diff options
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/capability.c | 111 | ||||
-rw-r--r-- | kernel/cgroup.c | 2 | ||||
-rw-r--r-- | kernel/cpuset.c | 10 | ||||
-rw-r--r-- | kernel/exit.c | 7 | ||||
-rw-r--r-- | kernel/kgdb.c | 16 | ||||
-rw-r--r-- | kernel/module.c | 18 | ||||
-rw-r--r-- | kernel/relay.c | 2 | ||||
-rw-r--r-- | kernel/sched.c | 468 | ||||
-rw-r--r-- | kernel/sched_clock.c | 18 | ||||
-rw-r--r-- | kernel/sched_debug.c | 5 | ||||
-rw-r--r-- | kernel/sched_fair.c | 254 | ||||
-rw-r--r-- | kernel/sched_rt.c | 4 | ||||
-rw-r--r-- | kernel/sched_stats.h | 1 | ||||
-rw-r--r-- | kernel/sched_trace.h | 41 | ||||
-rw-r--r-- | kernel/signal.c | 51 | ||||
-rw-r--r-- | kernel/stop_machine.c | 7 | ||||
-rw-r--r-- | kernel/sys.c | 6 | ||||
-rw-r--r-- | kernel/trace/Makefile | 1 | ||||
-rw-r--r-- | kernel/trace/ftrace.c | 457 | ||||
-rw-r--r-- | kernel/trace/trace.c | 153 | ||||
-rw-r--r-- | kernel/trace/trace.h | 18 | ||||
-rw-r--r-- | kernel/trace/trace_functions.c | 4 | ||||
-rw-r--r-- | kernel/trace/trace_irqsoff.c | 16 | ||||
-rw-r--r-- | kernel/trace/trace_mmiotrace.c | 295 | ||||
-rw-r--r-- | kernel/trace/trace_sched_switch.c | 45 | ||||
-rw-r--r-- | kernel/trace/trace_sched_wakeup.c | 67 | ||||
-rw-r--r-- | kernel/trace/trace_selftest.c | 1 |
27 files changed, 1213 insertions, 865 deletions
diff --git a/kernel/capability.c b/kernel/capability.c index 39e8193b41ea..cfbe44299488 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -53,6 +53,69 @@ static void warn_legacy_capability_use(void) } /* + * Version 2 capabilities worked fine, but the linux/capability.h file + * that accompanied their introduction encouraged their use without + * the necessary user-space source code changes. As such, we have + * created a version 3 with equivalent functionality to version 2, but + * with a header change to protect legacy source code from using + * version 2 when it wanted to use version 1. If your system has code + * that trips the following warning, it is using version 2 specific + * capabilities and may be doing so insecurely. + * + * The remedy is to either upgrade your version of libcap (to 2.10+, + * if the application is linked against it), or recompile your + * application with modern kernel headers and this warning will go + * away. + */ + +static void warn_deprecated_v2(void) +{ + static int warned; + + if (!warned) { + char name[sizeof(current->comm)]; + + printk(KERN_INFO "warning: `%s' uses deprecated v2" + " capabilities in a way that may be insecure.\n", + get_task_comm(name, current)); + warned = 1; + } +} + +/* + * Version check. Return the number of u32s in each capability flag + * array, or a negative value on error. + */ +static int cap_validate_magic(cap_user_header_t header, unsigned *tocopy) +{ + __u32 version; + + if (get_user(version, &header->version)) + return -EFAULT; + + switch (version) { + case _LINUX_CAPABILITY_VERSION_1: + warn_legacy_capability_use(); + *tocopy = _LINUX_CAPABILITY_U32S_1; + break; + case _LINUX_CAPABILITY_VERSION_2: + warn_deprecated_v2(); + /* + * fall through - v3 is otherwise equivalent to v2. + */ + case _LINUX_CAPABILITY_VERSION_3: + *tocopy = _LINUX_CAPABILITY_U32S_3; + break; + default: + if (put_user((u32)_KERNEL_CAPABILITY_VERSION, &header->version)) + return -EFAULT; + return -EINVAL; + } + + return 0; +} + +/* * For sys_getproccap() and sys_setproccap(), any of the three * capability set pointers may be NULL -- indicating that that set is * uninteresting and/or not to be changed. @@ -71,27 +134,13 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) { int ret = 0; pid_t pid; - __u32 version; struct task_struct *target; unsigned tocopy; kernel_cap_t pE, pI, pP; - if (get_user(version, &header->version)) - return -EFAULT; - - switch (version) { - case _LINUX_CAPABILITY_VERSION_1: - warn_legacy_capability_use(); - tocopy = _LINUX_CAPABILITY_U32S_1; - break; - case _LINUX_CAPABILITY_VERSION_2: - tocopy = _LINUX_CAPABILITY_U32S_2; - break; - default: - if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) - return -EFAULT; - return -EINVAL; - } + ret = cap_validate_magic(header, &tocopy); + if (ret != 0) + return ret; if (get_user(pid, &header->pid)) return -EFAULT; @@ -118,7 +167,7 @@ out: spin_unlock(&task_capability_lock); if (!ret) { - struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S]; + struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; unsigned i; for (i = 0; i < tocopy; i++) { @@ -128,7 +177,7 @@ out: } /* - * Note, in the case, tocopy < _LINUX_CAPABILITY_U32S, + * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S, * we silently drop the upper capabilities here. This * has the effect of making older libcap * implementations implicitly drop upper capability @@ -240,30 +289,16 @@ static inline int cap_set_all(kernel_cap_t *effective, */ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) { - struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S]; + struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; unsigned i, tocopy; kernel_cap_t inheritable, permitted, effective; - __u32 version; struct task_struct *target; int ret; pid_t pid; - if (get_user(version, &header->version)) - return -EFAULT; - - switch (version) { - case _LINUX_CAPABILITY_VERSION_1: - warn_legacy_capability_use(); - tocopy = _LINUX_CAPABILITY_U32S_1; - break; - case _LINUX_CAPABILITY_VERSION_2: - tocopy = _LINUX_CAPABILITY_U32S_2; - break; - default: - if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) - return -EFAULT; - return -EINVAL; - } + ret = cap_validate_magic(header, &tocopy); + if (ret != 0) + return ret; if (get_user(pid, &header->pid)) return -EFAULT; @@ -281,7 +316,7 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) permitted.cap[i] = kdata[i].permitted; inheritable.cap[i] = kdata[i].inheritable; } - while (i < _LINUX_CAPABILITY_U32S) { + while (i < _KERNEL_CAPABILITY_U32S) { effective.cap[i] = 0; permitted.cap[i] = 0; inheritable.cap[i] = 0; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index fbc6fc8949b4..15ac0e1e4f4d 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2903,7 +2903,7 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) cg = tsk->cgroups; parent = task_cgroup(tsk, subsys->subsys_id); - snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "node_%d", tsk->pid); + snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "%d", tsk->pid); /* Pin the hierarchy */ atomic_inc(&parent->root->sb->s_active); diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 86ea9e34e326..039baa4cd90c 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -797,8 +797,10 @@ static int update_cpumask(struct cpuset *cs, char *buf) retval = cpulist_parse(buf, trialcs.cpus_allowed); if (retval < 0) return retval; + + if (!cpus_subset(trialcs.cpus_allowed, cpu_online_map)) + return -EINVAL; } - cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map); retval = validate_change(cs, &trialcs); if (retval < 0) return retval; @@ -932,9 +934,11 @@ static int update_nodemask(struct cpuset *cs, char *buf) retval = nodelist_parse(buf, trialcs.mems_allowed); if (retval < 0) goto done; + + if (!nodes_subset(trialcs.mems_allowed, + node_states[N_HIGH_MEMORY])) + return -EINVAL; } - nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, - node_states[N_HIGH_MEMORY]); oldmem = cs->mems_allowed; if (nodes_equal(oldmem, trialcs.mems_allowed)) { retval = 0; /* Too easy - nothing to do */ diff --git a/kernel/exit.c b/kernel/exit.c index 1510f78a0ffa..8f6185e69b69 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -126,6 +126,12 @@ static void __exit_signal(struct task_struct *tsk) __unhash_process(tsk); + /* + * Do this under ->siglock, we can race with another thread + * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals. + */ + flush_sigqueue(&tsk->pending); + tsk->signal = NULL; tsk->sighand = NULL; spin_unlock(&sighand->siglock); @@ -133,7 +139,6 @@ static void __exit_signal(struct task_struct *tsk) __cleanup_sighand(sighand); clear_tsk_thread_flag(tsk,TIF_SIGPENDING); - flush_sigqueue(&tsk->pending); if (sig) { flush_sigqueue(&sig->shared_pending); taskstats_tgid_free(sig); diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 14787de568b3..79e3c90113c2 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -52,6 +52,7 @@ #include <asm/byteorder.h> #include <asm/atomic.h> #include <asm/system.h> +#include <asm/unaligned.h> static int kgdb_break_asap; @@ -227,8 +228,6 @@ void __weak kgdb_disable_hw_debug(struct pt_regs *regs) * GDB remote protocol parser: */ -static const char hexchars[] = "0123456789abcdef"; - static int hex(char ch) { if ((ch >= 'a') && (ch <= 'f')) @@ -316,8 +315,8 @@ static void put_packet(char *buffer) } kgdb_io_ops->write_char('#'); - kgdb_io_ops->write_char(hexchars[checksum >> 4]); - kgdb_io_ops->write_char(hexchars[checksum & 0xf]); + kgdb_io_ops->write_char(hex_asc_hi(checksum)); + kgdb_io_ops->write_char(hex_asc_lo(checksum)); if (kgdb_io_ops->flush) kgdb_io_ops->flush(); @@ -478,8 +477,8 @@ static void error_packet(char *pkt, int error) { error = -error; pkt[0] = 'E'; - pkt[1] = hexchars[(error / 10)]; - pkt[2] = hexchars[(error % 10)]; + pkt[1] = hex_asc[(error / 10)]; + pkt[2] = hex_asc[(error % 10)]; pkt[3] = '\0'; } @@ -510,10 +509,7 @@ static void int_to_threadref(unsigned char *id, int value) scan = (unsigned char *)id; while (i--) *scan++ = 0; - *scan++ = (value >> 24) & 0xff; - *scan++ = (value >> 16) & 0xff; - *scan++ = (value >> 8) & 0xff; - *scan++ = (value & 0xff); + put_unaligned_be32(value, scan); } static struct task_struct *getthread(struct pt_regs *regs, int tid) diff --git a/kernel/module.c b/kernel/module.c index f5e9491ef7ac..5f80478b746d 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1337,7 +1337,19 @@ out_unreg: kobject_put(&mod->mkobj.kobj); return err; } -#endif + +static void mod_sysfs_fini(struct module *mod) +{ + kobject_put(&mod->mkobj.kobj); +} + +#else /* CONFIG_SYSFS */ + +static void mod_sysfs_fini(struct module *mod) +{ +} + +#endif /* CONFIG_SYSFS */ static void mod_kobject_remove(struct module *mod) { @@ -1345,7 +1357,7 @@ static void mod_kobject_remove(struct module *mod) module_param_sysfs_remove(mod); kobject_put(mod->mkobj.drivers_dir); kobject_put(mod->holders_dir); - kobject_put(&mod->mkobj.kobj); + mod_sysfs_fini(mod); } /* @@ -1780,7 +1792,7 @@ static struct module *load_module(void __user *umod, /* Sanity checks against insmoding binaries or wrong arch, weird elf version */ - if (memcmp(hdr->e_ident, ELFMAG, 4) != 0 + if (memcmp(hdr->e_ident, ELFMAG, SELFMAG) != 0 || hdr->e_type != ET_REL || !elf_check_arch(hdr) || hdr->e_shentsize != sizeof(*sechdrs)) { diff --git a/kernel/relay.c b/kernel/relay.c index bc24dcdc570f..7de644cdec43 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1191,7 +1191,7 @@ static ssize_t relay_file_splice_read(struct file *in, ret = 0; spliced = 0; - while (len) { + while (len && !spliced) { ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret); if (ret < 0) break; diff --git a/kernel/sched.c b/kernel/sched.c index e2e985eeee78..801abd319355 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -137,7 +137,7 @@ static inline void sg_inc_cpu_power(struct