From e162b39a368f0401e41b558f430c354d12a85b37 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Thu, 15 Jan 2009 11:08:40 -0800 Subject: softlockup: decouple hung tasks check from softlockup detection Decoupling allows: * hung tasks check to happen at very low priority * hung tasks check and softlockup to be enabled/disabled independently at compile and/or run-time * individual panic settings to be enabled disabled independently at compile and/or run-time * softlockup threshold to be reduced without increasing hung tasks poll frequency (hung task check is expensive relative to softlock watchdog) * hung task check to be zero over-head when disabled at run-time Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/Makefile | 1 + kernel/hung_task.c | 198 ++++++++++++++++++++++++++++++++++++++++++++++++++++ kernel/softlockup.c | 100 -------------------------- kernel/sysctl.c | 15 +++- 4 files changed, 213 insertions(+), 101 deletions(-) create mode 100644 kernel/hung_task.c (limited to 'kernel') diff --git a/kernel/Makefile b/kernel/Makefile index 2aebc4cd7878..979745f1b4bc 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -74,6 +74,7 @@ obj-$(CONFIG_AUDIT_TREE) += audit_tree.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o +obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o diff --git a/kernel/hung_task.c b/kernel/hung_task.c new file mode 100644 index 000000000000..ba5a77cad3bb --- /dev/null +++ b/kernel/hung_task.c @@ -0,0 +1,198 @@ +/* + * Detect Hung Task + * + * kernel/hung_task.c - kernel thread for detecting tasks stuck in D state + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Have a reasonable limit on the number of tasks checked: + */ +unsigned long __read_mostly sysctl_hung_task_check_count = 1024; + +/* + * Zero means infinite timeout - no checking done: + */ +unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120; +static unsigned long __read_mostly hung_task_poll_jiffies; + +unsigned long __read_mostly sysctl_hung_task_warnings = 10; + +static int __read_mostly did_panic; + +static struct task_struct *watchdog_task; + +/* + * Should we panic (and reboot, if panic_timeout= is set) when a + * hung task is detected: + */ +unsigned int __read_mostly sysctl_hung_task_panic = + CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE; + +static int __init hung_task_panic_setup(char *str) +{ + sysctl_hung_task_panic = simple_strtoul(str, NULL, 0); + + return 1; +} +__setup("hung_task_panic=", hung_task_panic_setup); + +static int +hung_task_panic(struct notifier_block *this, unsigned long event, void *ptr) +{ + did_panic = 1; + + return NOTIFY_DONE; +} + +static struct notifier_block panic_block = { + .notifier_call = hung_task_panic, +}; + +/* + * Returns seconds, approximately. We don't need nanosecond + * resolution, and we don't need to waste time with a big divide when + * 2^30ns == 1.074s. + */ +static unsigned long get_timestamp(void) +{ + int this_cpu = raw_smp_processor_id(); + + return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ +} + +static void check_hung_task(struct task_struct *t, unsigned long now) +{ + unsigned long switch_count = t->nvcsw + t->nivcsw; + + if (t->flags & PF_FROZEN) + return; + + if (switch_count != t->last_switch_count || !t->last_switch_timestamp) { + t->last_switch_count = switch_count; + t->last_switch_timestamp = now; + return; + } + if ((long)(now - t->last_switch_timestamp) < + sysctl_hung_task_timeout_secs) + return; + if (!sysctl_hung_task_warnings) + return; + sysctl_hung_task_warnings--; + + /* + * Ok, the task did not get scheduled for more than 2 minutes, + * complain: + */ + printk(KERN_ERR "INFO: task %s:%d blocked for more than " + "%ld seconds.\n", t->comm, t->pid, + sysctl_hung_task_timeout_secs); + printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" + " disables this message.\n"); + sched_show_task(t); + __debug_show_held_locks(t); + + t->last_switch_timestamp = now; + touch_nmi_watchdog(); + + if (sysctl_hung_task_panic) + panic("hung_task: blocked tasks"); +} + +/* + * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for + * a really long time (120 seconds). If that happens, print out + * a warning. + */ +static void check_hung_uninterruptible_tasks(void) +{ + int max_count = sysctl_hung_task_check_count; + unsigned long now = get_timestamp(); + struct task_struct *g, *t; + + /* + * If the system crashed already then all bets are off, + * do not report extra hung tasks: + */ + if (test_taint(TAINT_DIE) || did_panic) + return; + + read_lock(&tasklist_lock); + do_each_thread(g, t) { + if (!--max_count) + goto unlock; + /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ + if (t->state == TASK_UNINTERRUPTIBLE) + check_hung_task(t, now); + } while_each_thread(g, t); + unlock: + read_unlock(&tasklist_lock); +} + +static void update_poll_jiffies(void) +{ + /* timeout of 0 will disable the watchdog */ + if (sysctl_hung_task_timeout_secs == 0) + hung_task_poll_jiffies = MAX_SCHEDULE_TIMEOUT; + else + hung_task_poll_jiffies = sysctl_hung_task_timeout_secs * HZ / 2; +} + +/* + * Process updating of timeout sysctl + */ +int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, + struct file *filp, void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + int ret; + + ret = proc_doulongvec_minmax(table, write, filp, buffer, lenp, ppos); + + if (ret || !write) + goto out; + + update_poll_jiffies(); + + wake_up_process(watchdog_task); + + out: + return ret; +} + +/* + * kthread which checks for tasks stuck in D state + */ +static int watchdog(void *dummy) +{ + set_user_nice(current, 0); + update_poll_jiffies(); + + for ( ; ; ) { + while (schedule_timeout_interruptible(hung_task_poll_jiffies)); + check_hung_uninterruptible_tasks(); + } + + return 0; +} + +static int __init hung_task_init(void) +{ + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); + watchdog_task = kthread_run(watchdog, NULL, "khungtaskd"); + + return 0; +} + +module_init(hung_task_init); diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 85d5a2455103..88796c330838 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -165,98 +165,12 @@ void softlockup_tick(void) panic("softlockup: hung tasks"); } -/* - * Have a reasonable limit on the number of tasks checked: - */ -unsigned long __read_mostly sysctl_hung_task_check_count = 1024; - -/* - * Zero means infinite timeout - no checking done: - */ -unsigned long __read_mostly sysctl_hung_task_timeout_secs = 480; - -unsigned long __read_mostly sysctl_hung_task_warnings = 10; - -/* - * Only do the hung-tasks check on one CPU: - */ -static int check_cpu __read_mostly = -1; - -static void check_hung_task(struct task_struct *t, unsigned long now) -{ - unsigned long switch_count = t->nvcsw + t->nivcsw; - - if (t->flags & PF_FROZEN) - return; - - if (switch_count != t->last_switch_count || !t->last_switch_timestamp) { - t->last_switch_count = switch_count; - t->last_switch_timestamp = now; - return; - } - if ((long)(now - t->last_switch_timestamp) < - sysctl_hung_task_timeout_secs) - return; - if (!sysctl_hung_task_warnings) - return; - sysctl_hung_task_warnings--; - - /* - * Ok, the task did not get scheduled for more than 2 minutes, - * complain: - */ - printk(KERN_ERR "INFO: task %s:%d blocked for more than " - "%ld seconds.\n", t->comm, t->pid, - sysctl_hung_task_timeout_secs); - printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" - " disables this message.\n"); - sched_show_task(t); - __debug_show_held_locks(t); - - t->last_switch_timestamp = now; - touch_nmi_watchdog(); - - if (softlockup_panic) - panic("softlockup: blocked tasks"); -} - -/* - * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for - * a really long time (120 seconds). If that happens, print out - * a warning. - */ -static void check_hung_uninterruptible_tasks(int this_cpu) -{ - int max_count = sysctl_hung_task_check_count; - unsigned long now = get_timestamp(this_cpu); - struct task_struct *g, *t; - - /* - * If the system crashed already then all bets are off, - * do not report extra hung tasks: - */ - if (test_taint(TAINT_DIE) || did_panic) - return; - - read_lock(&tasklist_lock); - do_each_thread(g, t) { - if (!--max_count) - goto unlock; - /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ - if (t->state == TASK_UNINTERRUPTIBLE) - check_hung_task(t, now); - } while_each_thread(g, t); - unlock: - read_unlock(&tasklist_lock); -} - /* * The watchdog thread - runs every second and touches the timestamp. */ static int watchdog(void *__bind_cpu) { struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; - int this_cpu = (long)__bind_cpu; sched_setscheduler(current, SCHED_FIFO, ¶m); @@ -276,11 +190,6 @@ static int watchdog(void *__bind_cpu) if (kthread_should_stop()) break; - if (this_cpu == check_cpu) { - if (sysctl_hung_task_timeout_secs) - check_hung_uninterruptible_tasks(this_cpu); - } - set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); @@ -312,18 +221,9 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) break; case CPU_ONLINE: case CPU_ONLINE_FROZEN: - check_cpu = cpumask_any(cpu_online_mask); wake_up_process(per_cpu(watchdog_task, hotcpu)); break; #ifdef CONFIG_HOTPLUG_CPU - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - if (hotcpu == check_cpu) { - /* Pick any other online cpu. */ - check_cpu = cpumask_any_but(cpu_online_mask, hotcpu); - } - break; - case CPU_UP_CANCELED: case CPU_UP_CANCELED_FROZEN: if (!per_cpu(watchdog_task, hotcpu)) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 596dc31a7116..2481ed30d2b5 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -805,6 +805,19 @@ static struct ctl_table kern_table[] = { .extra1 = &neg_one, .extra2 = &sixty, }, +#endif +#ifdef CONFIG_DETECT_HUNG_TASK + { + .ctl_name = CTL_UNNUMBERED, + .procname = "hung_task_panic", + .data = &sysctl_hung_task_panic, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &zero, + .extra2 = &one, + }, { .ctl_name = CTL_UNNUMBERED, .procname = "hung_task_check_count", @@ -820,7 +833,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_hung_task_timeout_secs, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = &proc_dohung_task_timeout_secs, .strategy = &sysctl_intvec, }, { -- cgit v1.2.3 From af432eb1cc3178ec7109aca2283aafb1c12ccac1 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Fri, 16 Jan 2009 09:09:38 -0800 Subject: softlockup: fix to allow compiling with !DETECT_HUNG_TASK Fixes the following compile error: kernel/fork.c:1049: error: 'struct task_struct' has no member named 'last_switch_count' kernel/fork.c:1050: error: 'struct task_struct' has no member named 'last_switch_timestamp' Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/fork.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 1d68f1255dd8..fb9444282836 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1041,7 +1041,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->default_timer_slack_ns = current->timer_slack_ns; -#ifdef CONFIG_DETECT_SOFTLOCKUP +#ifdef CONFIG_DETECT_HUNG_TASK p->last_switch_count = 0; p->last_switch_timestamp = 0; #endif -- cgit v1.2.3 From 603a148f434742fff08273207ffa44176cad13a1 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Sat, 17 Jan 2009 10:31:48 -0800 Subject: softlockup: fix potential race in hung_task when resetting timeout Impact: fix potential false panic A potential race exists if sysctl_hung_task_timeout_secs is reset to 0 while inside check_hung_uniterruptible_tasks(). If check_task() is entered, a comparison with 0 will result in a false hung_task being detected. If sysctl_hung_task_panic is set, the system will panic. Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/hung_task.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/hung_task.c b/kernel/hung_task.c index ba5a77cad3bb..ba8ccd432963 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -72,7 +72,8 @@ static unsigned long get_timestamp(void) return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ } -static void check_hung_task(struct task_struct *t, unsigned long now) +static void check_hung_task(struct task_struct *t, unsigned long now, + unsigned long timeout) { unsigned long switch_count = t->nvcsw + t->nivcsw; @@ -84,8 +85,7 @@ static void check_hung_task(struct task_struct *t, unsigned long now) t->last_switch_timestamp = now; return; } - if ((long)(now - t->last_switch_timestamp) < - sysctl_hung_task_timeout_secs) + if ((long)(now - t->last_switch_timestamp) < timeout) return; if (!sysctl_hung_task_warnings) return; @@ -96,8 +96,7 @@ static void check_hung_task(struct task_struct *t, unsigned long now) * complain: */ printk(KERN_ERR "INFO: task %s:%d blocked for more than " - "%ld seconds.\n", t->comm, t->pid, - sysctl_hung_task_timeout_secs); + "%ld seconds.\n", t->comm, t->pid, timeout); printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" " disables this message.\n"); sched_show_task(t); @@ -115,7 +114,7 @@ static void check_hung_task(struct task_struct *t, unsigned long now) * a really long time (120 seconds). If that happens, print out * a warning. */ -static void check_hung_uninterruptible_tasks(void) +static void check_hung_uninterruptible_tasks(unsigned long timeout) { int max_count = sysctl_hung_task_check_count; unsigned long now = get_timestamp(); @@ -134,7 +133,7 @@ static void check_hung_uninterruptible_tasks(void) goto unlock; /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ if (t->state == TASK_UNINTERRUPTIBLE) - check_hung_task(t, now); + check_hung_task(t, now, timeout); } while_each_thread(g, t); unlock: read_unlock(&tasklist_lock); @@ -180,8 +179,17 @@ static int watchdog(void *dummy) update_poll_jiffies(); for ( ; ; ) { + unsigned long timeout; + while (schedule_timeout_interruptible(hung_task_poll_jiffies)); - check_hung_uninterruptible_tasks(); + + /* + * Need to cache timeout here to avoid timeout being set + * to 0 via sysctl while inside check_hung_*_tasks(). + */ + timeout = sysctl_hung_task_timeout_secs; + if (timeout) + check_hung_uninterruptible_tasks(timeout); } return 0; -- cgit v1.2.3 From ce9dbe244bf2063c41792e40dae7745957b118e0 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Wed, 4 Feb 2009 20:35:48 -0800 Subject: softlockup: check all tasks in hung_task MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Impact: extend the scope of hung-task checks Changed the default value of hung_task_check_count to PID_MAX_LIMIT. hung_task_batch_count added to put an upper bound on the critical section. Every hung_task_batch_count checks, the rcu lock is never held for a too long time. Keeping the critical section small minimizes time preemption is disabled and keeps rcu grace periods small. To prevent following a stale pointer, get_task_struct is called on g and t. To verify that g and t have not been unhashed while outside the critical section, the task states are checked. The design was proposed by Frédéric Weisbecker. Signed-off-by: Mandeep Singh Baines Suggested-by: Frédéric Weisbecker Acked-by: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/hung_task.c | 39 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 37 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/hung_task.c b/kernel/hung_task.c index ba8ccd432963..481ca8b5c2bc 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -17,9 +17,18 @@ #include /* - * Have a reasonable limit on the number of tasks checked: + * The number of tasks checked: */ -unsigned long __read_mostly sysctl_hung_task_check_count = 1024; +unsigned long __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT; + +/* + * Limit number of tasks checked in a batch. + * + * This value controls the preemptibility of khungtaskd since preemption + * is disabled during the critical section. It also controls the size of + * the RCU grace period. So it needs to be upper-bound. + */ +#define HUNG_TASK_BATCHING 1024 /* * Zero means infinite timeout - no checking done: @@ -109,6 +118,24 @@ static void check_hung_task(struct task_struct *t, unsigned long now, panic("hung_task: blocked tasks"); } +/* + * To avoid extending the RCU grace period for an unbounded amount of time, + * periodically exit the critical section and enter a new one. + * + * For preemptible RCU it is sufficient to call rcu_read_unlock in order + * exit the grace period. For classic RCU, a reschedule is required. + */ +static void rcu_lock_break(struct task_struct *g, struct task_struct *t) +{ + get_task_struct(g); + get_task_struct(t); + rcu_read_unlock(); + cond_resched(); + rcu_read_lock(); + put_task_struct(t); + put_task_struct(g); +} + /* * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for * a really long time (120 seconds). If that happens, print out @@ -117,6 +144,7 @@ static void check_hung_task(struct task_struct *t, unsigned long now, static void check_hung_uninterruptible_tasks(unsigned long timeout) { int max_count = sysctl_hung_task_check_count; + int batch_count = HUNG_TASK_BATCHING; unsigned long now = get_timestamp(); struct task_struct *g, *t; @@ -131,6 +159,13 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) do_each_thread(g, t) { if (!--max_count) goto unlock; + if (!--batch_count) { + batch_count = HUNG_TASK_BATCHING; + rcu_lock_break(g, t); + /* Exit if t or g was unhashed during refresh. */ + if (t->state == TASK_DEAD || g->state == TASK_DEAD) + goto unlock; + } /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ if (t->state == TASK_UNINTERRUPTIBLE) check_hung_task(t, now, timeout); -- cgit v1.2.3 From 94be52dc075a32af4aa73d7e10f68734d62d6af2 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Thu, 5 Feb 2009 09:56:08 -0800 Subject: softlockup: convert read_lock in hung_task to rcu_read_lock Since the tasklist is protected by rcu list operations, it is safe to convert the read_lock()s to rcu_read_lock(). Suggested-by: Peter Zijlstra Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/hung_task.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 481ca8b5c2bc..3951a80e7cbe 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -155,7 +155,7 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) if (test_taint(TAINT_DIE) || did_panic) return; - read_lock(&tasklist_lock); + rcu_read_lock(); do_each_thread(g, t) { if (!--max_count) goto unlock; @@ -171,7 +171,7 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) check_hung_task(t, now, timeout); } while_each_thread(g, t); unlock: - read_unlock(&tasklist_lock); + rcu_read_unlock(); } static void update_poll_jiffies(void) -- cgit v1.2.3 From 17406b82d621930cca8ccc1272cdac9a7dae8e40 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Fri, 6 Feb 2009 15:37:47 -0800 Subject: softlockup: remove timestamp checking from hung_task Impact: saves sizeof(long) bytes per task_struct By guaranteeing that sysctl_hung_task_timeout_secs have elapsed between tasklist scans we can avoid using timestamps. Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/fork.c | 8 +++----- kernel/hung_task.c | 48 +++++++++--------------------------------------- 2 files changed, 12 insertions(+), 44 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index fb9444282836..bf582f75014b 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -639,6 +639,9 @@ static int copy_mm(unsigned long clone_flags, struct task_struct * tsk) tsk->min_flt = tsk->maj_flt = 0; tsk->nvcsw = tsk->nivcsw = 0; +#ifdef CONFIG_DETECT_HUNG_TASK + tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw; +#endif tsk->mm = NULL; tsk->active_mm = NULL; @@ -1041,11 +1044,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->default_timer_slack_ns = current->timer_slack_ns; -#ifdef CONFIG_DETECT_HUNG_TASK - p->last_switch_count = 0; - p->last_switch_timestamp = 0; -#endif - task_io_accounting_init(&p->ioac); acct_clear_integrals(p); diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 3951a80e7cbe..0c924de58cb2 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -34,7 +34,6 @@ unsigned long __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT; * Zero means infinite timeout - no checking done: */ unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120; -static unsigned long __read_mostly hung_task_poll_jiffies; unsigned long __read_mostly sysctl_hung_task_warnings = 10; @@ -69,33 +68,17 @@ static struct notifier_block panic_block = { .notifier_call = hung_task_panic, }; -/* - * Returns seconds, approximately. We don't need nanosecond - * resolution, and we don't need to waste time with a big divide when - * 2^30ns == 1.074s. - */ -static unsigned long get_timestamp(void) -{ - int this_cpu = raw_smp_processor_id(); - - return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ -} - -static void check_hung_task(struct task_struct *t, unsigned long now, - unsigned long timeout) +static void check_hung_task(struct task_struct *t, unsigned long timeout) { unsigned long switch_count = t->nvcsw + t->nivcsw; if (t->flags & PF_FROZEN) return; - if (switch_count != t->last_switch_count || !t->last_switch_timestamp) { + if (switch_count != t->last_switch_count) { t->last_switch_count = switch_count; - t->last_switch_timestamp = now; return; } - if ((long)(now - t->last_switch_timestamp) < timeout) - return; if (!sysctl_hung_task_warnings) return; sysctl_hung_task_warnings--; @@ -111,7 +94,6 @@ static void check_hung_task(struct task_struct *t, unsigned long now, sched_show_task(t); __debug_show_held_locks(t); - t->last_switch_timestamp = now; touch_nmi_watchdog(); if (sysctl_hung_task_panic) @@ -145,7 +127,6 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) { int max_count = sysctl_hung_task_check_count; int batch_count = HUNG_TASK_BATCHING; - unsigned long now = get_timestamp(); struct task_struct *g, *t; /* @@ -168,19 +149,16 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) } /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ if (t->state == TASK_UNINTERRUPTIBLE) - check_hung_task(t, now, timeout); + check_hung_task(t, timeout); } while_each_thread(g, t); unlock: rcu_read_unlock(); } -static void update_poll_jiffies(void) +static unsigned long timeout_jiffies(unsigned long timeout) { /* timeout of 0 will disable the watchdog */ - if (sysctl_hung_task_timeout_secs == 0) - hung_task_poll_jiffies = MAX_SCHEDULE_TIMEOUT; - else - hung_task_poll_jiffies = sysctl_hung_task_timeout_secs * HZ / 2; + return timeout ? timeout * HZ : MAX_SCHEDULE_TIMEOUT; } /* @@ -197,8 +175,6 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, if (ret || !write) goto out; - update_poll_jiffies(); - wake_up_process(watchdog_task); out: @@ -211,20 +187,14 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, static int watchdog(void *dummy) { set_user_nice(current, 0); - update_poll_jiffies(); for ( ; ; ) { - unsigned long timeout; + unsigned long timeout = sysctl_hung_task_timeout_secs; - while (schedule_timeout_interruptible(hung_task_poll_jiffies)); + while (schedule_timeout_interruptible(timeout_jiffies(timeout))) + timeout = sysctl_hung_task_timeout_secs; - /* - * Need to cache timeout here to avoid timeout being set - * to 0 via sysctl while inside check_hung_*_tasks(). - */ - timeout = sysctl_hung_task_timeout_secs; - if (timeout) - check_hung_uninterruptible_tasks(timeout); + check_hung_uninterruptible_tasks(timeout); } return 0; -- cgit v1.2.3 From cf2592f59c0e8ed4308adbdb2e0a88655379d579 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 10 Feb 2009 16:52:37 +0100 Subject: softlockup: ensure the task has been switched out once When we check if a task has been switched out since the last scan, we might have a race condition on the following scenario: - the task is freshly created and scheduled - it puts its state to TASK_UNINTERRUPTIBLE and is not yet switched out - check_hung_task() scans this task and will report a false positive because t->nvcsw + t->nivcsw == t->last_switch_count == 0 Add a check for such cases. Signed-off-by: Frederic Weisbecker Acked-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- kernel/hung_task.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 0c924de58cb2..022a4927b785 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -72,7 +72,13 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout) { unsigned long switch_count = t->nvcsw + t->nivcsw; - if (t->flags & PF_FROZEN) + /* + * Ensure the task is not frozen. + * Also, when a freshly created task is scheduled once, changes + * its state to TASK_UNINTERRUPTIBLE without having ever been + * switched out once, it musn't be checked. + */ + if (unlikely(t->flags & PF_FROZEN || !switch_count)) return; if (switch_count != t->last_switch_count) { -- cgit v1.2.3 From 9f8d979f082a3ee1b27f32b7e0811b51c3ad1d15 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 12 Feb 2009 13:10:17 +0100 Subject: softlockup: move 'one' to the softlockup section in sysctl.c CONFIG_SOFTLOCKUP=y || CONFIG_DETECT_HUNG_TASKS=y is now the only user of the 'one' constant in kernel/sysctl.c. Move it to the softlockup block of constants. This fixes a GCC warning. Signed-off-by: Ingo Molnar --- kernel/sysctl.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 3b6b54c8ac0d..6d2aeff92b3d 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -90,6 +90,9 @@ extern int rcutorture_runnable; #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */ /* Constants used for minimum and maximum */ +#if defined(CONFIG_DETECT_HUNG_TASK) || defined(CONFIG_DETECT_SOFTLOCKUP) || defined(CONFIG_HIGHMEM) +static int one = 1; +#endif #ifdef CONFIG_DETECT_SOFTLOCKUP static int sixty = 60; static int neg_one = -1; @@ -100,7 +103,6 @@ static int two = 2; #endif static int zero; -static int one = 1; static unsigned long one_ul = 1; static int one_hundred = 100; -- cgit v1.2.3 From 3aa551c9b4c40018f0e261a178e3d25478dc04a9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 23 Mar 2009 18:28:15 +0100 Subject: genirq: add threaded interrupt handler support Add support for threaded interrupt handlers: A device driver can request that its main interrupt handler runs in a thread. To achive this the device driver requests the interrupt with request_threaded_irq() and provides additionally to the handler a thread function. The handler function is called in hard interrupt context and needs to check whether the interrupt originated from the device. If the interrupt originated from the device then the handler can either return IRQ_HANDLED or IRQ_WAKE_THREAD. IRQ_HANDLED is returned when no further action is required. IRQ_WAKE_THREAD causes the genirq code to invoke the threaded (main) handler. When IRQ_WAKE_THREAD is returned handler must have disabled the interrupt on the device level. This is mandatory for shared interrupt handlers, but we need to do it as well for obscure x86 hardware where disabling an interrupt on the IO_APIC level redirects the interrupt to the legacy PIC interrupt lines. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- kernel/exit.c | 2 + kernel/irq/handle.c | 31 ++++++++- kernel/irq/manage.c | 192 ++++++++++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 211 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/exit.c b/kernel/exit.c index 167e1e3ad7c6..ca0b3488c4a9 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -1037,6 +1037,8 @@ NORET_TYPE void do_exit(long code) schedule(); } + exit_irq_thread(); + exit_signals(tsk); /* sets PF_EXITING */ /* * tsk->flags are checked in the futex code to protect against diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 9ebf77968871..fe8f45374e86 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -357,8 +357,37 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) do { ret = action->handler(irq, action->dev_id); - if (ret == IRQ_HANDLED) + + switch (ret) { + case IRQ_WAKE_THREAD: + /* + * Wake up the handler thread for this + * action. In case the thread crashed and was + * killed we just pretend that we handled the + * interrupt. The hardirq handler above has + * disabled the device interrupt, so no irq + * storm is lurking. + */ + if (likely(!test_bit(IRQTF_DIED, + &action->thread_flags))) { + set_bit(IRQTF_RUNTHREAD, &action->thread_flags); + wake_up_process(action->thread); + } + + /* + * Set it to handled so the spurious check + * does not trigger. + */ + ret = IRQ_HANDLED; + /* Fall through to add to randomness */ + case IRQ_HANDLED: status |= action->flags; + break; + + default: + break; + } + retval |= ret; action = action->next; } while (action); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 6458e99984c0..a4c1ab86cd25 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -8,16 +8,15 @@ */ #include +#include #include #include #include #include +#include #include "internals.h" -#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) -cpumask_var_t irq_default_affinity; - /** * synchronize_irq - wait for pending IRQ handlers (on other CPUs) * @irq: interrupt number to wait for @@ -53,9 +52,18 @@ void synchronize_irq(unsigned int irq) /* Oops, that failed? */ } while (status & IRQ_INPROGRESS); + + /* + * We made sure that no hardirq handler is running. Now verify + * that no threaded handlers are active. + */ + wait_event(desc->wait_for_threads, !atomic_read(&desc->threads_active)); } EXPORT_SYMBOL(synchronize_irq); +#ifdef CONFIG_SMP +cpumask_var_t irq_default_affinity; + /** * irq_can_set_affinity - Check if the affinity of a given irq can be set * @irq: Interrupt to check @@ -72,6 +80,18 @@ int irq_can_set_affinity(unsigned int irq) return 1; } +static void +irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask) +{ + struct irqaction *action = desc->action; + + while (action) { + if (action->thread) + set_cpus_allowed_ptr(action->thread, cpumask); + action = action->next; + } +} + /** * irq_set_affinity - Set the irq affinity of a given irq * @irq: Interrupt to set affinity @@ -100,6 +120,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) cpumask_copy(desc->affinity, cpumask); desc->chip->set_affinity(irq, cpumask); #endif + irq_set_thread_affinity(desc, cpumask); desc->status |= IRQ_AFFINITY_SET; spin_unlock_irqrestore(&desc->lock, flags); return 0; @@ -150,6 +171,8 @@ int irq_select_affinity_usr(unsigned int irq) spin_lock_irqsave(&desc->lock, flags); ret = setup_affinity(irq, desc); + if (!ret) + irq_set_thread_affinity(desc, desc->affinity); spin_unlock_irqrestore(&desc->lock, flags); return ret; @@ -384,6 +407,93 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, return ret; } +static inline int irq_thread_should_run(struct irqaction *action) +{ + return test_and_clear_bit(IRQTF_RUNTHREAD, &action->thread_flags); +} + +static int irq_wait_for_interrupt(struct irqaction *action) +{ + while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); + if (irq_thread_should_run(action)) { + __set_current_state(TASK_RUNNING); + return 0; + } else + schedule(); + } + return -1; +} + +/* + * Interrupt handler thread + */ +static int irq_thread(void *data) +{ + struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO/2, }; + struct irqaction *action = data; + struct irq_desc *desc = irq_to_desc(action->irq); + int wake; + + sched_setscheduler(current, SCHED_FIFO, ¶m); + current->irqaction = action; + + while (!irq_wait_for_interrupt(action)) { + + atomic_inc(&desc->threads_active); + + spin_lock_irq(&desc->lock); + if (unlikely(desc->status & IRQ_DISABLED)) { + /* + * CHECKME: We might need a dedicated + * IRQ_THREAD_PENDING flag here, which + * retriggers the thread in check_irq_resend() + * but AFAICT IRQ_PENDING should be fine as it + * retriggers the interrupt itself --- tglx + */ + desc->status |= IRQ_PENDING; + spin_unlock_irq(&desc->lock); + } else { + spin_unlock_irq(&desc->lock); + + action->thread_fn(action->irq, action->dev_id); + } + + wake = atomic_dec_and_test(&desc->threads_active); + + if (wake && waitqueue_active(&desc->wait_for_threads)) + wake_up(&desc->wait_for_threads); + } + + /* + * Clear irqaction. Otherwise exit_irq_thread() would make + * fuzz about an active irq thread going into nirvana. + */ + current->irqaction = NULL; + return 0; +} + +/* + * Called from do_exit() + */ +void exit_irq_thread(void) +{ + struct task_struct *tsk = current; + + if (!tsk->irqaction) + return; + + printk(KERN_ERR + "exiting task \"%s\" (%d) is an active IRQ thread (irq %d)\n", + tsk->comm ? tsk->comm : "", tsk->pid, tsk->irqaction->irq); + + /* + * Set the THREAD DIED flag to prevent further wakeups of the + * soon to be gone threaded handler. + */ + set_bit(IRQTF_DIED, &tsk->irqaction->flags); +} + /* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. @@ -419,6 +529,26 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) rand_initialize_irq(irq); } + /* + * Threaded handler ? + */ + if (new->thread_fn) { + struct task_struct *t; + + t = kthread_create(irq_thread, new, "irq/%d-%s", irq, + new->name); + if (IS_ERR(t)) + return PTR_ERR(t); + /* + * We keep the reference to the task struct even if + * the thread dies to avoid that the interrupt code + * references an already freed task_struct. + */ + get_task_struct(t); + new->thread = t; + wake_up_process(t); + } + /* * The following block of code has to be executed atomically */ @@ -456,15 +586,15 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (!shared) { irq_chip_set_defaults(desc->chip); + init_waitqueue_head(&desc->wait_for_threads); + /* Setup the type (level, edge polarity) if configured: */ if (new->flags & IRQF_TRIGGER_MASK) { ret = __irq_set_trigger(desc, irq, new->flags & IRQF_TRIGGER_MASK); - if (ret) { - spin_unlock_irqrestore(&desc->lock, flags); - return ret; - } + if (ret) + goto out_thread; } else compat_irq_chip_set_default_handler(desc); #if defined(CONFIG_IRQ_PER_CPU) @@ -532,8 +662,19 @@ mismatch: dump_stack(); } #endif + ret = -EBUSY; + +out_thread: spin_unlock_irqrestore(&desc->lock, flags); - return -EBUSY; + if (new->thread) { + struct task_struct *t = new->thread; + + new->thread = NULL; + if (likely(!test_bit(IRQTF_DIED, &new->thread_flags))) + kthread_stop(t); + put_task_struct(t); + } + return ret; } /** @@ -559,6 +700,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) { struct irq_desc *desc = irq_to_desc(irq); struct irqaction *action, **action_ptr; + struct task_struct *irqthread; unsigned long flags; WARN(in_interrupt(), "Trying to free IRQ %d from IRQ context!\n", irq); @@ -605,6 +747,10 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) else desc->chip->disable(irq); } + + irqthread = action->thread; + action->thread = NULL; + spin_unlock_irqrestore(&desc->lock, flags); unregister_handler_proc(irq, action); @@ -612,6 +758,12 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) /* Make sure it's not being used on another CPU: */ synchronize_irq(irq); + if (irqthread) { + if (!test_bit(IRQTF_DIED, &action->thread_flags)) + kthread_stop(irqthread); + put_task_struct(irqthread); + } + #ifdef CONFIG_DEBUG_SHIRQ /* * It's a shared IRQ -- the driver ought to be prepared for an IRQ @@ -664,9 +816,12 @@ void free_irq(unsigned int irq, void *dev_id) EXPORT_SYMBOL(free_irq); /** - * request_irq - allocate an interrupt line + * request_threaded_irq - allocate an interrupt line * @irq: Interrupt line to allocate - * @handler: Function to be called when the IRQ occurs + * @handler: Function to be called when the IRQ occurs. + * Primary handler for threaded interrupts + * @thread_fn: Function called from the irq handler thread + * If NULL, no irq thread is created * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function @@ -678,6 +833,15 @@ EXPORT_SYMBOL(free_irq); * raises, you must take care both to initialise your hardware * and to set up the interrupt handler in the right order. * + * If you want to set up a threaded irq handler for your device + * then you need to supply @handler and @thread_fn. @handler ist + * still called in hard interrupt context and has to check + * whether the interrupt originates from the device. If yes it + * needs to disable the interrupt on the device and return + * IRQ_THREAD_WAKE which will wake up the handler thread and run + * @thread_fn. This split handler design is necessary to support + * shared interrupts. + * * Dev_id must be globally unique. Normally the address of the * device data structure is used as the cookie. Since the handler * receives this value it makes sense to use it. @@ -693,8 +857,9 @@ EXPORT_SYMBOL(free_irq); * IRQF_TRIGGER_* Specify active edge(s) or level * */ -int request_irq(unsigned int irq, irq_handler_t handler, - unsigned long irqflags, const char *devname, void *dev_id) +int request_threaded_irq(unsigned int irq, irq_handler_t handler, + irq_handler_t thread_fn, unsigned long irqflags, + const char *devname, void *dev_id) { struct irqaction *action; struct irq_desc *desc; @@ -742,6 +907,7 @@ int request_irq(unsigned int irq, irq_handler_t handler, return -ENOMEM; action->handler = handler; + action->thread_fn = thread_fn; action->flags = irqflags; action->name = devname; action->dev_id = dev_id; @@ -771,4 +937,4 @@ int request_irq(unsigned int irq, irq_handler_t handler, #endif return retval; } -EXPORT_SYMBOL(request_irq); +EXPORT_SYMBOL(request_threaded_irq); -- cgit v1.2.3 From 935bd5b971f0df7c06d214d022cf8392e2f37952 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Mon, 23 Mar 2009 18:28:16 +0100 Subject: genirq: add support for threaded interrupts to devres Some devices use devres_request_irq() for to install their interrupt handler. Add support for threaded interrupts to devres as well. [tglx - simplified and adapted to latest threadirq version] Signed-off-by: Arjan van de Ven Signed-off-by: Thomas Gleixner --- kernel/irq/devres.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index 38a25b8d8bff..d06df9c41cba 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -26,10 +26,12 @@ static int devm_irq_match(struct device *dev, void *res, void *data) } /** - * devm_request_irq - allocate an interrupt line for a managed device + * devm_request_threaded_irq - allocate an interrupt line for a managed device * @dev: device to request interrupt for * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs + * @thread_fn: function to be called in a threaded interrupt context. NULL + * for devices which handle everything in @handler * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function @@ -42,9 +44,10 @@ static int devm_irq_match(struct device *dev, void *res, void *data) * If an IRQ allocated with this function needs to be freed * separately, dev_free_irq() must be used. */ -int devm_request_irq(struct device *dev, unsigned int irq, - irq_handler_t handler, unsigned long irqflags, - const char *devname, void *dev_id) +int devm_request_threaded_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, irq_handler_t thread_fn, + unsigned long irqflags, const char *devname, + void *dev_id) { struct irq_devres *dr; int rc; @@ -54,7 +57,8 @@ int devm_request_irq(struct device *dev, unsigned int irq, if (!dr) return -ENOMEM; - rc = request_irq(irq, handler, irqflags, devname, dev_id); + rc = request_threaded_irq(irq, handler, thread_fn, irqflags, devname, + dev_id); if (rc) { devres_free(dr); return rc; @@ -66,7 +70,7 @@ int devm_request_irq(struct device *dev, unsigned int irq, return 0; } -EXPORT_SYMBOL(devm_request_irq); +EXPORT_SYMBOL(devm_request_threaded_irq); /** * devm_free_irq - free an interrupt -- cgit v1.2.3 From f48fe81e5b032914183e9a17052313720c2cac56 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 24 Mar 2009 11:46:22 +0100 Subject: genirq: threaded irq handlers review fixups Delta patch to address the review comments. - Implement warning when IRQ_WAKE_THREAD is requested and no thread handler installed - coding style fixes Pointed-out-by: Christoph Hellwig Signed-off-by: Thomas Gleixner --- kernel/irq/handle.c | 29 ++++++++++++++++++++++++----- kernel/irq/manage.c | 17 +++++++---------- 2 files changed, 31 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index fe8f45374e86..38b49a9e508a 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -338,6 +338,15 @@ irqreturn_t no_action(int cpl, void *dev_id) return IRQ_NONE; } +static void warn_no_thread(unsigned int irq, struct irqaction *action) +{ + if (test_and_set_bit(IRQTF_WARNED, &action->thread_flags)) + return; + + printk(KERN_WARNING "IRQ %d device %s returned IRQ_WAKE_THREAD " + "but no thread function available.", irq, action->name); +} + /** * handle_IRQ_event - irq action chain handler * @irq: the interrupt number @@ -360,6 +369,21 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) switch (ret) { case IRQ_WAKE_THREAD: + /* + * Set result to handled so the spurious check + * does not trigger. + */ + ret = IRQ_HANDLED; + + /* + * Catch drivers which return WAKE_THREAD but + * did not set up a thread function + */ + if (unlikely(!action->thread_fn)) { + warn_no_thread(irq, action); + break; + } + /* * Wake up the handler thread for this * action. In case the thread crashed and was @@ -374,11 +398,6 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) wake_up_process(action->thread); } - /* - * Set it to handled so the spurious check - * does not trigger. - */ - ret = IRQ_HANDLED; /* Fall through to add to randomness */ case IRQ_HANDLED: status |= action->flags; diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index a4c1ab86cd25..a3eb7baf1e46 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -407,20 +407,17 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, return ret; } -static inline int irq_thread_should_run(struct irqaction *action) -{ - return test_and_clear_bit(IRQTF_RUNTHREAD, &action->thread_flags); -} - static int irq_wait_for_interrupt(struct irqaction *action) { while (!kthread_should_stop()) { set_current_state(TASK_INTERRUPTIBLE); - if (irq_thread_should_run(action)) { + + if (test_and_clear_bit(IRQTF_RUNTHREAD, + &action->thread_flags)) { __set_current_state(TASK_RUNNING); return 0; - } else - schedule(); + } + schedule(); } return -1; } @@ -820,8 +817,8 @@ EXPORT_SYMBOL(free_irq); * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs. * Primary handler for threaded interrupts - * @thread_fn: Function called from the irq handler thread - * If NULL, no irq thread is created + * @thread_fn: Function called from