register_module_notifier(&trace_module_nb); if (ret) pr_warning("Failed to register trace events module notifier\n"); @@ -1392,10 +1451,10 @@ static __init void event_trace_self_test_with_function(void) static __init int event_trace_self_tests_init(void) { - - event_trace_self_tests(); - - event_trace_self_test_with_function(); + if (!tracing_selftest_disabled) { + event_trace_self_tests(); + event_trace_self_test_with_function(); + } return 0; } diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index f32dc9d1ea7b..490337abed75 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -176,11 +176,13 @@ static int filter_pred_string(struct filter_pred *pred, void *event, static int filter_pred_strloc(struct filter_pred *pred, void *event, int val1, int val2) { - unsigned short str_loc = *(unsigned short *)(event + pred->offset); + u32 str_item = *(u32 *)(event + pred->offset); + int str_loc = str_item & 0xffff; + int str_len = str_item >> 16; char *addr = (char *)(event + str_loc); int cmp, match; - cmp = strncmp(addr, pred->str_val, pred->str_len); + cmp = strncmp(addr, pred->str_val, str_len); match = (!cmp) ^ pred->not; @@ -418,24 +420,29 @@ oom: } EXPORT_SYMBOL_GPL(init_preds); -static void filter_free_subsystem_preds(struct event_subsystem *system) +enum { + FILTER_DISABLE_ALL, + FILTER_INIT_NO_RESET, + FILTER_SKIP_NO_RESET, +}; + +static void filter_free_subsystem_preds(struct event_subsystem *system, + int flag) { - struct event_filter *filter = system->filter; struct ftrace_event_call *call; - int i; - - if (filter->n_preds) { - for (i = 0; i < filter->n_preds; i++) - filter_free_pred(filter->preds[i]); - kfree(filter->preds); - filter->preds = NULL; - filter->n_preds = 0; - } list_for_each_entry(call, &ftrace_events, list) { if (!call->define_fields) continue; + if (flag == FILTER_INIT_NO_RESET) { + call->filter->no_reset = false; + continue; + } + + if (flag == FILTER_SKIP_NO_RESET && call->filter->no_reset) + continue; + if (!strcmp(call->system, system->name)) { filter_disable_preds(call); remove_filter_string(call->filter); @@ -537,7 +544,8 @@ static filter_pred_fn_t select_comparison_fn(int op, int field_size, static int filter_add_pred(struct filter_parse_state *ps, struct ftrace_event_call *call, - struct filter_pred *pred) + struct filter_pred *pred, + bool dry_run) { struct ftrace_event_field *field; filter_pred_fn_t fn; @@ -549,10 +557,12 @@ static int filter_add_pred(struct filter_parse_state *ps, if (pred->op == OP_AND) { pred->pop_n = 2; - return filter_add_pred_fn(ps, call, pred, filter_pred_and); + fn = filter_pred_and; + goto add_pred_fn; } else if (pred->op == OP_OR) { pred->pop_n = 2; - return filter_add_pred_fn(ps, call, pred, filter_pred_or); + fn = filter_pred_or; + goto add_pred_fn; } field = find_event_field(call, pred->field_name); @@ -575,9 +585,6 @@ static int filter_add_pred(struct filter_parse_state *ps, else fn = filter_pred_strloc; pred->str_len = field->size; - if (pred->op == OP_NE) - pred->not = 1; - return filter_add_pred_fn(ps, call, pred, fn); } else { if (field->is_signed) ret = strict_strtoll(pred->str_val, 0, &val); @@ -588,41 +595,33 @@ static int filter_add_pred(struct filter_parse_state *ps, return -EINVAL; } pred->val = val; - } - fn = select_comparison_fn(pred->op, field->size, field->is_signed); - if (!fn) { - parse_error(ps, FILT_ERR_INVALID_OP, 0); - return -EINVAL; + fn = select_comparison_fn(pred->op, field->size, + field->is_signed); + if (!fn) { + parse_error(ps, FILT_ERR_INVALID_OP, 0); + return -EINVAL; + } } if (pred->op == OP_NE) pred->not = 1; - return filter_add_pred_fn(ps, call, pred, fn); +add_pred_fn: + if (!dry_run) + return filter_add_pred_fn(ps, call, pred, fn); + return 0; } static int filter_add_subsystem_pred(struct filter_parse_state *ps, struct event_subsystem *system, struct filter_pred *pred, - char *filter_string) + char *filter_string, + bool dry_run) { - struct event_filter *filter = system->filter; struct ftrace_event_call *call; int err = 0; - - if (!filter->preds) { - filter->preds = kzalloc(MAX_FILTER_PRED * sizeof(pred), - GFP_KERNEL); - - if (!filter->preds) - return -ENOMEM; - } - - if (filter->n_preds == MAX_FILTER_PRED) { - parse_error(ps, FILT_ERR_TOO_MANY_PREDS, 0); - return -ENOSPC; - } + bool fail = true; list_for_each_entry(call, &ftrace_events, list) { @@ -632,19 +631,24 @@ static int filter_add_subsystem_pred(struct filter_parse_state *ps, if (strcmp(call->system, system->name)) continue; - err = filter_add_pred(ps, call, pred); - if (err) { - filter_free_subsystem_preds(system); - parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0); - goto out; - } - replace_filter_string(call->filter, filter_string); + if (call->filter->no_reset) + continue; + + err = filter_add_pred(ps, call, pred, dry_run); + if (err) + call->filter->no_reset = true; + else + fail = false; + + if (!dry_run) + replace_filter_string(call->filter, filter_string); } - filter->preds[filter->n_preds] = pred; - filter->n_preds++; -out: - return err; + if (fail) { + parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0); + return err; + } + return 0; } static void parse_init(struct filter_parse_state *ps, @@ -1003,12 +1007,14 @@ static int check_preds(struct filter_parse_state *ps) static int replace_preds(struct event_subsystem *system, struct ftrace_event_call *call, struct filter_parse_state *ps, - char *filter_string) + char *filter_string, + bool dry_run) { char *operand1 = NULL, *operand2 = NULL; struct filter_pred *pred; struct postfix_elt *elt; int err; + int n_preds = 0; err = check_preds(ps); if (err) @@ -1027,24 +1033,14 @@ static int replace_preds(struct event_subsystem *system, continue; } + if (n_preds++ == MAX_FILTER_PRED) { + parse_error(ps, FILT_ERR_TOO_MANY_PREDS, 0); + return -ENOSPC; + } + if (elt->op == OP_AND || elt->op == OP_OR) { pred = create_logical_pred(elt->op); - if (!pred) - return -ENOMEM; - if (call) { - err = filter_add_pred(ps, call, pred); - filter_free_pred(pred); - } else { - err = filter_add_subsystem_pred(ps, system, - pred, filter_string); - if (err) - filter_free_pred(pred); - } - if (err) - return err; - - operand1 = operand2 = NULL; - continue; + goto add_pred; } if (!operand1 || !operand2) { @@ -1053,17 +1049,15 @@ static int replace_preds(struct event_subsystem *system, } pred = create_pred(elt->op, operand1, operand2); +add_pred: if (!pred) return -ENOMEM; - if (call) { - err = filter_add_pred(ps, call, pred); - filter_free_pred(pred); - } else { + if (call) + err = filter_add_pred(ps, call, pred, false); + else err = filter_add_subsystem_pred(ps, system, pred, - filter_string); - if (err) - filter_free_pred(pred); - } + filter_string, dry_run); + filter_free_pred(pred); if (err) return err; @@ -1103,7 +1097,7 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) goto out; } - err = replace_preds(NULL, call, ps, filter_string); + err = replace_preds(NULL, call, ps, filter_string, false); if (err) append_filter_err(ps, call->filter); @@ -1127,7 +1121,7 @@ int apply_subsystem_event_filter(struct event_subsystem *system, mutex_lock(&event_mutex); if (!strcmp(strstrip(filter_string), "0")) { - filter_free_subsystem_preds(system); + filter_free_subsystem_preds(system, FILTER_DISABLE_ALL); remove_filter_string(system->filter); mutex_unlock(&event_mutex); return 0; @@ -1138,7 +1132,6 @@ int apply_subsystem_event_filter(struct event_subsystem *system, if (!ps) goto out_unlock; - filter_free_subsystem_preds(system); replace_filter_string(system->filter, filter_string); parse_init(ps, filter_ops, filter_string); @@ -1148,9 +1141,23 @@ int apply_subsystem_event_filter(struct event_subsystem *system, goto out; } - err = replace_preds(system, NULL, ps, filter_string); - if (err) + filter_free_subsystem_preds(system, FILTER_INIT_NO_RESET); + + /* try to see the filter can be applied to which events */ + err = replace_preds(system, NULL, ps, filter_string, true); + if (err) { append_filter_err(ps, system->filter); + goto out; + } + + filter_free_subsystem_preds(system, FILTER_SKIP_NO_RESET); + + /* really apply the filter to the events */ + err = replace_preds(system, NULL, ps, filter_string, false); + if (err) { + append_filter_err(ps, system->filter); + filter_free_subsystem_preds(system, 2); + } out: filter_opstack_clear(ps); diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 75ef000613c3..5b01b94518fc 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -288,11 +288,9 @@ static int ftrace_trace_onoff_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - char str[KSYM_SYMBOL_LEN]; long count = (long)data; - kallsyms_lookup(ip, NULL, NULL, NULL, str); - seq_printf(m, "%s:", str); + seq_printf(m, "%pf:", (void *)ip); if (ops == &traceon_probe_ops) seq_printf(m, "traceon"); diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 420ec3487579..3f4a251b7d16 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -52,7 +52,7 @@ static struct tracer_flags tracer_flags = { .opts = trace_opts }; -/* pid on the last trace processed */ +static struct trace_array *graph_array; /* Add a function return address to the trace stack on thread info.