sched_group *sg, u32 val) static inline int rt_policy(int policy) { - if (unlikely(policy == SCHED_FIFO) || unlikely(policy == SCHED_RR)) + if (unlikely(policy == SCHED_FIFO || policy == SCHED_RR)) return 1; return 0; } @@ -399,43 +399,6 @@ struct cfs_rq { */ struct list_head leaf_cfs_rq_list; struct task_group *tg; /* group that "owns" this runqueue */ - -#ifdef CONFIG_SMP - unsigned long task_weight; - unsigned long shares; - /* - * We need space to build a sched_domain wide view of the full task - * group tree, in order to avoid depending on dynamic memory allocation - * during the load balancing we place this in the per cpu task group - * hierarchy. This limits the load balancing to one instance per cpu, - * but more should not be needed anyway. - */ - struct aggregate_struct { - /* - * load = weight(cpus) * f(tg) - * - * Where f(tg) is the recursive weight fraction assigned to - * this group. - */ - unsigned long load; - - /* - * part of the group weight distributed to this span. - */ - unsigned long shares; - - /* - * The sum of all runqueue weights within this span. - */ - unsigned long rq_weight; - - /* - * Weight contributed by tasks; this is the part we can - * influence by moving tasks around. - */ - unsigned long task_weight; - } aggregate; -#endif #endif }; @@ -633,6 +596,8 @@ static inline void update_rq_clock(struct rq *rq) rq->clock = sched_clock_cpu(cpu_of(rq)); } +#include "sched_trace.h" + /* * Tunables that become constants when CONFIG_SCHED_DEBUG is off: */ @@ -884,7 +849,7 @@ static unsigned long long __cpu_clock(int cpu) * For kernel-internal use: high-speed (but slightly incorrect) per-cpu * clock constructed from sched_clock(): */ -unsigned long long cpu_clock(int cpu) +unsigned long long notrace cpu_clock(int cpu) { unsigned long long prev_cpu_time, time, delta_time; unsigned long flags; @@ -1387,9 +1352,6 @@ static void __resched_task(struct task_struct *p, int tif_bit) */ #define SRR(x, y) (((x) + (1UL << ((y) - 1))) >> (y)) -/* - * delta *= weight / lw - */ static unsigned long calc_delta_mine(unsigned long delta_exec, unsigned long weight, struct load_weight *lw) @@ -1412,6 +1374,12 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight, return (unsigned long)min(tmp, (u64)(unsigned long)LONG_MAX); } +static inline unsigned long +calc_delta_fair(unsigned long delta_exec, struct load_weight *lw) +{ + return calc_delta_mine(delta_exec, NICE_0_LOAD, lw); +} + static inline void update_load_add(struct load_weight *lw, unsigned long inc) { lw->weight += inc; @@ -1524,326 +1492,6 @@ static unsigned long source_load(int cpu, int type); static unsigned long target_load(int cpu, int type); static unsigned long cpu_avg_load_per_task(int cpu); static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); - -#ifdef CONFIG_FAIR_GROUP_SCHED - -/* - * Group load balancing. - * - * We calculate a few balance domain wide aggregate numbers; load and weight. - * Given the pictures below, and assuming each item has equal weight: - * - * root 1 - thread - * / | \ A - group - * A 1 B - * /|\ / \ - * C 2 D 3 4 - * | | - * 5 6 - * - * load: - * A and B get 1/3-rd of the total load. C and D get 1/3-rd of A's 1/3-rd, - * which equals 1/9-th of the total load. - * - * shares: - * The weight of this group on the selected cpus. - * - * rq_weight: - * Direct sum of all the cpu's their rq weight, e.g. A would get 3 while - * B would get 2. - * - * task_weight: - * Part of the rq_weight contributed by tasks; all groups except B would - * get 1, B gets 2. - */ - -static inline struct aggregate_struct * -aggregate(struct task_group *tg, struct sched_domain *sd) -{ - return &tg->cfs_rq[sd->first_cpu]->aggregate; -} - -typedef void (*aggregate_func)(struct task_group *, struct sched_domain *); - -/* - * Iterate the full tree, calling @down when first entering a node and @up when - * leaving it for the final time. - */ -static -void aggregate_walk_tree(aggregate_func down, aggregate_func up, - struct sched_domain *sd) -{ - struct task_group *parent, *child; - - rcu_read_lock(); - parent = &root_task_group; -down: - (*down)(parent, sd); - list_for_each_entry_rcu(child, &parent->children, siblings) { - parent = child; - goto down; - -up: - continue; - } - (*up)(parent, sd); - - child = parent; - parent = parent->parent; - if (parent) - goto up; - rcu_read_unlock(); -} - -/* - * Calculate the aggregate runqueue weight. - */ -static -void aggregate_group_weight(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long rq_weight = 0; - unsigned long task_weight = 0; - int i; - - for_each_cpu_mask(i, sd->span) { - rq_weight += tg->cfs_rq[i]->load.weight; - task_weight += tg->cfs_rq[i]->task_weight; - } - - aggregate(tg, sd)->rq_weight = rq_weight; - aggregate(tg, sd)->task_weight = task_weight; -} - -/* - * Compute the weight of this group on the given cpus. - */ -static -void aggregate_group_shares(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long shares = 0; - int i; - - for_each_cpu_mask(i, sd->span) - shares += tg->cfs_rq[i]->shares; - - if ((!shares && aggregate(tg, sd)->rq_weight) || shares > tg->shares) - shares = tg->shares; - - aggregate(tg, sd)->shares = shares; -} - -/* - * Compute the load fraction assigned to this group, relies on the aggregate - * weight and this group's parent's load, i.e. top-down. - */ -static -void aggregate_group_load(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long load; - - if (!tg->parent) { - int i; - - load = 0; - for_each_cpu_mask(i, sd->span) - load += cpu_rq(i)->load.weight; - - } else { - load = aggregate(tg->parent, sd)->load; - - /* - * shares is our weight in the parent's rq so - * shares/parent->rq_weight gives our fraction of the load - */ - load *= aggregate(tg, sd)->shares; - load /= aggregate(tg->parent, sd)->rq_weight + 1; - } - - aggregate(tg, sd)->load = load; -} - -static void __set_se_shares(struct sched_entity *se, unsigned long shares); - -/* - * Calculate and set the cpu's group shares. - */ -static void -__update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd, - int tcpu) -{ - int boost = 0; - unsigned long shares; - unsigned long rq_weight; - - if (!tg->se[tcpu]) - return; - - rq_weight = tg->cfs_rq[tcpu]->load.weight; - - /* - * If there are currently no tasks on the cpu pretend there is one of - * average load so that when a new task gets to run here it will not - * get delayed by group starvation. - */ - if (!rq_weight) { - boost = 1; - rq_weight = NICE_0_LOAD; - } - - /* - * \Sum shares * rq_weight - * shares = ----------------------- - * \Sum rq_weight - * - */ - shares = aggregate(tg, sd)->shares * rq_weight; - shares /= aggregate(tg, sd)->rq_weight + 1; - - /* - * record the actual number of shares, not the boosted amount. - */ - tg->cfs_rq[tcpu]->shares = boost ? 0 : shares; - - if (shares < MIN_SHARES) - shares = MIN_SHARES; - else if (shares > MAX_SHARES) - shares = MAX_SHARES; - - __set_se_shares(tg->se[tcpu], shares); -} - -/* - * Re-adjust the weights on the cpu the task came from and on the cpu the - * task went to. - */ -static void -__move_group_shares(struct task_group *tg, struct sched_domain *sd, - int scpu, int dcpu) -{ - unsigned long shares; - - shares = tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares; - - __update_group_shares_cpu(tg, sd, scpu); - __update_group_shares_cpu(tg, sd, dcpu); - - /* - * ensure we never loose shares due to rounding errors in the - * above redistribution. - */ - shares -= tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares; - if (shares) - tg->cfs_rq[dcpu]->shares += shares; -} - -/* - * Because changing a group's shares changes the weight of the super-group - * we need to walk up the tree and change all shares until we hit the root. - */ -static void -move_group_shares(struct task_group *tg, struct sched_domain *sd, - int scpu, int dcpu) -{ - while (tg) { - __move_group_shares(tg, sd, scpu, dcpu); - tg = tg->parent; - } -} - -static -void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long shares = aggregate(tg, sd)->shares; - int i; - - for_each_cpu_mask(i, sd->span) { - struct rq *rq = cpu_rq(i); - unsigned long flags; - - spin_lock_irqsave(&rq->lock, flags); - __update_group_shares_cpu(tg, sd, i); - spin_unlock_irqrestore(&rq->lock, flags); - } - - aggregate_group_shares(tg, sd); - - /* - * ensure we never loose shares due to rounding errors in the - * above redistribution. - */ - shares -= aggregate(tg, sd)->shares; - if (shares) { - tg->cfs_rq[sd->first_cpu]->shares += shares; - aggregate(tg, sd)->shares += shares; - } -} - -/* - * Calculate the accumulative weight and recursive load of each task group - * while walking down the tree. - */ -static -void aggregate_get_down(struct task_group *tg, struct sched_domain *sd) -{ - aggregate_group_weight(tg, sd); - aggregate_group_shares(tg, sd); - aggregate_group_load(tg, sd); -} - -/* - * Rebalance the cpu shares while walking back up the tree. - */ -static -void aggregate_get_up(struct task_group *tg, struct sched_domain *sd) -{ - aggregate_group_set_shares(tg, sd); -} - -static DEFINE_PER_CPU(spinlock_t, aggregate_lock); - -static void __init init_aggregate(void) -{ - int i; - - for_each_possible_cpu(i) - spin_lock_init(&per_cpu(aggregate_lock, i)); -} - -static int get_aggregate(struct sched_domain *sd) -{ - if (!spin_trylock(&per_cpu(aggregate_lock, sd->first_cpu))) - return 0; - - aggregate_walk_tree(aggregate_get_down, aggregate_get_up, sd); - return 1; -} - -static void put_aggregate(struct sched_domain *sd) -{ - spin_unlock(&per_cpu(aggregate_lock, sd->first_cpu)); -} - -static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) -{ - cfs_rq->shares = shares; -} - -#else - -static inline void init_aggregate(void) -{ -} - -static inline int get_aggregate(struct sched_domain *sd) -{ - return 0; -} - -static inline void put_aggregate(struct sched_domain *sd) -{ -} -#endif - #else /* CONFIG_SMP */ #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1864,14 +1512,26 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) #define sched_class_highest (&rt_sched_class) -static void inc_nr_running(struct rq *rq) +static inline void inc_load(struct rq *rq, const struct task_struct *p) +{ + update_load_add(&rq->load, p->se.load.weight); +} + +static inline void dec_load(struct rq *rq, const struct task_struct *p) +{ + update_load_sub(&rq->load, p->se.load.weight); +} + +static void inc_nr_running(struct task_struct *p, struct rq *rq) { rq->nr_running++; + inc_load(rq, p); } -static void dec_nr_running(struct rq *rq) +static void dec_nr_running(struct task_struct *p, struct rq *rq) { rq->nr_running--; + dec_load(rq, p); } static void set_load_weight(struct task_struct *p) @@ -1963,7 +1623,7 @@ static void activate_task(struct rq *rq, struct task_struct *p, int wakeup) rq->nr_uninterruptible--; enqueue_task(rq, p, wakeup); - inc_nr_running(rq); + inc_nr_running(p, rq); } /* @@ -1975,7 +1635,7 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep) rq->nr_uninterruptible++; dequeue_task(rq, p, sleep); - dec_nr_running(rq); + dec_nr_running(p, rq); } /** @@ -2157,6 +1817,7 @@ void wait_task_inactive(struct task_struct *p) * just go back and repeat. */ rq = task_rq_lock(p, &flags); + trace_kernel_sched_wait(p); running = task_running(rq, p); on_rq = p->se.on_rq; task_rq_unlock(rq, &flags); @@ -2500,9 +2161,7 @@ out_activate: success = 1; out_running: - trace_mark(kernel_sched_wakeup, - "pid %d state %ld ## rq %p task %p rq->curr %p", - p->pid, p->state, rq, p, rq->curr); + trace_kernel_sched_wakeup(rq, p); check_preempt_curr(rq, p); p->state = TASK_RUNNING; @@ -2631,11 +2290,9 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) * management (if any): */ p->sched_class->task_new(rq, p); - inc_nr_running(rq); + inc_nr_running(p, rq); } - trace_mark(kernel_sched_wakeup_new, - "pid %d state %ld ## rq %p task %p rq->curr %p", - p->pid, p->state, rq, p, rq->curr); + trace_kernel_sched_wakeup_new(rq, p); check_preempt_curr(rq, p); #ifdef CONFIG_SMP if (p->sched_class->task_wake_up) @@ -2808,11 +2465,8 @@ context_switch(struct rq *rq, struct task_struct *prev, struct mm_struct *mm, *oldmm; prepare_task_switch(rq, prev, next); - trace_mark(kernel_sched_schedule, - "prev_pid %d next_pid %d prev_state %ld " - "## rq %p prev %p next %p", - prev->pid, next->pid, prev->state, - rq, prev, next); + + trace_kernel_sched_switch(rq, prev, next); mm = next->mm; oldmm = prev->active_mm; /* @@ -3045,6 +2699,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu) || unlikely(cpu_is_offline(dest_cpu))) goto out; + trace_kernel_sched_migrate_task(p, cpu_of(rq), dest_cpu); /* force the process onto the specified CPU */ if (migrate_task(p, dest_cpu, &req)) { /* Need to wait for migration thread (might exit: take ref). */ @@ -3630,12 +3285,9 @@ static int load_balance(int this_cpu, struct rq *this_rq, unsigned long imbalance; struct rq *busiest; unsigned long flags; - int unlock_aggregate; cpus_setall(*cpus); - unlock_aggregate = get_aggregate(sd); - /* * When power savings policy is enabled for the parent domain, idle * sibling can pick up load irrespective of busy siblings. In this case, @@ -3751,9 +3403,8 @@ redo: if (!ld_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER && !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE)) - ld_moved = -1; - - goto out; + return -1; + return ld_moved; out_balanced: schedstat_inc(sd, lb_balanced[idle]); @@ -3768,13 +3419,8 @@ out_one_pinned: if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE)) - ld_moved = -1; - else - ld_moved = 0; -out: - if (unlock_aggregate) - put_aggregate(sd); - return ld_moved; + return -1; + return 0; } /* @@ -4481,7 +4127,7 @@ static inline void schedule_debug(struct task_struct *prev) * schedule() atomically, we ignore that path for now. * Otherwise, whine if we are scheduling when we should not be. */ - if (unlikely(in_atomic_preempt_off()) && unlikely(!prev->exit_state)) + if (unlikely(in_atomic_preempt_off() && !prev->exit_state)) __schedule_bug(prev); profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -4982,8 +4628,10 @@ void set_user_nice(struct task_struct *p, long nice) goto out_unlock; } on_rq = p->se.on_rq; - if (on_rq) + if (on_rq) { dequeue_task(rq, p, 0); + dec_load(rq, p); + } p->static_prio = NICE_TO_PRIO(nice); set_load_weight(p); @@ -4993,6 +4641,7 @@ void set_user_nice(struct task_struct *p, long nice) if (on_rq) { enqueue_task(rq, p, 0); + inc_load(rq, p); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -7367,7 +7016,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, ALLNODES); set_domain_attribute(sd, attr); sd->span = *cpu_map; - sd->first_cpu = first_cpu(sd->span); cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask); p = sd; sd_allnodes = 1; @@ -7378,7 +7026,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, NODE); set_domain_attribute(sd, attr); sched_domain_node_span(cpu_to_node(i), &sd->span); - sd->first_cpu = first_cpu(sd->span); sd->parent = p; if (p) p->child = sd; @@ -7390,7 +7037,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, CPU); set_domain_attribute(sd, attr); sd->span = *nodemask; - sd->first_cpu = first_cpu(sd->span); sd->parent = p; if (p) p->child = sd; @@ -7402,7 +7048,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, MC); set_domain_attribute(sd, attr); sd->span = cpu_coregroup_map(i); - sd->first_cpu = first_cpu(sd->span); cpus_and(sd->span, sd->span, *cpu_map); sd->parent = p; p->child = sd; @@ -7415,7 +7060,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, SIBLING); set_domain_attribute(sd, attr); sd->span = per_cpu(cpu_sibling_map, i); - sd->first_cpu = first_cpu(sd->span); cpus_and(sd->span, sd->span, *cpu_map); sd->parent = p; p->child = sd; @@ -7619,8 +7263,8 @@ static int build_sched_domains(const cpumask_t *cpu_map) static cpumask_t *doms_cur; /* current sched domains */ static int ndoms_cur; /* number of sched domains in 'doms_cur' */ -static struct sched_domain_attr *dattr_cur; /* attribues of custom domains - in 'doms_cur' */ +static struct sched_domain_attr *dattr_cur; + /* attribues of custom domains in 'doms_cur' */ /* * Special case: If a kmalloc of a doms_cur partition (array of @@ -8085,7 +7729,6 @@ void __init sched_init(void) } #ifdef CONFIG_SMP - init_aggregate(); init_defrootdomain(); #endif @@ -8650,11 +8293,14 @@ void sched_move_task(struct task_struct *tsk) #endif #ifdef CONFIG_FAIR_GROUP_SCHED -static void __set_se_shares(struct sched_entity *se, unsigned long shares) +static void set_se_shares(struct sched_entity *se, unsigned long shares) { struct cfs_rq *cfs_rq = se->cfs_rq; + struct rq *rq = cfs_rq->rq; int on_rq; + spin_lock_irq(&rq->lock); + on_rq = se->on_rq; if (on_rq) dequeue_entity(cfs_rq, se, 0); @@ -8664,17 +8310,8 @@ static void __set_se_shares(struct sched_entity *se, unsigned long shares) if (on_rq) enqueue_entity(cfs_rq, se, 0); -} -static void set_se_shares(struct sched_entity *se, unsigned long shares) -{ - struct cfs_rq *cfs_rq = se->cfs_rq; - struct rq *rq = cfs_rq->rq; - unsigned long flags; - - spin_lock_irqsave(&rq->lock, flags); - __set_se_shares(se, shares); - spin_unlock_irqrestore(&rq->lock, flags); + spin_unlock_irq(&rq->lock); } static DEFINE_MUTEX(shares_mutex); @@ -8713,13 +8350,8 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) * w/o tripping rebalance_share or load_balance_fair. */ tg->shares = shares; - for_each_possible_cpu(i) { - /* - * force a rebalance - */ - cfs_rq_set_shares(tg->cfs_rq[i], 0); + for_each_possible_cpu(i) set_se_shares(tg->se[i], shares); - } /* * Enable load balance activity on this group, by inserting it back on diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 9c597e37f7de..ce05271219ab 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -59,22 +59,26 @@ static inline struct sched_clock_data *cpu_sdc(int cpu) return &per_cpu(sched_clock_data, cpu); } +static __read_mostly int sched_clock_running; + void sched_clock_init(void) { u64 ktime_now = ktime_to_ns(ktime_get()); - u64 now = 0; + unsigned long now_jiffies = jiffies; int cpu; for_each_possible_cpu(cpu) { struct sched_clock_data *scd = cpu_sdc(cpu); scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; - scd->prev_jiffies = jiffies; - scd->prev_raw = now; - scd->tick_raw = now; + scd->prev_jiffies = now_jiffies; + scd->prev_raw = 0; + scd->tick_raw = 0; scd->tick_gtod = ktime_now; scd->clock = ktime_now; } + + sched_clock_running = 1; } /* @@ -136,6 +140,9 @@ u64 sched_clock_cpu(int cpu) struct sched_clock_data *scd = cpu_sdc(cpu); u64 now, clock; + if (unlikely(!sched_clock_running)) + return 0ull; + WARN_ON_ONCE(!irqs_disabled()); now = sched_clock(); @@ -174,6 +181,9 @@ void sched_clock_tick(void) struct sched_clock_data *scd = this_scd(); u64 now, now_gtod; + if (unlikely(!sched_clock_running)) + return; + WARN_ON_ONCE(!irqs_disabled()); now = sched_clock(); diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 5f06118fbc31..8bb713040ac9 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -167,11 +167,6 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) #endif SEQ_printf(m, " .%-30s: %ld\n", "nr_spread_over", cfs_rq->nr_spread_over); -#ifdef CONFIG_FAIR_GROUP_SCHED -#ifdef CONFIG_SMP - SEQ_printf(m, " .%-30s: %lu\n", "shares", cfs_rq->shares); -#endif -#endif } static void print_cpu(struct seq_file *m, int cpu) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index e24ecd39c4b8..08ae848b71d4 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -334,34 +334,6 @@ int sched_nr_latency_handler(struct ctl_table *table, int write, #endif /* - * delta *= w / rw - */ -static inline unsigned long -calc_delta_weight(unsigned long delta, struct sched_entity *se) -{ - for_each_sched_entity(se) { - delta = calc_delta_mine(delta, - se->load.weight, &cfs_rq_of(se)->load); - } - - return delta; -} - -/* - * delta *= rw / w - */ -static inline unsigned long -calc_delta_fair(unsigned long delta, struct sched_entity *se) -{ - for_each_sched_entity(se) { - delta = calc_delta_mine(delta, - cfs_rq_of(se)->load.weight, &se->load); - } - - return delta; -} - -/* * The idea is to set a period in which each task runs once. * * When there are too many tasks (sysctl_sched_nr_latency) we have to stretch @@ -390,54 +362,47 @@ static u64 __sched_period(unsigned long nr_running) */ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return calc_delta_weight(__sched_period(cfs_rq->nr_running), se); + u64 slice = __sched_period(cfs_rq->nr_running); + + for_each_sched_entity(se) { + cfs_rq = cfs_rq_of(se); + + slice *= se->load.weight; + do_div(slice, cfs_rq->load.weight); + } + + + return slice; } /* * We calculate the vruntime slice of a to be inserted task * - * vs = s*rw/w = p + * vs = s/w = p/rw */ static u64 sched_vslice_add(struct cfs_rq *cfs_rq, struct sched_entity *se) { unsigned long nr_running = cfs_rq->nr_running; + unsigned long weight; + u64 vslice; if (!se->on_rq) nr_running++; - return __sched_period(nr_running); -} - -/* - * The goal of calc_delta_asym() is to be asymmetrically around NICE_0_LOAD, in - * that it favours >=0 over <0. - * - * -20 | - * | - * 0 --------+------- - * .' - * 19 .' - * - */ -static unsigned long -calc_delta_asym(unsigned long delta, struct sched_entity *se) -{ - struct load_weight lw = { - .weight = NICE_0_LOAD, - .inv_weight = 1UL << (WMULT_SHIFT-NICE_0_SHIFT) - }; + vslice = __sched_period(nr_running); for_each_sched_entity(se) { - struct load_weight *se_lw = &se->load; + cfs_rq = cfs_rq_of(se); - if (se->load.weight < NICE_0_LOAD) - se_lw = &lw; + weight = cfs_rq->load.weight; + if (!se->on_rq) + weight += se->load.weight; - delta = calc_delta_mine(delta, - cfs_rq_of(se)->load.weight, se_lw); + vslice *= NICE_0_LOAD; + do_div(vslice, weight); } - return delta; + return vslice; } /* @@ -454,7 +419,11 @@ __update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr, curr->sum_exec_runtime += delta_exec; schedstat_add(cfs_rq, exec_clock, delta_exec); - delta_exec_weighted = calc_delta_fair(delta_exec, curr); + delta_exec_weighted = delta_exec; + if (unlikely(curr->load.weight != NICE_0_LOAD)) { + delta_exec_weighted = calc_delta_fair(delta_exec_weighted, + &curr->load); + } curr->vruntime += delta_exec_weighted; } @@ -541,27 +510,10 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se) * Scheduling class queueing methods: */ -#if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED -static void -add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight) -{ - cfs_rq->task_weight += weight; -} -#else -static inline void -add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight) -{ -} -#endif - static void account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_add(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - inc_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) - add_cfs_task_weight(cfs_rq, se->load.weight); cfs_rq->nr_running++; se->on_rq = 1; list_add(&se->group_node, &cfs_rq->tasks); @@ -571,10 +523,6 @@ static void account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_sub(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - dec_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) - add_cfs_task_weight(cfs_rq, -se->load.weight); cfs_rq->nr_running--; se->on_rq = 0; list_del_init(&se->group_node); @@ -661,17 +609,8 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) if (!initial) { /* sleeps upto a single latency don't count. */ - if (sched_feat(NEW_FAIR_SLEEPERS)) { - unsigned long thresh = sysctl_sched_latency; - - /* - * convert the sleeper threshold into virtual time - */ - if (sched_feat(NORMALIZED_SLEEPER)) - thresh = calc_delta_fair(thresh, se); - - vruntime -= thresh; - } + if (sched_feat(NEW_FAIR_SLEEPERS)) + vruntime -= sysctl_sched_latency; /* ensure we never gain time by being placed backwards. */ vruntime = max_vruntime(se->vruntime, vruntime); @@ -1057,16 +996,27 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, struct task_struct *curr = this_rq->curr; unsigned long tl = this_load; unsigned long tl_per_task; + int balanced; - if (!