the irq handler thread + * If NULL, no irq thread is created * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function -- cgit v1.2.3 From 7f1e2ca9f04b02794597f60e7b1d43f0a1317939 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 13 Mar 2009 12:21:27 +0100 Subject: hrtimer: fix rq->lock inversion (again) It appears I inadvertly introduced rq->lock recursion to the hrtimer_start() path when I delegated running already expired timers to softirq context. This patch fixes it by introducing a __hrtimer_start_range_ns() method that will not use raise_softirq_irqoff() but __raise_softirq_irqoff() which avoids the wakeup. It then also changes schedule() to check for pending softirqs and do the wakeup then, I'm not quite sure I like this last bit, nor am I convinced its really needed. Signed-off-by: Peter Zijlstra Cc: Peter Zijlstra Cc: paulus@samba.org LKML-Reference: <20090313112301.096138802@chello.nl> Signed-off-by: Ingo Molnar --- kernel/hrtimer.c | 55 ++++++++++++++++++++++++++++++++++--------------------- kernel/sched.c | 14 +++++++++++--- kernel/softirq.c | 2 +- 3 files changed, 46 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index f394d2a42ca3..cb8a15c19583 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -651,14 +651,20 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer) * and expiry check is done in the hrtimer_interrupt or in the softirq. */ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) + struct hrtimer_clock_base *base, + int wakeup) { if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) { - spin_unlock(&base->cpu_base->lock); - raise_softirq_irqoff(HRTIMER_SOFTIRQ); - spin_lock(&base->cpu_base->lock); + if (wakeup) { + spin_unlock(&base->cpu_base->lock); + raise_softirq_irqoff(HRTIMER_SOFTIRQ); + spin_lock(&base->cpu_base->lock); + } else + __raise_softirq_irqoff(HRTIMER_SOFTIRQ); + return 1; } + return 0; } @@ -703,7 +709,8 @@ static inline int hrtimer_is_hres_enabled(void) { return 0; } static inline int hrtimer_switch_to_hres(void) { return 0; } static inline void hrtimer_force_reprogram(struct hrtimer_cpu_base *base) { } static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) + struct hrtimer_clock_base *base, + int wakeup) { return 0; } @@ -886,20 +893,9 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) return 0; } -/** - * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU - * @timer: the timer to be added - * @tim: expiry time - * @delta_ns: "slack" range for the timer - * @mode: expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL) - * - * Returns: - * 0 on success - * 1 when the timer was active - */ -int -hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_ns, - const enum hrtimer_mode mode) +int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, + unsigned long delta_ns, const enum hrtimer_mode mode, + int wakeup) { struct hrtimer_clock_base *base, *new_base; unsigned long flags; @@ -940,12 +936,29 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n * XXX send_remote_softirq() ? */ if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases)) - hrtimer_enqueue_reprogram(timer, new_base); + hrtimer_enqueue_reprogram(timer, new_base, wakeup); unlock_hrtimer_base(timer, &flags); return ret; } + +/** + * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU + * @timer: the timer to be added + * @tim: expiry time + * @delta_ns: "slack" range for the timer + * @mode: expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL) + * + * Returns: + * 0 on success + * 1 when the timer was active + */ +int hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, + unsigned long delta_ns, const enum hrtimer_mode mode) +{ + return __hrtimer_start_range_ns(timer, tim, delta_ns, mode, 1); +} EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); /** @@ -961,7 +974,7 @@ EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); int hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode) { - return hrtimer_start_range_ns(timer, tim, 0, mode); + return __hrtimer_start_range_ns(timer, tim, 0, mode, 1); } EXPORT_SYMBOL_GPL(hrtimer_start); diff --git a/kernel/sched.c b/kernel/sched.c index 196d48babbef..63256e3ede2a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -231,13 +231,20 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) spin_lock(&rt_b->rt_runtime_lock); for (;;) { + unsigned long delta; + ktime_t soft, hard; + if (hrtimer_active(&rt_b->rt_period_timer)) break; now = hrtimer_cb_get_time(&rt_b->rt_period_timer); hrtimer_forward(&rt_b->rt_period_timer, now, rt_b->rt_period); - hrtimer_start_expires(&rt_b->rt_period_timer, - HRTIMER_MODE_ABS); + + soft = hrtimer_get_softexpires(&rt_b->rt_period_timer); + hard = hrtimer_get_expires(&rt_b->rt_period_timer); + delta = ktime_to_ns(ktime_sub(hard, soft)); + __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta, + HRTIMER_MODE_ABS, 0); } spin_unlock(&rt_b->rt_runtime_lock); } @@ -1146,7 +1153,8 @@ static __init void init_hrtick(void) */ static void hrtick_start(struct rq *rq, u64 delay) { - hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), HRTIMER_MODE_REL); + __hrtimer_start_range_ns(&rq->hrtick_timer, ns_to_ktime(delay), 0, + HRTIMER_MODE_REL, 0); } static inline void init_hrtick(void) diff --git a/kernel/softirq.c b/kernel/softirq.c index 487751604300..accc85197c49 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -58,7 +58,7 @@ static DEFINE_PER_CPU(struct task_struct *, ksoftirqd); * to the pending events, so lets the scheduler to balance * the softirq load for us. */ -static inline void wakeup_softirqd(void) +void wakeup_softirqd(void) { /* Interrupts are disabled: no need to stop preemption */ struct task_struct *tsk = __get_cpu_var(ksoftirqd); -- cgit v1.2.3 From eedeeabdeeadb016b8c783e3620d06b98d0cb4e1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 18 Mar 2009 12:38:47 +0100 Subject: lockdep: add stack dumps to asserts Have a better idea about exactly which loc causes a lockdep limit overflow. Often it's a bug or inefficiency in that subsystem. Signed-off-by: Peter Zijlstra LKML-Reference: <1237376327.5069.253.camel@laptop> Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 981cd4854281..a288ae107b50 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -792,6 +792,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) printk("BUG: MAX_LOCKDEP_KEYS too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return NULL; } class = lock_classes + nr_lock_classes++; @@ -855,6 +856,7 @@ static struct lock_list *alloc_list_entry(void) printk("BUG: MAX_LOCKDEP_ENTRIES too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return NULL; } return list_entries + nr_list_entries++; @@ -1681,6 +1683,7 @@ cache_hit: printk("BUG: MAX_LOCKDEP_CHAINS too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } chain = lock_chains + nr_lock_chains++; @@ -2540,6 +2543,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, debug_locks_off(); printk("BUG: MAX_LOCKDEP_SUBCLASSES too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } @@ -2636,6 +2640,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, debug_locks_off(); printk("BUG: MAX_LOCK_DEPTH too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } -- cgit v1.2.3 From a18b83b7ef3c98cd8b4bb885e4a649a8f30fb7b0 Mon Sep 17 00:00:00 2001 From: Bharata B Rao Date: Mon, 23 Mar 2009 10:02:53 +0530 Subject: cpuacct: make cpuacct hierarchy walk in cpuacct_charge() safe when rcupreempt is used -v2 Impact: fix cgroups race under rcu-preempt cpuacct_charge() obtains task's ca and does a hierarchy walk upwards. This can race with the task's movement between cgroups. This race can cause an access to freed ca pointer in cpuacct_charge() or access to invalid cgroups pointer of the task. This will not happen with rcu or tree rcu as cpuacct_charge() is called with preemption disabled. However if rcupreempt is used, the race is seen. Thanks to Li Zefan for explaining this. Fix this race by explicitly protecting ca and the hierarchy walk with rcu_read_lock(). Changes for v2: - Update patch descrition (as per Li Zefan's review comments). - Remove comments in cpuacct_charge() which explained why rcu_read_lock() was needed (as per Peter Zijlstra's review comments). Signed-off-by: Bharata B Rao Cc: Dhaval Giani Cc: Li Zefan Cc: Paul Menage Cc: KAMEZAWA Hiroyuki Acked-by: Peter Zijlstra Acked-by: Balbir Singh Tested-by: Balbir Singh Signed-off-by: Ingo Molnar --- kernel/sched.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 186c6fd08acf..cc397aae5eae 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -9654,12 +9654,17 @@ static void cpuacct_charge(struct task_struct *tsk, u64 cputime) return; cpu = task_cpu(tsk); + + rcu_read_lock(); + ca = task_ca(tsk); for (; ca; ca = ca->parent) { u64 *cpuusage = percpu_ptr(ca->cpuusage, cpu); *cpuusage += cputime; } + + rcu_read_unlock(); } struct cgroup_subsys cpuacct_subsys = { -- cgit v1.2.3 From 13b8bd0a5713bdf05659019badd7c0407984ece1 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Wed, 25 Mar 2009 15:01:22 +1030 Subject: sched_rt: don't allocate cpumask in fastpath Impact: cleanup As pointed out by Steven Rostedt. Since the arg in question is unused, we simply change cpupri_find() to accept NULL. Reported-by: Steven Rostedt Signed-off-by: Rusty Russell LKML-Reference: <200903251501.22664.rusty@rustcorp.com.au> Signed-off-by: Ingo Molnar --- kernel/sched_cpupri.c | 5 +++-- kernel/sched_rt.c | 15 ++++----------- 2 files changed, 7 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 1e00bfacf9b8..cdd3c89574cd 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -55,7 +55,7 @@ static int convert_prio(int prio) * cpupri_find - find the best (lowest-pri) CPU in the system * @cp: The cpupri context * @p: The task - * @lowest_mask: A mask to fill in with selected CPUs + * @lowest_mask: A mask to fill in with selected CPUs (or NULL) * * Note: This function returns the recommended CPUs as calculated during the * current invokation. By the time the call returns, the CPUs may have in @@ -81,7 +81,8 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; - cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); + if (lowest_mask) + cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); return 1; } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index bac1061cea2f..fbec5a58ff10 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -805,20 +805,15 @@ static int select_task_rq_rt(struct task_struct *p, int sync) static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) { - cpumask_var_t mask; - if (rq->curr->rt.nr_cpus_allowed == 1) return; - if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) - return; - if (p->rt.nr_cpus_allowed != 1 - && cpupri_find(&rq->rd->cpupri, p, mask)) - goto free; + && cpupri_find(&rq->rd->cpupri, p, NULL)) + return; - if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask)) - goto free; + if (!cpupri_find(&rq->rd->cpupri, rq->curr, NULL)) + return; /* * There appears to be other cpus that can accept @@ -827,8 +822,6 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) */ requeue_task_rt(rq, p, 1); resched_task(rq->curr); -free: - free_cpumask_var(mask); } #endif /* CONFIG_SMP */ -- cgit v1.2.3 From c5f8d99585d7b5b7e857fabf8aefd0174903a98c Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Tue, 31 Mar 2009 16:56:03 +0900 Subject: posixtimers, sched: Fix posix clock monotonicity Impact: Regression fix (against clock_gettime() backwarding bug) This patch re-introduces a couple of functions, task_sched_runtime and thread_group_sched_runtime, which was once removed at the time of 2.6.28-rc1. These functions protect the sampling of thread/process clock with rq lock. This rq lock is required not to update rq->clock during the sampling. i.e. The clock_gettime() may return ((accounted runtime before update) + (delta after update)) that is less than what it should be. v2 -> v3: - Rename static helper function __task_delta_exec() to do_task_delta_exec() since -tip tree already has a __task_delta_exec() of different version. v1 -> v2: - Revises comments of function and patch description. - Add note about accuracy of thread group's runtime. Signed-off-by: Hidetoshi Seto Acked-by: Peter Zijlstra Cc: stable@kernel.org [2.6.28.x][2.6.29.x] LKML-Reference: <49D1CC93.4080401@jp.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/posix-cpu-timers.c | 7 ++--- kernel/sched.c | 65 +++++++++++++++++++++++++++++++++++++++++------ 2 files changed, 61 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index fa07da94d7be..4318c3085788 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -224,7 +224,7 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, cpu->cpu = virt_ticks(p); break; case CPUCLOCK_SCHED: - cpu->sched = p->se.sum_exec_runtime + task_delta_exec(p); + cpu->sched = task_sched_runtime(p); break; } return 0; @@ -240,18 +240,19 @@ static int cpu_clock_sample_group(const clockid_t which_clock, { struct task_cputime cputime; - thread_group_cputime(p, &cputime); switch (CPUCLOCK_WHICH(which_clock)) { default: return -EINVAL; case CPUCLOCK_PROF: + thread_group_cputime(p, &cputime); cpu->cpu = cputime_add(cputime.utime, cputime.stime); break; case CPUCLOCK_VIRT: + thread_group_cputime(p, &cputime); cpu->cpu = cputime.utime; break; case CPUCLOCK_SCHED: - cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p); + cpu->sched = thread_group_sched_runtime(p); break; } return 0; diff --git a/kernel/sched.c b/kernel/sched.c index cc397aae5eae..c8d7f17bd036 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4139,9 +4139,25 @@ DEFINE_PER_CPU(struct kernel_stat, kstat); EXPORT_PER_CPU_SYMBOL(kstat); /* - * Return any ns on the sched_clock that have not yet been banked in + * Return any ns on the sched_clock that have not yet been accounted in * @p in case that task is currently running. + * + * Called with task_rq_lock() held on @rq. */ +static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) +{ + u64 ns = 0; + + if (task_current(rq, p)) { + update_rq_clock(rq); + ns = rq->clock - p->se.exec_start; + if ((s64)ns < 0) + ns = 0; + } + + return ns; +} + unsigned long long task_delta_exec(struct task_struct *p) { unsigned long flags; @@ -4149,16 +4165,49 @@ unsigned long long task_delta_exec(struct task_struct *p) u64 ns = 0; rq = task_rq_lock(p, &flags); + ns = do_task_delta_exec(p, rq); + task_rq_unlock(rq, &flags); - if (task_current(rq, p)) { - u64 delta_exec; + return ns; +} - update_rq_clock(rq); - delta_exec = rq->clock - p->se.exec_start; - if ((s64)delta_exec > 0) - ns = delta_exec; - } +/* + * Return accounted runtime for the task. + * In case the task is currently running, return the runtime plus current's + * pending runtime that have not been accounted yet. + */ +unsigned long long task_sched_runtime(struct task_struct *p) +{ + unsigned long flags; + struct rq *rq; + u64 ns = 0; + + rq = task_rq_lock(p, &flags); + ns = p->se.sum_exec_runtime + do_task_delta_exec(p, rq); + task_rq_unlock(rq, &flags); + + return ns; +} +/* + * Return sum_exec_runtime for the thread group. + * In case the task is currently running, return the sum plus current's + * pending runtime that have not been accounted yet. + * + * Note that the thread group might have other running tasks as well, + * so the return value not includes other pending runtime that other + * running tasks might have. + */ +unsigned long long thread_group_sched_runtime(struct task_struct *p) +{ + struct task_cputime totals; + unsigned long flags; + struct rq *rq; + u64 ns; + + rq = task_rq_lock(p, &flags); + thread_group_cputime(p, &totals); + ns = totals.sum_exec_runtime + do_task_delta_exec(p, rq); task_rq_unlock(rq, &flags); return ns; -- cgit v1.2.3 From ef12fefabf94b6a902ad3abd3eb124b00560c445 Mon Sep 17 00:00:00 2001 From: Bharata B Rao Date: Tue, 31 Mar 2009 10:02:22 +0530 Subject: cpuacct: add per-cgroup utime/stime statistics Add per-cgroup cpuacct controller statistics like the system and user time consumed by the group of tasks. Changelog: v7 - Changed the name of the statistic from utime to user and from stime to system so that in future we could easily add other statistics like irq, softirq, steal times etc easily. v6 - Fixed a bug in the error path of cpuacct_create() (pointed by Li Zefan). v5 - In cpuacct_stats_show(), use cputime64_to_clock_t() since we are operating on a 64bit variable here. v4 - Remove comments in cpuacct_update_stats() which explained why rcu_read_lock() was needed (as per Peter Zijlstra's review comments). - Don't say that percpu_counter_read() is broken in Documentation/cpuacct.txt as per KAMEZAWA Hiroyuki's review comments. v3 - Fix a small race in the cpuacct hierarchy walk. v2 - stime and utime now exported in clock_t units instead of msecs. - Addressed the code review comments from Balbir and Li Zefan. - Moved to -tip tree. v1 - Moved the stime/utime accounting to cpuacct controller. Earlier versions - http://lkml.org/lkml/2009/2/25/129 Signed-off-by: Bharata B Rao Signed-off-by: Balaji Rao Cc: Dhaval Giani Cc: Paul Menage Cc: Andrew Morton Cc: KAMEZAWA Hiroyuki Reviewed-by: Li Zefan Acked-by: Peter Zijlstra Acked-by: Balbir Singh Tested-by: Balbir Singh LKML-Reference: <20090331043222.GA4093@in.ibm.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 87 ++++++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 81 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index c8d7f17bd036..8d1bdbe8aafc 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1393,10 +1393,22 @@ iter_move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest, struct rq_iterator *iterator); #endif +/* Time spent by the tasks of the cpu accounting group executing in ... */ +enum cpuacct_stat_index { + CPUACCT_STAT_USER, /* ... user mode */ + CPUACCT_STAT_SYSTEM, /* ... kernel mode */ + + CPUACCT_STAT_NSTATS, +}; + #ifdef CONFIG_CGROUP_CPUACCT static void cpuacct_charge(struct task_struct *tsk, u64 cputime); +static void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val); #else static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {} +static inline void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val) {} #endif static inline void inc_cpu_load(struct rq *rq, unsigned long load) @@ -4236,6 +4248,8 @@ void account_user_time(struct task_struct *p, cputime_t cputime, cpustat->nice = cputime64_add(cpustat->nice, tmp); else cpustat->user = cputime64_add(cpustat->user, tmp); + + cpuacct_update_stats(p, CPUACCT_STAT_USER, cputime); /* Account for user time used */ acct_update_integrals(p); } @@ -4297,6 +4311,8 @@ void account_system_time(struct task_struct *p, int hardirq_offset, else cpustat->system = cputime64_add(cpustat->system, tmp); + cpuacct_update_stats(p, CPUACCT_STAT_SYSTEM, cputime); + /* Account for system time used */ acct_update_integrals(p); } @@ -9539,6 +9555,7 @@ struct cpuacct { struct cgroup_subsys_state css; /* cpuusage holds pointer to a u64-type object on every cpu */ u64 *cpuusage; + struct percpu_counter cpustat[CPUACCT_STAT_NSTATS]; struct cpuacct *parent; }; @@ -9563,20 +9580,32 @@ static struct cgroup_subsys_state *cpuacct_create( struct cgroup_subsys *ss, struct cgroup *cgrp) { struct cpuacct *ca = kzalloc(sizeof(*ca), GFP_KERNEL); + int i; if (!ca) - return ERR_PTR(-ENOMEM); + goto out; ca->cpuusage = alloc_percpu(u64); - if (!ca->cpuusage) { - kfree(ca); - return ERR_PTR(-ENOMEM); - } + if (!ca->cpuusage) + goto out_free_ca; + + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) + if (percpu_counter_init(&ca->cpustat[i], 0)) + goto out_free_counters; if (cgrp->parent) ca->parent = cgroup_ca(cgrp->parent); return &ca->css; + +out_free_counters: + while (--i >= 0) + percpu_counter_destroy(&ca->cpustat[i]); + free_percpu(ca->cpuusage); +out_free_ca: + kfree(ca); +out: + return ERR_PTR(-ENOMEM); } /* destroy an existing cpu accounting group */ @@ -9584,7 +9613,10 @@ static void cpuacct_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp) { struct cpuacct *ca = cgroup_ca(cgrp); + int i; + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) + percpu_counter_destroy(&ca->cpustat[i]); free_percpu(ca->cpuusage); kfree(ca); } @@ -9671,6 +9703,25 @@ static int cpuacct_percpu_seq_read(struct cgroup *cgroup, struct cftype *cft, return 0; } +static const char *cpuacct_stat_desc[] = { + [CPUACCT_STAT_USER] = "user", + [CPUACCT_STAT_SYSTEM] = "system", +}; + +static int cpuacct_stats_show(struct cgroup *cgrp, struct cftype *cft, + struct cgroup_map_cb *cb) +{ + struct cpuacct *ca = cgroup_ca(cgrp); + int i; + + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) { + s64 val = percpu_counter_read(&ca->cpustat[i]); + val = cputime64_to_clock_t(val); + cb->fill(cb, cpuacct_stat_desc[i], val); + } + return 0; +} + static struct cftype files[] = { { .name = "usage", @@ -9681,7 +9732,10 @@ static struct cftype files[] = { .name = "usage_percpu", .read_seq_string = cpuacct_percpu_seq_read, }, - + { + .name = "stat", + .read_map = cpuacct_stats_show, + }, }; static int cpuacct_populate(struct cgroup_subsys *ss, struct cgroup *cgrp) @@ -9716,6 +9770,27 @@ static void cpuacct_charge(struct task_struct *tsk, u64 cputime) rcu_read_unlock(); } +/* + * Charge the system/user time to the task's accounting group. + */ +static void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val) +{ + struct cpuacct *ca; + + if (unlikely(!cpuacct_subsys.active)) + return; + + rcu_read_lock(); + ca = task_ca(tsk); + + do { + percpu_counter_add(&ca->cpustat[idx], val); + ca = ca->parent; + } while (ca); + rcu_read_unlock(); +} + struct cgroup_subsys cpuacct_subsys = { .name = "cpuacct", .create = cpuacct_create, -- cgit v1.2.3 From 46e0bb9c12f4bab539736f1714cbf16600f681ec Mon Sep 17 00:00:00 2001 From: Gautham R Shenoy Date: Mon, 30 Mar 2009 10:25:20 +0530 Subject: sched: Print sched_group::__cpu_power in sched_domain_debug Impact: extend debug info /proc/sched_debug If the user changes the value of the sched_mc/smt_power_savings sysfs tunable, it'll trigger a rebuilding of the whole sched_domain tree, with the SD_POWERSAVINGS_BALANCE flag set at certain levels. As a result, there would be a change in the __cpu_power of sched_groups in the sched_domain hierarchy. Print the __cpu_power values for each sched_group in sched_domain_debug to help verify this change and correlate it with the change in the load-balancing behavior. Signed-off-by: Gautham R Shenoy Cc: Peter Zijlstra LKML-Reference: <20090330045520.2869.24777.stgit@sofia.in.ibm.