*/ @@ -166,10 +166,121 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer) return ret; } +static int __trace_graph_entry(struct trace_array *tr, + struct ftrace_graph_ent *trace, + unsigned long flags, + int pc) +{ + struct ftrace_event_call *call = &event_funcgraph_entry; + struct ring_buffer_event *event; + struct ftrace_graph_ent_entry *entry; + + if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + return 0; + + event = trace_buffer_lock_reserve(tr, TRACE_GRAPH_ENT, + sizeof(*entry), flags, pc); + if (!event) + return 0; + entry = ring_buffer_event_data(event); + entry->graph_ent = *trace; + if (!filter_current_check_discard(call, entry, event)) + ring_buffer_unlock_commit(tr->buffer, event); + + return 1; +} + +int trace_graph_entry(struct ftrace_graph_ent *trace) +{ + struct trace_array *tr = graph_array; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int ret; + int cpu; + int pc; + + if (unlikely(!tr)) + return 0; + + if (!ftrace_trace_task(current)) + return 0; + + if (!ftrace_graph_addr(trace->func)) + return 0; + + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (likely(disabled == 1)) { + pc = preempt_count(); + ret = __trace_graph_entry(tr, trace, flags, pc); + } else { + ret = 0; + } + /* Only do the atomic if it is not already set */ + if (!test_tsk_trace_graph(current)) + set_tsk_trace_graph(current); + + atomic_dec(&data->disabled); + local_irq_restore(flags); + + return ret; +} + +static void __trace_graph_return(struct trace_array *tr, + struct ftrace_graph_ret *trace, + unsigned long flags, + int pc) +{ + struct ftrace_event_call *call = &event_funcgraph_exit; + struct ring_buffer_event *event; + struct ftrace_graph_ret_entry *entry; + + if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + return; + + event = trace_buffer_lock_reserve(tr, TRACE_GRAPH_RET, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->ret = *trace; + if (!filter_current_check_discard(call, entry, event)) + ring_buffer_unlock_commit(tr->buffer, event); +} + +void trace_graph_return(struct ftrace_graph_ret *trace) +{ + struct trace_array *tr = graph_array; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (likely(disabled == 1)) { + pc = preempt_count(); + __trace_graph_return(tr, trace, flags, pc); + } + if (!trace->depth) + clear_tsk_trace_graph(current); + atomic_dec(&data->disabled); + local_irq_restore(flags); +} + static int graph_trace_init(struct trace_array *tr) { - int ret = register_ftrace_graph(&trace_graph_return, - &trace_graph_entry); + int ret; + + graph_array = tr; + ret = register_ftrace_graph(&trace_graph_return, + &trace_graph_entry); if (ret) return ret; tracing_start_cmdline_record(); @@ -177,49 +288,30 @@ static int graph_trace_init(struct trace_array *tr) return 0; } +void set_graph_array(struct trace_array *tr) +{ + graph_array = tr; +} + static void graph_trace_reset(struct trace_array *tr) { tracing_stop_cmdline_record(); unregister_ftrace_graph(); } -static inline int log10_cpu(int nb) -{ - if (nb / 100) - return 3; - if (nb / 10) - return 2; - return 1; -} +static int max_bytes_for_cpu; static enum print_line_t print_graph_cpu(struct trace_seq *s, int cpu) { - int i; int ret; - int log10_this = log10_cpu(cpu); - int log10_all = log10_cpu(cpumask_weight(cpu_online_mask)); - /* * Start with a space character - to make it stand out * to the right a bit when trace output is pasted into * email: */ - ret = trace_seq_printf(s, " "); - - /* - * Tricky - we space the CPU field according to the max - * number of online CPUs. On a 2-cpu system it would take - * a maximum of 1 digit - on a 128 cpu system it would - * take up to 3 digits: - */ - for (i = 0; i < log10_all - log10_this; i++) { - ret = trace_seq_printf(s, " "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } - ret = trace_seq_printf(s, "%d) ", cpu); + ret = trace_seq_printf(s, " %*d) ", max_bytes_for_cpu, cpu); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -565,11 +657,7 @@ print_graph_entry_leaf(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } - ret = seq_print_ip_sym(s, call->func, 0); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, "();\n"); + ret = trace_seq_printf(s, "%pf();\n", (void *)call->func); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -612,11 +700,7 @@ print_graph_entry_nested(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } - ret = seq_print_ip_sym(s, call->func, 0); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, "() {\n"); + ret = trace_seq_printf(s, "%pf() {\n", (void *)call->func); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -934,6 +1018,8 @@ static struct tracer graph_trace __read_mostly = { static __init int init_graph_trace(void) { + max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1); + return register_tracer(&graph_trace); } diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index a98106dd979c..e1285d7b5488 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -20,6 +20,34 @@ static int sched_ref; static DEFINE_MUTEX(sched_register_mutex); static int sched_stopped; + +void +tracing_sched_switch_trace(struct trace_array *tr, + struct task_struct *prev, + struct task_struct *next, + unsigned long flags, int pc) +{ + struct ftrace_event_call *call = &event_context_switch; + struct ring_buffer_event *event; + struct ctx_switch_entry *entry; + + event = trace_buffer_lock_reserve(tr, TRACE_CTX, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->prev_pid = prev->pid; + entry->prev_prio = prev->prio; + entry->prev_state = prev->state; + entry->next_pid = next->pid; + entry->next_prio = next->prio; + entry->next_state = next->state; + entry->next_cpu = task_cpu(next); + + if (!filter_check_discard(call, entry, tr->buffer, event)) + trace_buffer_unlock_commit(tr, event, flags, pc); +} + static void probe_sched_switch(struct rq *__rq, struct task_struct *prev, struct task_struct *next) @@ -49,6 +77,35 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev, local_irq_restore(flags); } +void +tracing_sched_wakeup_trace(struct trace_array *tr, + struct task_struct *wakee, + struct task_struct *curr, + unsigned long flags, int pc) +{ + struct ftrace_event_call *call = &event_wakeup; + struct ring_buffer_event *event; + struct ctx_switch_entry *entry; + + event = trace_buffer_lock_reserve(tr, TRACE_WAKE, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->prev_pid = curr->pid; + entry->prev_prio = curr->prio; + entry->prev_state = curr->state; + entry->next_pid = wakee->pid; + entry->next_prio = wakee->prio; + entry->next_state = wakee->state; + entry->next_cpu = task_cpu(wakee); + + if (!filter_check_discard(call, entry, tr->buffer, event)) + ring_buffer_unlock_commit(tr->buffer, event); + ftrace_trace_stack(tr, flags, 6, pc); + ftrace_trace_userstack(tr, flags, pc); +} + static void probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success) { diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 00dd6485bdd7..d2cdbabb4ead 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -288,6 +288,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, * to detect and recover from possible hangs */ tracing_reset_online_cpus(tr); + set_graph_array(tr); ret = register_ftrace_graph(&trace_graph_return, &trace_graph_entry_watchdog); if (ret) { diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 6a2a9d484cd6..0da1cff08d67 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -234,15 +234,8 @@ static void t_stop(struct seq_file *m, void *p) static int trace_lookup_stack(struct seq_file *m, long i) { unsigned long addr = stack_dump_trace[i]; -#ifdef CONFIG_KALLSYMS - char str[KSYM_SYMBOL_LEN]; - sprint_symbol(str, addr); - - return seq_printf(m, "%s\n", str); -#else - return seq_printf(m, "%p\n", (void*)addr); -#endif + return seq_printf(m, "%pF\n", (void *)addr); } static void print_disabled(struct seq_file *m) diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c index aea321c82fa0..07c60b09258f 100644 --- a/kernel/trace/trace_stat.c +++ b/kernel/trace/trace_stat.c @@ -49,7 +49,8 @@ static struct dentry *stat_dir; * but it will at least advance closer to the next one * to be released. */ -static struct rb_node *release_next(struct rb_node *node) +static struct rb_node *release_next(struct tracer_stat *ts, + struct rb_node *node) { struct stat_node *snode; struct rb_node *parent = rb_parent(node); @@ -67,6 +68,8 @@ static struct rb_node *release_next(struct rb_node *node) parent->rb_right = NULL; snode = container_of(node, struct stat_node, node); + if (ts->stat_release) + ts->stat_release(snode->stat); kfree(snode); return parent; @@ -78,7 +81,7 @@ static void __reset_stat_session(struct stat_session *session) struct rb_node *node = session->stat_root.rb_node; while (node) - node = release_next(node); + node = release_next(session->ts, node); session->stat_root = RB_ROOT; } diff --git a/kernel/trace/trace_stat.h b/kernel/trace/trace_stat.h index f3546a2cd826..8f03914b9a6a 100644 --- a/kernel/trace/trace_stat.h +++ b/kernel/trace/trace_stat.h @@ -18,6 +18,8 @@ struct tracer_stat { int (*stat_cmp)(void *p1, void *p2); /* Print a stat entry */ int (*stat_show)(struct seq_file *s, void *p); + /* Release an entry */ + void (*stat_release)(void *stat); /* Print the headers of your stat entries */ int (*stat_headers)(struct seq_file *s); }; diff --git a/kernel/trace/trace_workqueue.c b/kernel/trace/trace_workqueue.c index 97fcea4acce1..40cafb07dffd 100644 --- a/kernel/trace/trace_workqueue.c +++ b/kernel/trace/trace_workqueue.c @@ -9,6 +9,7 @@ #include <trace/events/workqueue.h> #include <linux/list.h> #include <linux/percpu.h> +#include <linux/kref.h> #include "trace_stat.h" #include "trace.h" @@ -16,6 +17,7 @@ /* A cpu workqueue thread */ struct cpu_workqueue_stats { struct list_head list; + struct kref kref; int cpu; pid_t pid; /* Can be inserted from interrupt or user context, need to be atomic */ @@ -39,6 +41,11 @@ struct workqueue_global_stats { static DEFINE_PER_CPU(struct workqueue_global_stats, all_workqueue_stat); #define workqueue_cpu_stat(cpu) (&per_cpu(all_workqueue_stat, cpu)) +static void cpu_workqueue_stat_free(struct kref *kref) +{ + kfree(container_of(kref, struct cpu_workqueue_stats, kref)); +} + /* Insertion of a work */ static void probe_workqueue_insertion(struct task_struct *wq_thread, @@ -96,8 +103,8 @@ static void probe_workqueue_creation(struct task_struct *wq_thread, int cpu) return; } INIT_LIST_HEAD(&cws->list); + kref_init(&cws->kref); cws->cpu = cpu; - cws->pid = wq_thread->pid; spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); @@ -118,7 +125,7 @@ static void probe_workqueue_destruction(struct task_struct *wq_thread) list) { if (node->pid == wq_thread->pid) { list_del(&node->list); - kfree(node); + kref_put(&node->kref, cpu_workqueue_stat_free); goto found; } } @@ -137,9 +144,11 @@ static struct cpu_workqueue_stats *workqueue_stat_start_cpu(int cpu) spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); - if (!list_empty(&workqueue_cpu_stat(cpu)->list)) + if (!list_empty(&workqueue_cpu_stat(cpu)->list)) { ret = list_entry(workqueue_cpu_stat(cpu)->list.next, struct cpu_workqueue_stats, list); + kref_get(&ret->kref); + } spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); @@ -162,9 +171,9 @@ static void *workqueue_stat_start(struct tracer_stat *trace) static void *workqueue_stat_next(void *prev, int idx) { struct cpu_workqueue_stats *prev_cws = prev; + struct cpu_workqueue_stats *ret; int cpu = prev_cws->cpu; unsigned long flags; - void *ret = NULL; spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); if (list_is_last(&prev_cws->list, &workqueue_cpu_stat(cpu)->list)) { @@ -175,11 +184,14 @@ static void *workqueue_stat_next(void *prev, int idx) return NULL; } while (!