(this_sd->flags & SD_WAKE_AFFINE)) + if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS)) return 0; /* + * If sync wakeup then subtract the (maximum possible) + * effect of the currently running task from the load + * of the current CPU: + */ + if (sync) + tl -= current->se.load.weight; + + balanced = 100*(tl + p->se.load.weight) <= imbalance*load; + + /* * If the currently running task will sleep within * a reasonable amount of time then attract this newly * woken task: */ - if (sync && curr->sched_class == &fair_sched_class) { + if (sync && balanced && curr->sched_class == &fair_sched_class) { if (curr->se.avg_overlap < sysctl_sched_migration_cost && p->se.avg_overlap < sysctl_sched_migration_cost) return 1; @@ -1075,16 +1025,8 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, schedstat_inc(p, se.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); - /* - * If sync wakeup then subtract the (maximum possible) - * effect of the currently running task from the load - * of the current CPU: - */ - if (sync) - tl -= current->se.load.weight; - if ((tl <= load && tl + target_load(prev_cpu, idx) <= tl_per_task) || - 100*(tl + p->se.load.weight) <= imbalance*load) { + balanced) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and @@ -1169,10 +1111,11 @@ static unsigned long wakeup_gran(struct sched_entity *se) unsigned long gran = sysctl_sched_wakeup_granularity; /* - * More easily preempt - nice tasks, while not making it harder for - * + nice tasks. + * More easily preempt - nice tasks, while not making + * it harder for + nice tasks. */ - gran = calc_delta_asym(sysctl_sched_wakeup_granularity, se); + if (unlikely(se->load.weight > NICE_0_LOAD)) + gran = calc_delta_fair(gran, &se->load); return gran; } @@ -1366,90 +1309,75 @@ static struct task_struct *load_balance_next_fair(void *arg) return __load_balance_iterator(cfs_rq, cfs_rq->balance_iterator); } -static unsigned long -__load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, - unsigned long max_load_move, struct sched_domain *sd, - enum cpu_idle_type idle, int *all_pinned, int *this_best_prio, - struct cfs_rq *cfs_rq) +#ifdef CONFIG_FAIR_GROUP_SCHED +static int cfs_rq_best_prio(struct cfs_rq *cfs_rq) { - struct rq_iterator cfs_rq_iterator; + struct sched_entity *curr; + struct task_struct *p; - cfs_rq_iterator.start = load_balance_start_fair; - cfs_rq_iterator.next = load_balance_next_fair; - cfs_rq_iterator.arg = cfs_rq; + if (!cfs_rq->nr_running || !first_fair(cfs_rq)) + return MAX_PRIO; + + curr = cfs_rq->curr; + if (!curr) + curr = __pick_next_entity(cfs_rq); + + p = task_of(curr); - return balance_tasks(this_rq, this_cpu, busiest, - max_load_move, sd, idle, all_pinned, - this_best_prio, &cfs_rq_iterator); + return p->prio; } +#endif -#ifdef CONFIG_FAIR_GROUP_SCHED static unsigned long load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, unsigned long max_load_move, struct sched_domain *sd, enum cpu_idle_type idle, int *all_pinned, int *this_best_prio) { + struct cfs_rq *busy_cfs_rq; long rem_load_move = max_load_move; - int busiest_cpu = cpu_of(busiest); - struct task_group *tg; - - rcu_read_lock(); - list_for_each_entry(tg, &task_groups, list) { - long imbalance; - unsigned long this_weight, busiest_weight; - long rem_load, max_load, moved_load; - - /* - * empty group - */ - if (!aggregate(tg, sd)->task_weight) - continue; - - rem_load = rem_load_move * aggregate(tg, sd)->rq_weight; - rem_load /= aggregate(tg, sd)->load + 1; - - this_weight = tg->cfs_rq[this_cpu]->task_weight; - busiest_weight = tg->cfs_rq[busiest_cpu]->task_weight; + struct rq_iterator cfs_rq_iterator; - imbalance = (busiest_weight - this_weight) / 2; + cfs_rq_iterator.start = load_balance_start_fair; + cfs_rq_iterator.next = load_balance_next_fair; - if (imbalance < 0) - imbalance = busiest_weight; + for_each_leaf_cfs_rq(busiest, busy_cfs_rq) { +#ifdef CONFIG_FAIR_GROUP_SCHED + struct cfs_rq *this_cfs_rq; + long imbalance; + unsigned long maxload; - max_load = max(rem_load, imbalance); - moved_load = __load_balance_fair(this_rq, this_cpu, busiest, - max_load, sd, idle, all_pinned, this_best_prio, - tg->cfs_rq[busiest_cpu]); + this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu); - if (!moved_load) + imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight; + /* Don't pull if this_cfs_rq has more load than busy_cfs_rq */ + if (imbalance <= 0) continue; - move_group_shares(tg, sd, busiest_cpu, this_cpu); + /* Don't pull more than imbalance/2 */ + imbalance /= 2; + maxload = min(rem_load_move, imbalance); - moved_load *= aggregate(tg, sd)->load; - moved_load /= aggregate(tg, sd)->rq_weight + 1; + *this_best_prio = cfs_rq_best_prio(this_cfs_rq); +#else +# define maxload rem_load_move +#endif + /* + * pass busy_cfs_rq argument into + * load_balance_[start|next]_fair iterators + */ + cfs_rq_iterator.arg = busy_cfs_rq; + rem_load_move -= balance_tasks(this_rq, this_cpu, busiest, + maxload, sd, idle, all_pinned, + this_best_prio, + &cfs_rq_iterator); - rem_load_move -= moved_load; - if (rem_load_move < 0) + if (rem_load_move <= 0) break; } - rcu_read_unlock(); return max_load_move - rem_load_move; } -#else -static unsigned long -load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, - unsigned long max_load_move, - struct sched_domain *sd, enum cpu_idle_type idle, - int *all_pinned, int *this_best_prio) -{ - return __load_balance_fair(this_rq, this_cpu, busiest, - max_load_move, sd, idle, all_pinned, - this_best_prio, &busiest->cfs); -} -#endif static int move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 060e87b0cb1c..3432d573205d 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -513,8 +513,6 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup) */ for_each_sched_rt_entity(rt_se) enqueue_rt_entity(rt_se); - - inc_cpu_load(rq, p->se.load.weight); } static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) @@ -534,8 +532,6 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) if (rt_rq && rt_rq->rt_nr_running) enqueue_rt_entity(rt_se); } - - dec_cpu_load(rq, p->se.load.weight); } /* diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 5bae2e0c3ff2..a38878e0e49d 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -67,6 +67,7 @@ static int show_schedstat(struct seq_file *seq, void *v) preempt_enable(); #endif } + kfree(mask_str); return 0; } diff --git a/kernel/sched_trace.h b/kernel/sched_trace.h new file mode 100644 index 000000000000..29b48f34fd02 --- /dev/null +++ b/kernel/sched_trace.h @@ -0,0 +1,41 @@ +#include <linux/marker.h> + +static inline void trace_kernel_sched_wait(struct task_struct *p) +{ + trace_mark(kernel_sched_wait_task, "pid %d state %ld", + p->pid, p->state); +} + +static inline +void trace_kernel_sched_wakeup(struct rq *rq, struct task_struct *p) +{ + trace_mark(kernel_sched_wakeup, + "pid %d state %ld ## rq %p task %p rq->curr %p", + p->pid, p->state, rq, p, rq->curr); +} + +static inline +void trace_kernel_sched_wakeup_new(struct rq *rq, struct task_struct *p) +{ + trace_mark(kernel_sched_wakeup_new, + "pid %d state %ld ## rq %p task %p rq->curr %p", + p->pid, p->state, rq, p, rq->curr); +} + +static inline void trace_kernel_sched_switch(struct rq *rq, + struct task_struct *prev, struct task_struct *next) +{ + trace_mark(kernel_sched_schedule, + "prev_pid %d next_pid %d prev_state %ld " + "## rq %p prev %p next %p", + prev->pid, next->pid, prev->state, + rq, prev, next); +} + +static inline void +trace_kernel_sched_migrate_task(struct task_struct *p, int src, int dst) +{ + trace_mark(kernel_sched_migrate_task, + "pid %d state %ld dest_cpu %d", + p->pid, p->state, dst); +} diff --git a/kernel/signal.c b/kernel/signal.c index 72bb4f51f963..6c0958e52ea7 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -231,6 +231,40 @@ void flush_signals(struct task_struct *t) spin_unlock_irqrestore(&t->sighand->siglock, flags); } +static void __flush_itimer_signals(struct sigpending *pending) +{ + sigset_t signal, retain; + struct sigqueue *q, *n; + + signal = pending->signal; + sigemptyset(&retain); + + list_for_each_entry_safe(q, n, &pending->list, list) { + int sig = q->info.si_signo; + + if (likely(q->info.si_code != SI_TIMER)) { + sigaddset(&retain, sig); + } else { + sigdelset(&signal, sig); + list_del_init(&q->list); + __sigqueue_free(q); + } + } + + sigorsets(&pending->signal, &signal, &retain); +} + +void flush_itimer_signals(void) +{ + struct task_struct *tsk = current; + unsigned long flags; + + spin_lock_irqsave(&tsk->sighand->siglock, flags); + __flush_itimer_signals(&tsk->pending); + __flush_itimer_signals(&tsk->signal->shared_pending); + spin_unlock_irqrestore(&tsk->sighand->siglock, flags); +} + void ignore_signals(struct task_struct *t) { int i; @@ -1240,17 +1274,22 @@ void sigqueue_free(struct sigqueue *q) BUG_ON(!