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 8d1bdbe8aafc..6234d10c6a79 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6963,7 +6963,8 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpumask_or(groupmask, groupmask, sched_group_cpus(group)); cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); - printk(KERN_CONT " %s", str); + printk(KERN_CONT " %s (__cpu_power = %d)", str, + group->__cpu_power); group = group->next; } while (group != sd->groups); -- cgit v1.2.3 From 633fe795b80693a8198e7d82f66538a72d2bbba2 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 1 Apr 2009 17:47:23 -0700 Subject: timers: add missing kernel-doc Add missing kernel-doc parameter notation and change function name to its new name: Warning(kernel/timer.c:543): No description found for parameter 'name' Warning(kernel/timer.c:543): No description found for parameter 'key' Signed-off-by: Randy Dunlap Cc: akpm Cc: Johannes Berg LKML-Reference: <20090401174723.f0bea0eb.randy.dunlap@oracle.com> Signed-off-by: Ingo Molnar --- kernel/timer.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/timer.c b/kernel/timer.c index 9b77fc9a9ac8..3af9a0bc4292 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -524,10 +524,13 @@ static void __init_timer(struct timer_list *timer) } /** - * init_timer - initialize a timer. + * init_timer_key - initialize a timer * @timer: the timer to be initialized + * @name: name of the timer + * @key: lockdep class key of the fake lock used for tracking timer + * sync lock dependencies * - * init_timer() must be done to a timer prior calling *any* of the + * init_timer_key() must be done to a timer prior calling *any* of the * other timer functions. */ void init_timer(struct timer_list *timer) -- cgit v1.2.3 From cd84a42f315e50edd454c27a3da3951ccd3d735a Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Thu, 2 Apr 2009 14:19:38 -0700 Subject: futex: comment requeue key reference semantics We've tripped over the futex_requeue drop_count refering to key2 instead of key1. The code is actually correct, but is non-intuitive. This patch adds an explicit comment explaining the requeue. Signed-off-by: Darren Hart Cc: Peter Zijlstra Cc: Nick Piggin Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/futex.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 6b50a024bca2..eef8cd26b5e5 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -883,7 +883,12 @@ retry_private: out_unlock: double_unlock_hb(hb1, hb2); - /* drop_futex_key_refs() must be called outside the spinlocks. */ + /* + * drop_futex_key_refs() must be called outside the spinlocks. During + * the requeue we moved futex_q's from the hash bucket at key1 to the + * one at key2 and updated their key pointer. We no longer need to + * hold the references to key1. + */ while (--drop_count >= 0) drop_futex_key_refs(&key1); -- cgit v1.2.3 From b1f77b0581b8fd837acb4a973f7d5496cae6efee Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 13 Mar 2009 03:20:49 +0100 Subject: kmemtrace, rcu: fix linux/rcutree.h and linux/rcuclassic.h dependencies Impact: build fix for all non-x86 architectures We want to remove percpu.h from rcuclassic.h/rcutree.h (for upcoming kmemtrace changes) but that would break the DECLARE_PER_CPU based declarations in these files. Move the quiescent counter management functions to their respective RCU implementation .c files - they were slightly above the inlining limit anyway. Cc: Pekka Enberg Cc: Eduard - Gabriel Munteanu Cc: paulmck@linux.vnet.ibm.com LKML-Reference: <1237898630.25315.83.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- kernel/rcuclassic.c | 23 +++++++++++++++++++++-- kernel/rcutree.c | 28 ++++++++++++++++++++++++---- 2 files changed, 45 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index 654c640a6b9c..0f2b0b311304 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -65,6 +65,7 @@ static struct rcu_ctrlblk rcu_ctrlblk = { .lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock), .cpumask = CPU_BITS_NONE, }; + static struct rcu_ctrlblk rcu_bh_ctrlblk = { .cur = -300, .completed = -300, @@ -73,8 +74,26 @@ static struct rcu_ctrlblk rcu_bh_ctrlblk = { .cpumask = CPU_BITS_NONE, }; -DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L }; -DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L }; +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; diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 97ce31579ec0..a2015edfe167 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -72,11 +72,31 @@ EXPORT_SYMBOL_GPL(rcu_lock_map); .n_force_qs_ngp = 0, \ } -struct rcu_state rcu_state = RCU_STATE_INITIALIZER(rcu_state); -DEFINE_PER_CPU(struct rcu_data, rcu_data); +static struct rcu_state rcu_state = RCU_STATE_INITIALIZER(rcu_state); +static DEFINE_PER_CPU(struct rcu_data, rcu_data); -struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); -DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); +static struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); +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; + rdp->passed_quiesc_completed = rdp->completed; +} + +void rcu_bh_qsctr_inc(int cpu) +{ + struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu); + rdp->passed_quiesc = 1; + rdp->passed_quiesc_completed = rdp->completed; +} #ifdef CONFIG_NO_HZ DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { -- cgit v1.2.3 From 6258c4fb59e77d748f7efc2c137ad420372edd07 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 25 Mar 2009 16:42:24 +0100 Subject: kmemtrace, rcu: fix rcu_tree_trace.c data structure dependencies Impact: cleanup We want to remove rcutree internals from the public rcutree.h file for upcoming kmemtrace changes - but kernel/rcutree_trace.c depends on them. Introduce kernel/rcutree.h for internal definitions. (Probably all the other data types from include/linux/rcutree.h could be moved here too - except rcu_data.) Cc: Pekka Enberg Cc: Eduard - Gabriel Munteanu Cc: paulmck@linux.vnet.ibm.com LKML-Reference: <1237898630.25315.83.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 8 ++++---- kernel/rcutree.h | 10 ++++++++++ kernel/rcutree_trace.c | 2 ++ 3 files changed, 16 insertions(+), 4 deletions(-) create mode 100644 kernel/rcutree.h (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a2015edfe167..7f3266922572 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -72,11 +72,11 @@ EXPORT_SYMBOL_GPL(rcu_lock_map); .n_force_qs_ngp = 0, \ } -static struct rcu_state rcu_state = RCU_STATE_INITIALIZER(rcu_state); -static DEFINE_PER_CPU(struct rcu_data, rcu_data); +struct rcu_state rcu_state = RCU_STATE_INITIALIZER(rcu_state); +DEFINE_PER_CPU(struct rcu_data, rcu_data); -static struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); -static DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); +struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); +DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); /* * Increment the quiescent state counter. diff --git a/kernel/rcutree.h b/kernel/rcutree.h new file mode 100644 index 000000000000..5e872bbf07f5 --- /dev/null +++ b/kernel/rcutree.h @@ -0,0 +1,10 @@ + +/* + * RCU implementation internal declarations: + */ +extern struct rcu_state rcu_state; +DECLARE_PER_CPU(struct rcu_data, rcu_data); + +extern struct rcu_state rcu_bh_state; +DECLARE_PER_CPU(struct rcu_data, rcu_bh_data); + diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index d6db3e837826..4ee954f6a8d5 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -43,6 +43,8 @@ #include #include +#include "rcutree.h" + static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) { if (!rdp->beenonline) -- cgit v1.2.3 From a979241c532f07c201fe94e0a632107268f02578 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 25 Mar 2009 16:42:24 +0100 Subject: kmemtrace, rcu: fix rcupreempt.c data structure dependencies Impact: cleanup We want to remove percpu.h from rcupreempt.h, but if we do that the percpu primitives there wont build anymore. Move them to the .c file instead. Cc: Pekka Enberg Cc: Eduard - Gabriel Munteanu Cc: paulmck@linux.vnet.ibm.com LKML-Reference: <1237898630.25315.83.camel@penberg-laptop> Signed-off-by: Ingo Molnar --- kernel/rcupreempt.c | 48 ++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 44 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index 5d59e850fb71..ce97a4df64d3 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -147,7 +147,51 @@ struct rcu_ctrlblk { wait_queue_head_t sched_wq; /* Place for rcu_sched to sleep. */ }; +struct rcu_dyntick_sched { + int dynticks; + int dynticks_snap; + int sched_qs; + int sched_qs_snap; + int sched_dynticks_snap; +}; + +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_dyntick_sched, rcu_dyntick_sched) = { + .dynticks = 1, +}; + +void rcu_qsctr_inc(int cpu) +{ + struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); + + rdssp->sched_qs++; +} + +#ifdef CONFIG_NO_HZ + +void rcu_enter_nohz(void) +{ + static DEFINE_RATELIMIT_STATE(rs, 10 * HZ, 1); + + smp_mb(); /* CPUs seeing ++ must see prior RCU read-side crit sects */ + __get_cpu_var(rcu_dyntick_sched).dynticks++; + WARN_ON_RATELIMIT(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1, &rs); +} + +void rcu_exit_nohz(void) +{ + static DEFINE_RATELIMIT_STATE(rs, 10 * HZ, 1); + + __get_cpu_var(rcu_dyntick_sched).dynticks++; + smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */ + WARN_ON_RATELIMIT(!(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1), + &rs); +} + +#endif /* CONFIG_NO_HZ */ + + static DEFINE_PER_CPU(struct rcu_data, rcu_data); + static struct rcu_ctrlblk rcu_ctrlblk = { .fliplock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.fliplock), .completed = 0, @@ -427,10 +471,6 @@ static void __rcu_advance_callbacks(struct rcu_data *rdp) } } -DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_dyntick_sched, rcu_dyntick_sched) = { - .dynticks = 1, -}; - #ifdef CONFIG_NO_HZ static DEFINE_PER_CPU(int, rcu_update_flag); -- cgit v1.2.3 From ca2b84cb3c4a0d4d2143b46ec072cdff5d1b3b87 Mon Sep 17 00:00:00 2001 From: Eduard - Gabriel Munteanu Date: Mon, 23 Mar 2009 15:12:24 +0200 Subject: kmemtrace: use tracepoints kmemtrace now uses tracepoints instead of markers. We no longer need to use format specifiers to pass arguments. Signed-off-by: Eduard - Gabriel Munteanu [ folded: Use the new TP_PROTO and TP_ARGS to fix the build. ] [ folded: fix build when CONFIG_KMEMTRACE is disabled. ] [ folded: define tracepoints when CONFIG_TRACEPOINTS is enabled. ] Signed-off-by: Pekka Enberg LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/trace/kmemtrace.c | 206 +++++++++++++++++++++++++++++++++-------------- kernel/trace/trace.h | 6 ++ 2 files changed, 151 insertions(+), 61 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index ae201b3eda89..4f7b5db5dd06 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include "trace.h" @@ -29,10 +30,150 @@ static struct tracer_flags kmem_tracer_flags = { .opts = kmem_opts }; - -static bool kmem_tracing_enabled __read_mostly; static struct trace_array *kmemtrace_array; +/* Trace allocations */ +static inline void kmemtrace_alloc(enum kmemtrace_type_id type_id, + unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + struct ring_buffer_event *event; + struct kmemtrace_alloc_entry *entry; + struct trace_array *tr = kmemtrace_array; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); + if (!event) + return; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, 0, 0); + + entry->ent.type = TRACE_KMEM_ALLOC; + entry->call_site = call_site; + entry->ptr = ptr; + entry->bytes_req = bytes_req; + entry->bytes_alloc = bytes_alloc; + entry->gfp_flags = gfp_flags; + entry->node = node; + + ring_buffer_unlock_commit(tr->buffer, event); + + trace_wake_up(); +} + +static inline void kmemtrace_free(enum kmemtrace_type_id type_id, + unsigned long call_site, + const void *ptr) +{ + struct ring_buffer_event *event; + struct kmemtrace_free_entry *entry; + struct trace_array *tr = kmemtrace_array; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); + if (!event) + return; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, 0, 0); + + entry->ent.type = TRACE_KMEM_FREE; + entry->type_id = type_id; + entry->call_site = call_site; + entry->ptr = ptr; + + ring_buffer_unlock_commit(tr->buffer, event); + + trace_wake_up(); +} + +static void kmemtrace_kmalloc(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_KMALLOC, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, -1); +} + +static void kmemtrace_kmem_cache_alloc(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_CACHE, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, -1); +} + +static void kmemtrace_kmalloc_node(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_KMALLOC, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, node); +} + +static void kmemtrace_kmem_cache_alloc_node(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_CACHE, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, node); +} + +static void kmemtrace_kfree(unsigned long call_site, const void *ptr) +{ + kmemtrace_free(KMEMTRACE_TYPE_KMALLOC, call_site, ptr); +} + +static void kmemtrace_kmem_cache_free(unsigned long call_site, const void *ptr) +{ + kmemtrace_free(KMEMTRACE_TYPE_CACHE, call_site, ptr); +} + +static int kmemtrace_start_probes(void) +{ + int err; + + err = register_trace_kmalloc(kmemtrace_kmalloc); + if (err) + return err; + err = register_trace_kmem_cache_alloc(kmemtrace_kmem_cache_alloc); + if (err) + return err; + err = register_trace_kmalloc_node(kmemtrace_kmalloc_node); + if (err) + return err; + err = register_trace_kmem_cache_alloc_node(kmemtrace_kmem_cache_alloc_node); + if (err) + return err; + err = register_trace_kfree(kmemtrace_kfree); + if (err) + return err; + err = register_trace_kmem_cache_free(kmemtrace_kmem_cache_free); + + return err; +} + +static void kmemtrace_stop_probes(void) +{ + unregister_trace_kmalloc(kmemtrace_kmalloc); + unregister_trace_kmem_cache_alloc(kmemtrace_kmem_cache_alloc); + unregister_trace_kmalloc_node(kmemtrace_kmalloc_node); + unregister_trace_kmem_cache_alloc_node(kmemtrace_kmem_cache_alloc_node); + unregister_trace_kfree(kmemtrace_kfree); + unregister_trace_kmem_cache_free(kmemtrace_kmem_cache_free); +} + static int kmem_trace_init(struct trace_array *tr) { int cpu; @@ -41,14 +182,14 @@ static int kmem_trace_init(struct trace_array *tr) for_each_cpu_mask(cpu, cpu_possible_map) tracing_reset(tr, cpu); - kmem_tracing_enabled = true; + kmemtrace_start_probes(); return 0; } static void kmem_trace_reset(struct trace_array *tr) { - kmem_tracing_enabled = false; + kmemtrace_stop_probes(); } static void kmemtrace_headers(struct seq_file *s) @@ -260,63 +401,6 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) } } -/* Trace allocations */ -void kmemtrace_mark_alloc_node(enum kmemtrace_type_id type_id, - unsigned long call_site, - const void *ptr, - size_t bytes_req, - size_t bytes_alloc, - gfp_t gfp_flags, - int node) -{ - struct ring_buffer_event *event; - struct kmemtrace_alloc_entry *entry; - struct trace_array *tr = kmemtrace_array; - - if (!kmem_tracing_enabled) - return; - - event = trace_buffer_lock_reserve(tr, TRACE_KMEM_ALLOC, - sizeof(*entry), 0, 0); - if (!event) - return; - entry = ring_buffer_event_data(event); - - entry->call_site = call_site; - entry->ptr = ptr; - entry->bytes_req = bytes_req; - entry->bytes_alloc = bytes_alloc; - entry->gfp_flags = gfp_flags; - entry->node = node; - - trace_buffer_unlock_commit(tr, event, 0, 0); -} -EXPORT_SYMBOL(kmemtrace_mark_alloc_node); - -void kmemtrace_mark_free(enum kmemtrace_type_id type_id, - unsigned long call_site, - const void *ptr) -{ - struct ring_buffer_event *event; - struct kmemtrace_free_entry *entry; - struct trace_array *tr = kmemtrace_array; - - if (!kmem_tracing_enabled) - return; - - event = trace_buffer_lock_reserve(tr, TRACE_KMEM_FREE, - sizeof(*entry), 0, 0); - if (!event) - return; - entry = ring_buffer_event_data(event); - entry->type_id = type_id; - entry->call_site = call_site; - entry->ptr = ptr; - - trace_buffer_unlock_commit(tr, event, 0, 0); -} -EXPORT_SYMBOL(kmemtrace_mark_free); - static struct tracer kmem_tracer __read_mostly = { .name = "kmemtrace", .init = kmem_trace_init, diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index cb0ce3fc36d3..cbc168f1e43d 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -182,6 +182,12 @@ struct trace_power { struct power_trace state_data; }; +enum kmemtrace_type_id { + KMEMTRACE_TYPE_KMALLOC = 0, /* kmalloc() or kfree(). */ + KMEMTRACE_TYPE_CACHE, /* kmem_cache_*(). */ + KMEMTRACE_TYPE_PAGES, /* __get_free_pages() and friends. */ +}; + struct kmemtrace_alloc_entry { struct trace_entry ent; enum kmemtrace_type_id type_id; -- cgit v1.2.3 From da2635a9854423b4aa3a5f0e4e6efcc39ac99004 Mon Sep 17 00:00:00 2001 From: Eduard - Gabriel Munteanu Date: Mon, 23 Mar 2009 15:12:25 +0200 Subject: kmemtrace: kmemtrace_alloc() must fill type_id Impact: fix trace output kmemtrace_alloc() was not filling type_id, which allowed garbage to make it into tracing data. Signed-off-by: Eduard - Gabriel Munteanu LKML-Reference: <284dba2732a144849d5aa82258fe0de2ad8dcb0b.1237813499.git.eduard.munteanu@linux360.ro> Signed-off-by: Ingo Molnar --- kernel/trace/kmemtrace.