(ret = workqueue_stat_start_cpu(cpu))); return ret; + } else { + ret = list_entry(prev_cws->list.next, + struct cpu_workqueue_stats, list); + kref_get(&ret->kref); } spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); - return list_entry(prev_cws->list.next, struct cpu_workqueue_stats, - list); + return ret; } static int workqueue_stat_show(struct seq_file *s, void *p) @@ -203,6 +215,13 @@ static int workqueue_stat_show(struct seq_file *s, void *p) return 0; } +static void workqueue_stat_release(void *stat) +{ + struct cpu_workqueue_stats *node = stat; + + kref_put(&node->kref, cpu_workqueue_stat_free); +} + static int workqueue_stat_headers(struct seq_file *s) { seq_printf(s, "# CPU INSERTED EXECUTED NAME\n"); @@ -215,6 +234,7 @@ struct tracer_stat workqueue_stats __read_mostly = { .stat_start = workqueue_stat_start, .stat_next = workqueue_stat_next, .stat_show = workqueue_stat_show, + .stat_release = workqueue_stat_release, .stat_headers = workqueue_stat_headers }; diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 0668795d8818..3c44b56b0da7 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -600,7 +600,12 @@ static struct workqueue_struct *keventd_wq __read_mostly; * schedule_work - put work task in global workqueue * @work: job to be done * - * This puts a job in the kernel-global workqueue. + * Returns zero if @work was already on the kernel-global workqueue and + * non-zero otherwise. + * + * This puts a job in the kernel-global workqueue if it was not already + * queued and leaves it in the same position on the kernel-global + * workqueue otherwise. */ int schedule_work(struct work_struct *work) { diff --git a/lib/swiotlb.c b/lib/swiotlb.c index bffe6d7ef9d9..9e2fe3e1804c 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -114,46 +114,11 @@ setup_io_tlb_npages(char *str) __setup("swiotlb=", setup_io_tlb_npages); /* make io_tlb_overflow tunable too? */ -void * __weak __init swiotlb_alloc_boot(size_t size, unsigned long nslabs) -{ - return alloc_bootmem_low_pages(size); -} - -void * __weak swiotlb_alloc(unsigned order, unsigned long nslabs) -{ - return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order); -} - -dma_addr_t __weak swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr) -{ - return paddr; -} - -phys_addr_t __weak swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr) -{ - return baddr; -} - +/* Note that this doesn't work with highmem page */ static dma_addr_t swiotlb_virt_to_bus(struct device *hwdev, volatile void *address) { - return swiotlb_phys_to_bus(hwdev, virt_to_phys(address)); -} - -void * __weak swiotlb_bus_to_virt(struct device *hwdev, dma_addr_t address) -{ - return phys_to_virt(swiotlb_bus_to_phys(hwdev, address)); -} - -int __weak swiotlb_arch_address_needs_mapping(struct device *hwdev, - dma_addr_t addr, size_t size) -{ - return !is_buffer_dma_capable(dma_get_mask(hwdev), addr, size); -} - -int __weak swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size) -{ - return 0; + return phys_to_dma(hwdev, virt_to_phys(address)); } static void swiotlb_print_info(unsigned long bytes) @@ -189,7 +154,7 @@ swiotlb_init_with_default_size(size_t default_size) /* * Get IO TLB memory from the low pages */ - io_tlb_start = swiotlb_alloc_boot(bytes, io_tlb_nslabs); + io_tlb_start = alloc_bootmem_low_pages(bytes); if (!io_tlb_start) panic("Cannot allocate SWIOTLB buffer"); io_tlb_end = io_tlb_start + bytes; @@ -245,7 +210,8 @@ swiotlb_late_init_with_default_size(size_t default_size) bytes = io_tlb_nslabs << IO_TLB_SHIFT; while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) { - io_tlb_start = swiotlb_alloc(order, io_tlb_nslabs); + io_tlb_start = (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, + order); if (io_tlb_start) break; order--; @@ -315,20 +281,10 @@ cleanup1: return -ENOMEM; } -static inline int -address_needs_mapping(struct device *hwdev, dma_addr_t addr, size_t size) +static int is_swiotlb_buffer(phys_addr_t paddr) { - return swiotlb_arch_address_needs_mapping(hwdev, addr, size); -} - -static inline int range_needs_mapping(phys_addr_t paddr, size_t size) -{ - return swiotlb_force || swiotlb_arch_range_needs_mapping(paddr, size); -} - -static int is_swiotlb_buffer(char *addr) -{ - return addr >= io_tlb_start && addr < io_tlb_end; + return paddr >= virt_to_phys(io_tlb_start) && + paddr < virt_to_phys(io_tlb_end); } /* @@ -561,9 +517,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_mask = hwdev->coherent_dma_mask; ret = (void *)__get_free_pages(flags, order); - if (ret && - !is_buffer_dma_capable(dma_mask, swiotlb_virt_to_bus(hwdev, ret), - size)) { + if (ret && swiotlb_virt_to_bus(hwdev, ret) + size > dma_mask) { /* * The allocated memory isn't reachable by the device. */ @@ -585,7 +539,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, dev_addr = swiotlb_virt_to_bus(hwdev, ret); /* Confirm address can be DMA'd by device */ - if (!is_buffer_dma_capable(dma_mask, dev_addr, size)) { + if (dev_addr + size > dma_mask) { printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n", (unsigned long long)dma_mask, (unsigned long long)dev_addr); @@ -601,11 +555,13 @@ EXPORT_SYMBOL(swiotlb_alloc_coherent); void swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr, - dma_addr_t dma_handle) + dma_addr_t dev_addr) { + phys_addr_t paddr = dma_to_phys(hwdev, dev_addr); + WARN_ON(irqs_disabled()); - if (!is_swiotlb_buffer(vaddr)) - free_pages((unsigned long) vaddr, get_order(size)); + if (!is_swiotlb_buffer(paddr)) + free_pages((unsigned long)vaddr, get_order(size)); else /* DMA_TO_DEVICE to avoid memcpy in unmap_single */ do_unmap_single(hwdev, vaddr, size, DMA_TO_DEVICE); @@ -646,7 +602,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page, struct dma_attrs *attrs) { phys_addr_t phys = page_to_phys(page) + offset; - dma_addr_t dev_addr = swiotlb_phys_to_bus(dev, phys); + dma_addr_t dev_addr = phys_to_dma(dev, phys); void *map; BUG_ON(dir == DMA_NONE); @@ -655,8 +611,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page, * we can safely return the device addr and not worry about bounce * buffering it. */ - if (!address_needs_mapping(dev, dev_addr, size) && - !range_needs_mapping(phys, size)) + if (dma_capable(dev, dev_addr, size) && !swiotlb_force) return dev_addr; /* @@ -673,7 +628,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page, /* * Ensure that the address returned is DMA'ble */ - if (address_needs_mapping(dev, dev_addr, size)) + if (!dma_capable(dev, dev_addr, size)) panic("map_single: bounce buffer is not DMA'ble"); return dev_addr; @@ -691,19 +646,25 @@ EXPORT_SYMBOL_GPL(swiotlb_map_page); static void unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir) { - char *dma_addr = swiotlb_bus_to_virt(hwdev, dev_addr); + phys_addr_t paddr = dma_to_phys(hwdev, dev_addr); BUG_ON(dir == DMA_NONE); - if (is_swiotlb_buffer(dma_addr)) { - do_unmap_single(hwdev, dma_addr, size, dir); + if (is_swiotlb_buffer(paddr)) { + do_unmap_single(hwdev, phys_to_virt(paddr), size, dir); return; } if (dir != DMA_FROM_DEVICE) return; - dma_mark_clean(dma_addr, size); + /* + * phys_to_virt doesn't work with hihgmem page but we could + * call dma_mark_clean() with hihgmem page here. However, we + * are fine since dma_mark_clean() is null on POWERPC. We can + * make dma_mark_clean() take a physical address if necessary. + */ + dma_mark_clean(phys_to_virt(paddr), size); } void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, @@ -728,19 +689,19 @@ static void swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir, int target) { - char *dma_addr = swiotlb_bus_to_virt(hwdev, dev_addr); + phys_addr_t paddr = dma_to_phys(hwdev, dev_addr); BUG_ON(dir == DMA_NONE); - if (is_swiotlb_buffer(dma_addr)) { - sync_single(hwdev, dma_addr, size, dir, target); + if (is_swiotlb_buffer(paddr)) { + sync_single(hwdev, phys_to_virt(paddr), size, dir, target); return; } if (dir != DMA_FROM_DEVICE) return; - dma_mark_clean(dma_addr, size); + dma_mark_clean(phys_to_virt(paddr), size); } void @@ -817,10 +778,10 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems, for_each_sg(sgl, sg, nelems, i) { phys_addr_t paddr = sg_phys(sg); - dma_addr_t dev_addr = swiotlb_phys_to_bus(hwdev, paddr); + dma_addr_t dev_addr = phys_to_dma(hwdev, paddr); - if (range_needs_mapping(paddr, sg->length) || - address_needs_mapping(hwdev, dev_addr, sg->length)) { + if (swiotlb_force || + !dma_capable(hwdev, dev_addr, sg->length)) { void *map = map_single(hwdev, sg_phys(sg), sg->length, dir); if (!map) { diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl index 911ba7ffab84..090d300d7394 100755 --- a/scripts/recordmcount.pl +++ b/scripts/recordmcount.pl @@ -57,7 +57,6 @@ # call mcount (offset: 0x5) # [...] # ret -# .globl my_func # other_func: # [...] # call mcount (offset: 0x1b) diff --git a/tools/perf/Documentation/perf-examples.txt b/tools/perf/Documentation/perf-examples.txt new file mode 100644 index 000000000000..8eb6c489fb15 --- /dev/null +++ b/tools/perf/Documentation/perf-examples.txt @@ -0,0 +1,225 @@ + + ------------------------------ + ****** perf by examples ****** + ------------------------------ + +[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ] + + +First, discovery/enumeration of available counters can be done via +'perf list': + +titan:~> perf list + [...] + kmem:kmalloc [Tracepoint event] + kmem:kmem_cache_alloc [Tracepoint event] + kmem:kmalloc_node [Tracepoint event] + kmem:kmem_cache_alloc_node [Tracepoint event] + kmem:kfree [Tracepoint event] + kmem:kmem_cache_free [Tracepoint event] + kmem:mm_page_free_direct [Tracepoint event] + kmem:mm_pagevec_free [Tracepoint event] + kmem:mm_page_alloc [Tracepoint event] + kmem:mm_page_alloc_zone_locked [Tracepoint event] + kmem:mm_page_pcpu_drain [Tracepoint event] + kmem:mm_page_alloc_extfrag [Tracepoint event] + +Then any (or all) of the above event sources can be activated and +measured. For example the page alloc/free properties of a 'hackbench +run' are: + + titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc + -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10 + Time: 0.575 + + Performance counter stats for './hackbench 10': + + 13857 kmem:mm_page_pcpu_drain + 27576 kmem:mm_page_alloc + 6025 kmem:mm_pagevec_free + 20934 kmem:mm_page_free_direct + + 0.613972165 seconds time elapsed + +You can observe the statistical properties as well, by using the +'repeat the workload N times' feature of perf stat: + + titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct ./hackbench 10 + Time: 0.627 + Time: 0.644 + Time: 0.564 + Time: 0.559 + Time: 0.626 + + Performance counter stats for './hackbench 10' (5 runs): + + 12920 kmem:mm_page_pcpu_drain ( +- 3.359% ) + 25035 kmem:mm_page_alloc ( +- 3.783% ) + 6104 kmem:mm_pagevec_free ( +- 0.934% ) + 18376 kmem:mm_page_free_direct ( +- 4.941% ) + + 0.643954516 seconds time elapsed ( +- 2.363% ) + +Furthermore, these tracepoints can be used to sample the workload as +well. For example the page allocations done by a 'git gc' can be +captured the following way: + + titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ] + +To check which functions generated page allocations: + + titan:~/git> perf report + # Samples: 10646 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.57% git-repack /lib64/libc-2.5.so + 21.81% git /lib64/libc-2.5.so + 14.59% git ./git + 11.79% git-repack ./git + 7.12% git /lib64/ld-2.5.so + 3.16% git-repack /lib64/libpthread-2.5.so + 2.09% git-repack /bin/bash + 1.97% rm /lib64/libc-2.5.so + 1.39% mv /lib64/ld-2.5.so + 1.37% mv /lib64/libc-2.5.so + 1.12% git-repack /lib64/ld-2.5.so + 0.95% rm /lib64/ld-2.5.so + 0.90% git-update-serv /lib64/libc-2.5.so + 0.73% git-update-serv /lib64/ld-2.5.so + 0.68% perf /lib64/libpthread-2.5.so + 0.64% git-repack /usr/lib64/libz.so.1.2.3 + +Or to see it on a more finegrained level: + +titan:~/git> perf report --sort comm,dso,symbol +# Samples: 10646 +# +# Overhead Command Shared Object Symbol +# ........ ............... .......................... ...... +# + 9.35% git-repack ./git [.] insert_obj_hash + 9.12% git ./git [.] insert_obj_hash + 7.31% git /lib64/libc-2.5.so [.] memcpy + 6.34% git-repack /lib64/libc-2.5.so [.] _int_malloc + 6.24% git-repack /lib64/libc-2.5.so [.] memcpy + 5.82% git-repack /lib64/libc-2.5.so [.] __GI___fork + 5.47% git /lib64/libc-2.5.so [.] _int_malloc + 2.99% git /lib64/libc-2.5.so [.] memset + +Furthermore, call-graph sampling can be done too, of page +allocations - to see precisely what kind of page allocations there +are: + + titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ] + + titan:~/git> perf report -g + # Samples: 10686 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.25% git-repack /lib64/libc-2.5.so + | + |--50.00%-- _int_free + | + |--37.50%-- __GI___fork + | make_child + | + |--12.50%-- ptmalloc_unlock_all2 + | make_child + | + --6.25%-- __GI_strcpy + 21.61% git /lib64/libc-2.5.so + | + |--30.00%-- __GI_read + | | + | --83.33%-- git_config_from_file + | git_config + | | + [...] + +Or you can observe the whole system's page allocations for 10 +seconds: + +titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e +kmem:mm_page_alloc -e kmem:mm_pagevec_free -e +kmem:mm_page_free_direct sleep 10 + + Performance counter stats for 'sleep 10': + + 171585 kmem:mm_page_pcpu_drain + 322114 kmem:mm_page_alloc + 73623 kmem:mm_pagevec_free + 254115 kmem:mm_page_free_direct + + 10.000591410 seconds time elapsed + +Or observe how fluctuating the page allocations are, via statistical +analysis done over ten 1-second intervals: + + titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct sleep 1 + + Performance counter stats for 'sleep 1' (10 runs): + + 17254 kmem:mm_page_pcpu_drain ( +- 3.709% ) + 34394 kmem:mm_page_alloc ( +- 4.617% ) + 7509 kmem:mm_pagevec_free ( +- 4.820% ) + 25653 kmem:mm_page_free_direct ( +- 3.672% ) + + 1.058135029 seconds time elapsed ( +- 3.089% ) + +Or you can annotate the recorded 'git gc' run on a per symbol basis +and check which instructions/source-code generated page allocations: + + titan:~/git> perf annotate __GI___fork + ------------------------------------------------ + Percent | Source code & Disassembly of libc-2.5.so + ------------------------------------------------ + : + : + : Disassembly of section .plt: + : Disassembly of section .text: + : + : 00000031a2e95560 <__fork>: + [...] + 0.00 : 31a2e95602: b8 38 00 00 00 mov $0x38,%eax + 0.00 : 31a2e95607: 0f 05 syscall + 83.42 : 31a2e95609: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax + 0.00 : 31a2e9560f: 0f 87 4d 01 00 00 ja 31a2e95762 <__fork+0x202> + 0.00 : 31a2e95615: 85 c0 test %eax,%eax + +( this shows that 83.42% of __GI___fork's page allocations come from + the 0x38 system call it performs. ) + +etc. etc. - a lot more is possible. I could list a dozen of +other different usecases straight away - neither of which is +possible via /proc/vmstat. + +/proc/vmstat is not in the same league really, in terms of +expressive power of system analysis and performance +analysis. + +All that the above results needed were those new tracepoints +in include/tracing/events/kmem.h. + + Ingo + + diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index e72e93110782..370344afb5b2 100644 --- a/tools/perf/Documentation/perf-report.txt +++ b/tools/perf/Documentation/perf-report.txt @@ -27,6 +27,9 @@ OPTIONS -n --show-nr-samples Show the number of samples for each symbol +-T +--threads + Show per-thread event counters -C:: --comms=:: Only consider symbols in these comms. CSV that understands diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index 0d74346d21ab..484080dd5b6f 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -40,7 +40,7 @@ OPTIONS -a:: system-wide collection --S:: +-c:: scale counter values EXAMPLES diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt index 539d01289725..4a7d558dc309 100644 --- a/tools/perf/Documentation/perf-top.txt +++ b/tools/perf/Documentation/perf-top.txt @@ -3,36 +3,122 @@ perf-top(1) NAME ---- -perf-top - Run a command and profile it +perf-top - System profiling tool. SYNOPSIS -------- [verse] -'perf top' [-e <EVENT> | --event=EVENT] [-l] [-a] <command> +'perf top' [-e <EVENT> | --event=EVENT] [<options>] DESCRIPTION ----------- -This command runs a command and gathers a performance counter profile -from it. +This command generates and displays a performance counter profile in realtime. OPTIONS ------- -<command>...:: - Any command you can specify in a shell. +-a:: +--all-cpus:: + System-wide collection. (default) + +-c <count>:: +--count=<count>:: + Event period to sample. + +-C <cpu>:: +--CPU=<cpu>:: + CPU to profile. + +-d <seconds>:: +--delay=<seconds>:: + Number of seconds to delay between refreshes. --e:: ---event=:: +-e <event>:: +--event=<event>:: Select the PMU event. Selection can be a symbolic event name (use 'perf list' to list all events) or a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a - hexadecimal event descriptor. + hexadecimal event descriptor. --a:: - system-wide collection +-E <entries>:: +--entries=<entries>:: + Display this many functions. + +-f <count>:: +--count-filter=<count>:: + Only display functions with more events than this. + +-F <freq>:: +--freq=<freq>:: + Profile at this frequency. + +-i:: +--inherit:: + Child tasks inherit counters, only makes sens with -p option. + +-k <path>:: +--vmlinux=<path>:: + Path to vmlinux. Required for annotation functionality. + +-m <pages>:: +--mmap-pages=<pages>:: + Number of mmapped data pages. + +-p <pid>:: +--pid=<pid>:: + Profile events on existing pid. + +-r <priority>:: +--realtime=<priority>:: + Collect data with this RT SCHED_FIFO priority. + +-s <symbol>:: +--sym-annotate=<symbol>:: + Annotate this symbol. Requires -k option. + +-v:: +--verbose:: + Be more verbose (show counter open errors, etc). + +-z:: +--zero:: + Zero history across display updates. + +INTERACTIVE PROMPTING KEYS +-------------------------- + +[d]:: + Display refresh delay. + +[e]:: + Number of entries to display. + +[E]:: + Event to display when multiple counters are active. + +[f]:: + Profile display filter (>= hit count). + +[F]:: + Annotation display filter (>= % of total). + +[s]:: + Annotate symbol. + +[S]:: + Stop annotation, return to full profile display. + +[w]:: + Toggle between weighted sum and individual count[E]r profile. + +[z]:: + Toggle event count zeroing across display updates. + +[qQ]:: + Quit. + +Pressing any unmapped key displays a menu, and prompts for input. --l:: - scale counter values SEE ALSO -------- diff --git a/tools/perf/Makefile b/tools/perf/Makefile index 1916e44b9bb0..de7beac1095e 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -310,6 +310,7 @@ LIB_H += util/sigchain.h LIB_H += util/symbol.h LIB_H += util/module.h LIB_H += util/color.h +LIB_H += util/values.h LIB_OBJS += util/abspath.o LIB_OBJS += util/alias.o @@ -337,6 +338,7 @@ 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 BUILTIN_OBJS += builtin-annotate.o BUILTIN_OBJS += builtin-help.o @@ -387,10 +389,14 @@ else has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y") + has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y") + ifeq ($(has_bfd),y) EXTLIBS += -lbfd else ifeq ($(has_bfd_iberty),y) EXTLIBS += -lbfd -liberty + else ifeq ($(has_bfd_iberty_z),y) + EXTLIBS += -lbfd -liberty -lz else msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling) BASIC_CFLAGS += -DNO_DEMANGLE diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 90c98082af10..0345aad8eba5 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -525,10 +525,14 @@ static int __cmd_record(int argc, const char **argv) signal(SIGCHLD, sig_handler); signal(SIGINT, sig_handler); - if (!stat(output_name, &st) && !force && !append_file) { - fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n", - output_name); - exit(-1); + if (!stat(output_name, &st) && st.st_size) { + if (!force && !append_file) { + fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n", + output_name); + exit(-1); + } + } else { + append_file = 0; } flags = O_CREAT|O_RDWR; diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 8cb58d68a006..41639182fb3f 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -17,6 +17,7 @@ #include "util/string.h" #include "util/callchain.h" #include "util/strlist.h" +#include "util/values.h" #include "perf.h" #include "util/header.h" @@ -53,6 +54,9 @@ static int modules; static int full_paths; static int show_nr_samples; +static int show_threads; +static struct perf_read_values show_threads_values; + static unsigned long page_size; static unsigned long mmap_window = 32; @@ -68,7 +72,7 @@ static int callchain; static struct callchain_param callchain_param = { - .mode = CHAIN_GRAPH_ABS, + .mode = CHAIN_GRAPH_REL, .min_percent = 0.5 }; @@ -112,7 +116,9 @@ struct read_event { struct perf_event_header header; u32 pid,tid; u64 value; - u64 format[3]; + u64 time_enabled; + u64 time_running; + u64 id; }; typedef union event_union { @@ -698,7 +704,8 @@ sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used) size_t ret = 0; if (verbose) - ret += repsep_fprintf(fp, "%#018llx ", (u64)self->ip); + ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip, + dso__symtab_origin(self->dso)); ret += repsep_fprintf(fp, "[%c] ", self->level); if (self->sym) { @@ -888,6 +895,21 @@ ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth, return ret; } +static struct symbol *rem_sq_bracket; +static struct callchain_list rem_hits; + +static void init_rem_hits(void) +{ + rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6); + if (!rem_sq_bracket) { + fprintf(stderr, "Not enough memory to display remaining hits\n"); + return; + } + + strcpy(rem_sq_bracket->name, "[...]"); + rem_hits.sym = rem_sq_bracket; +} + static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self, u64 total_samples, int depth, int depth_mask) @@ -897,25 +919,34 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, struct callchain_list *chain; int new_depth_mask = depth_mask; u64 new_total; + u64 remaining; size_t ret = 0; int i; if (callchain_param.mode == CHAIN_GRAPH_REL) - new_total = self->cumul_hit; + new_total = self->children_hit; else new_total = total_samples; + remaining = new_total; + node = rb_first(&self->rb_root); while (node) { + u64 cumul; + child = rb_entry(node, struct callchain_node, rb_node); + cumul = cumul_hits(child); + remaining -= cumul; /* * The depth mask manages the output of pipes that show * the depth. We don't want to keep the pipes of the current - * level for the last child of this depth + * level for the last child of this depth. + * Except if we have remaining filtered hits. They will + * supersede the last child */ next = rb_next(node); - if (!next) + if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining)) new_depth_mask &= ~(1 << (depth - 1)); /* @@ -930,7 +961,7 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, ret += ipchain__fprintf_graph(fp, chain, depth, new_depth_mask, i++, new_total, - child->cumul_hit); + cumul); } ret += callchain__fprintf_graph(fp, child, new_total, depth + 1, @@ -938,6 +969,19 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, node = next; } + if (callchain_param.mode == CHAIN_GRAPH_REL && + remaining && remaining != new_total) { + + if (!rem_sq_bracket) + return ret; + + new_depth_mask &= ~(1 << (depth - 1)); + + ret += ipchain__fprintf_graph(fp, &rem_hits, depth, + new_depth_mask, 0, new_total, + remaining); + } + return ret; } @@ -1358,6 +1402,8 @@ static size_t output__fprintf(FILE *fp, u64 total_samples) unsigned int width; char *col_width = col_width_list_str; + init_rem_hits(); + fprintf(fp, "# Samples: %Ld\n", (u64)total_samples); fprintf(fp, "#\n"); @@ -1429,6 +1475,11 @@ print_entries: } fprintf(fp, "\n"); + free(rem_sq_bracket); + + if (show_threads) + perf_read_values_display(fp, &show_threads_values); + return ret; } @@ -1690,14 +1741,47 @@ static void trace_event(event_t *event) dprintf(".\n"); } +static struct perf_header *header; + +static struct perf_counter_attr *perf_header__find_attr(u64 id) +{ + int i; + + for (i = 0; i < header->attrs; i++) { + struct perf_header_attr *attr = header->attr[i]; + int j; + + for (j = 0; j < attr->ids; j++) { + if (attr->id[j] == id) + return &attr->attr; + } + } + + return NULL; +} + static int process_read_event(event_t *event, unsigned long offset, unsigned long head) { - dprintf("%p [%p]: PERF_EVENT_READ: %d %d %Lu\n", + struct perf_counter_attr *attr = perf_header__find_attr(event->read.id); + + if (show_threads) { + char *name = attr ? __event_name(attr->type, attr->config) + : "unknown"; + perf_read_values_add_value(&show_threads_values, + event->read.pid, event->read.tid, + event->read.id, + name, + event->read.value); + } + + dprintf("%p [%p]: PERF_EVENT_READ: %d %d %s %Lu\n", (void *)(offset + head), (void *)(long)(event->header.size), event->read.pid, event->read.tid, + attr ? __event_name(attr->type, attr->config) + : "FAIL", event->read.value); return 0; @@ -1743,8 +1827,6 @@ process_event(event_t *event, unsigned long offset, unsigned long head) return 0; } -static struct perf_header *header; - static u64 perf_header__sample_type(void) { u64 sample_type = 0; @@ -1774,6 +1856,9 @@ static int __cmd_report(void) register_idle_thread(); + if (show_threads) + perf_read_values_init(&show_threads_values); + input = open(input_name, O_RDONLY); if (input < 0) { fprintf(stderr, " failed to open file: %s", input_name); @@ -1812,6 +1897,13 @@ static int __cmd_report(void) " -g?\n"); exit(-1); } + } else if (callchain_param.mode != CHAIN_NONE && !callchain) { + callchain = 1; + if (register_callchain_param(&callchain_param) < 0) { + fprintf(stderr, "Can't register callchain" + " params\n"); + exit(-1); + } } if (load_kernel() < 0) { @@ -1921,6 +2013,9 @@ done: output__resort(total); output__fprintf(stdout, total); + if (show_threads) + perf_read_values_destroy(&show_threads_values); + return rc; } @@ -1950,6 +2045,13 @@ parse_callchain_opt(const struct option *opt __used, const char *arg, else if (!strncmp(tok, "fractal", strlen(arg))) callchain_param.mode = CHAIN_GRAPH_REL; + else if (!strncmp(tok, "none", strlen(arg))) { + callchain_param.mode = CHAIN_NONE; + callchain = 0; + + return 0; + } + else return -1; @@ -1987,6 +2089,8 @@ static const struct option options[] = { "load module symbols - WARNING: use only with -k and LIVE kernel"), OPT_BOOLEAN('n', "show-nr-samples", &show_nr_samples, "Show a column with the number of samples"), + OPT_BOOLEAN('T', "threads", &show_threads, + "Show per-thread event counters"), OPT_STRING('s', "sort", &sort_order, "key[,key2...]", "sort by key(s): pid, comm, dso, symbol, parent"), OPT_BOOLEAN('P', "full-paths", &full_paths, diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index f9510eeeb6c7..b4b06c7903e1 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -496,7 +496,7 @@ static const struct option options[] = { "stat events on existing pid"), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), - OPT_BOOLEAN('S', "scale", &scale, + OPT_BOOLEAN('c', "scale", &scale, "scale/normalize counters"), OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show counter open errors, etc)"), diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index f139f1ab9333..7de28ce9ca26 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -31,6 +31,8 @@ #include <fcntl.h> #include <stdio.h> +#include <termios.h> +#include <unistd.h> #include <errno.h> #include <time.h> @@ -54,7 +56,7 @@ static int system_wide = 0; static int default_interval = 100000; -static u64 count_filter = 5; +static int count_filter = 5; static int print_entries = 15; static int target_pid = -1; @@ -69,15 +71,28 @@ static int freq = 0; static int verbose = 0; static char *vmlinux = NULL; -static char *sym_filter; -static unsigned long filter_start; -static unsigned long filter_end; - static int delay_secs = 2; static int zero; static int dump_symtab; /* + * Source + */ + +struct source_line { + u64 eip; + unsigned long count[MAX_COUNTERS]; + char *line; + struct source_line *next; +}; + +static char *sym_filter = NULL; +struct sym_entry *sym_filter_entry = NULL; +static int sym_pcnt_filter = 5; +static int sym_counter = 0; +static int display_weighted = -1; + +/* * Symbols */ @@ -91,9 +106,237 @@ struct sym_entry { unsigned long snap_count; double weight; int skip; + struct source_line *source; + struct source_line *lines; + struct source_line **lines_tail; + pthread_mutex_t source_lock; }; -struct sym_entry *sym_filter_entry; +/* + * Source functions + */ + +static void parse_source(struct sym_entry *syme) +{ + struct symbol *sym; + struct module *module; + struct section *section = NULL; + FILE *file; + char command[PATH_MAX*2], *path = vmlinux; + u64 start, end, len; + + if (!syme) + return; + + if (syme->lines) { + pthread_mutex_lock(&syme->source_lock); + goto out_assign; + } + + sym = (struct symbol *)(syme + 1); + module = sym->module; + + if (module) + path = module->path; + if (!path) + return; + + start = sym->obj_start; + if (!start) + start = sym->start; + + if (module) { + section = module->sections->find_section(module->sections, ".text"); + if (section) + start -= section->vma; + } + + end = start + sym->end - sym->start + 1; + len = sym->end - sym->start; + + sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", start, end, path); + + file = popen(command, "r"); + if (!file) + return; + + pthread_mutex_lock(&syme->source_lock); + syme->lines_tail = &syme->lines; + while (!feof(file)) { + struct source_line *src; + size_t dummy = 0; + char *c; + + src = malloc(sizeof(struct source_line)); + assert(src != NULL); + memset(src, 0, sizeof(struct source_line)); + + if (getline(&src->line, &dummy, file) < 0) + break; + if (!src->line) + break; + + c = strchr(src->line, '\n'); + if (c) + *c = 0; + + src->next = NULL; + *syme->lines_tail = src; + syme->lines_tail = &src->next; + + if (strlen(src->line)>8 && src->line[8] == ':') { + src->eip = strtoull(src->line, NULL, 16); + if (section) + src->eip += section->vma; + } + if (strlen(src->line)>8 && src->line[16] == ':') { + src->eip = strtoull(src->line, NULL, 16); + if (section) + src->eip += section->vma; + } + } + pclose(file); +out_assign: + sym_filter_entry = syme; + pthread_mutex_unlock(&syme->source_lock); +} + +static void __zero_source_counters(struct sym_entry *syme) +{ + int i; + struct source_line *line; + + line = syme->lines; + while (line) { + for (i = 0; i < nr_counters; i++) + line->count[i] = 0; + line = line->next; + } +} + +static void record_precise_ip(struct sym_entry *syme, int counter, u64 ip) +{ + struct source_line *line; + + if (syme != sym_filter_entry) + return; + + if (pthread_mutex_trylock(&syme->source_lock)) + return; + + if (!syme->source) + goto out_unlock; + + for (line = syme->lines; line; line = line->next) { + if (line->eip == ip) { + line->count[counter]++; + break; + } + if (line->eip > ip) + break; + } +out_unlock: + pthread_mutex_unlock(&syme->source_lock); +} + +static void lookup_sym_source(struct sym_entry *syme) +{ + struct symbol *symbol = (struct symbol *)(syme + 1); + struct source_line *line; + char pattern[PATH_MAX]; + char *idx; + + sprintf(pattern, "<%s>:", symbol->name); + + if (symbol->module) { + idx = strstr(pattern, "\t"); + if (idx) + *idx = 0; + } + + pthread_mutex_lock(&syme->source_lock); + for (line = syme->lines; line; line = line->next) { + if (strstr(line->line, pattern)) { + syme->source = line; + break; + } + } + pthread_mutex_unlock(&syme->source_lock); +} + +static void show_lines(struct source_line *queue, int count, int total) +{ + int i; + struct source_line *line; + + line = queue; + for (i = 0; i < count; i++) { + float pcnt = 100.0*(float)line->count[sym_counter]/(float)total; + + printf("%8li %4.1f%%\t%s\n", line->count[sym_counter], pcnt, line->line); + line = line->next; + } +} + +#define TRACE_COUNT 3 + +static void show_details(struct sym_entry *syme) +{ + struct symbol *symbol; + struct source_line *line; + struct source_line *line_queue = NULL; + int displayed = 0; + int line_queue_count = 0, total = 0, more = 0; + + if (!syme) + return; + + if (!syme->source) + lookup_sym_source(syme); + + if (!syme->source) + return; + + symbol = (struct symbol *)(syme + 1); + printf("Showing %s for %s\n", event_name(sym_counter), symbol->name); + printf(" Events Pcnt (>=%d%%)\n", sym_pcnt_filter); + + pthread_mutex_lock(&syme->source_lock); + line = syme->source; + while (line) { + total += line->count[sym_counter]; + line = line->next; + } + + line = syme->source; + while (line) { + float pcnt = 0.0; + + if (!line_queue_count) + line_queue = line; + line_queue_count++; + + if (line->count[sym_counter]) + pcnt = 100.0 * line->count[sym_counter] / (float)total; + if (pcnt >= (float)sym_pcnt_filter) { + if (displayed <= print_entries) + show_lines(line_queue, line_queue_count, total); + else more++; + displayed += line_queue_count; + line_queue_count = 0; + line_queue = NULL; + } else if (line_queue_count > TRACE_COUNT) { + line_queue = line_queue->next; + line_queue_count--; + } + + line->count[sym_counter] = zero ? 0 : line->count[sym_counter] * 7 / 8; + line = line->next; + } + pthread_mutex_unlock(&syme->source_lock); + if (more) + printf("%d lines not displayed, maybe increase display entries [e]\n", more); +} struct dso *kernel_dso; @@ -112,6 +355,9 @@ static double sym_weight(const struct sym_entry *sym) double weight = sym->snap_count; int counter; + if (!display_weighted) + return weight; + for (counter = 1; counter < nr_counters-1; counter++) weight *= sym->count[counter]; @@ -159,7 +405,7 @@ static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se) static void print_sym_table(void) { int printed = 0, j; - int counter; + 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 sum_ksamples = 0.0; @@ -175,7 +421,7 @@ static void print_sym_table(void) pthread_mutex_unlock(&active_symbols_lock); list_for_each_entry_safe_from(syme, n, &active_symbols, node) { - syme->snap_count = syme->count[0]; + syme->snap_count = syme->count[snap]; if (syme->snap_count != 0) { syme->weight = sym_weight(syme); rb_insert_active_sym(&tmp, syme); @@ -195,7 +441,7 @@ static void print_sym_table(void) samples_per_sec, 100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec))); - if (nr_counters == 1) { + if (nr_counters == 1 || !display_weighted) { printf("%Ld", (u64)attrs[0].sample_period); if (freq) printf("Hz "); @@ -203,7 +449,9 @@ static void print_sym_table(void) printf(" "); } - for (counter = 0; counter < nr_counters; counter++) { + if (!display_weighted) + printf("%s", event_name(sym_counter)); + else for (counter = 0; counter < nr_counters; counter++) { if (counter) printf("/"); @@ -228,6 +476,11 @@ static void print_sym_table(void) printf("------------------------------------------------------------------------------\n\n"); + if (sym_filter_entry) { + show_details(sym_filter_entry); + return; + } + if (nr_counters == 1) printf(" samples pcnt"); else @@ -242,13 +495,13 @@ static void print_sym_table(void) struct symbol *sym = (struct symbol *)(syme + 1); double pcnt; - if (++printed > print_entries || syme->snap_count < count_filter) + if (++printed > print_entries || (int)syme->snap_count < count_filter) continue; pcnt = 100.0 - (100.0 * ((sum_ksamples - syme->snap_count) / sum_ksamples)); - if (nr_counters == 1) + if (nr_counters == 1 || !display_weighted) printf("%20.2f - ", syme->weight); else printf("%9.1f %10ld - ", syme->weight, syme->snap_count); @@ -261,19 +514,250 @@ static void print_sym_table(void) } } +static void prompt_integer(int *target, const char *msg) +{ + char *buf = malloc(0), *p; + size_t dummy = 0; + int tmp; + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + return; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + p = buf; + while(*p) { + if (!isdigit(*p)) + goto out_free; + p++; + } + tmp = strtoul(buf, NULL, 10); + *target = tmp; +out_free: + free(buf); +} + +static void prompt_percent(int *target, const char *msg) +{ + int tmp = 0; + + prompt_integer(&tmp, msg); + if (tmp >= 0 && tmp <= 100) + *target = tmp; +} + +static void prompt_symbol(struct sym_entry **target, const char *msg) +{ + char *buf = malloc(0), *p; + struct sym_entry *syme = *target, *n, *found = NULL; + size_t dummy = 0; + + /* zero counters of active symbol */ + if (syme) { + pthread_mutex_lock(&syme->source_lock); + __zero_source_counters(syme); + *target = NULL; + pthread_mutex_unlock(&syme->source_lock); + } + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + goto out_free; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + pthread_mutex_lock(&active_symbols_lock); + syme = list_entry(active_symbols.next, struct sym_entry, node); + pthread_mutex_unlock(&active_symbols_lock); + + list_for_each_entry_safe_from(syme, n, &active_symbols, node) { + struct symbol *sym = (struct symbol *)(syme + 1); + + if (!strcmp(buf, sym->name)) { + found = syme; + break; + } + } + + if (!found) { + fprintf(stderr, "Sorry, %s is not active.\n", sym_filter); + sleep(1); + return; + } else + parse_source(found); + +out_free: + free(buf); +} + +static void print_mapped_keys(void) +{ + char *name = NULL; + + if (sym_filter_entry) { + struct symbol *sym = (struct symbol *)(sym_filter_entry+1); + name = sym->name; + } + + fprintf(stdout, "\nMapped keys:\n"); + fprintf(stdout, "\t[d] display refresh delay. \t(%d)\n", delay_secs); + fprintf(stdout, "\t[e] display entries (lines). \t(%d)\n", print_entries); + + if (nr_counters > 1) + fprintf(stdout, "\t[E] active event counter. \t(%s)\n", event_name(sym_counter)); + + fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", count_filter); + + if (vmlinux) { + fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter); + fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL"); + fprintf(stdout, "\t[S] stop annotation.\n"); + } + + if (nr_counters > 1) + fprintf(stdout, "\t[w] toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0); + + fprintf(stdout, "\t[z] toggle sample zeroing. \t(%d)\n", zero ? 1 : 0); + fprintf(stdout, "\t[qQ] quit.\n"); +} + +static int key_mapped(int c) +{ + switch (c) { + case 'd': + case 'e': + case 'f': + case 'z': + case 'q': + case 'Q': + return 1; + case 'E': + case 'w': + return nr_counters > 1 ? 1 : 0; + case 'F': + case 's': + case 'S': + return vmlinux ? 1 : 0; + } + + return 0; +} + +static void handle_keypress(int c) +{ + if (!key_mapped(c)) { + struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; + struct termios tc, save; + + print_mapped_keys(); + fprintf(stdout, "\nEnter selection, or unmapped key to continue: "); + fflush(stdout); + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; + tcsetattr(0, TCSANOW, &tc); + + poll(&stdin_poll, 1, -1); + c = getc(stdin); + + tcsetattr(0, TCSAFLUSH, &save); + if (!key_mapped(c)) + return; + } + + switch (c) { + case 'd': + prompt_integer(&delay_secs, "Enter display delay"); + break; + case 'e': + prompt_integer(&print_entries, "Enter display entries (lines)"); + break; + case 'E': + if (nr_counters > 1) { + int i; + + fprintf(stderr, "\nAvailable events:"); + for (i = 0; i < nr_counters; i++) + fprintf(stderr, "\n\t%d %s", i, event_name(i)); + + prompt_integer(&sym_counter, "Enter details event counter"); + + if (sym_counter >= nr_counters) { + fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(0)); + sym_counter = 0; + sleep(1); + } + } else sym_counter = 0; + break; + case 'f': + prompt_integer(&count_filter, "Enter display event count filter"); + break; + case 'F': + prompt_percent(&sym_pcnt_filter, "Enter details display event filter (percent)"); + break; + case 'q': + case 'Q': + printf("exiting.