(q->flags & SIGQUEUE_PREALLOC)); /* - * If the signal is still pending remove it from the - * pending queue. We must hold ->siglock while testing - * q->list to serialize with collect_signal(). + * We must hold ->siglock while testing q->list + * to serialize with collect_signal() or with + * __exit_signal()->flush_sigqueue(). */ spin_lock_irqsave(lock, flags); + q->flags &= ~SIGQUEUE_PREALLOC; + /* + * If it is queued it will be freed when dequeued, + * like the "regular" sigqueue. + */ if (!list_empty(&q->list)) - list_del_init(&q->list); + q = NULL; spin_unlock_irqrestore(lock, flags); - q->flags &= ~SIGQUEUE_PREALLOC; - __sigqueue_free(q); + if (q) + __sigqueue_free(q); } int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group) diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 0101aeef7ed7..b7350bbfb076 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -62,8 +62,7 @@ static int stopmachine(void *cpu) * help our sisters onto their CPUs. */ if (!prepared && !irqs_disabled) yield(); - else - cpu_relax(); + cpu_relax(); } /* Ack: we are exiting. */ @@ -106,8 +105,10 @@ static int stop_machine(void) } /* Wait for them all to come to life. */ - while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) + while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) { yield(); + cpu_relax(); + } /* If some failed, kill them all. */ if (ret < 0) { diff --git a/kernel/sys.c b/kernel/sys.c index 895d2d4c9493..14e97282eb6c 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1652,7 +1652,7 @@ asmlinkage long sys_umask(int mask) asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5) { - long uninitialized_var(error); + long error = 0; if (security_task_prctl(option, arg2, arg3, arg4, arg5, &error)) return error; @@ -1701,9 +1701,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = PR_TIMING_STATISTICAL; break; case PR_SET_TIMING: - if (arg2 == PR_TIMING_STATISTICAL) - error = 0; - else + if (arg2 != PR_TIMING_STATISTICAL) error = -EINVAL; break; diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 7aec123ec1d8..71d17de17288 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -19,5 +19,6 @@ obj-$(CONFIG_FTRACE) += trace_functions.o obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o +obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o libftrace-y := ftrace.o diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 89bd9a6f52ec..0118979e211f 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -151,8 +151,6 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) #ifdef CONFIG_DYNAMIC_FTRACE static struct task_struct *ftraced_task; -static DECLARE_WAIT_QUEUE_HEAD(ftraced_waiters); -static unsigned long ftraced_iteration_counter; enum { FTRACE_ENABLE_CALLS = (1 << 0), @@ -170,7 +168,7 @@ static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu); static DEFINE_SPINLOCK(ftrace_shutdown_lock); static DEFINE_MUTEX(ftraced_lock); -static DEFINE_MUTEX(ftrace_filter_lock); +static DEFINE_MUTEX(ftrace_regex_lock); struct ftrace_page { struct ftrace_page *next; @@ -189,6 +187,7 @@ static struct ftrace_page *ftrace_pages; static int ftraced_trigger; static int ftraced_suspend; +static int ftraced_stop; static int ftrace_record_suspend; @@ -201,7 +200,7 @@ ftrace_ip_in_hash(unsigned long ip, unsigned long key) struct hlist_node *t; int found = 0; - hlist_for_each_entry(p, t, &ftrace_hash[key], node) { + hlist_for_each_entry_rcu(p, t, &ftrace_hash[key], node) { if (p->ip == ip) { found = 1; break; @@ -214,7 +213,13 @@ ftrace_ip_in_hash(unsigned long ip, unsigned long key) static inline void ftrace_add_hash(struct dyn_ftrace *node, unsigned long key) { - hlist_add_head(&node->node, &ftrace_hash[key]); + hlist_add_head_rcu(&node->node, &ftrace_hash[key]); +} + +/* called from kstop_machine */ +static inline void ftrace_del_hash(struct dyn_ftrace *node) +{ + hlist_del(&node->node); } static void ftrace_free_rec(struct dyn_ftrace *rec) @@ -301,13 +306,6 @@ ftrace_record_ip(unsigned long ip) if (ftrace_ip_in_hash(ip, key)) goto out_unlock; - /* - * There's a slight race that the ftraced will update the - * hash and reset here. If it is already converted, skip it. - */ - if (ftrace_ip_converted(ip)) - goto out_unlock; - node = ftrace_alloc_dyn_node(ip); if (!node) goto out_unlock; @@ -333,17 +331,15 @@ ftrace_record_ip(unsigned long ip) #define FTRACE_ADDR ((long)(ftrace_caller)) #define MCOUNT_ADDR ((long)(mcount)) -static void +static int __ftrace_replace_code(struct dyn_ftrace *rec, unsigned char *old, unsigned char *new, int enable) { - unsigned long ip; - int failed; + unsigned long ip, fl; ip = rec->ip; if (ftrace_filtered && enable) { - unsigned long fl; /* * If filtering is on: * @@ -356,14 +352,17 @@ __ftrace_replace_code(struct dyn_ftrace *rec, * If this record is not set to be filtered * and it is not enabled do nothing. * + * If this record is set not to trace then + * do nothing. + * * If this record is not set to be filtered and * it is enabled, disable it. */ fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED); if ((fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) || - (fl == 0)) - return; + (fl == 0) || (rec->flags & FTRACE_FL_NOTRACE)) + return 0; /* * If it is enabled disable it, @@ -380,41 +379,39 @@ __ftrace_replace_code(struct dyn_ftrace *rec, } } else { - if (enable) + if (enable) { + /* + * If this record is set not to trace and is + * not enabled, do nothing. + */ + fl = rec->flags & (FTRACE_FL_NOTRACE | FTRACE_FL_ENABLED); + if (fl == FTRACE_FL_NOTRACE) + return 0; + new = ftrace_call_replace(ip, FTRACE_ADDR); - else + } else old = ftrace_call_replace(ip, FTRACE_ADDR); if (enable) { if (rec->flags & FTRACE_FL_ENABLED) - return; + return 0; rec->flags |= FTRACE_FL_ENABLED; } else { if (!(rec->flags & FTRACE_FL_ENABLED)) - return; + return 0; rec->flags &= ~FTRACE_FL_ENABLED; } } - failed = ftrace_modify_code(ip, old, new); - if (failed) { - unsigned long key; - /* It is possible that the function hasn't been converted yet */ - key = hash_long(ip, FTRACE_HASHBITS); - if (!ftrace_ip_in_hash(ip, key)) { - rec->flags |= FTRACE_FL_FAILED; - ftrace_free_rec(rec); - } - - } + return ftrace_modify_code(ip, old, new); } static void ftrace_replace_code(int enable) { + int i, failed; unsigned char *new = NULL, *old = NULL; struct dyn_ftrace *rec; struct ftrace_page *pg; - int i; if (enable) old = ftrace_nop_replace(); @@ -429,7 +426,15 @@ static void ftrace_replace_code(int enable) if (rec->flags & FTRACE_FL_FAILED) continue; - __ftrace_replace_code(rec, old, new, enable); + failed = __ftrace_replace_code(rec, old, new, enable); + if (failed && (rec->flags & FTRACE_FL_CONVERTED)) { + rec->flags |= FTRACE_FL_FAILED; + if ((system_state == SYSTEM_BOOTING) || + !core_kernel_text(rec->ip)) { + ftrace_del_hash(rec); + ftrace_free_rec(rec); + } + } } } } @@ -443,7 +448,7 @@ static void ftrace_shutdown_replenish(void) ftrace_pages->next = (void *)get_zeroed_page(GFP_KERNEL); } -static void +static int ftrace_code_disable(struct dyn_ftrace *rec) { unsigned long ip; @@ -458,18 +463,26 @@ ftrace_code_disable(struct dyn_ftrace *rec) failed = ftrace_modify_code(ip, call, nop); if (failed) { rec->flags |= FTRACE_FL_FAILED; - ftrace_free_rec(rec); + return 0; } + return 1; } +static int __ftrace_update_code(void *ignore); + static int __ftrace_modify_code(void *data) { unsigned long addr; int *command = data; - if (*command & FTRACE_ENABLE_CALLS) + if (*command & FTRACE_ENABLE_CALLS) { + /* + * Update any recorded ips now that we have the + * machine stopped + */ + __ftrace_update_code(NULL); ftrace_replace_code(1); - else if (*command & FTRACE_DISABLE_CALLS) + } else if (*command & FTRACE_DISABLE_CALLS) ftrace_replace_code(0); if (*command & FTRACE_UPDATE_TRACE_FUNC) @@ -491,6 +504,25 @@ static void ftrace_run_update_code(int command) stop_machine_run(__ftrace_modify_code, &command, NR_CPUS); } +void ftrace_disable_daemon(void) +{ + /* Stop the daemon from calling kstop_machine */ + mutex_lock(&ftraced_lock); + ftraced_stop = 1; + mutex_unlock(&ftraced_lock); + + ftrace_force_update(); +} + +void ftrace_enable_daemon(void) +{ + mutex_lock(&ftraced_lock); + ftraced_stop = 0; + mutex_unlock(&ftraced_lock); + + ftrace_force_update(); +} + static ftrace_func_t saved_ftrace_func; static void ftrace_startup(void) @@ -584,13 +616,13 @@ unsigned long ftrace_update_tot_cnt; static int __ftrace_update_code(void *ignore) { struct dyn_ftrace *p; - struct hlist_head head; - struct hlist_node *t; + struct hlist_node *t, *n; int save_ftrace_enabled; cycle_t start, stop; int i; /* Don't be recording funcs now */ + ftrace_record_suspend++; save_ftrace_enabled = ftrace_enabled; ftrace_enabled = 0; @@ -599,35 +631,54 @@ static int __ftrace_update_code(void *ignore) /* No locks needed, the machine is stopped! */ for (i = 0; i < FTRACE_HASHSIZE; i++) { - if (hlist_empty(&ftrace_hash[i])) - continue; + /* all CPUS are stopped, we are safe to modify code */ + hlist_for_each_entry_safe(p, t, n, &ftrace_hash[i], node) { + /* Skip over failed records which have not been + * freed. */ + if (p->flags & FTRACE_FL_FAILED) + continue; - head = ftrace_hash[i]; - INIT_HLIST_HEAD(&ftrace_hash[i]); + /* Unconverted records are always at the head of the + * hash bucket. Once we encounter a converted record, + * simply skip over to the next bucket. Saves ftraced + * some processor cycles (ftrace does its bid for + * global warming :-p ). */ + if (p->flags & (FTRACE_FL_CONVERTED)) + break; - /* all CPUS are stopped, we are safe to modify code */ - hlist_for_each_entry(p, t, &head, node) { - ftrace_code_disable(p); - ftrace_update_cnt++; + if (ftrace_code_disable(p)) { + p->flags |= FTRACE_FL_CONVERTED; + ftrace_update_cnt++; + } else { + if ((system_state == SYSTEM_BOOTING) || + !core_kernel_text(p->ip)) { + ftrace_del_hash(p); + ftrace_free_rec(p); + } + } } - } stop = ftrace_now(raw_smp_processor_id()); ftrace_update_time = stop - start; ftrace_update_tot_cnt += ftrace_update_cnt; + ftraced_trigger = 0; ftrace_enabled = save_ftrace_enabled; + ftrace_record_suspend--; return 0; } -static void ftrace_update_code(void) +static int ftrace_update_code(void) { - if (unlikely(ftrace_disabled)) - return; + if (unlikely(ftrace_disabled) || + !ftrace_enabled || !ftraced_trigger) + return 0; stop_machine_run(__ftrace_update_code, NULL, NR_CPUS); + + return 1; } static int ftraced(void *ignore) @@ -646,14 +697,13 @@ static int ftraced(void *ignore) mutex_lock(&ftrace_sysctl_lock); mutex_lock(&ftraced_lock); - if (ftrace_enabled && ftraced_trigger && !ftraced_suspend) { - ftrace_record_suspend++; - ftrace_update_code(); + if (!ftraced_suspend && !ftraced_stop && + ftrace_update_code()) { usecs = nsecs_to_usecs(ftrace_update_time); if (ftrace_update_tot_cnt > 100000) { ftrace_update_tot_cnt = 0; pr_info("hm, dftrace overflow: %lu change%s" - " (%lu total) in %lu usec%s\n", + " (%lu total) in %lu usec%s\n", ftrace_update_cnt, ftrace_update_cnt != 1 ? "s" : "", ftrace_update_tot_cnt, @@ -661,15 +711,10 @@ static int ftraced(void *ignore) ftrace_disabled = 1; WARN_ON_ONCE(1); } - ftraced_trigger = 0; - ftrace_record_suspend--; } - ftraced_iteration_counter++; mutex_unlock(&ftraced_lock); mutex_unlock(&ftrace_sysctl_lock); - wake_up_interruptible(&ftraced_waiters); - ftrace_shutdown_replenish(); } __set_current_state(TASK_RUNNING); @@ -721,6 +766,8 @@ static int __init ftrace_dyn_table_alloc(void) enum { FTRACE_ITER_FILTER = (1 << 0), FTRACE_ITER_CONT = (1 << 1), + FTRACE_ITER_NOTRACE = (1 << 2), + FTRACE_ITER_FAILURES = (1 << 3), }; #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */ @@ -752,9 +799,18 @@ t_next(struct seq_file *m, void *v, loff_t *pos) } } else { rec = &iter->pg->records[iter->idx++]; - if ((rec->flags & FTRACE_FL_FAILED) || + if ((!(iter->flags & FTRACE_ITER_FAILURES) && + (rec->flags & FTRACE_FL_FAILED)) || + + ((iter->flags & FTRACE_ITER_FAILURES) && + (!