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index 4f7b5db5dd06..ae259f04ee39 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -52,6 +52,7 @@ static inline void kmemtrace_alloc(enum kmemtrace_type_id type_id, tracing_generic_entry_update(&entry->ent, 0, 0); entry->ent.type = TRACE_KMEM_ALLOC; + entry->type_id = type_id; entry->call_site = call_site; entry->ptr = ptr; entry->bytes_req = bytes_req; -- cgit v1.2.3 From 42af9054c0eeed09ec58d13ec8bf52d225ebcfcc Mon Sep 17 00:00:00 2001 From: Eduard - Gabriel Munteanu Date: Mon, 23 Mar 2009 15:12:26 +0200 Subject: kmemtrace: restore original tracing data binary format, improve ABI When kmemtrace was ported to ftrace, the marker strings were taken as an indication of how the traced data was being exposed to the userspace. However, the actual format had been binary, not text. This restores the original binary format, while also adding an origin CPU field (since ftrace doesn't expose the data per-CPU to userspace), and re-adding the timestamp field. It also drops arch-independent field sizing where it didn't make sense, so pointers won't always be 64 bits wide like they used to. Signed-off-by: Eduard - Gabriel Munteanu LKML-Reference: <161be9ca8a27b432c4a6ab79f47788c4521652ae.1237813499.git.eduard.munteanu@linux360.ro> Signed-off-by: Ingo Molnar --- kernel/trace/kmemtrace.c | 82 ++++++++++++++++++++++++++++++++++-------------- 1 file changed, 58 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index ae259f04ee39..d8c2d0c91b4c 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -208,47 +208,81 @@ static void kmemtrace_headers(struct seq_file *s) } /* - * The two following functions give the original output from kmemtrace, - * or something close to....perhaps they need some missing things + * The following functions give the original output from kmemtrace, + * plus the origin CPU, since reordering occurs in-kernel now. */ + +#define KMEMTRACE_USER_ALLOC 0 +#define KMEMTRACE_USER_FREE 1 + +struct kmemtrace_user_event { + u8 event_id; + u8 type_id; + u16 event_size; + u32 cpu; + u64 timestamp; + unsigned long call_site; + unsigned long ptr; +}; + +struct kmemtrace_user_event_alloc { + size_t bytes_req; + size_t bytes_alloc; + unsigned gfp_flags; + int node; +}; + static enum print_line_t -kmemtrace_print_alloc_original(struct trace_iterator *iter, - struct kmemtrace_alloc_entry *entry) +kmemtrace_print_alloc_user(struct trace_iterator *iter, + struct kmemtrace_alloc_entry *entry) { struct trace_seq *s = &iter->seq; - int ret; - - /* Taken from the old linux/kmemtrace.h */ - ret = trace_seq_printf(s, "type_id %d call_site %lu ptr %lu " - "bytes_req %lu bytes_alloc %lu gfp_flags %lu node %d\n", - entry->type_id, entry->call_site, (unsigned long) entry->ptr, - (unsigned long) entry->bytes_req, (unsigned long) entry->bytes_alloc, - (unsigned long) entry->gfp_flags, entry->node); + struct kmemtrace_user_event *ev; + struct kmemtrace_user_event_alloc *ev_alloc; - if (!ret) + ev = trace_seq_reserve(s, sizeof(*ev)); + if (!ev) + return TRACE_TYPE_PARTIAL_LINE; + ev->event_id = KMEMTRACE_USER_ALLOC; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev) + sizeof(*ev_alloc); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long) entry->ptr; + + ev_alloc = trace_seq_reserve(s, sizeof(*ev_alloc)); + if (!ev_alloc) return TRACE_TYPE_PARTIAL_LINE; + ev_alloc->bytes_req = entry->bytes_req; + ev_alloc->bytes_alloc = entry->bytes_alloc; + ev_alloc->gfp_flags = entry->gfp_flags; + ev_alloc->node = entry->node; return TRACE_TYPE_HANDLED; } static enum print_line_t -kmemtrace_print_free_original(struct trace_iterator *iter, - struct kmemtrace_free_entry *entry) +kmemtrace_print_free_user(struct trace_iterator *iter, + struct kmemtrace_free_entry *entry) { struct trace_seq *s = &iter->seq; - int ret; + struct kmemtrace_user_event *ev; - /* Taken from the old linux/kmemtrace.h */ - ret = trace_seq_printf(s, "type_id %d call_site %lu ptr %lu\n", - entry->type_id, entry->call_site, (unsigned long) entry->ptr); - - if (!ret) + ev = trace_seq_reserve(s, sizeof(*ev)); + if (!ev) return TRACE_TYPE_PARTIAL_LINE; + ev->event_id = KMEMTRACE_USER_FREE; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long) entry->ptr; return TRACE_TYPE_HANDLED; } - /* The two other following provide a more minimalistic output */ static enum print_line_t kmemtrace_print_alloc_compress(struct trace_iterator *iter, @@ -385,7 +419,7 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) if (kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL) return kmemtrace_print_alloc_compress(iter, field); else - return kmemtrace_print_alloc_original(iter, field); + return kmemtrace_print_alloc_user(iter, field); } case TRACE_KMEM_FREE: { @@ -394,7 +428,7 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) if (kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL) return kmemtrace_print_free_compress(iter, field); else - return kmemtrace_print_free_original(iter, field); + return kmemtrace_print_free_user(iter, field); } default: -- cgit v1.2.3 From c826e3cd0c931d60d548f2468122da570d145556 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 23 Mar 2009 16:14:13 +0100 Subject: kmemtrace: small cleanups Cc: Eduard - Gabriel Munteanu LKML-Reference: <161be9ca8a27b432c4a6ab79f47788c4521652ae.1237813499.git.eduard.munteanu@linux360.ro> Signed-off-by: Ingo Molnar --- kernel/trace/kmemtrace.c | 128 +++++++++++++++++++++++++---------------------- 1 file changed, 67 insertions(+), 61 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index d8c2d0c91b4c..5011f4d91e37 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -6,15 +6,16 @@ * Copyright (C) 2008 Frederic Weisbecker */ -#include +#include +#include #include +#include #include -#include -#include + #include -#include "trace.h" #include "trace_output.h" +#include "trace.h" /* Select an alternative, minimalistic output than the original one */ #define TRACE_KMEM_OPT_MINIMAL 0x1 @@ -26,8 +27,8 @@ static struct tracer_opt kmem_opts[] = { }; static struct tracer_flags kmem_tracer_flags = { - .val = 0, - .opts = kmem_opts + .val = 0, + .opts = kmem_opts }; static struct trace_array *kmemtrace_array; @@ -41,24 +42,25 @@ static inline void kmemtrace_alloc(enum kmemtrace_type_id type_id, gfp_t gfp_flags, int node) { - struct ring_buffer_event *event; - struct kmemtrace_alloc_entry *entry; struct trace_array *tr = kmemtrace_array; + struct kmemtrace_alloc_entry *entry; + struct ring_buffer_event *event; event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); if (!event) return; - entry = ring_buffer_event_data(event); + + entry = ring_buffer_event_data(event); tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_KMEM_ALLOC; - entry->type_id = type_id; - entry->call_site = call_site; - entry->ptr = ptr; - entry->bytes_req = bytes_req; - entry->bytes_alloc = bytes_alloc; - entry->gfp_flags = gfp_flags; - entry->node = node; + entry->ent.type = TRACE_KMEM_ALLOC; + entry->type_id = type_id; + entry->call_site = call_site; + entry->ptr = ptr; + entry->bytes_req = bytes_req; + entry->bytes_alloc = bytes_alloc; + entry->gfp_flags = gfp_flags; + entry->node = node; ring_buffer_unlock_commit(tr->buffer, event); @@ -69,9 +71,9 @@ static inline void kmemtrace_free(enum kmemtrace_type_id type_id, unsigned long call_site, const void *ptr) { - struct ring_buffer_event *event; - struct kmemtrace_free_entry *entry; struct trace_array *tr = kmemtrace_array; + struct kmemtrace_free_entry *entry; + struct ring_buffer_event *event; event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); if (!event) @@ -79,10 +81,10 @@ static inline void kmemtrace_free(enum kmemtrace_type_id type_id, entry = ring_buffer_event_data(event); tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_KMEM_FREE; - entry->type_id = type_id; - entry->call_site = call_site; - entry->ptr = ptr; + entry->ent.type = TRACE_KMEM_FREE; + entry->type_id = type_id; + entry->call_site = call_site; + entry->ptr = ptr; ring_buffer_unlock_commit(tr->buffer, event); @@ -216,48 +218,50 @@ static void kmemtrace_headers(struct seq_file *s) #define KMEMTRACE_USER_FREE 1 struct kmemtrace_user_event { - u8 event_id; - u8 type_id; - u16 event_size; - u32 cpu; - u64 timestamp; - unsigned long call_site; - unsigned long ptr; + u8 event_id; + u8 type_id; + u16 event_size; + u32 cpu; + u64 timestamp; + unsigned long call_site; + unsigned long ptr; }; struct kmemtrace_user_event_alloc { - size_t bytes_req; - size_t bytes_alloc; - unsigned gfp_flags; - int node; + size_t bytes_req; + size_t bytes_alloc; + unsigned gfp_flags; + int node; }; static enum print_line_t kmemtrace_print_alloc_user(struct trace_iterator *iter, struct kmemtrace_alloc_entry *entry) { + struct kmemtrace_user_event_alloc *ev_alloc; struct trace_seq *s = &iter->seq; struct kmemtrace_user_event *ev; - struct kmemtrace_user_event_alloc *ev_alloc; ev = trace_seq_reserve(s, sizeof(*ev)); if (!ev) return TRACE_TYPE_PARTIAL_LINE; - ev->event_id = KMEMTRACE_USER_ALLOC; - ev->type_id = entry->type_id; - ev->event_size = sizeof(*ev) + sizeof(*ev_alloc); - ev->cpu = iter->cpu; - ev->timestamp = iter->ts; - ev->call_site = entry->call_site; - ev->ptr = (unsigned long) entry->ptr; + + ev->event_id = KMEMTRACE_USER_ALLOC; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev) + sizeof(*ev_alloc); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long)entry->ptr; ev_alloc = trace_seq_reserve(s, sizeof(*ev_alloc)); if (!ev_alloc) return TRACE_TYPE_PARTIAL_LINE; - ev_alloc->bytes_req = entry->bytes_req; - ev_alloc->bytes_alloc = entry->bytes_alloc; - ev_alloc->gfp_flags = entry->gfp_flags; - ev_alloc->node = entry->node; + + ev_alloc->bytes_req = entry->bytes_req; + ev_alloc->bytes_alloc = entry->bytes_alloc; + ev_alloc->gfp_flags = entry->gfp_flags; + ev_alloc->node = entry->node; return TRACE_TYPE_HANDLED; } @@ -272,13 +276,14 @@ kmemtrace_print_free_user(struct trace_iterator *iter, ev = trace_seq_reserve(s, sizeof(*ev)); if (!ev) return TRACE_TYPE_PARTIAL_LINE; - ev->event_id = KMEMTRACE_USER_FREE; - ev->type_id = entry->type_id; - ev->event_size = sizeof(*ev); - ev->cpu = iter->cpu; - ev->timestamp = iter->ts; - ev->call_site = entry->call_site; - ev->ptr = (unsigned long) entry->ptr; + + ev->event_id = KMEMTRACE_USER_FREE; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long)entry->ptr; return TRACE_TYPE_HANDLED; } @@ -354,7 +359,7 @@ kmemtrace_print_alloc_compress(struct trace_iterator *iter, static enum print_line_t kmemtrace_print_free_compress(struct trace_iterator *iter, - struct kmemtrace_free_entry *entry) + struct kmemtrace_free_entry *entry) { struct trace_seq *s = &iter->seq; int ret; @@ -415,6 +420,7 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) 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); @@ -424,6 +430,7 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) 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); @@ -437,12 +444,12 @@ static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) } static struct tracer kmem_tracer __read_mostly = { - .name = "kmemtrace", - .init = kmem_trace_init, - .reset = kmem_trace_reset, - .print_line = kmemtrace_print_line, - .print_header = kmemtrace_headers, - .flags = &kmem_tracer_flags + .name = "kmemtrace", + .init = kmem_trace_init, + .reset = kmem_trace_reset, + .print_line = kmemtrace_print_line, + .print_header = kmemtrace_headers, + .flags = &kmem_tracer_flags }; void kmemtrace_init(void) @@ -454,5 +461,4 @@ static int __init init_kmem_tracer(void) { return register_tracer(&kmem_tracer); } - device_initcall(init_kmem_tracer); -- cgit v1.2.3 From a4b3ada83d06554d307dd54abdc62b2e5648264a Mon Sep 17 00:00:00 2001 From: Carl Henrik Lunde Date: Fri, 3 Apr 2009 14:27:15 +0200 Subject: blktrace: NUL-terminate user space messages Impact: fix corrupted blkparse output Make sure messages from user space are NUL-terminated strings, otherwise we could dump random memory to the block trace file. Additionally, I've limited the message to BLK_TN_MAX_MSG-1 characters, because the last character would be stripped by vscnprintf anyway. Signed-off-by: Carl Henrik Lunde Cc: Li Zefan Cc: Arnaldo Carvalho de Melo Cc: "Alan D. Brunelle" Cc: Steven Rostedt LKML-Reference: <20090403122714.GT5178@kernel.dk> Signed-off-by: Ingo Molnar --- kernel/trace/blktrace.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 947c5b3f90c4..a400b861fad3 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -327,10 +327,10 @@ static ssize_t blk_msg_write(struct file *filp, const char __user *buffer, char *msg; struct blk_trace *bt; - if (count > BLK_TN_MAX_MSG) + if (count > BLK_TN_MAX_MSG - 1) return -EINVAL; - msg = kmalloc(count, GFP_KERNEL); + msg = kmalloc(count + 1, GFP_KERNEL); if (msg == NULL) return -ENOMEM; @@ -339,6 +339,7 @@ static ssize_t blk_msg_write(struct file *filp, const char __user *buffer, return -EFAULT; } + msg[count] = '\0'; bt = filp->private_data; __trace_note_message(bt, "%s", msg); kfree(msg); -- cgit v1.2.3 From 7635b03adf3d7b84da7649b81efa91e6ebf11b85 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Fri, 3 Apr 2009 15:31:34 +0800 Subject: blktrace: small cleanup in blk_msg_write() Signed-off-by: Li Zefan Cc: Arnaldo Carvalho de Melo Cc: "Alan D. Brunelle" Cc: Jens Axboe LKML-Reference: <49D5BB56.7000807@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/blktrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index a400b861fad3..73d7860b72e2 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -327,7 +327,7 @@ static ssize_t blk_msg_write(struct file *filp, const char __user *buffer, char *msg; struct blk_trace *bt; - if (count > BLK_TN_MAX_MSG - 1) + if (count >= BLK_TN_MAX_MSG) return -EINVAL; msg = kmalloc(count + 1, GFP_KERNEL); -- cgit v1.2.3 From e2494e1b42ebac402324105d57646489d19e2b01 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 2 Apr 2009 13:43:26 +0800 Subject: blktrace: fix pdu_len when tracing packet command requests Impact: output all of packet commands - not just the first 4 / 8 bytes Since commit d7e3c3249ef23b4617393c69fe464765b4ff1645 ("block: add large command support"), struct request->cmd has been changed from unsinged char cmd[BLK_MAX_CDB] to unsigned char *cmd. v1 -> v2: by: FUJITA Tomonori - make sure rq->cmd_len is always intialized, and then we can use rq->cmd_len instead of BLK_MAX_CDB. Signed-off-by: Li Zefan Acked-by: FUJITA Tomonori Cc: Arnaldo Carvalho de Melo Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Jens Axboe LKML-Reference: <49D4507E.2060602@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/blktrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 73d7860b72e2..b32ff446c3fb 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -643,7 +643,7 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, if (blk_pc_request(rq)) { what |= BLK_TC_ACT(BLK_TC_PC); __blk_add_trace(bt, 0, rq->data_len, rw, what, rq->errors, - sizeof(rq->cmd), rq->cmd); + rq->cmd_len, rq->cmd); } else { what |= BLK_TC_ACT(BLK_TC_FS); __blk_add_trace(bt, rq->hard_sector, rq->hard_nr_sectors << 9, -- cgit v1.2.3 From 9756b15e1b58453a6fd54b85c1ad8515209e10bb Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 30 Mar 2009 20:37:20 -0700 Subject: irq: fix cpumask memory leak on offstack cpumask kernels Need to free the old cpumask for affinity and pending_mask. Signed-off-by: Yinghai Lu Acked-by: Rusty Russell LKML-Reference: <49D18FF0.50707@kernel.org> Signed-off-by: Ingo Molnar --- kernel/irq/numa_migrate.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 243d6121e50e..44bbdcbaf8d2 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -54,6 +54,7 @@ static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc, static void free_one_irq_desc(struct irq_desc *old_desc, struct irq_desc *desc) { free_kstat_irqs(old_desc, desc); + free_desc_masks(old_desc, desc); arch_free_chip_data(old_desc, desc); } -- cgit v1.2.3 From cd5f9a4c3199b090e91ea0064cb110985ba54814 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 6 Apr 2009 13:38:46 -0700 Subject: kernel/sysctl.c: avoid annoying warnings Some of the limit constants are used only depending on some complex configuration dependencies, yet it's not worth making the simple variables depend on those configuration details. Just mark them as perhaps not being unused, and avoid the warning. Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 82350f8f04f6..b125e3387568 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -97,8 +97,8 @@ static int neg_one = -1; #endif static int zero; -static int one = 1; -static int two = 2; +static int __maybe_unused one = 1; +static int __maybe_unused two = 2; static unsigned long one_ul = 1; static int one_hundred = 100; -- cgit v1.2.3 From 432870dab85a2f69dc417022646cb9a70acf7f94 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 6 Apr 2009 16:16:02 +0200 Subject: exit_notify: kill the wrong capable(CAP_KILL) check The CAP_KILL check in exit_notify() looks just wrong, kill it. Whatever logic we have to reset ->exit_signal, the malicious user can bypass it if it execs the setuid application before exiting. Signed-off-by: Oleg Nesterov Acked-by: Serge Hallyn Acked-by: Roland McGrath Signed-off-by: Linus Torvalds --- kernel/exit.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/exit.c b/kernel/exit.c index 6686ed1e4aa3..32cbf2607cb0 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -837,8 +837,7 @@ static void exit_notify(struct task_struct *tsk, int group_dead) */ if (tsk->exit_signal != SIGCHLD && !task_detached(tsk) && (tsk->parent_exec_id != tsk->real_parent->self_exec_id || - tsk->self_exec_id != tsk->parent_exec_id) && - !capable(CAP_KILL)) + tsk->self_exec_id != tsk->parent_exec_id)) tsk->exit_signal = SIGCHLD; signal = tracehook_notify_death(tsk, &cookie, group_dead); -- cgit v1.2.3 From 2e45e77787c9d0720b046eb69856edf43b17e33e Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 7 Apr 2009 17:12:43 +0930 Subject: Revert "module: remove the SHF_ALLOC flag on the __versions section." This reverts commit 9cb610d8e35fe3ec95a2fe2030b02f85aeea83c1. This was an impressively stupid patch. Firstly, we reset the SHF_ALLOC flag lower down in the same function, so the patch was useless. Even better, find_sec() ignores sections with SHF_ALLOC not set, so it breaks CONFIG_MODVERSIONS=y with CONFIG_MODULE_FORCE_LOAD=n, which refuses to load the module since it can't find the __versions section. Signed-off-by: Rusty Russell --- kernel/module.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index c268a771595c..05f014efa32c 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1952,9 +1952,6 @@ static noinline struct module *load_module(void __user *umod, if (strstarts(secstrings+sechdrs[i].sh_name, ".exit")) sechdrs[i].sh_flags &= ~(unsigned long)SHF_ALLOC; #endif - /* Don't keep __versions around; it's just for loading. */ - if (strcmp(secstrings + sechdrs[i].sh_name, "__versions") == 0) - sechdrs[i].sh_flags &= ~(unsigned long)SHF_ALLOC; } modindex = find_sec(hdr, sechdrs, secstrings, -- cgit v1.2.3 From 301fd748e2c81e78e74edbc694a64caa7b95dda2 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 3 Apr 2009 11:12:23 -0400 Subject: tracing: remove CALLER_ADDR2 from wakeup tracer Maneesh Soni was getting a crash when running the wakeup tracer. We debugged it down to the recording of the function with the CALLER_ADDR2 macro. This is used to get the location of the caller to schedule. But the problem comes when schedule is called by assmebly. In the case that Maneesh had, retint_careful would call schedule. But retint_careful does not set up a proper frame pointer. CALLER_ADDR2 is defined as __builtin_return_address(2). This produces the following assembly in the wakeup tracer code. mov 0x0(%rbp),%rcx <--- get the frame pointer of the caller mov %r14d,%r8d mov 0xf2de8e(%rip),%rdi mov 0x8(%rcx),%rsi <-- this is __builtin_return_address(1) mov 0x28(%rdi,%rax,8),%rbx mov (%rcx),%rax <-- get the frame pointer of the caller's caller mov %r12,%rcx mov 0x8(%rax),%rdx <-- this is __builtin_return_address(2) At the reading of 0x8(%rax) Maneesh's machine would take a fault. The reason is that retint_careful did not set up the return address and the content of %rax here was zero. To verify this, I sent Maneesh a patch to create a frame pointer in retint_careful. He ran the test again but this time he would take the same type of fault from sysret_careful. The retint_careful was no longer an issue, but there are other callers that still have issues. Instead of adding frame pointers for all callers to schedule (in possibly all archs), it is much safer to simply not use CALLER_ADDR2. This loses out on knowing what called schedule, but the function tracer will help there if needed. Reported-by: Maneesh Soni Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace_sched_wakeup.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 3c5ad6b2ec84..5bc00e8f153e 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -154,7 +154,7 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, if (unlikely(!tracer_enabled || next != wakeup_task)) goto out_unlock; - trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); + trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); /* @@ -257,6 +257,12 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) data = wakeup_trace->data[wakeup_cpu]; data->preempt_timestamp = ftrace_now(cpu); tracing_sched_wakeup_trace(wakeup_trace, p, current, flags, pc); + + /* + * We must be careful in using CALLER_ADDR2. But since wake_up + * is not called by an assembly function (where as schedule is) + * it should be safe to use it here. + */ trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); out_locked: -- cgit v1.2.3 From cf8e3474654f20433aab9aa35826d43b5f245008 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Mon, 30 Mar 2009 13:48:00 +0800 Subject: tracing: fix incorrect return type of ns2usecs() Impact: fix time output bug in 32bits system ns2usecs() returns 'long', it's incorrect. (In i386) ... -0 [000] 521.442100: _spin_lock <-tick_do_update_jiffies64 -0 [000] 521.442101: do_timer <-tick_do_update_jiffies64 -0 [000] 521.442102: update_wall_time <-do_timer -0 [000] 521.442102: update_xtime_cache <-update_wall_time .... (It always print the time less than 2200 seconds besides ...) Because 'long' is 32bits in i386. ( (1<<31) useconds is about 2200 seconds) ... -0 [001] 4154502640.134759: rcu_bh_qsctr_inc <-__do_softirq -0 [001] 4154502640.134760: _local_bh_enable <-__do_softirq -0 [001] 4154502640.134761: idle_cpu <-irq_exit ... (very large value) Because 'long' is a signed type and it is 32bits in i386. Changes in v2: return 'unsigned long long' instead of 'cycle_t' Signed-off-by: Lai Jiangshan LKML-Reference: <49D05D10.4030009@cn.fujitsu.com> Reported-by: Li Zefan Acked-by: Frederic Weisbecker Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 3 +-- kernel/trace/trace.h | 2 +- kernel/trace/trace_output.c | 2 +- 3 files changed, 3 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index a0174a40c563..457dd8c97e0d 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -147,8 +147,7 @@ static int __init set_ftrace_dump_on_oops(char *str) } __setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops); -long -ns2usecs(cycle_t nsec) +unsigned long long ns2usecs(cycle_t nsec) { nsec += 500; do_div(nsec, 1000); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index cb0ce3fc36d3..0d81a4a2a4a5 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -596,7 +596,7 @@ extern int trace_selftest_startup_branch(struct tracer *trace, #endif /* CONFIG_FTRACE_STARTUP_TEST */ extern void *head_page(struct trace_array_cpu *data); -extern long ns2usecs(cycle_t nsec); +extern unsigned long long ns2usecs(cycle_t nsec); extern int trace_vbprintk(unsigned long ip, const char *fmt, va_list args); extern int diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index d72b9a63b247..64b54a59c55b 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -423,7 +423,7 @@ int trace_print_lat_context(struct trace_iterator *iter) trace_find_cmdline(entry->pid, comm); - ret = trace_seq_printf(s, "%16s %5d %3d %d %08x %08lx [%08lx]" + ret = trace_seq_printf(s, "%16s %5d %3d %d %08x %08lx [%08llx]" " %ld.