\n"); + exit(0); + case 's': + prompt_symbol(&sym_filter_entry, "Enter details symbol"); + break; + case 'S': + if (!sym_filter_entry) + break; + else { + struct sym_entry *syme = sym_filter_entry; + + pthread_mutex_lock(&syme->source_lock); + sym_filter_entry = NULL; + __zero_source_counters(syme); + pthread_mutex_unlock(&syme->source_lock); + } + break; + case 'w': + display_weighted = ~display_weighted; + break; + case 'z': + zero = ~zero; + break; + } +} + static void *display_thread(void *arg __used) { struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; - int delay_msecs = delay_secs * 1000; + struct termios tc, save; + int delay_msecs, c; + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; - printf("PerfTop refresh period: %d seconds\n", delay_secs); +repeat: + delay_msecs = delay_secs * 1000; + tcsetattr(0, TCSANOW, &tc); + /* trash return*/ + getc(stdin); do { print_sym_table(); } while (!poll(&stdin_poll, 1, delay_msecs) == 1); - printf("key pressed - exiting.\n"); - exit(0); + c = getc(stdin); + tcsetattr(0, TCSAFLUSH, &save); + + handle_keypress(c); + goto repeat; return NULL; } @@ -293,7 +777,6 @@ static const char *skip_symbols[] = { static int symbol_filter(struct dso *self, struct symbol *sym) { - static int filter_match; struct sym_entry *syme; const char *name = sym->name; int i; @@ -315,6 +798,10 @@ static int symbol_filter(struct dso *self, struct symbol *sym) return 1; syme = dso__sym_priv(self, sym); + pthread_mutex_init(&syme->source_lock, NULL); + if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter)) + sym_filter_entry = syme; + for (i = 0; skip_symbols[i]; i++) { if (!strcmp(skip_symbols[i], name)) { syme->skip = 1; @@ -322,29 +809,6 @@ static int symbol_filter(struct dso *self, struct symbol *sym) } } - if (filter_match == 1) { - filter_end = sym->start; - filter_match = -1; - if (filter_end - filter_start > 10000) { - fprintf(stderr, - "hm, too large filter symbol <%s> - skipping.\n", - sym_filter); - fprintf(stderr, "symbol filter start: %016lx\n", - filter_start); - fprintf(stderr, " end: %016lx\n", - filter_end); - filter_end = filter_start = 0; - sym_filter = NULL; - sleep(1); - } - } - - if (filter_match == 0 && sym_filter && !strcmp(name, sym_filter)) { - filter_match = 1; - filter_start = sym->start; - } - - return 0; } @@ -380,8 +844,6 @@ out_delete_dso: return -1; } -#define TRACE_COUNT 3 - /* * Binary search in the histogram table and record the hit: */ @@ -394,6 +856,7 @@ static void record_ip(u64 ip, int counter) if (!syme->skip) { syme->count[counter]++; + record_precise_ip(syme, counter, ip); pthread_mutex_lock(&active_symbols_lock); if (list_empty(&syme->node) || !syme->node.next) __list_insert_active_sym(syme); @@ -690,8 +1153,8 @@ static const struct option options[] = { "put the counters into a counter group"), OPT_BOOLEAN('i', "inherit", &inherit, "child tasks inherit counters"), - OPT_STRING('s', "sym-filter", &sym_filter, "pattern", - "only display symbols matchig this pattern"), + OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name", + "symbol to annotate - requires -k option"), OPT_BOOLEAN('z', "zero", &zero, "zero history across updates"), OPT_INTEGER('F', "freq", &freq, @@ -734,6 +1197,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) delay_secs = 1; parse_symbols(); + parse_source(sym_filter_entry); /* * Fill in the ones not specifically initialized via -c: diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c index 9d3c8141b8c1..011473411642 100644 --- a/tools/perf/util/callchain.c +++ b/tools/perf/util/callchain.c @@ -13,6 +13,7 @@ #include <stdio.h> #include <stdbool.h> #include <errno.h> +#include <math.h> #include "callchain.h" @@ -26,10 +27,14 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain, struct rb_node **p = &root->rb_node; struct rb_node *parent = NULL; struct callchain_node *rnode; + u64 chain_cumul = cumul_hits(chain); while (*p) { + u64 rnode_cumul; + parent = *p; rnode = rb_entry(parent, struct callchain_node, rb_node); + rnode_cumul = cumul_hits(rnode); switch (mode) { case CHAIN_FLAT: @@ -40,7 +45,7 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain, break; case CHAIN_GRAPH_ABS: /* Falldown */ case CHAIN_GRAPH_REL: - if (rnode->cumul_hit < chain->cumul_hit) + if (rnode_cumul < chain_cumul) p = &(*p)->rb_left; else p = &(*p)->rb_right; @@ -87,7 +92,7 @@ static void __sort_chain_graph_abs(struct callchain_node *node, chain_for_each_child(child, node) { __sort_chain_graph_abs(child, min_hit); - if (child->cumul_hit >= min_hit) + if (cumul_hits(child) >= min_hit) rb_insert_callchain(&node->rb_root, child, CHAIN_GRAPH_ABS); } @@ -108,11 +113,11 @@ static void __sort_chain_graph_rel(struct callchain_node *node, u64 min_hit; node->rb_root = RB_ROOT; - min_hit = node->cumul_hit * min_percent / 100.0; + min_hit = ceil(node->children_hit * min_percent); chain_for_each_child(child, node) { __sort_chain_graph_rel(child, min_percent); - if (child->cumul_hit >= min_hit) + if (cumul_hits(child) >= min_hit) rb_insert_callchain(&node->rb_root, child, CHAIN_GRAPH_REL); } @@ -122,7 +127,7 @@ static void sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root, u64 min_hit __used, struct callchain_param *param) { - __sort_chain_graph_rel(chain_root, param->min_percent); + __sort_chain_graph_rel(chain_root, param->min_percent / 100.0); rb_root->rb_node = chain_root->rb_root.rb_node; } @@ -211,7 +216,8 @@ add_child(struct callchain_node *parent, struct ip_callchain *chain, new = create_child(parent, false); fill_node(new, chain, start, syms); - new->cumul_hit = new->hit = 1; + new->children_hit = 0; + new->hit = 1; } /* @@ -241,7 +247,8 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, /* split the hits */ new->hit = parent->hit; - new->cumul_hit = parent->cumul_hit; + new->children_hit = parent->children_hit; + parent->children_hit = cumul_hits(new); new->val_nr = parent->val_nr - idx_local; parent->val_nr = idx_local; @@ -249,6 +256,7 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, if (idx_total < chain->nr) { parent->hit = 0; add_child(parent, chain, idx_total, syms); + parent->children_hit++; } else { parent->hit = 1; } @@ -269,13 +277,13 @@ __append_chain_children(struct callchain_node *root, struct ip_callchain *chain, unsigned int ret = __append_chain(rnode, chain, start, syms); if (!ret) - goto cumul; + goto inc_children_hit; } /* nothing in children, add to the current node */ add_child(root, chain, start, syms); -cumul: - root->cumul_hit++; +inc_children_hit: + root->children_hit++; } static int @@ -317,8 +325,6 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, /* we match 100% of the path, increment the hit */ if (i - start == root->val_nr && i == chain->nr) { root->hit++; - root->cumul_hit++; - return 0; } @@ -331,5 +337,7 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, void append_chain(struct callchain_node *root, struct ip_callchain *chain, struct symbol **syms) { + if (!chain->nr) + return; __append_chain_children(root, chain, syms, 0); } diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h index 7812122bea1d..a926ae4f5a16 100644 --- a/tools/perf/util/callchain.h +++ b/tools/perf/util/callchain.h @@ -7,6 +7,7 @@ #include "symbol.h" enum chain_mode { + CHAIN_NONE, CHAIN_FLAT, CHAIN_GRAPH_ABS, CHAIN_GRAPH_REL @@ -21,7 +22,7 @@ struct callchain_node { struct rb_root rb_root; /* sorted tree of children */ unsigned int val_nr; u64 hit; - u64 cumul_hit; /* hit + hits of children */ + u64 children_hit; }; struct callchain_param; @@ -48,6 +49,11 @@ static inline void callchain_init(struct callchain_node *node) INIT_LIST_HEAD(&node->val); } +static inline u64 cumul_hits(struct callchain_node *node) +{ + return node->hit + node->children_hit; +} + int register_callchain_param(struct callchain_param *param); void append_chain(struct callchain_node *root, struct ip_callchain *chain, struct symbol **syms); diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 450384b3bbe5..b92a457ca32e 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -185,6 +185,8 @@ static void do_read(int fd, void *buf, size_t size) if (ret < 0) die("failed to read"); + if (ret == 0) + die("failed to read: missing data"); size -= ret; buf += ret; @@ -213,9 +215,10 @@ struct perf_header *perf_header__read(int fd) for (i = 0; i < nr_attrs; i++) { struct perf_header_attr *attr; - off_t tmp = lseek(fd, 0, SEEK_CUR); + off_t tmp; do_read(fd, &f_attr, sizeof(f_attr)); + tmp = lseek(fd, 0, SEEK_CUR); attr = perf_header_attr__new(&f_attr.attr); diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 7bdad8df22a6..4858d83b3b67 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -121,13 +121,29 @@ static unsigned long hw_cache_stat[C(MAX)] = { (strcmp(sys_dirent.d_name, ".")) && \ (strcmp(sys_dirent.d_name, ".."))) +static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir) +{ + char evt_path[MAXPATHLEN]; + int fd; + + snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path, + sys_dir->d_name, evt_dir->d_name); + fd = open(evt_path, O_RDONLY); + if (fd < 0) + return -EINVAL; + close(fd); + + return 0; +} + #define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next, file, st) \ while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next) \ if (snprintf(file, MAXPATHLEN, "%s/%s/%s", debugfs_path, \ sys_dirent.d_name, evt_dirent.d_name) && \ (!stat(file, &st)) && (S_ISDIR(st.st_mode)) && \ (strcmp(evt_dirent.d_name, ".")) && \ - (strcmp(evt_dirent.d_name, ".."))) + (strcmp(evt_dirent.d_name, "..")) && \ + (!tp_event_has_id(&sys_dirent, &evt_dirent))) #define MAX_EVENT_LENGTH 30 @@ -223,9 +239,15 @@ char *event_name(int counter) { u64 config = attrs[counter].config; int type = attrs[counter].type; + + return __event_name(type, config); +} + +char *__event_name(int type, u64 config) +{ static char