(rec->flags & FTRACE_FL_FAILED) || + (rec->flags & FTRACE_FL_FREE))) || + ((iter->flags & FTRACE_ITER_FILTER) && - !(rec->flags & FTRACE_FL_FILTER))) { + !(rec->flags & FTRACE_FL_FILTER)) || + + ((iter->flags & FTRACE_ITER_NOTRACE) && + !(rec->flags & FTRACE_FL_NOTRACE))) { rec = NULL; goto retry; } @@ -847,22 +903,42 @@ int ftrace_avail_release(struct inode *inode, struct file *file) return 0; } -static void ftrace_filter_reset(void) +static int +ftrace_failures_open(struct inode *inode, struct file *file) +{ + int ret; + struct seq_file *m; + struct ftrace_iterator *iter; + + ret = ftrace_avail_open(inode, file); + if (!ret) { + m = (struct seq_file *)file->private_data; + iter = (struct ftrace_iterator *)m->private; + iter->flags = FTRACE_ITER_FAILURES; + } + + return ret; +} + + +static void ftrace_filter_reset(int enable) { struct ftrace_page *pg; struct dyn_ftrace *rec; + unsigned long type = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; unsigned i; /* keep kstop machine from running */ preempt_disable(); - ftrace_filtered = 0; + if (enable) + ftrace_filtered = 0; pg = ftrace_pages_start; while (pg) { for (i = 0; i < pg->index; i++) { rec = &pg->records[i]; if (rec->flags & FTRACE_FL_FAILED) continue; - rec->flags &= ~FTRACE_FL_FILTER; + rec->flags &= ~type; } pg = pg->next; } @@ -870,7 +946,7 @@ static void ftrace_filter_reset(void) } static int -ftrace_filter_open(struct inode *inode, struct file *file) +ftrace_regex_open(struct inode *inode, struct file *file, int enable) { struct ftrace_iterator *iter; int ret = 0; @@ -882,15 +958,16 @@ ftrace_filter_open(struct inode *inode, struct file *file) if (!iter) return -ENOMEM; - mutex_lock(&ftrace_filter_lock); + mutex_lock(&ftrace_regex_lock); if ((file->f_mode & FMODE_WRITE) && !(file->f_flags & O_APPEND)) - ftrace_filter_reset(); + ftrace_filter_reset(enable); if (file->f_mode & FMODE_READ) { iter->pg = ftrace_pages_start; iter->pos = -1; - iter->flags = FTRACE_ITER_FILTER; + iter->flags = enable ? FTRACE_ITER_FILTER : + FTRACE_ITER_NOTRACE; ret = seq_open(file, &show_ftrace_seq_ops); if (!ret) { @@ -900,13 +977,25 @@ ftrace_filter_open(struct inode *inode, struct file *file) kfree(iter); } else file->private_data = iter; - mutex_unlock(&ftrace_filter_lock); + mutex_unlock(&ftrace_regex_lock); return ret; } +static int +ftrace_filter_open(struct inode *inode, struct file *file) +{ + return ftrace_regex_open(inode, file, 1); +} + +static int +ftrace_notrace_open(struct inode *inode, struct file *file) +{ + return ftrace_regex_open(inode, file, 0); +} + static ssize_t -ftrace_filter_read(struct file *file, char __user *ubuf, +ftrace_regex_read(struct file *file, char __user *ubuf, size_t cnt, loff_t *ppos) { if (file->f_mode & FMODE_READ) @@ -916,7 +1005,7 @@ ftrace_filter_read(struct file *file, char __user *ubuf, } static loff_t -ftrace_filter_lseek(struct file *file, loff_t offset, int origin) +ftrace_regex_lseek(struct file *file, loff_t offset, int origin) { loff_t ret; @@ -936,13 +1025,14 @@ enum { }; static void -ftrace_match(unsigned char *buff, int len) +ftrace_match(unsigned char *buff, int len, int enable) { char str[KSYM_SYMBOL_LEN]; char *search = NULL; struct ftrace_page *pg; struct dyn_ftrace *rec; int type = MATCH_FULL; + unsigned long flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; unsigned i, match = 0, search_len = 0; for (i = 0; i < len; i++) { @@ -966,7 +1056,8 @@ ftrace_match(unsigned char *buff, int len) /* keep kstop machine from running */ preempt_disable(); - ftrace_filtered = 1; + if (enable) + ftrace_filtered = 1; pg = ftrace_pages_start; while (pg) { for (i = 0; i < pg->index; i++) { @@ -997,7 +1088,7 @@ ftrace_match(unsigned char *buff, int len) break; } if (matched) - rec->flags |= FTRACE_FL_FILTER; + rec->flags |= flag; } pg = pg->next; } @@ -1005,8 +1096,8 @@ ftrace_match(unsigned char *buff, int len) } static ssize_t -ftrace_filter_write(struct file *file, const char __user *ubuf, - size_t cnt, loff_t *ppos) +ftrace_regex_write(struct file *file, const char __user *ubuf, + size_t cnt, loff_t *ppos, int enable) { struct ftrace_iterator *iter; char ch; @@ -1016,7 +1107,7 @@ ftrace_filter_write(struct file *file, const char __user *ubuf, if (!cnt || cnt < 0) return 0; - mutex_lock(&ftrace_filter_lock); + mutex_lock(&ftrace_regex_lock); if (file->f_mode & FMODE_READ) { struct seq_file *m = file->private_data; @@ -1045,7 +1136,6 @@ ftrace_filter_write(struct file *file, const char __user *ubuf, cnt--; } - if (isspace(ch)) { file->f_pos += read; ret = read; @@ -1072,7 +1162,7 @@ ftrace_filter_write(struct file *file, const char __user *ubuf, if (isspace(ch)) { iter->filtered++; iter->buffer[iter->buffer_idx] = 0; - ftrace_match(iter->buffer, iter->buffer_idx); + ftrace_match(iter->buffer, iter->buffer_idx, enable); iter->buffer_idx = 0; } else iter->flags |= FTRACE_ITER_CONT; @@ -1082,11 +1172,39 @@ ftrace_filter_write(struct file *file, const char __user *ubuf, ret = read; out: - mutex_unlock(&ftrace_filter_lock); + mutex_unlock(&ftrace_regex_lock); return ret; } +static ssize_t +ftrace_filter_write(struct file *file, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + return ftrace_regex_write(file, ubuf, cnt, ppos, 1); +} + +static ssize_t +ftrace_notrace_write(struct file *file, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + return ftrace_regex_write(file, ubuf, cnt, ppos, 0); +} + +static void +ftrace_set_regex(unsigned char *buf, int len, int reset, int enable) +{ + if (unlikely(ftrace_disabled)) + return; + + mutex_lock(&ftrace_regex_lock); + if (reset) + ftrace_filter_reset(enable); + if (buf) + ftrace_match(buf, len, enable); + mutex_unlock(&ftrace_regex_lock); +} + /** * ftrace_set_filter - set a function to filter on in ftrace * @buf - the string that holds the function filter text. @@ -1098,24 +1216,31 @@ ftrace_filter_write(struct file *file, const char __user *ubuf, */ void ftrace_set_filter(unsigned char *buf, int len, int reset) { - if (unlikely(ftrace_disabled)) - return; + ftrace_set_regex(buf, len, reset, 1); +} - mutex_lock(&ftrace_filter_lock); - if (reset) - ftrace_filter_reset(); - if (buf) - ftrace_match(buf, len); - mutex_unlock(&ftrace_filter_lock); +/** + * ftrace_set_notrace - set a function to not trace in ftrace + * @buf - the string that holds the function notrace text. + * @len - the length of the string. + * @reset - non zero to reset all filters before applying this filter. + * + * Notrace Filters denote which functions should not be enabled when tracing + * is enabled. If @buf is NULL and reset is set, all functions will be enabled + * for tracing. + */ +void ftrace_set_notrace(unsigned char *buf, int len, int reset) +{ + ftrace_set_regex(buf, len, reset, 0); } static int -ftrace_filter_release(struct inode *inode, struct file *file) +ftrace_regex_release(struct inode *inode, struct file *file, int enable) { struct seq_file *m = (struct seq_file *)file->private_data; struct ftrace_iterator *iter; - mutex_lock(&ftrace_filter_lock); + mutex_lock(&ftrace_regex_lock); if (file->f_mode & FMODE_READ) { iter = m->private; @@ -1126,7 +1251,7 @@ ftrace_filter_release(struct inode *inode, struct file *file) if (iter->buffer_idx) { iter->filtered++; iter->buffer[iter->buffer_idx] = 0; - ftrace_match(iter->buffer, iter->buffer_idx); + ftrace_match(iter->buffer, iter->buffer_idx, enable); } mutex_lock(&ftrace_sysctl_lock); @@ -1137,10 +1262,71 @@ ftrace_filter_release(struct inode *inode, struct file *file) mutex_unlock(&ftrace_sysctl_lock); kfree(iter); - mutex_unlock(&ftrace_filter_lock); + mutex_unlock(&ftrace_regex_lock); return 0; } +static int +ftrace_filter_release(struct inode *inode, struct file *file) +{ + return ftrace_regex_release(inode, file, 1); +} + +static int +ftrace_notrace_release(struct inode *inode, struct file *file) +{ + return ftrace_regex_release(inode, file, 0); +} + +static ssize_t +ftraced_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + /* don't worry about races */ + char *buf = ftraced_stop ? "disabled\n" : "enabled\n"; + int r = strlen(buf); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +static ssize_t +ftraced_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[64]; + long val; + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + if (strncmp(buf, "enable", 6) == 0) + val = 1; + else if (strncmp(buf, "disable", 7) == 0) + val = 0; + else { + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + val = !!val; + } + + if (val) + ftrace_enable_daemon(); + else + ftrace_disable_daemon(); + + filp->f_pos += cnt; + + return cnt; +} + static struct file_operations ftrace_avail_fops = { .open = ftrace_avail_open, .read = seq_read, @@ -1148,59 +1334,57 @@ static struct file_operations ftrace_avail_fops = { .release = ftrace_avail_release, }; +static struct file_operations ftrace_failures_fops = { + .open = ftrace_failures_open, + .read = seq_read, + .llseek = seq_lseek, + .release = ftrace_avail_release, +}; + static struct file_operations ftrace_filter_fops = { .open = ftrace_filter_open, - .read = ftrace_filter_read, + .read = ftrace_regex_read, .write = ftrace_filter_write, - .llseek = ftrace_filter_lseek, + .llseek = ftrace_regex_lseek, .release = ftrace_filter_release, }; +static struct file_operations ftrace_notrace_fops = { + .open = ftrace_notrace_open, + .read = ftrace_regex_read, + .write = ftrace_notrace_write, + .llseek = ftrace_regex_lseek, + .release = ftrace_notrace_release, +}; + +static struct file_operations ftraced_fops = { + .open = tracing_open_generic, + .read = ftraced_read, + .write = ftraced_write, +}; + /** * ftrace_force_update - force an update to all recording ftrace functions - * - * The ftrace dynamic update daemon only wakes up once a second. - * There may be cases where an update needs to be done immediately - * for tests or internal kernel tracing to begin. This function - * wakes the daemon to do an update and will not return until the - * update is complete. */ int ftrace_force_update(void) { - unsigned long last_counter; - DECLARE_WAITQUEUE(wait, current); int ret = 0; if (unlikely(ftrace_disabled)) return -ENODEV; + mutex_lock(&ftrace_sysctl_lock); mutex_lock(&ftraced_lock); - last_counter = ftraced_iteration_counter; - - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&ftraced_waiters, &wait); - - if (unlikely(!ftraced_task)) { - ret = -ENODEV; - goto out; - } - do { - mutex_unlock(&ftraced_lock); - wake_up_process(ftraced_task); - schedule(); - mutex_lock(&ftraced_lock); - if (signal_pending(current)) { - ret = -EINTR; - break; - } - set_current_state(TASK_INTERRUPTIBLE); - } while (last_counter == ftraced_iteration_counter); + /* + * If ftraced_trigger is not set, then there is nothing + * to update. + */ + if (ftraced_trigger && !ftrace_update_code()) + ret = -EBUSY; - out: mutex_unlock(&ftraced_lock); - remove_wait_queue(&ftraced_waiters, &wait); - set_current_state(TASK_RUNNING); + mutex_unlock(&ftrace_sysctl_lock); return ret; } @@ -1234,11 +1418,28 @@ static __init int ftrace_init_debugfs(void) pr_warning("Could not create debugfs " "'available_filter_functions' entry\n"); + entry = debugfs_create_file("failures", 0444, + d_tracer, NULL, &ftrace_failures_fops); + if (!entry) + pr_warning("Could