%03ldms (+%ld.%03ldms): ", comm, entry->pid, iter->cpu, entry->flags, entry->preempt_count, iter->idx, -- cgit v1.2.3 From 5f0c6c03c5fee91c02c696bc9bf4c0d41392abe7 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Fri, 27 Mar 2009 14:22:10 +0100 Subject: tracing/ftrace: fix missing include string.h Building a kernel with tracing can raise the following warning on tip/master: kernel/trace/trace.c:1249: error: implicit declaration of function 'vbin_printf' We are missing an include to string.h Reported-by: Ingo Molnar Signed-off-by: Frederic Weisbecker LKML-Reference: <1238160130-7437-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 457dd8c97e0d..2230b46f9e1c 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From 8bcae09b93e7f96f700b6bb372c2b3f2b36636dc Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Tue, 31 Mar 2009 15:24:51 +0800 Subject: ftrace: Add check of sched_stopped for probe_sched_wakeup The wakeup tracing in sched_switch does not stop when a user disables tracing. This is because the probe_sched_wakeup() is missing the check to prevent the wakeup from being traced. Signed-off-by: Zhao Lei LKML-Reference: <49D1C543.3010307@cn.fujitsu.com> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace_sched_switch.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index de35f200abd3..9117cea6f1ae 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -62,6 +62,9 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success) pc = preempt_count(); tracing_record_cmdline(current); + if (sched_stopped) + return; + local_irq_save(flags); cpu = raw_smp_processor_id(); data = ctx_trace->data[cpu]; -- cgit v1.2.3 From b0dfa978c7a1699fb3506fbfcba0b6a5c4bd17ae Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 1 Apr 2009 22:53:08 +0200 Subject: tracing/ftrace: alloc the started cpumask for the trace file Impact: fix a crash while cat trace file Currently we are using a cpumask to remind each cpu where a trace occured. It lets us notice the user that a cpu just had its first trace. But on latest -tip we have the following crash once we cat the trace file: IP: [] print_trace_fmt+0x45/0xe7 *pde = 00000000 Oops: 0000 [#1] PREEMPT SMP last sysfs file: /sys/class/net/eth0/carrier Pid: 3897, comm: cat Not tainted (2.6.29-tip-02825-g0f22972-dirty #81) EIP: 0060:[] EFLAGS: 00010297 CPU: 0 EIP is at print_trace_fmt+0x45/0xe7 EAX: 00000000 EBX: 00000000 ECX: c12d9e98 EDX: ccdb7010 ESI: d31f4000 EDI: 00322401 EBP: d31f3f10 ESP: d31f3efc DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Process cat (pid: 3897, ti=d31f2000 task=d3b3cf20 task.ti=d31f2000) Stack: d31f4080 ccdb7010 d31f4000 d691fe70 ccdb7010 d31f3f24 c0270e5c d31f4000 d691fe70 d31f4000 d31f3f34 c02718e8 c12d9e98 d691fe70 d31f3f70 c02bfc33 00001000 09130000 d3b46e00 d691fe98 00000000 00000079 00000001 00000000 Call Trace: [] ? print_trace_line+0x170/0x17c [] ? s_show+0xa7/0xbd [] ? seq_read+0x24a/0x327 [] ? seq_read+0x0/0x327 [] ? vfs_read+0x86/0xe1 [] ? sys_read+0x40/0x65 [] ? sysenter_do_call+0x12/0x3c Code: 00 00 00 89 45 ec f7 c7 00 20 00 00 89 55 f0 74 4e f6 86 98 10 00 00 02 74 45 8b 86 8c 10 00 00 8b 9e a8 10 00 00 e8 52 f3 ff ff <0f> a3 03 19 c0 85 c0 75 2b 8b 86 8c 10 00 00 8b 9e a8 10 00 00 EIP: [] print_trace_fmt+0x45/0xe7 SS:ESP 0068:d31f3efc CR2: 0000000000000000 ---[ end trace aa9cf38e5ebed9dd ]--- This is because we alloc the iter->started cpumask on tracing_pipe_open but not on tracing_open. It hadn't been noticed until now because we need to have ring buffer overruns to activate the starting of cpu buffer detection. Also, we need a check to not print the messagge for the first trace on the file. Signed-off-by: Frederic Weisbecker LKML-Reference: <1238619188-6109-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 2230b46f9e1c..fc8c7d66832b 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1632,7 +1632,11 @@ static void test_cpu_buff_start(struct trace_iterator *iter) return; cpumask_set_cpu(iter->cpu, iter->started); - trace_seq_printf(s, "##### CPU %u buffer started ####\n", iter->cpu); + + /* Don't print started cpu buffer for the first entry of the trace */ + if (iter->idx > 1) + trace_seq_printf(s, "##### CPU %u buffer started ####\n", + iter->cpu); } static enum print_line_t print_trace_fmt(struct trace_iterator *iter) @@ -1867,6 +1871,11 @@ __tracing_open(struct inode *inode, struct file *file) if (current_trace) *iter->trace = *current_trace; + if (!alloc_cpumask_var(&iter->started, GFP_KERNEL)) + goto fail; + + cpumask_clear(iter->started); + if (current_trace && current_trace->print_max) iter->tr = &max_tr; else @@ -1917,6 +1926,7 @@ __tracing_open(struct inode *inode, struct file *file) if (iter->buffer_iter[cpu]) ring_buffer_read_finish(iter->buffer_iter[cpu]); } + free_cpumask_var(iter->started); fail: mutex_unlock(&trace_types_lock); kfree(iter->trace); @@ -1960,6 +1970,7 @@ static int tracing_release(struct inode *inode, struct file *file) seq_release(inode, file); mutex_destroy(&iter->mutex); + free_cpumask_var(iter->started); kfree(iter->trace); kfree(iter); return 0; -- cgit v1.2.3 From bc2b6871c17b3aff79fb14e1a1c06c5f5a187f76 Mon Sep 17 00:00:00 2001 From: Nikanth Karthikesan Date: Mon, 23 Mar 2009 11:58:31 +0530 Subject: Update /debug/tracing/README Some of the tracers have been renamed, which was not updated in the in-kernel run-time README file. Update it. Signed-off-by: Nikanth Karthikesan LKML-Reference: <200903231158.32151.knikanth@suse.de> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index fc8c7d66832b..9d28476a9851 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2369,9 +2369,9 @@ static const char readme_msg[] = "# mkdir /debug\n" "# mount -t debugfs nodev /debug\n\n" "# cat /debug/tracing/available_tracers\n" - "wakeup preemptirqsoff preemptoff irqsoff ftrace sched_switch none\n\n" + "wakeup preemptirqsoff preemptoff irqsoff function sched_switch nop\n\n" "# cat /debug/tracing/current_tracer\n" - "none\n" + "nop\n" "# echo sched_switch > /debug/tracing/current_tracer\n" "# cat /debug/tracing/current_tracer\n" "sched_switch\n" -- cgit v1.2.3 From 1bbe2a83ab68e5cf8c66c372c7cb3b51910c2cfe Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Fri, 3 Apr 2009 18:24:46 +0800 Subject: ftrace: Correct a text align for event format output If we cat debugfs/tracing/events/ftrace/bprint/format, we'll see: name: bprint ID: 6 format: field:unsigned char common_type; offset:0; size:1; field:unsigned char common_flags; offset:1; size:1; field:unsigned char common_preempt_count; offset:2; size:1; field:int common_pid; offset:4; size:4; field:int common_tgid; offset:8; size:4; field:unsigned long ip; offset:12; size:4; field:char * fmt; offset:16; size:4; field: char buf; offset:20; size:0; print fmt: "%08lx (%d) fmt:%p %s" There is an inconsistent blank before char buf. Signed-off-by: Zhao Lei LKML-Reference: <49D5E3EE.70201@cn.fujitsu.com> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace_export.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index 4d9952d3df50..07a22c33ebf3 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -40,7 +40,7 @@ #undef TRACE_FIELD_ZERO_CHAR #define TRACE_FIELD_ZERO_CHAR(item) \ - ret = trace_seq_printf(s, "\tfield: char " #item ";\t" \ + ret = trace_seq_printf(s, "\tfield:char " #item ";\t" \ "offset:%u;\tsize:0;\n", \ (unsigned int)offsetof(typeof(field), item)); \ if (!ret) \ -- cgit v1.2.3 From fafd688e4c0c34da0f3de909881117d374e4c7af Mon Sep 17 00:00:00 2001 From: Peter W Morreale Date: Mon, 6 Apr 2009 19:00:29 -0700 Subject: mm: add /proc controls for pdflush threads Add /proc entries to give the admin the ability to control the minimum and maximum number of pdflush threads. This allows finer control of pdflush on both large and small machines. The rationale is simply one size does not fit all. Admins on large and/or small systems may want to tune the min/max pdflush thread count to best suit their needs. Right now the min/max is hardcoded to 2/8. While probably a fair estimate for smaller machines, large machines with large numbers of CPUs and large numbers of filesystems/block devices may benefit from larger numbers of threads working on different block devices. Even if the background flushing algorithm is radically changed, it is still likely that multiple threads will be involved and admins would still desire finer control on the min/max other than to have to recompile the kernel. The patch adds '/proc/sys/vm/nr_pdflush_threads_min' and '/proc/sys/vm/nr_pdflush_threads_max' with r/w permissions. The minimum value for nr_pdflush_threads_min is 1 and the maximum value is the current value of nr_pdflush_threads_max. This minimum is required since additional thread creation is performed in a pdflush thread itself. The minimum value for nr_pdflush_threads_max is the current value of nr_pdflush_threads_min and the maximum value can be 1000. Documentation/sysctl/vm.txt is also updated. [akpm@linux-foundation.org: fix comment, fix whitespace, use __read_mostly] Signed-off-by: Peter W Morreale Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index b125e3387568..72eb1a41dcab 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -101,6 +101,7 @@ static int __maybe_unused one = 1; static int __maybe_unused two = 2; static unsigned long one_ul = 1; static int one_hundred = 100; +static int one_thousand = 1000; /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ static int maxolduid = 65535; @@ -1026,6 +1027,28 @@ static struct ctl_table vm_table[] = { .mode = 0444 /* read-only*/, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nr_pdflush_threads_min", + .data = &nr_pdflush_threads_min, + .maxlen = sizeof nr_pdflush_threads_min, + .mode = 0644 /* read-write */, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &one, + .extra2 = &nr_pdflush_threads_max, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nr_pdflush_threads_max", + .data = &nr_pdflush_threads_max, + .maxlen = sizeof nr_pdflush_threads_max, + .mode = 0644 /* read-write */, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &nr_pdflush_threads_min, + .extra2 = &one_thousand, + }, { .ctl_name = VM_SWAPPINESS, .procname = "swappiness", -- cgit v1.2.3 From b918e5e60d775549478e4268155142156a95aa17 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 6 Apr 2009 19:00:58 -0700 Subject: kprobes: cleanup aggr_kprobe related code Currently, kprobes can disable all probes at once, but can't disable it individually (not unregister, just disable an kprobe, because unregistering needs to wait for scheduler synchronization). These patches introduce APIs for on-the-fly per-probe disabling and re-enabling by dis-arming/re-arming its breakpoint instruction. This patch: Change old_p to ap in add_new_kprobe() for readability, copy flags member in add_aggr_kprobe(), and simplify the code flow of register_aggr_kprobe(). Signed-off-by: Masami Hiramatsu Acked-by: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Cc: David S. Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kprobes.c | 60 ++++++++++++++++++++++++++++---------------------------- 1 file changed, 30 insertions(+), 30 deletions(-) (limited to 'kernel') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 5016bfb682b9..a55bfadfd766 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -518,20 +518,20 @@ static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p) } /* -* Add the new probe to old_p->list. Fail if this is the +* Add the new probe to ap->list. Fail if this is the * second jprobe at the address - two jprobes can't coexist */ -static int __kprobes add_new_kprobe(struct kprobe *old_p, struct kprobe *p) +static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) { if (p->break_handler) { - if (old_p->break_handler) + if (ap->break_handler) return -EEXIST; - list_add_tail_rcu(&p->list, &old_p->list); - old_p->break_handler = aggr_break_handler; + list_add_tail_rcu(&p->list, &ap->list); + ap->break_handler = aggr_break_handler; } else - list_add_rcu(&p->list, &old_p->list); - if (p->post_handler && !old_p->post_handler) - old_p->post_handler = aggr_post_handler; + list_add_rcu(&p->list, &ap->list); + if (p->post_handler && !ap->post_handler) + ap->post_handler = aggr_post_handler; return 0; } @@ -544,6 +544,7 @@ static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p) copy_kprobe(p, ap); flush_insn_slot(ap); ap->addr = p->addr; + ap->flags = p->flags; ap->pre_handler = aggr_pre_handler; ap->fault_handler = aggr_fault_handler; /* We don't care the kprobe which has gone. */ @@ -566,44 +567,43 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p) { int ret = 0; - struct kprobe *ap; + struct kprobe *ap = old_p; - if (kprobe_gone(old_p)) { + if (old_p->pre_handler != aggr_pre_handler) { + /* If old_p is not an aggr_probe, create new aggr_kprobe. */ + ap = kzalloc(sizeof(struct kprobe), GFP_KERNEL); + if (!ap) + return -ENOMEM; + add_aggr_kprobe(ap, old_p); + } + + if (kprobe_gone(ap)) { /* * Attempting to insert new probe at the same location that * had a probe in the module vaddr area which already * freed. So, the instruction slot has already been * released. We need a new slot for the new probe. */ - ret = arch_prepare_kprobe(old_p); + ret = arch_prepare_kprobe(ap); if (ret) + /* + * Even if fail to allocate new slot, don't need to + * free aggr_probe. It will be used next time, or + * freed by unregister_kprobe. + */ return ret; - } - if (old_p->pre_handler == aggr_pre_handler) { - copy_kprobe(old_p, p); - ret = add_new_kprobe(old_p, p); - ap = old_p; - } else { - ap = kzalloc(sizeof(struct kprobe), GFP_KERNEL); - if (!ap) { - if (kprobe_gone(old_p)) - arch_remove_kprobe(old_p); - return -ENOMEM; - } - add_aggr_kprobe(ap, old_p); - copy_kprobe(ap, p); - ret = add_new_kprobe(ap, p); - } - if (kprobe_gone(old_p)) { + /* Clear gone flag to prevent allocating new slot again. */ + ap->flags &= ~KPROBE_FLAG_GONE; /* * If the old_p has gone, its breakpoint has been disarmed. * We have to arm it again after preparing real kprobes. */ - ap->flags &= ~KPROBE_FLAG_GONE; if (kprobe_enabled) arch_arm_kprobe(ap); } - return ret; + + copy_kprobe(ap, p); + return add_new_kprobe(ap, p); } static int __kprobes in_kprobes_functions(unsigned long addr) -- cgit v1.2.3 From 99081ab553d6a88dd6b3774af5eef94dbb8faad7 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 6 Apr 2009 19:00:59 -0700 Subject: kprobes: move EXPORT_SYMBOL_GPL just after function definitions Clean up positions of EXPORT_SYMBOL_GPL in kernel/kprobes.c according to checkpatch.pl. Signed-off-by: Masami Hiramatsu Acked-by: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Cc: David S. Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kprobes.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index a55bfadfd766..ca4c22d78cfd 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -722,6 +722,7 @@ out: return ret; } +EXPORT_SYMBOL_GPL(register_kprobe); /* * Unregister a kprobe without a scheduler synchronization. @@ -803,11 +804,13 @@ int __kprobes register_kprobes(struct kprobe **kps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_kprobes); void __kprobes unregister_kprobe(struct kprobe *p) { unregister_kprobes(&p, 1); } +EXPORT_SYMBOL_GPL(unregister_kprobe); void __kprobes unregister_kprobes(struct kprobe **kps, int num) { @@ -826,6 +829,7 @@ void __kprobes unregister_kprobes(struct kprobe **kps, int num) if (kps[i]->addr) __unregister_kprobe_bottom(kps[i]); } +EXPORT_SYMBOL_GPL(unregister_kprobes); static struct notifier_block kprobe_exceptions_nb = { .notifier_call = kprobe_exceptions_notify, @@ -865,16 +869,19 @@ int __kprobes register_jprobes(struct jprobe **jps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_jprobes); int __kprobes register_jprobe(struct jprobe *jp) { return register_jprobes(&jp, 1); } +EXPORT_SYMBOL_GPL(register_jprobe); void __kprobes unregister_jprobe(struct jprobe *jp) { unregister_jprobes(&jp, 1); } +EXPORT_SYMBOL_GPL(unregister_jprobe); void __kprobes unregister_jprobes(struct jprobe **jps, int num) { @@ -894,6 +901,7 @@ void __kprobes unregister_jprobes(struct jprobe **jps, int num) __unregister_kprobe_bottom(&jps[i]->kp); } } +EXPORT_SYMBOL_GPL(unregister_jprobes); #ifdef CONFIG_KRETPROBES /* @@ -987,6 +995,7 @@ int __kprobes register_kretprobe(struct kretprobe *rp) free_rp_inst(rp); return ret; } +EXPORT_SYMBOL_GPL(register_kretprobe); int __kprobes register_kretprobes(struct kretprobe **rps, int num) { @@ -1004,11 +1013,13 @@ int __kprobes register_kretprobes(struct kretprobe **rps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_kretprobes); void __kprobes unregister_kretprobe(struct kretprobe *rp) { unregister_kretprobes(&rp, 1); } +EXPORT_SYMBOL_GPL(unregister_kretprobe); void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) { @@ -1030,24 +1041,30 @@ void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) } } } +EXPORT_SYMBOL_GPL(unregister_kretprobes); #else /* CONFIG_KRETPROBES */ int __kprobes register_kretprobe(struct kretprobe *rp) { return -ENOSYS; } +EXPORT_SYMBOL_GPL(register_kretprobe); int __kprobes register_kretprobes(struct kretprobe **rps, int num) { return -ENOSYS; } +EXPORT_SYMBOL_GPL(register_kretprobes); + void __kprobes unregister_kretprobe(struct kretprobe *rp) { } +EXPORT_SYMBOL_GPL(unregister_kretprobe); void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) { } +EXPORT_SYMBOL_GPL(unregister_kretprobes); static int __kprobes pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs) @@ -1418,16 +1435,5 @@ late_initcall(debugfs_kprobe_init); module_init(init_kprobes); -EXPORT_SYMBOL_GPL(register_kprobe); -EXPORT_SYMBOL_GPL(unregister_kprobe); -EXPORT_SYMBOL_GPL(register_kprobes); -EXPORT_SYMBOL_GPL(unregister_kprobes); -EXPORT_SYMBOL_GPL(register_jprobe); -EXPORT_SYMBOL_GPL(unregister_jprobe); -EXPORT_SYMBOL_GPL(register_jprobes); -EXPORT_SYMBOL_GPL(unregister_jprobes); +/* defined in arch/.../kernel/kprobes.c */ EXPORT_SYMBOL_GPL(jprobe_return); -EXPORT_SYMBOL_GPL(register_kretprobe); -EXPORT_SYMBOL_GPL(unregister_kretprobe); -EXPORT_SYMBOL_GPL(register_kretprobes); -EXPORT_SYMBOL_GPL(unregister_kretprobes); -- cgit v1.2.3 From e579abeb58eb4b8d7321c6eb44dd9e2d0cbaebaa Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 6 Apr 2009 19:01:01 -0700 Subject: kprobes: rename kprobe_enabled to kprobes_all_disarmed Rename kprobe_enabled to kprobes_all_disarmed and invert logic due to avoiding naming confusion from per-probe disabling. Signed-off-by: Masami Hiramatsu Acked-by: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Cc: David S. Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kprobes.