buf[32]; - if (attrs[counter].type == PERF_TYPE_RAW) { + if (type == PERF_TYPE_RAW) { sprintf(buf, "raw 0x%llx", config); return buf; } diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index 1ea5d09b6eb1..192a962e3a0f 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -10,6 +10,7 @@ extern int nr_counters; extern struct perf_counter_attr attrs[MAX_COUNTERS]; extern char *event_name(int ctr); +extern char *__event_name(int type, u64 config); extern int parse_events(const struct option *opt, const char *str, int unset); diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 16ddca202948..f1dcede14307 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -24,6 +24,16 @@ const char *sym_hist_filter; #define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ #endif +enum dso_origin { + DSO__ORIG_KERNEL = 0, + DSO__ORIG_JAVA_JIT, + DSO__ORIG_FEDORA, + DSO__ORIG_UBUNTU, + DSO__ORIG_BUILDID, + DSO__ORIG_DSO, + DSO__ORIG_NOT_FOUND, +}; + static struct symbol *symbol__new(u64 start, u64 len, const char *name, unsigned int priv_size, u64 obj_start, int verbose) @@ -81,6 +91,7 @@ struct dso *dso__new(const char *name, unsigned int sym_priv_size) self->sym_priv_size = sym_priv_size; self->find_symbol = dso__find_symbol; self->slen_calculated = 0; + self->origin = DSO__ORIG_NOT_FOUND; } return self; @@ -710,7 +721,7 @@ static char *dso__read_build_id(struct dso *self, int verbose) ++raw; bid += 2; } - if (verbose) + if (verbose >= 2) printf("%s(%s): %s\n", __func__, self->name, build_id); out_elf_end: elf_end(elf); @@ -720,11 +731,26 @@ out: return build_id; } +char dso__symtab_origin(const struct dso *self) +{ + static const char origin[] = { + [DSO__ORIG_KERNEL] = 'k', + [DSO__ORIG_JAVA_JIT] = 'j', + [DSO__ORIG_FEDORA] = 'f', + [DSO__ORIG_UBUNTU] = 'u', + [DSO__ORIG_BUILDID] = 'b', + [DSO__ORIG_DSO] = 'd', + }; + + if (self == NULL || self->origin == DSO__ORIG_NOT_FOUND) + return '!'; + return origin[self->origin]; +} + int dso__load(struct dso *self, symbol_filter_t filter, int verbose) { int size = PATH_MAX; char *name = malloc(size), *build_id = NULL; - int variant = 0; int ret = -1; int fd; @@ -733,19 +759,26 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose) self->adjust_symbols = 0; - if (strncmp(self->name, "/tmp/perf-", 10) == 0) - return dso__load_perf_map(self, filter, verbose); + if (strncmp(self->name, "/tmp/perf-", 10) == 0) { + ret = dso__load_perf_map(self, filter, verbose); + self->origin = ret > 0 ? DSO__ORIG_JAVA_JIT : + DSO__ORIG_NOT_FOUND; + return ret; + } + + self->origin = DSO__ORIG_FEDORA - 1; more: do { - switch (variant) { - case 0: /* Fedora */ + self->origin++; + switch (self->origin) { + case DSO__ORIG_FEDORA: snprintf(name, size, "/usr/lib/debug%s.debug", self->name); break; - case 1: /* Ubuntu */ + case DSO__ORIG_UBUNTU: snprintf(name, size, "/usr/lib/debug%s", self->name); break; - case 2: + case DSO__ORIG_BUILDID: build_id = dso__read_build_id(self, verbose); if (build_id != NULL) { snprintf(name, size, @@ -754,16 +787,15 @@ more: free(build_id); break; } - variant++; + self->origin++; /* Fall thru */ - case 3: /* Sane people */ + case DSO__ORIG_DSO: snprintf(name, size, "%s", self->name); break; default: goto out; } - variant++; fd = open(name, O_RDONLY); } while (fd < 0); @@ -899,6 +931,9 @@ int dso__load_kernel(struct dso *self, const char *vmlinux, if (err <= 0) err = dso__load_kallsyms(self, filter, verbose); + if (err > 0) + self->origin = DSO__ORIG_KERNEL; + return err; } diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 2f92b21c712d..1e003ec2f4b1 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -26,6 +26,7 @@ struct dso { unsigned int sym_priv_size; unsigned char adjust_symbols; unsigned char slen_calculated; + unsigned char origin; char name[0]; }; @@ -49,6 +50,7 @@ int dso__load_modules(struct dso *self, symbol_filter_t filter, int verbose); int dso__load(struct dso *self, symbol_filter_t filter, int verbose); size_t dso__fprintf(struct dso *self, FILE *fp); +char dso__symtab_origin(const struct dso *self); void symbol__init(void); #endif /* _PERF_SYMBOL_ */ diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c new file mode 100644 index 000000000000..8551c0b8b233 --- /dev/null +++ b/tools/perf/util/values.c @@ -0,0 +1,171 @@ +#include <stdlib.h> + +#include "util.h" +#include "values.h" + +void perf_read_values_init(struct perf_read_values *values) +{ + values->threads_max = 16; + values->pid = malloc(values->threads_max * sizeof(*values->pid)); + values->tid = malloc(values->threads_max * sizeof(*values->tid)); + values->value = malloc(values->threads_max * sizeof(*values->value)); + if (!values->pid || !values->tid || !values->value) + die("failed to allocate read_values threads arrays"); + values->threads = 0; + + values->counters_max = 16; + values->counterrawid = malloc(values->counters_max + * sizeof(*values->counterrawid)); + values->countername = malloc(values->counters_max + * sizeof(*values->countername)); + if (!values->counterrawid || !values->countername) + die("failed to allocate read_values counters arrays"); + values->counters = 0; +} + +void perf_read_values_destroy(struct perf_read_values *values) +{ + int i; + + if (!values->threads_max || !values->counters_max) + return; + + for (i = 0; i < values->threads; i++) + free(values->value[i]); + free(values->pid); + free(values->tid); + free(values->counterrawid); + for (i = 0; i < values->counters; i++) + free(values->countername[i]); + free(values->countername); +} + +static void perf_read_values__enlarge_threads(struct perf_read_values *values) +{ + values->threads_max *= 2; + values->pid = realloc(values->pid, + values->threads_max * sizeof(*values->pid)); + values->tid = realloc(values->tid, + values->threads_max * sizeof(*values->tid)); + values->value = realloc(values->value, + values->threads_max * sizeof(*values->value)); + if (!values->pid || !values->tid || !values->value) + die("failed to enlarge read_values threads arrays"); +} + +static int perf_read_values__findnew_thread(struct perf_read_values *values, + u32 pid, u32 tid) +{ + int i; + + for (i = 0; i < values->threads; i++) + if (values->pid[i] == pid && values->tid[i] == tid) + return i; + + if (values->threads == values->threads_max) + perf_read_values__enlarge_threads(values); + + i = values->threads++; + values->pid[i] = pid; + values->tid[i] = tid; + values->value[i] = malloc(values->counters_max * sizeof(**values->value)); + if (!values->value[i]) + die("failed to allocate read_values counters array"); + + return i; +} + +static void perf_read_values__enlarge_counters(struct perf_read_values *values) +{ + int i; + + values->counters_max *= 2; + values->counterrawid = realloc(values->counterrawid, + values->counters_max * sizeof(*values->counterrawid)); + values->countername = realloc(values->countername, + values->counters_max * sizeof(*values->countername)); + if (!values->counterrawid || !values->countername) + die("failed to enlarge read_values counters arrays"); + + for (i = 0; i < values->threads; i++) { + values->value[i] = realloc(values->value[i], + values->counters_max * sizeof(**values->value)); + if (!values->value[i]) + die("failed to enlarge read_values counters arrays"); + } +} + +static int perf_read_values__findnew_counter(struct perf_read_values *values, + u64 rawid, char *name) +{ + int i; + + for (i = 0; i < values->counters; i++) + if (values->counterrawid[i] == rawid) + return i; + + if (values->counters == values->counters_max) + perf_read_values__enlarge_counters(values); + + i = values->counters++; + values->counterrawid[i] = rawid; + values->countername[i] = strdup(name); + + return i; +} + +void perf_read_values_add_value(struct perf_read_values *values, + u32 pid, u32 tid, + u64 rawid, char *name, u64 value) +{ + int tindex, cindex; + + tindex = perf_read_values__findnew_thread(values, pid, tid); + cindex = perf_read_values__findnew_counter(values, rawid, name); + + values->value[tindex][cindex] = value; +} + +void perf_read_values_display(FILE *fp, struct perf_read_values *values) +{ + int i, j; + int pidwidth, tidwidth; + int *counterwidth; + + counterwidth = malloc(values->counters * sizeof(*counterwidth)); + if (!counterwidth) + die("failed to allocate counterwidth array"); + tidwidth = 3; + pidwidth = 3; + for (j = 0; j < values->counters; j++) + counterwidth[j] = strlen(values->countername[j]); + for (i = 0; i < values->threads; i++) { + int width; + + width = snprintf(NULL, 0, "%d", values->pid[i]); + if (width > pidwidth) + pidwidth = width; + width = snprintf(NULL, 0, "%d", values->tid[i]); + if (width > tidwidth) + tidwidth = width; + for (j = 0; j < values->counters; j++) { + width = snprintf(NULL, 0, "%Lu", values->value[i][j]); + if (width > counterwidth[j]) + counterwidth[j] = width; + } + } + + fprintf(fp, "# %*s %*s", pidwidth, "PID", tidwidth, "TID"); + for (j = 0; j < values->counters; j++) + fprintf(fp, " %*s", counterwidth[j], values->countername[j]); + fprintf(fp, "\n"); + + for (i = 0; i < values->threads; i++) { + fprintf(fp, " %*d %*d", pidwidth, values->pid[i], + tidwidth, values->tid[i]); + for (j = 0; j < values->counters; j++) + fprintf(fp, " %*Lu", + counterwidth[j], values->value[i][j]); + fprintf(fp, "\n"); + } +} diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h new file mode 100644 index 000000000000..e41be5e86e6b --- /dev/null +++ b/tools/perf/util/values.h @@ -0,0 +1,26 @@ +#ifndef _PERF_VALUES_H +#define _PERF_VALUES_H + +#include "types.h" + +struct perf_read_values { + int threads; + int threads_max; + u32 *pid, *tid; + int counters; + int counters_max; + u64 *counterrawid; + char **countername; + u64 **value; +}; + +void perf_read_values_init(struct perf_read_values *values); +void perf_read_values_destroy(struct perf_read_values *values); + +void perf_read_values_add_value(struct perf_read_values *values, + u32 pid, u32 tid, + u64 rawid, char *name, u64 value); + +void perf_read_values_display(FILE *fp, struct perf_read_values *values); + +#endif /* _PERF_VALUES_H */ |