not create debugfs 'failures' entry\n"); + entry = debugfs_create_file("set_ftrace_filter", 0644, d_tracer, NULL, &ftrace_filter_fops); if (!entry) pr_warning("Could not create debugfs " "'set_ftrace_filter' entry\n"); + + entry = debugfs_create_file("set_ftrace_notrace", 0644, d_tracer, + NULL, &ftrace_notrace_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'set_ftrace_notrace' entry\n"); + + entry = debugfs_create_file("ftraced_enabled", 0644, d_tracer, + NULL, &ftraced_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'ftraced_enabled' entry\n"); return 0; } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 95b7c48a9a1d..2c6ffd6aad47 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -27,6 +27,7 @@ #include <linux/poll.h> #include <linux/gfp.h> #include <linux/fs.h> +#include <linux/kprobes.h> #include <linux/writeback.h> #include <linux/stacktrace.h> @@ -42,11 +43,6 @@ static cpumask_t __read_mostly tracing_buffer_mask; #define for_each_tracing_cpu(cpu) \ for_each_cpu_mask(cpu, tracing_buffer_mask) -/* dummy trace to disable tracing */ -static struct tracer no_tracer __read_mostly = { - .name = "none", -}; - static int trace_alloc_page(void); static int trace_free_page(void); @@ -134,6 +130,23 @@ static DECLARE_WAIT_QUEUE_HEAD(trace_wait); /* trace_flags holds iter_ctrl options */ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT; +static notrace void no_trace_init(struct trace_array *tr) +{ + int cpu; + + if(tr->ctrl) + for_each_online_cpu(cpu) + tracing_reset(tr->data[cpu]); + tracer_enabled = 0; +} + +/* dummy trace to disable tracing */ +static struct tracer no_tracer __read_mostly = { + .name = "none", + .init = no_trace_init +}; + + /** * trace_wake_up - wake up tasks waiting for trace input * @@ -249,24 +262,32 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) tracing_record_cmdline(current); } +#define CHECK_COND(cond) \ + if (unlikely(cond)) { \ + tracing_disabled = 1; \ + WARN_ON(1); \ + return -1; \ + } + /** * check_pages - integrity check of trace buffers * * As a safty measure we check to make sure the data pages have not - * been corrupted. TODO: configure to disable this because it adds - * a bit of overhead. + * been corrupted. */ -void check_pages(struct trace_array_cpu *data) +int check_pages(struct trace_array_cpu *data) { struct page *page, *tmp; - BUG_ON(data->trace_pages.next->prev != &data->trace_pages); - BUG_ON(data->trace_pages.prev->next != &data->trace_pages); + CHECK_COND(data->trace_pages.next->prev != &data->trace_pages); + CHECK_COND(data->trace_pages.prev->next != &data->trace_pages); list_for_each_entry_safe(page, tmp, &data->trace_pages, lru) { - BUG_ON(page->lru.next->prev != &page->lru); - BUG_ON(page->lru.prev->next != &page->lru); + CHECK_COND(page->lru.next->prev != &page->lru); + CHECK_COND(page->lru.prev->next != &page->lru); } + + return 0; } /** @@ -280,7 +301,6 @@ void *head_page(struct trace_array_cpu *data) { struct page *page; - check_pages(data); if (list_empty(&data->trace_pages)) return NULL; @@ -645,9 +665,6 @@ static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN]; static int cmdline_idx; static DEFINE_SPINLOCK(trace_cmdline_lock); -/* trace in all context switches */ -atomic_t trace_record_cmdline_enabled __read_mostly; - /* temporary disable recording */ atomic_t trace_record_cmdline_disabled __read_mostly; @@ -831,6 +848,48 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data, trace_function(tr, data, ip, parent_ip, flags); } +#ifdef CONFIG_MMIOTRACE +void __trace_mmiotrace_rw(struct trace_array *tr, struct trace_array_cpu *data, + struct mmiotrace_rw *rw) +{ + struct trace_entry *entry; + unsigned long irq_flags; + + raw_local_irq_save(irq_flags); + __raw_spin_lock(&data->lock); + + entry = tracing_get_trace_entry(tr, data); + tracing_generic_entry_update(entry, 0); + entry->type = TRACE_MMIO_RW; + entry->mmiorw = *rw; + + __raw_spin_unlock(&data->lock); + raw_local_irq_restore(irq_flags); + + trace_wake_up(); +} + +void __trace_mmiotrace_map(struct trace_array *tr, struct trace_array_cpu *data, + struct mmiotrace_map *map) +{ + struct trace_entry *entry; + unsigned long irq_flags; + + raw_local_irq_save(irq_flags); + __raw_spin_lock(&data->lock); + + entry = tracing_get_trace_entry(tr, data); + tracing_generic_entry_update(entry, 0); + entry->type = TRACE_MMIO_MAP; + entry->mmiomap = *map; + + __raw_spin_unlock(&data->lock); + raw_local_irq_restore(irq_flags); + + trace_wake_up(); +} +#endif + void __trace_stack(struct trace_array *tr, struct trace_array_cpu *data, unsigned long flags, @@ -934,6 +993,30 @@ tracing_sched_wakeup_trace(struct trace_array *tr, trace_wake_up(); } +void +ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) +{ + struct trace_array *tr = &global_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + + if (tracing_disabled || current_trace == &no_tracer || !tr->ctrl) + return; + + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + + if (likely(disabled == 1)) + __trace_special(tr, data, arg1, arg2, arg3); + + atomic_dec(&data->disabled); + local_irq_restore(flags); +} + #ifdef CONFIG_FTRACE static void function_trace_call(unsigned long ip, unsigned long parent_ip) @@ -1171,6 +1254,20 @@ static void s_stop(struct seq_file *m, void *p) mutex_unlock(&trace_types_lock); } +#define KRETPROBE_MSG "[unknown/kretprobe'd]" + +#ifdef CONFIG_KRETPROBES +static inline int kretprobed(unsigned long addr) +{ + return addr == (unsigned long)kretprobe_trampoline; +} +#else +static inline int kretprobed(unsigned long addr) +{ + return 0; +} +#endif /* CONFIG_KRETPROBES */ + static int seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address) { @@ -1406,7 +1503,10 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) case TRACE_FN: seq_print_ip_sym(s, entry->fn.ip, sym_flags); trace_seq_puts(s, " ("); - seq_print_ip_sym(s, entry->fn.parent_ip, sym_flags); + if (kretprobed(entry->fn.parent_ip)) + trace_seq_puts(s, KRETPROBE_MSG); + else + seq_print_ip_sym(s, entry->fn.parent_ip, sym_flags); trace_seq_puts(s, ")\n"); break; case TRACE_CTX: @@ -1486,8 +1586,11 @@ static int print_trace_fmt(struct trace_iterator *iter) ret = trace_seq_printf(s, " <-"); if (!ret) return 0; - ret = seq_print_ip_sym(s, entry->fn.parent_ip, - sym_flags); + if (kretprobed(entry->fn.parent_ip)) + ret = trace_seq_puts(s, KRETPROBE_MSG); + else + ret = seq_print_ip_sym(s, entry->fn.parent_ip, + sym_flags); if (!ret) return 0; } @@ -2566,7 +2669,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, { unsigned long val; char buf[64]; - int ret; + int i, ret; if (cnt >= sizeof(buf)) return -EINVAL; @@ -2635,8 +2738,15 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, trace_free_page(); } + /* check integrity */ + for_each_tracing_cpu(i) + check_pages(global_trace.data[i]); + filp->f_pos += cnt; + /* If check pages failed, return ENOMEM */ + if (tracing_disabled) + cnt = -ENOMEM; out: max_tr.entries = global_trace.entries; mutex_unlock(&trace_types_lock); @@ -2930,8 +3040,6 @@ __init static int tracer_alloc_buffers(void) int ret = -ENOMEM; int i; - global_trace.ctrl = tracer_enabled; - /* TODO: make the number of buffers hot pluggable with CPUS */ tracing_nr_buffers = num_possible_cpus(); tracing_buffer_mask = cpu_possible_map; @@ -3001,6 +3109,7 @@ __init static int tracer_alloc_buffers(void) current_trace = &no_tracer; /* All seems OK, enable tracing */ + global_trace.ctrl = tracer_enabled; tracing_disabled = 0; return 0; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index b7f85d9c80d7..8cb215b239d5 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -5,6 +5,7 @@ #include <asm/atomic.h> #include <linux/sched.h> #include <linux/clocksource.h> +#include <linux/mmiotrace.h> enum trace_type { __TRACE_FIRST_TYPE = 0, @@ -14,6 +15,8 @@ enum trace_type { TRACE_WAKE, TRACE_STACK, TRACE_SPECIAL, + TRACE_MMIO_RW, + TRACE_MMIO_MAP, __TRACE_LAST_TYPE }; @@ -75,6 +78,8 @@ struct trace_entry { struct ctx_switch_entry ctx; struct special_entry special; struct stack_entry stack; + struct mmiotrace_rw mmiorw; + struct mmiotrace_map mmiomap; }; }; @@ -220,6 +225,8 @@ void trace_function(struct trace_array *tr, void tracing_start_function_trace(void); void tracing_stop_function_trace(void); +void tracing_start_cmdline_record(void); +void tracing_stop_cmdline_record(void); int register_tracer(struct tracer *type); void unregister_tracer(struct tracer *type); @@ -228,8 +235,6 @@ extern unsigned long nsecs_to_usecs(unsigned long nsecs); extern unsigned long tracing_max_latency; extern unsigned long tracing_thresh; -extern atomic_t trace_record_cmdline_enabled; - 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); @@ -257,6 +262,15 @@ extern unsigned long ftrace_update_tot_cnt; extern int DYN_FTRACE_TEST_NAME(void); #endif +#ifdef CONFIG_MMIOTRACE +extern void __trace_mmiotrace_rw(struct trace_array *tr, + struct trace_array_cpu *data, + struct mmiotrace_rw *rw); +extern void __trace_mmiotrace_map(struct trace_array *tr, + struct trace_array_cpu *data, + struct mmiotrace_map *map); +#endif + #ifdef CONFIG_FTRACE_STARTUP_TEST #ifdef CONFIG_FTRACE extern int trace_selftest_startup_function(struct tracer *trace, diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 0a084656d7cf..7ee7dcd76b7d 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -29,14 +29,14 @@ static void function_reset(struct trace_array *tr) static void start_function_trace(struct trace_array *tr) { function_reset(tr); - atomic_inc(&trace_record_cmdline_enabled); + tracing_start_cmdline_record(); tracing_start_function_trace(); } static void stop_function_trace(struct trace_array *tr) { tracing_stop_function_trace(); - atomic_dec(&trace_record_cmdline_enabled); + tracing_stop_cmdline_record(); } static void function_trace_init(struct trace_array *tr) diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 761f3ec66c50..421d6fe3650e 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -165,22 +165,6 @@ check_critical_timing(struct trace_array *tr, update_max_tr_single(tr, current, cpu); - if (!runqueue_is_locked()) { - if (tracing_thresh) { - printk(KERN_INFO "(%16s-%-5d|#%d): %lu us critical" - " section violates %lu us threshold.\n", - current->comm, current->pid, - raw_smp_processor_id(), - latency, nsecs_to_usecs(tracing_thresh)); - } else { - printk(KERN_INFO "(%16s-%-5d|#%d): new %lu us" - " maximum-latency critical section.\n", - current->comm, current->pid, - raw_smp_processor_id(), - latency); - } - } - max_sequence++; out_unlock: diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c new file mode 100644 index 000000000000..b13dc19dcbb4 --- /dev/null +++ b/kernel/trace/trace_mmiotrace.c @@ -0,0 +1,295 @@ +/* + * Memory mapped I/O tracing + * + * Copyright (C) 2008 Pekka Paalanen <pq@iki.fi> + */ + +#define DEBUG 1 + +#include <linux/kernel.h> +#include <linux/mmiotrace.h> +#include <linux/pci.h> + +#include "trace.h" + +struct header_iter { + struct pci_dev *dev; +}; + +static struct trace_array *mmio_trace_array; +static bool overrun_detected; + +static void mmio_reset_data(struct trace_array *tr) +{ + int cpu; + + overrun_detected = false; + tr->time_start = ftrace_now(tr->cpu); + + for_each_online_cpu(cpu) + tracing_reset(tr->data[cpu]); +} + +static void mmio_trace_init(struct trace_array *tr) +{ + pr_debug("in %s\n", __func__); + mmio_trace_array = tr; + if (tr->ctrl) { + mmio_reset_data(tr); + enable_mmiotrace(); + } +} + +static void mmio_trace_reset(struct trace_array *tr) +{ + pr_debug("in %s\n", __func__); + if (tr->ctrl) + disable_mmiotrace(); + mmio_reset_data(tr); + mmio_trace_array = NULL; +} + +static void mmio_trace_ctrl_update(struct trace_array *tr) +{ + pr_debug("in %s\n", __func__); + if (tr->ctrl) { + mmio_reset_data(tr); + enable_mmiotrace(); + } else { + disable_mmiotrace(); + } +} + +static int mmio_print_pcidev(struct trace_seq *s, const struct pci_dev *dev) +{ + int ret = 0; + int i; + resource_size_t start, end; + const struct pci_driver *drv = pci_dev_driver(dev); + + /* XXX: incomplete checks for trace_seq_printf() return value */ + ret += trace_seq_printf(s, "PCIDEV %02x%02x %04x%04x %x", + dev->bus->number, dev->devfn, + dev->vendor, dev->device, dev->irq); + /* + * XXX: is pci_resource_to_user() appropriate, since we are + * supposed to interpret the __ioremap() phys_addr argument based on + * these printed values? + */ + for (i = 0; i < 7; i++) { + pci_resource_to_user(dev, i, &dev->resource[i], &start, &end); + ret += trace_seq_printf(s, " %llx", + (unsigned long long)(start | + (dev->resource[i].flags & PCI_REGION_FLAG_MASK))); + } + for (i = 0; i < 7; i++) { + pci_resource_to_user(dev, i, &dev->resource[i], &start, &end); + ret += trace_seq_printf(s, " %llx", + dev->resource[i].start < dev->resource[i].end ? + (unsigned long long)(end - start) + 1 : 0); + } + if (drv) + ret += trace_seq_printf(s, " %s\n", drv->name); + else + ret += trace_seq_printf(s, " \n"); + return ret; +} + +static void destroy_header_iter(struct header_iter *hiter) +{ + if (!hiter) + return; + pci_dev_put(hiter->dev); + kfree(hiter); +} + +static void mmio_pipe_open(struct trace_iterator *iter) +{ + struct header_iter *hiter; + struct trace_seq *s = &iter->seq; + + trace_seq_printf(s, "VERSION 20070824\n"); + + hiter = kzalloc(sizeof(*hiter), GFP_KERNEL); + if (!hiter) + return; + + hiter->dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, NULL); + iter->private = hiter; +} + +/* XXX: This is not called when the pipe is closed! */ +static void mmio_close(struct trace_iterator *iter) +{ + struct header_iter *hiter = iter->private; + destroy_header_iter(hiter); + iter->private = NULL; +} + +static unsigned long count_overruns(struct trace_iterator *iter) +{ + int cpu; + unsigned long cnt = 0; + for_each_online_cpu(cpu) { + cnt += iter->overrun[cpu]; + iter->overrun[cpu] = 0; + } + return cnt; +} + +static ssize_t mmio_read(struct trace_iterator *iter, struct file *filp, + char __user *ubuf, size_t cnt, loff_t *ppos) +{ + ssize_t ret; + struct header_iter *hiter = iter->private; + struct trace_seq *s = &iter->seq; + unsigned long n; + + n = count_overruns(iter); + if (n) { + /* XXX: This is later than where events were lost. */ + trace_seq_printf(s, "MARK 0.000000 Lost %lu events.\n", n); + if (!overrun_detected) + pr_warning("mmiotrace has lost events.\n"); + overrun_detected = true; + goto print_out; + } + + if (!hiter) + return 0; + + mmio_print_pcidev(s, hiter->dev); + hiter->dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, hiter->dev); + + if (!hiter->dev) { + destroy_header_iter(hiter); + iter->private = NULL; + } + +print_out: + ret = trace_seq_to_user(s, ubuf, cnt); + return (ret == -EBUSY) ? 0 : ret; +} + +static int mmio_print_rw(struct trace_iterator *iter) +{ + struct trace_entry *entry = iter->ent; + struct mmiotrace_rw *rw = &entry->mmiorw; + struct trace_seq *s = &iter->seq; + unsigned long long t = ns2usecs(entry->t); + unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned secs = (unsigned long)t; + int ret = 1; + + switch (entry->mmiorw.opcode) { + case MMIO_READ: + ret = trace_seq_printf(s, + "R %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", + rw->width, secs, usec_rem, rw->map_id, + (unsigned long long)rw->phys, + rw->value, rw->pc, 0); + break; + case MMIO_WRITE: + ret = trace_seq_printf(s, + "W %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", + rw->width, secs, usec_rem, rw->map_id, + (unsigned long long)rw->phys, + rw->value, rw->pc, 0); + break; + case MMIO_UNKNOWN_OP: + ret = trace_seq_printf(s, + "UNKNOWN %lu.%06lu %d 0x%llx %02x,%02x,%02x 0x%lx %d\n", + secs, usec_rem, rw->map_id, + (unsigned long long)rw->phys, + (rw->value >> 16) & 0xff, (rw->value >> 8) & 0xff, + (rw->value >> 0) & 0xff, rw->pc, 0); + break; + default: + ret = trace_seq_printf(s, "rw what?\n"); + break; + } + if (ret) + return 1; + return 0; +} + +static int mmio_print_map(struct trace_iterator *iter) +{ + struct trace_entry *entry = iter->ent; + struct mmiotrace_map *m = &entry->mmiomap; + struct trace_seq *s = &iter->seq; + unsigned long long t = ns2usecs(entry->t); + unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned secs = (unsigned long)t; + int ret = 1; + + switch (entry->mmiorw.opcode) { + case MMIO_PROBE: + ret = trace_seq_printf(s, + "MAP %lu.%06lu %d 0x%llx 0x%lx 0x%lx 0x%lx %d\n", + secs, usec_rem, m->map_id, + (unsigned long long)m->phys, m->virt, m->len, + 0UL, 0); + break; + case MMIO_UNPROBE: + ret = trace_seq_printf(s, + "UNMAP %lu.%06lu %d 0x%lx %d\n", + secs, usec_rem, m->map_id, 0UL, 0); + break; + default: + ret = trace_seq_printf(s, "map what?\n"); + break; + } + if (ret) + return 1; + return 0; +} + +/* return 0 to abort printing without consuming current entry in pipe mode */ +static int mmio_print_line(struct trace_iterator *iter) +{ + switch (iter->ent->type) { + case TRACE_MMIO_RW: + return mmio_print_rw(iter); + case TRACE_MMIO_MAP: + return mmio_print_map(iter); + default: + return 1; /* ignore unknown entries */ + } +} + +static struct tracer mmio_tracer __read_mostly = +{ + .name = "mmiotrace", + .init = mmio_trace_init, + .reset = mmio_trace_reset, + .pipe_open = mmio_pipe_open, + .close = mmio_close, + .read = mmio_read, + .ctrl_update = mmio_trace_ctrl_update, + .print_line = mmio_print_line, +}; + +__init static int init_mmio_trace(void) +{ + return register_tracer(&mmio_tracer); +} +device_initcall(init_mmio_trace); + +void mmio_trace_rw(struct mmiotrace_rw *rw) +{ + struct trace_array *tr = mmio_trace_array; + struct trace_array_cpu *data = tr->data[smp_processor_id()]; + __trace_mmiotrace_rw(tr, data, rw); +} + +void mmio_trace_mapping(struct mmiotrace_map *map) +{ + struct trace_array *tr = mmio_trace_array; + struct trace_array_cpu *data; + + preempt_disable(); + data = tr->data[smp_processor_id()]; + __trace_mmiotrace_map(tr, data, map); + preempt_enable(); +} diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index d25ffa5eaf2b..c16935d3bc5c 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -29,6 +29,9 @@ sched_switch_func(void *private, void *__rq, struct task_struct *prev, long disabled; int cpu; + tracing_record_cmdline(prev); + tracing_record_cmdline(next); + if (!tracer_enabled) return; @@ -63,8 +66,6 @@ sched_switch_callback(void *probe_data, void *call_data, prev = va_arg(*args, typeof(prev)); next = va_arg(*args, typeof(next)); - tracing_record_cmdline(prev); - /* * If tracer_switch_func only points to the local * switch func, it still needs the ptr passed to it. @@ -125,30 +126,6 @@ wake_up_callback(void *probe_data, void *call_data, wakeup_func(probe_data, __rq, task, curr); } -void -ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) -{ - struct trace_array *tr = ctx_trace; - struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int cpu; - - if (!tracer_enabled) - return; - - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - - if (likely(disabled == 1)) - __trace_special(tr, data, arg1, arg2, arg3); - - atomic_dec(&data->disabled); - local_irq_restore(flags); -} - static void sched_switch_reset(struct trace_array *tr) { int cpu; @@ -237,18 +214,26 @@ void tracing_stop_sched_switch(void) tracing_sched_unregister(); } +void tracing_start_cmdline_record(void) +{ + tracing_start_sched_switch(); +} + +void tracing_stop_cmdline_record(void) +{ + tracing_stop_sched_switch(); +} + static void start_sched_trace(struct trace_array *tr) { sched_switch_reset(tr); - atomic_inc(&trace_record_cmdline_enabled); tracer_enabled = 1; - tracing_start_sched_switch(); + tracing_start_cmdline_record(); } static void stop_sched_trace(struct trace_array *tr) { - tracing_stop_sched_switch(); - atomic_dec(&trace_record_cmdline_enabled); + tracing_stop_cmdline_record(); tracer_enabled = 0; } diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 5d2fb48e47f8..bf7e91caef57 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -30,6 +30,69 @@ static DEFINE_SPINLOCK(wakeup_lock); static void __wakeup_reset(struct trace_array *tr); +#ifdef CONFIG_FTRACE +/* + * irqsoff uses its own tracer function to keep the overhead down: + */ +static void +wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) +{ + struct trace_array *tr = wakeup_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int resched; + int cpu; + + if (likely(!wakeup_task)) + return; + + resched = need_resched(); + preempt_disable_notrace(); + + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (unlikely(disabled != 1)) + goto out; + + spin_lock_irqsave(&wakeup_lock, flags); + + if (unlikely(!wakeup_task)) + goto unlock; + + /* + * The task can't disappear because it needs to + * wake up first, and we have the wakeup_lock. + */ + if (task_cpu(wakeup_task) != cpu) + goto unlock; + + trace_function(tr, data, ip, parent_ip, flags); + + unlock: + spin_unlock_irqrestore(&wakeup_lock, flags); + + out: + atomic_dec(&data->disabled); + + /* + * To prevent recursion from the scheduler, if the + * resched flag was set before we entered, then + * don't reschedule. + */ + if (resched) + preempt_enable_no_resched_notrace(); + else + preempt_enable_notrace(); +} + +static struct ftrace_ops trace_ops __read_mostly = +{ + .func = wakeup_tracer_call, +}; +#endif /* CONFIG_FTRACE */ + /* * Should this new latency be reported/recorded? */ @@ -73,7 +136,7 @@ wakeup_sched_switch(void *private, void *rq, struct task_struct *prev, if (next != wakeup_task) return; - /* The task we are waitng for is waking up */ + /* The task we are waiting for is waking up */ data = tr->data[wakeup_cpu]; /* disable local data, not wakeup_cpu data */ @@ -290,6 +353,7 @@ static void start_wakeup_tracer(struct trace_array *tr) smp_wmb(); tracer_enabled = 1; + register_ftrace_function(&trace_ops); return; fail_deprobe_wake_new: @@ -305,6 +369,7 @@ fail_deprobe: static void stop_wakeup_tracer(struct trace_array *tr) { tracer_enabled = 0; + unregister_ftrace_function(&trace_ops); marker_probe_unregister("kernel_sched_schedule", sched_switch_callback, &wakeup_trace); diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 5588ecc40985..0911b7e073bf 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -28,6 +28,7 @@ trace_test_buffer_cpu(struct trace_array *tr, struct trace_array_cpu *data) page = list_entry(data->trace_pages.next, struct page, lru); entries = page_address(page); + check_pages(data); if (head_page(data) != entries) goto failed; |