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index ca4c22d78cfd..dae198b68e97 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -68,7 +68,7 @@ static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE]; static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE]; /* NOTE: change this value only with kprobe_mutex held */ -static bool kprobe_enabled; +static bool kprobes_all_disarmed; static DEFINE_MUTEX(kprobe_mutex); /* Protects kprobe_table */ static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL; @@ -598,7 +598,7 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p, * If the old_p has gone, its breakpoint has been disarmed. * We have to arm it again after preparing real kprobes. */ - if (kprobe_enabled) + if (!kprobes_all_disarmed) arch_arm_kprobe(ap); } @@ -709,7 +709,7 @@ int __kprobes register_kprobe(struct kprobe *p) hlist_add_head_rcu(&p->hlist, &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]); - if (kprobe_enabled) + if (!kprobes_all_disarmed) arch_arm_kprobe(p); out_unlock_text: @@ -751,7 +751,7 @@ valid_p: * enabled and not gone - otherwise, the breakpoint would * already have been removed. We save on flushing icache. */ - if (kprobe_enabled && !kprobe_gone(old_p)) { + if (!kprobes_all_disarmed && !kprobe_gone(old_p)) { mutex_lock(&text_mutex); arch_disarm_kprobe(p); mutex_unlock(&text_mutex); @@ -1190,8 +1190,8 @@ static int __init init_kprobes(void) } } - /* By default, kprobes are enabled */ - kprobe_enabled = true; + /* By default, kprobes are armed */ + kprobes_all_disarmed = false; err = arch_init_kprobes(); if (!err) @@ -1289,7 +1289,7 @@ static struct file_operations debugfs_kprobes_operations = { .release = seq_release, }; -static void __kprobes enable_all_kprobes(void) +static void __kprobes arm_all_kprobes(void) { struct hlist_head *head; struct hlist_node *node; @@ -1298,8 +1298,8 @@ static void __kprobes enable_all_kprobes(void) mutex_lock(&kprobe_mutex); - /* If kprobes are already enabled, just return */ - if (kprobe_enabled) + /* If kprobes are armed, just return */ + if (!kprobes_all_disarmed) goto already_enabled; mutex_lock(&text_mutex); @@ -1311,7 +1311,7 @@ static void __kprobes enable_all_kprobes(void) } mutex_unlock(&text_mutex); - kprobe_enabled = true; + kprobes_all_disarmed = false; printk(KERN_INFO "Kprobes globally enabled\n"); already_enabled: @@ -1319,7 +1319,7 @@ already_enabled: return; } -static void __kprobes disable_all_kprobes(void) +static void __kprobes disarm_all_kprobes(void) { struct hlist_head *head; struct hlist_node *node; @@ -1328,11 +1328,11 @@ static void __kprobes disable_all_kprobes(void) mutex_lock(&kprobe_mutex); - /* If kprobes are already disabled, just return */ - if (!kprobe_enabled) + /* If kprobes are already disarmed, just return */ + if (kprobes_all_disarmed) goto already_disabled; - kprobe_enabled = false; + kprobes_all_disarmed = true; printk(KERN_INFO "Kprobes globally disabled\n"); mutex_lock(&text_mutex); for (i = 0; i < KPROBE_TABLE_SIZE; i++) { @@ -1364,7 +1364,7 @@ static ssize_t read_enabled_file_bool(struct file *file, { char buf[3]; - if (kprobe_enabled) + if (!kprobes_all_disarmed) buf[0] = '1'; else buf[0] = '0'; @@ -1387,12 +1387,12 @@ static ssize_t write_enabled_file_bool(struct file *file, case 'y': case 'Y': case '1': - enable_all_kprobes(); + arm_all_kprobes(); break; case 'n': case 'N': case '0': - disable_all_kprobes(); + disarm_all_kprobes(); break; } -- cgit v1.2.3 From de5bd88d5a5cce3cacea904d3503e5ebdb3852a2 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 6 Apr 2009 19:01:02 -0700 Subject: kprobes: support per-kprobe disabling Add disable_kprobe() and enable_kprobe() to disable/enable kprobes temporarily. disable_kprobe() asynchronously disables probe handlers of specified kprobe. So, after calling it, some handlers can be called at a while. enable_kprobe() enables specified kprobe. aggr_pre_handler and aggr_post_handler check disabled probes. On the other hand aggr_break_handler and aggr_fault_handler don't check it because these handlers will be called while executing pre or post handlers and usually those help error handling. Signed-off-by: Masami Hiramatsu Acked-by: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Cc: David S. Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kprobes.c | 167 ++++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 141 insertions(+), 26 deletions(-) (limited to 'kernel') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index dae198b68e97..a5e74ddee0e2 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -328,7 +328,7 @@ static int __kprobes aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) struct kprobe *kp; list_for_each_entry_rcu(kp, &p->list, list) { - if (kp->pre_handler && !kprobe_gone(kp)) { + if (kp->pre_handler && likely(!kprobe_disabled(kp))) { set_kprobe_instance(kp); if (kp->pre_handler(kp, regs)) return 1; @@ -344,7 +344,7 @@ static void __kprobes aggr_post_handler(struct kprobe *p, struct pt_regs *regs, struct kprobe *kp; list_for_each_entry_rcu(kp, &p->list, list) { - if (kp->post_handler && !kprobe_gone(kp)) { + if (kp->post_handler && likely(!kprobe_disabled(kp))) { set_kprobe_instance(kp); kp->post_handler(kp, regs, flags); reset_kprobe_instance(); @@ -523,6 +523,7 @@ static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p) */ static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) { + BUG_ON(kprobe_gone(ap) || kprobe_gone(p)); if (p->break_handler) { if (ap->break_handler) return -EEXIST; @@ -532,6 +533,13 @@ static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) list_add_rcu(&p->list, &ap->list); if (p->post_handler && !ap->post_handler) ap->post_handler = aggr_post_handler; + + if (kprobe_disabled(ap) && !kprobe_disabled(p)) { + ap->flags &= ~KPROBE_FLAG_DISABLED; + if (!kprobes_all_disarmed) + /* Arm the breakpoint again. */ + arch_arm_kprobe(ap); + } return 0; } @@ -592,20 +600,36 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p, * freed by unregister_kprobe. */ return ret; - /* Clear gone flag to prevent allocating new slot again. */ - ap->flags &= ~KPROBE_FLAG_GONE; + /* - * If the old_p has gone, its breakpoint has been disarmed. - * We have to arm it again after preparing real kprobes. + * Clear gone flag to prevent allocating new slot again, and + * set disabled flag because it is not armed yet. */ - if (!kprobes_all_disarmed) - arch_arm_kprobe(ap); + ap->flags = (ap->flags & ~KPROBE_FLAG_GONE) + | KPROBE_FLAG_DISABLED; } copy_kprobe(ap, p); return add_new_kprobe(ap, p); } +/* Try to disable aggr_kprobe, and return 1 if succeeded.*/ +static int __kprobes try_to_disable_aggr_kprobe(struct kprobe *p) +{ + struct kprobe *kp; + + list_for_each_entry_rcu(kp, &p->list, list) { + if (!kprobe_disabled(kp)) + /* + * There is an active probe on the list. + * We can't disable aggr_kprobe. + */ + return 0; + } + p->flags |= KPROBE_FLAG_DISABLED; + return 1; +} + static int __kprobes in_kprobes_functions(unsigned long addr) { struct kprobe_blackpoint *kb; @@ -664,7 +688,9 @@ int __kprobes register_kprobe(struct kprobe *p) return -EINVAL; } - p->flags = 0; + /* User can pass only KPROBE_FLAG_DISABLED to register_kprobe */ + p->flags &= KPROBE_FLAG_DISABLED; + /* * Check if are we probing a module. */ @@ -709,7 +735,7 @@ int __kprobes register_kprobe(struct kprobe *p) hlist_add_head_rcu(&p->hlist, &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]); - if (!kprobes_all_disarmed) + if (!kprobes_all_disarmed && !kprobe_disabled(p)) arch_arm_kprobe(p); out_unlock_text: @@ -724,25 +750,37 @@ out: } EXPORT_SYMBOL_GPL(register_kprobe); -/* - * Unregister a kprobe without a scheduler synchronization. - */ -static int __kprobes __unregister_kprobe_top(struct kprobe *p) +/* Check passed kprobe is valid and return kprobe in kprobe_table. */ +static struct kprobe * __kprobes __get_valid_kprobe(struct kprobe *p) { struct kprobe *old_p, *list_p; old_p = get_kprobe(p->addr); if (unlikely(!old_p)) - return -EINVAL; + return NULL; if (p != old_p) { list_for_each_entry_rcu(list_p, &old_p->list, list) if (list_p == p) /* kprobe p is a valid probe */ - goto valid_p; - return -EINVAL; + goto valid; + return NULL; } -valid_p: +valid: + return old_p; +} + +/* + * Unregister a kprobe without a scheduler synchronization. + */ +static int __kprobes __unregister_kprobe_top(struct kprobe *p) +{ + struct kprobe *old_p, *list_p; + + old_p = __get_valid_kprobe(p); + if (old_p == NULL) + return -EINVAL; + if (old_p == p || (old_p->pre_handler == aggr_pre_handler && list_is_singular(&old_p->list))) { @@ -751,7 +789,7 @@ valid_p: * enabled and not gone - otherwise, the breakpoint would * already have been removed. We save on flushing icache. */ - if (!kprobes_all_disarmed && !kprobe_gone(old_p)) { + if (!kprobes_all_disarmed && !kprobe_disabled(old_p)) { mutex_lock(&text_mutex); arch_disarm_kprobe(p); mutex_unlock(&text_mutex); @@ -769,6 +807,11 @@ valid_p: } noclean: list_del_rcu(&p->list); + if (!kprobe_disabled(old_p)) { + try_to_disable_aggr_kprobe(old_p); + if (!kprobes_all_disarmed && kprobe_disabled(old_p)) + arch_disarm_kprobe(old_p); + } } return 0; } @@ -1078,6 +1121,7 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p, static void __kprobes kill_kprobe(struct kprobe *p) { struct kprobe *kp; + p->flags |= KPROBE_FLAG_GONE; if (p->pre_handler == aggr_pre_handler) { /* @@ -1219,12 +1263,18 @@ static void __kprobes report_probe(struct seq_file *pi, struct kprobe *p, else kprobe_type = "k"; if (sym) - seq_printf(pi, "%p %s %s+0x%x %s %s\n", p->addr, kprobe_type, - sym, offset, (modname ? modname : " "), - (kprobe_gone(p) ? "[GONE]" : "")); + seq_printf(pi, "%p %s %s+0x%x %s %s%s\n", + p->addr, kprobe_type, sym, offset, + (modname ? modname : " "), + (kprobe_gone(p) ? "[GONE]" : ""), + ((kprobe_disabled(p) && !kprobe_gone(p)) ? + "[DISABLED]" : "")); else - seq_printf(pi, "%p %s %p %s\n", p->addr, kprobe_type, p->addr, - (kprobe_gone(p) ? "[GONE]" : "")); + seq_printf(pi, "%p %s %p %s%s\n", + p->addr, kprobe_type, p->addr, + (kprobe_gone(p) ? "[GONE]" : ""), + ((kprobe_disabled(p) && !kprobe_gone(p)) ? + "[DISABLED]" : "")); } static void __kprobes *kprobe_seq_start(struct seq_file *f, loff_t *pos) @@ -1289,6 +1339,71 @@ static struct file_operations debugfs_kprobes_operations = { .release = seq_release, }; +/* Disable one kprobe */ +int __kprobes disable_kprobe(struct kprobe *kp) +{ + int ret = 0; + struct kprobe *p; + + mutex_lock(&kprobe_mutex); + + /* Check whether specified probe is valid. */ + p = __get_valid_kprobe(kp); + if (unlikely(p == NULL)) { + ret = -EINVAL; + goto out; + } + + /* If the probe is already disabled (or gone), just return */ + if (kprobe_disabled(kp)) + goto out; + + kp->flags |= KPROBE_FLAG_DISABLED; + if (p != kp) + /* When kp != p, p is always enabled. */ + try_to_disable_aggr_kprobe(p); + + if (!kprobes_all_disarmed && kprobe_disabled(p)) + arch_disarm_kprobe(p); +out: + mutex_unlock(&kprobe_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(disable_kprobe); + +/* Enable one kprobe */ +int __kprobes enable_kprobe(struct kprobe *kp) +{ + int ret = 0; + struct kprobe *p; + + mutex_lock(&kprobe_mutex); + + /* Check whether specified probe is valid. */ + p = __get_valid_kprobe(kp); + if (unlikely(p == NULL)) { + ret = -EINVAL; + goto out; + } + + if (kprobe_gone(kp)) { + /* This kprobe has gone, we couldn't enable it. */ + ret = -EINVAL; + goto out; + } + + if (!kprobes_all_disarmed && kprobe_disabled(p)) + arch_arm_kprobe(p); + + p->flags &= ~KPROBE_FLAG_DISABLED; + if (p != kp) + kp->flags &= ~KPROBE_FLAG_DISABLED; +out: + mutex_unlock(&kprobe_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(enable_kprobe); + static void __kprobes arm_all_kprobes(void) { struct hlist_head *head; @@ -1306,7 +1421,7 @@ static void __kprobes arm_all_kprobes(void) for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) - if (!kprobe_gone(p)) + if (!kprobe_disabled(p)) arch_arm_kprobe(p); } mutex_unlock(&text_mutex); @@ -1338,7 +1453,7 @@ static void __kprobes disarm_all_kprobes(void) for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) { - if (!arch_trampoline_kprobe(p) && !kprobe_gone(p)) + if (!arch_trampoline_kprobe(p) && !kprobe_disabled(p)) arch_disarm_kprobe(p); } } -- cgit v1.2.3 From 6279a751fe096a21dc7704e918d570d3ff06e769 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 27 Mar 2009 01:06:07 +0100 Subject: posix-timers: fix RLIMIT_CPU && fork() See http://bugzilla.kernel.org/show_bug.cgi?id=12911 copy_signal() copies signal->rlim, but RLIMIT_CPU is "lost". Because posix_cpu_timers_init_group() sets cputime_expires.prof_exp = 0 and thus fastpath_timer_check() returns false unless we have other expired cpu timers. Change copy_signal() to set cputime_expires.prof_exp if we have RLIMIT_CPU. Also, set cputimer.running = 1 in that case. This is not strictly necessary, but imho makes sense. Reported-by: Peter Lojkin Signed-off-by: Oleg Nesterov Acked-by: Peter Zijlstra Cc: Peter Lojkin Cc: Roland McGrath Cc: stable@kernel.org LKML-Reference: <20090327000607.GA10104@redhat.com> Signed-off-by: Ingo Molnar --- kernel/fork.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 4854c2c4a82e..9b51a1b190d4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -808,6 +808,12 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) sig->cputime_expires.virt_exp = cputime_zero; sig->cputime_expires.sched_exp = 0; + if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { + sig->cputime_expires.prof_exp = + secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur); + sig->cputimer.running = 1; + } + /* The timer lists. */ INIT_LIST_HEAD(&sig->cpu_timers[0]); INIT_LIST_HEAD(&sig->cpu_timers[1]); @@ -823,11 +829,8 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) atomic_inc(¤t->signal->live); return 0; } - sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); - - if (sig) - posix_cpu_timers_init_group(sig); + sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); tsk->signal = sig; if (!sig) return -ENOMEM; @@ -865,6 +868,8 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim); task_unlock(current->group_leader); + posix_cpu_timers_init_group(sig); + acct_init_pacct(&sig->pacct); tty_audit_fork(sig); -- cgit v1.2.3 From 8f2e586567b1bad72dac7c3810fe9a2ef7117506 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 27 Mar 2009 01:06:10 +0100 Subject: posix-timers: fix RLIMIT_CPU && setitimer(CPUCLOCK_PROF) update_rlimit_cpu() tries to optimize out set_process_cpu_timer() in case when we already have CPUCLOCK_PROF timer which should expire first. But it uses cputime_lt() instead of cputime_gt(). Test case: int main(void) { struct itimerval it = { .it_value = { .tv_sec = 1000 }, }; assert(!setitimer(ITIMER_PROF, &it, NULL)); struct rlimit rl = { .rlim_cur = 1, .rlim_max = 1, }; assert(!setrlimit(RLIMIT_CPU, &rl)); for (;;) ; return 0; } Without this patch, the task is not killed as RLIMIT_CPU demands. Signed-off-by: Oleg Nesterov Acked-by: Peter Zijlstra Cc: Peter Lojkin Cc: Roland McGrath Cc: stable@kernel.org LKML-Reference: <20090327000610.GA10108@redhat.com> Signed-off-by: Ingo Molnar --- kernel/posix-cpu-timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 8e5d9a68b022..bb53185d8c78 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -18,7 +18,7 @@ void update_rlimit_cpu(unsigned long rlim_new) cputime = secs_to_cputime(rlim_new); if (cputime_eq(current->signal->it_prof_expires, cputime_zero) || - cputime_lt(current->signal->it_prof_expires, cputime)) { + cputime_gt(current->signal->it_prof_expires, cputime)) { spin_lock_irq(¤t->sighand->siglock); set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL); spin_unlock_irq(¤t->sighand->siglock); -- cgit v1.2.3 From 3a709703538c471530405556dda136fd0d82b0dc Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Tue, 7 Apr 2009 23:21:06 -0700 Subject: ptrace: some checkpatch fixes This fixes all the checkpatch --file complaints about kernel/ptrace.c and also removes an unused #include. I've verified that there are no changes to the compiled code on x86_64. Signed-off-by: Roland McGrath [ Removed the parts that just split a line - Linus ] Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index aaad0ec34194..64191fa09b7e 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -21,9 +21,7 @@ #include #include #include - -#include -#include +#include /* @@ -48,7 +46,7 @@ void __ptrace_link(struct task_struct *child, struct task_struct *new_parent) list_add(&child->ptrace_entry, &new_parent->ptraced); child->parent = new_parent; } - + /* * Turn a tracing stop into a normal stop now, since with no tracer there * would be no way to wake it up with SIGCONT or SIGKILL. If there was a @@ -173,7 +171,7 @@ bool ptrace_may_access(struct task_struct *task, unsigned int mode) task_lock(task); err = __ptrace_may_access(task, mode); task_unlock(task); - return (!err ? true : false); + return !err; } int ptrace_attach(struct task_struct *task) @@ -358,7 +356,7 @@ int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst copied += retval; src += retval; dst += retval; - len -= retval; + len -= retval; } return copied; } @@ -383,7 +381,7 @@ int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long ds copied += retval; src += retval; dst += retval; - len -= retval; + len -= retval; } return copied; } @@ -496,9 +494,9 @@ static int ptrace_resume(struct task_struct *child, long request, long data) if (unlikely(!arch_has_single_step())) return -EIO; user_enable_single_step(child); - } - else + } else { user_disable_single_step(child); + } child->exit_code = data; wake_up_process(child); -- cgit v1.2.3 From 3217ab97f14c5c8f9f975ed8c40c351164b0b10e Mon Sep 17 00:00:00 2001 From: Vitaliy Gusev Date: Thu, 9 Apr 2009 09:50:35 -0600 Subject: kthread: Don't looking for a task in create_kthread() #2 Remove the unnecessary find_task_by_pid_ns(). kthread() can just use "current" to get the same result. Signed-off-by: Vitaliy Gusev Acked-by: Oleg Nesterov Signed-off-by: Rusty Russell --- kernel/kthread.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/kthread.c b/kernel/kthread.c index 84bbadd4d021..c013bf000ce6 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -76,6 +76,7 @@ static int kthread(void *_create) /* OK, tell user we're spawned, wait for stop or wakeup */ __set_current_state(TASK_UNINTERRUPTIBLE); + create->result = current; complete(&create->started); schedule(); @@ -101,9 +102,6 @@ static void create_kthread(struct kthread_create_info *create) } else { struct sched_param param = { .sched_priority = 0 }; wait_for_completion(&create->started); - read_lock(&tasklist_lock); - create->result = find_task_by_pid_ns(pid, &init_pid_ns); - read_unlock(&tasklist_lock); /* * root may have changed our (kthreadd's) priority or CPU mask. * The kernel thread should not inherit these properties. -- cgit v1.2.3 From 1c99315bb36b5d776210546d438ca928dc9b1f22 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 9 Apr 2009 09:50:36 -0600 Subject: kthread: move sched-realeted initialization from kthreadd context kthreadd is the single thread which implements ths "create" request, move sched_setscheduler/etc from create_kthread() to kthread_create() to improve the scalability. We should be careful with sched_setscheduler(), use _nochek helper. Signed-off-by: Oleg Nesterov Cc: Christoph Hellwig Cc: "Eric W. Biederman" Cc: Ingo Molnar Cc: Pavel Emelyanov Cc: Vitaliy Gusev Signed-off-by: Rusty Russell --- kernel/kthread.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/kthread.c b/kernel/kthread.c index c013bf000ce6..4ebaf8519abf 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -97,19 +97,10 @@ static void create_kthread(struct kthread_create_info *create) /* We want our own signal handler (we take no signals by default). */ pid = kernel_thread(kthread, create, CLONE_FS | CLONE_FILES | SIGCHLD); - if (pid < 0) { + if (pid < 0) create->result = ERR_PTR(pid); - } else { - struct sched_param param = { .sched_priority = 0 }; + else wait_for_completion(&create->started); - /* - * root may have changed our (kthreadd's) priority or CPU mask. - * The kernel thread should not inherit these properties. - */ - sched_setscheduler(create->result, SCHED_NORMAL, ¶m); - set_user_nice(create->result, KTHREAD_NICE_LEVEL); - set_cpus_allowed_ptr(create->result, cpu_all_mask); - } complete(&create->done); } @@ -152,11 +143,20 @@ struct task_struct *kthread_create(int (*threadfn)(void *data), wait_for_completion(&create.done); if (!IS_ERR(create.result)) { + struct sched_param param = { .sched_priority = 0 }; va_list args; + va_start(args, namefmt); vsnprintf(create.result->comm, sizeof(create.result->comm), namefmt, args); va_end(args); + /* + * root may have changed our (kthreadd's) priority or CPU mask. + * The kernel thread should not inherit these properties. + */ + sched_setscheduler_nocheck(create.result, SCHED_NORMAL, ¶m); + set_user_nice(create.result, KTHREAD_NICE_LEVEL); + set_cpus_allowed_ptr(create.result, cpu_all_mask); } return create.result; } -- cgit v1.2.3 From 6b44003e5ca66a3fffeb5bc90f40ada2c4340896 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 9 Apr 2009 09:50:37 -0600 Subject: work_on_cpu(): rewrite it to create a kernel thread on demand Impact: circular locking bugfix The various implemetnations and proposed implemetnations of work_on_cpu() are vulnerable to various deadlocks because they all used queues of some form. Unrelated pieces of kernel code thus gained dependencies wherein if one work_on_cpu() caller holds a lock which some other work_on_cpu() callback also takes, the kernel could rarely deadlock. Fix this by creating a short-lived kernel thread for each work_on_cpu() invokation. This is not terribly fast, but the only current caller of work_on_cpu() is pci_call_probe(). It would be nice to find some other way of doing the node-local allocations in the PCI probe code so that we can zap work_on_cpu() altogether. The code there is rather nasty. I can't think of anything simple at this time... Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Rusty Russell --- kernel/workqueue.c | 36 +++++++++++++++++++----------------- 1 file changed, 19 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index b6b966ce1451..f71fb2a08950 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -966,20 +966,20 @@ undo: } #ifdef CONFIG_SMP -static struct workqueue_struct *work_on_cpu_wq __read_mostly; struct work_for_cpu { - struct work_struct work; + struct completion completion; long (*fn)(void *); void *arg; long ret; }; -static void do_work_for_cpu(struct work_struct *w) +static int do_work_for_cpu(void *_wfc) { - struct work_for_cpu *wfc = container_of(w, struct work_for_cpu, work); - + struct work_for_cpu *wfc = _wfc; wfc->ret = wfc->fn(wfc->arg); + complete(&wfc->completion); + return 0; } /** @@ -990,17 +990,23 @@ static void do_work_for_cpu(struct work_struct *w) * * This will return the value @fn returns. * It is up to the caller to ensure that the cpu doesn't go offline. + * The caller must not hold any locks which would prevent @fn from completing. */ long work_on_cpu(unsigned int cpu, long (*fn)(void *), void *arg) { - struct work_for_cpu wfc; - - INIT_WORK(&wfc.work, do_work_for_cpu); - wfc.fn = fn; - wfc.arg = arg; - queue_work_on(cpu, work_on_cpu_wq, &wfc.work); - flush_work(&wfc.work); - + struct task_struct *sub_thread; + struct work_for_cpu wfc = { + .completion = COMPLETION_INITIALIZER_ONSTACK(wfc.completion), + .fn = fn, + .arg = arg, + }; + + sub_thread = kthread_create(do_work_for_cpu, &wfc, "work_for_cpu"); + if (IS_ERR(sub_thread)) + return PTR_ERR(sub_thread); + kthread_bind(sub_thread, cpu); + wake_up_process(sub_thread); + wait_for_completion(&wfc.completion); return wfc.ret; } EXPORT_SYMBOL_GPL(work_on_cpu); @@ -1016,8 +1022,4 @@ void __init init_workqueues(void) hotcpu_notifier(workqueue_cpu_callback, 0); keventd_wq = create_workqueue("events"); BUG_ON(!keventd_wq); -#ifdef CONFIG_SMP - work_on_cpu_wq = create_workqueue("work_on_cpu"); - BUG_ON(!work_on_cpu_wq); -#endif } -- cgit v1.2.3 From 47788c58e66c050982241d9a05eb690daceb05a9 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 8 Apr 2009 20:40:59 +0200 Subject: tracing/syscalls: use a dedicated file header Impact: fix build warnings and possibe compat misbehavior on IA64 Building a kernel on ia64 might trigger these ugly build warnings: CC arch/ia64/ia32/sys_ia32.o In file included from arch/ia64/ia32/sys_ia32.c:55: arch/ia64/ia32/ia32priv.h:290:1: warning: "elf_check_arch" redefined In file included from include/linux/elf.h:7, from include/linux/module.h:14, from include/linux/ftrace.h:8, from include/linux/syscalls.h:68, from arch/ia64/ia32/sys_ia32.c:18: arch/ia64/include/asm/elf.h:19:1: warning: this is the location of the previous definition [...] sys_ia32.c includes linux/syscalls.h which in turn includes linux/ftrace.h to import the syscalls tracing prototypes. But including ftrace.h can pull too much things for a low level file, especially on ia64 where the ia32 private headers conflict with higher level headers. Now we isolate the syscall tracing headers in their own lightweight file. Reported-by: Tony Luck Tested-by: Tony Luck Signed-off-by: Frederic Weisbecker Acked-by: Tony Luck Signed-off-by: Steven Rostedt Cc: Peter Zijlstra Cc: Jason Baron Cc: "Frank Ch. Eigler" Cc: Mathieu Desnoyers Cc: KOSAKI Motohiro Cc: Lai Jiangshan Cc: Jiaying Zhang Cc: Michael Rubin Cc: Martin Bligh Cc: Michael Davidson LKML-Reference: <20090408184058.GB6017@nowhere> Signed-off-by: Ingo Molnar --- kernel/trace/trace_syscalls.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index a2a3af29c943..5e579645ac86 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -1,5 +1,5 @@ +#include #include -#include #include #include "trace_output.h" -- cgit v1.2.3 From 9eb85125ce218a8b8d9a7c982510388e227adbec Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 9 Apr 2009 11:19:40 +0800 Subject: blktrace: pass the right pointer to kfree() Impact: fix kfree crash with non-standard act_mask string If passing a string with leading white spaces to strstrip(), the returned ptr != the original ptr. This bug was introduced by me. Signed-off-by: Li Zefan Cc: Jens Axboe Cc: Arnaldo Carvalho de Melo LKML-Reference: <49DD694C.8020902@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/blktrace.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index b32ff446c3fb..921ef5d1f0ba 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -1377,12 +1377,12 @@ static int blk_trace_str2mask(const char *str) { int i; int mask = 0; - char *s, *token; + char *buf, *s, *token; - s = kstrdup(str, GFP_KERNEL); - if (s == NULL) + buf = kstrdup(str, GFP_KERNEL); + if (buf == NULL) return -ENOMEM; - s = strstrip(s); + s = strstrip(buf); while (1) { token = strsep(&s, ","); @@ -1403,7 +1403,7 @@ static int blk_trace_str2mask(const char *str) break; } } - kfree(s); + kfree(buf); return mask; } -- cgit v1.2.3 From 36cd3c9f925b9307236505ae7ad1ad7ac4d4357c Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 9 Apr 2009 18:48:34 +0200 Subject: mutex: have non-spinning mutexes on s390 by default Impact: performance regression fix for s390 The adaptive spinning mutexes will not always do what one would expect on virtualized architectures like s390. Especially the cpu_relax() loop in mutex_spin_on_owner might hurt if the mutex holding cpu has been scheduled away by the hypervisor. We would end up in a cpu_relax() loop when there is no chance that the state of the mutex changes until the target cpu has been scheduled again by the hypervisor. For that reason we should change the default behaviour to no-spin on s390. We do have an instruction which allows to yield the current cpu in favour of a different target cpu. Also we have an instruction which allows us to figure out if the target cpu is physically backed. However we need to do some performance tests until we can come up with a solution that will do the right thing on s390. Signed-off-by: Heiko Carstens Acked-by: Peter Zijlstra Cc: Martin Schwidefsky Cc: Christian Borntraeger LKML-Reference: <20090409184834.7a0df7b2@osiris.boeblingen.de.ibm.com> Signed-off-by: Ingo Molnar --- kernel/mutex.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/mutex.c b/kernel/mutex.c index 5d79781394a3..507cf2b5e9f1 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -148,7 +148,8 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, preempt_disable(); mutex_acquire(&lock->dep_map, subclass, 0, ip); -#if defined(CONFIG_SMP) && !defined(CONFIG_DEBUG_MUTEXES) +#if defined(CONFIG_SMP) && !defined(CONFIG_DEBUG_MUTEXES) && \ + !defined(CONFIG_HAVE_DEFAULT_NO_SPIN_MUTEXES) /* * Optimistic spinning. * -- cgit v1.2.3 From d1e7e02f30be672c6f6ee40908be83877a0d49d1 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 2 Apr 2009 15:16:56 +0800 Subject: tracing: disable seeking for trace_pipe_raw Impact: disable pread() We set tracing_buffers_fops.llseek to no_llseek, but we can still perform pread() to read this file. That is not expected. This fix uses nonseekable_open() to disable it. tracing_buffers_fops.llseek is still set to no_llseek, it mark this file is a "non-seekable device" and is used by sys_splice(). See also do_splice() or manual of splice(2): ERRORS EINVAL Target file system doesn't support splicing; neither of the descriptors refers to a pipe; or offset given for non-seekable device. Signed-off-by: Lai Jiangshan Cc: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49D46668.8030806@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 9d28476a9851..24b0168f1a29 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3285,7 +3285,7 @@ static int tracing_buffers_open(struct inode *inode, struct file *filp) filp->private_data = info; - return 0; + return nonseekable_open(inode, filp); out: kfree(info); -- cgit v1.2.3 From ddd538f3e6a1a4bec2f6942f83a753263e6577b4 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 2 Apr 2009 15:16:59 +0800 Subject: tracing: allocate page when needed Impact: Cleanup Sometimes, we open trace_pipe_raw, but we don't read(2) it, we just splice(2) it, thus, the page is not used. Signed-off-by: Lai Jiangshan Cc: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49D4666B.4010608@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 24b0168f1a29..8e189ffb899a 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3277,19 +3277,13 @@ static int tracing_buffers_open(struct inode *inode, struct file *filp) info->tr = &global_trace; info->cpu = cpu; - info->spare = ring_buffer_alloc_read_page(info->tr->buffer); + info->spare = NULL; /* Force reading ring buffer for first read */ info->read = (unsigned int)-1; - if (!info->spare) - goto out; filp->private_data = info; return nonseekable_open(inode, filp); - - out: - kfree(info); - return -ENOMEM; } static ssize_t @@ -3304,6 +3298,11 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, if (!count) return 0; + if (!info->spare) + info->spare = ring_buffer_alloc_read_page(info->tr->buffer); + if (!info->spare) + return -ENOMEM; + /* Do we have previous read data to read? */ if (info->read < PAGE_SIZE) goto read; @@ -3342,7 +3341,8 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) { struct ftrace_buffer_info *info = file->private_data; - ring_buffer_free_read_page(info->tr->buffer, info->spare); + if (info->spare) + ring_buffer_free_read_page(info->tr->buffer, info->spare); kfree(info); return 0; -- cgit v1.2.3 From c7625a555f55d7ae49236cde551786c88f5a5ce1 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 2 Apr 2009 15:17:04 +0800 Subject: tracing: update file->f_pos when splice(2) it Impact: Cleanup These two lines: if (unlikely(*ppos)) return -ESPIPE; in tracing_buffers_splice_read() are not needed, VFS layer has disabled seek(2). We remove these two lines, and then we can update file->f_pos. And tracing_buffers_read() updates file->f_pos, this fix make tracing_buffers_splice_read() updates file->f_pos too. Signed-off-by: Lai Jiangshan Cc: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49D46670.4010503@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 8e189ffb899a..94629760dabf 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3428,13 +3428,6 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, int size, i; size_t ret; - /* - * We can't seek on a buffer input - */ - if (unlikely(*ppos)) - return -ESPIPE; - - for (i = 0; i < PIPE_BUFFERS && len; i++, len -= size) { struct page *page; int r; @@ -3474,6 +3467,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, spd.partial[i].offset = 0; spd.partial[i].private = (unsigned long)ref; spd.nr_pages++; + *ppos += size; } spd.nr_pages = i; -- cgit v1.2.3 From 93cfb3c9fd83d877a8f1ffad9ff862b617b32828 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 2 Apr 2009 15:17:08 +0800 Subject: tracing: fix splice return too large I got these from strace: splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 12288 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 12288 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 12288 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 16384 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 8192 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 8192 splice(0x3, 0, 0x5, 0, 0x1000, 0x1) = 8192 I wanted to splice_read 4096 bytes, but it returns 8192 or larger. It is because the return value of tracing_buffers_splice_read() does not include "zero out any left over data" bytes. But tracing_buffers_read() includes these bytes, we make them consistent. Signed-off-by: Lai Jiangshan Cc: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49D46674.9030804@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 94629760dabf..1ce5dc6372b8 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3428,7 +3428,19 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, int size, i; size_t ret; - for (i = 0; i < PIPE_BUFFERS && len; i++, len -= size) { + if (*ppos & (PAGE_SIZE - 1)) { + WARN_ONCE(1, "Ftrace: previous read must page-align\n"); + return -EINVAL; + } + + if (len & (PAGE_SIZE - 1)) { + WARN_ONCE(1, "Ftrace: splice_read should page-align\n"); + if (len < PAGE_SIZE) + return -EINVAL; + len &= PAGE_MASK; + } + + for (i = 0; i < PIPE_BUFFERS && len; i++, len -= PAGE_SIZE) { struct page *page; int r; @@ -3467,7 +3479,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, spd.partial[i].offset = 0; spd.partial[i].private = (unsigned long)ref; spd.nr_pages++; - *ppos += size; + *ppos += PAGE_SIZE; } spd.nr_pages = i; -- cgit v1.2.3 From 4d1f4372dbea068ba4ee3d98231133a4a4ee15bd Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Fri, 10 Apr 2009 08:48:36 +0800 Subject: tracing: fix document references When moving documents to Documentation/trace/, I forgot to grep Kconfig to find out those references. Signed-off-by: Li Zefan Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Pekka Enberg Cc: Pekka Paalanen Cc: eduard.munteanu@linux360.ro LKML-Reference: <49DE97EF.7080208@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 2246141bda4d..417d1985e299 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -312,7 +312,7 @@ config KMEMTRACE and profile kernel code. This requires an userspace application to use. See - Documentation/vm/kmemtrace.txt for more information. + Documentation/trace/kmemtrace.txt for more information. Saying Y will make the kernel somewhat larger and slower. However, if you disable kmemtrace at run-time or boot-time, the performance @@ -403,7 +403,7 @@ config MMIOTRACE implementation and works via page faults. Tracing is disabled by default and can be enabled at run-time. - See Documentation/tracers/mmiotrace.txt. + See Documentation/trace/mmiotrace.txt. If you are not helping to develop drivers, say N. config MMIOTRACE_TEST -- cgit v1.2.3 From 0462b5664b2bda5a18fef7efb5bb32ce36590c1a Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Wed, 8 Apr 2009 17:00:13 +0800 Subject: ftrace: Output REC->var instead of __entry->var for trace format print fmt: "irq=%d return=%s", __entry->irq, __entry->ret ? \"handled\" : \"unhandled\" "__entry" should be convert to "REC" by __stringify() macro. Signed-off-by: Zhao Lei Acked-by: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49DC679D.2090901@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_events_stage_2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_stage_2.h b/kernel/trace/trace_events_stage_2.h index 30743f7d4110..d363c6672c6c 100644 --- a/kernel/trace/trace_events_stage_2.h +++ b/kernel/trace/trace_events_stage_2.h @@ -105,10 +105,10 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags) \ return 0; #undef __entry -#define __entry "REC" +#define __entry REC #undef TP_printk -#define TP_printk(fmt, args...) "%s, %s\n", #fmt, #args +#define TP_printk(fmt, args...) "%s, %s\n", #fmt, __stringify(args) #undef TP_fast_assign #define TP_fast_assign(args...) args -- cgit v1.2.3 From d6de2c80e9d758d2e36c21699117db6178c0f517 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 10 Apr 2009 12:17:41 -0700 Subject: async: Fix module loading async-work regression Several drivers use asynchronous work to do device discovery, and we synchronize with them in the compiled-in case before we actually try to mount root filesystems etc. However, when compiled as modules, that synchronization is missing - the module loading completes, but the driver hasn't actually finished probing for devices, and that means that any user mode that expects to use the devices after the 'insmod' is now potentially broken. We already saw one case of a similar issue in the ACPI battery code, where the kernel itself expected the module to be all done, and unmapped the init memory - but the async device discovery was still running. That got hacked around by just removing the "__init" (see commit 5d38258ec026921a7b266f4047ebeaa75db358e5 "ACPI battery: fix async boot oops"), but the real fix is to just make the module loading wait for all async work to be completed. It will slow down module loading, but since common devices should be built in anyway, and since the bug is really annoying and hard to handle from user space (and caused several S3 resume regressions), the simple fix to wait is the right one. This fixes at least http://bugzilla.kernel.org/show_bug.cgi?id=13063 but probably a few other bugzilla entries too (12936, for example), and is confirmed to fix Rafael's storage driver breakage after resume bug report (no bugzilla entry). We should also be able to now revert that ACPI battery fix. Reported-and-tested-by: Rafael J. Wysocki Tested-by: Heinz Diehl Acked-by: Arjan van de Ven Signed-off-by: Linus Torvalds --- kernel/module.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index 05f014efa32c..e797812a4d95 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2388,6 +2388,9 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_LIVE, mod); + /* We need to finish all async code before the module init sequence is done */ + async_synchronize_full(); + mutex_lock(&module_mutex); /* Drop initial reference. */ module_put(mod); -- cgit v1.2.3 From 8433a40eb7f2c4883ad57f9900f63e4d59240eb7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sat, 11 Apr 2009 15:52:18 +0800 Subject: tracing/filters: NIL-terminate user input filter MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Make sure messages from user space are NIL-terminated strings, otherwise we could dump random memory while reading filter file. Try this: # echo 'parent_comm ==' > events/sched/sched_process_fork/filter # cat events/sched/sched_process_fork/filter parent_comm == � Signed-off-by: Li Zefan Acked-by: Tom Zanussi Acked-by: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49E04C32.6060508@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_events.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 64ec4d278ffb..054bc1802bcd 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -503,6 +503,7 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, if (copy_from_user(&buf, ubuf, cnt)) return -EFAULT; + buf[cnt] = '\0'; pred = kzalloc(sizeof(*pred), GFP_KERNEL); if (!pred) @@ -569,6 +570,7 @@ subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, if (copy_from_user(&buf, ubuf, cnt)) return -EFAULT; + buf[cnt] = '\0'; pred = kzalloc(sizeof(*pred), GFP_KERNEL); if (!pred) -- cgit v1.2.3 From bcabd91c271e50eebc0cb9220ac92700332b452e Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sat, 11 Apr 2009 15:52:35 +0800 Subject: tracing/filters: fix NULL pointer dereference Try this, and you'll see NULL pointer dereference bug: # echo -n 'parent_comm ==' > sched/sched_process_fork/filter Because we passed NULL ptr to simple_strtoull(). Signed-off-by: Li Zefan Acked-by: Tom Zanussi Acked-by: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49E04C43.1050504@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_events_filter.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 026be412f356..9d2162fd2305 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -410,6 +410,11 @@ int filter_parse(char **pbuf, struct filter_pred *pred) } } + if (!val_str) { + pred->field_name = NULL; + return -EINVAL; + } + pred->field_name = kstrdup(pred->field_name, GFP_KERNEL); if (!pred->field_name) return -ENOMEM; -- cgit v1.2.3 From a3e0ab050774117d4a6173087c8bf3888662a83f Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sat, 11 Apr 2009 15:52:51 +0800 Subject: tracing/filters: allow user input integer to be oct or hex Before patch: # echo 'parent_pid == 0x10' > events/sched/sched_process_fork/filter # cat sched/sched_process_fork/filter parent_pid == 0 After patch: # cat sched/sched_process_fork/filter parent_pid == 16 Also check the input more strictly. Signed-off-by: Li Zefan Acked-by: Tom Zanussi Acked-by: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49E04C53.4010600@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_events_filter.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 9d2162fd2305..49b3ef54ec46 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -419,12 +419,13 @@ int filter_parse(char **pbuf, struct filter_pred *pred) if (!pred->field_name) return -ENOMEM; - pred->val = simple_strtoull(val_str, &tmp, 10); + pred->val = simple_strtoull(val_str, &tmp, 0); if (tmp == val_str) { pred->str_val = kstrdup(val_str, GFP_KERNEL); if (!pred->str_val) return -ENOMEM; - } + } else if (*tmp != '\0') + return -EINVAL; return 0; } -- cgit v1.2.3 From 44e9c8b7adc52079f0535f9de0c2c2477831389b Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sat, 11 Apr 2009 15:55:28 +0800 Subject: tracing/filters: return proper error code when writing filter file - propagate return value of filter_add_pred() to the user - return -ENOSPC but not -ENOMEM or -EINVAL when the filter array is full Signed-off-by: Li Zefan Acked-by: Tom Zanussi Acked-by: Frederic Weisbecker Cc: Steven Rostedt LKML-Reference: <49E04CF0.3010105@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_events.c | 10 ++++++---- kernel/trace/trace_events_filter.c | 4 ++-- 2 files changed, 8 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 054bc1802bcd..576f4fa2af0d 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -521,9 +521,10 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, return cnt; } - if (filter_add_pred(call, pred)) { + err = filter_add_pred(call, pred); + if (err < 0) { filter_free_pred(pred); - return -EINVAL; + return err; } *ppos += cnt; @@ -588,10 +589,11 @@ subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, return cnt; } - if (filter_add_subsystem_pred(system, pred)) { + err = filter_add_subsystem_pred(system, pred); + if (err < 0) { filter_free_subsystem_preds(system); filter_free_pred(pred); - return -EINVAL; + return err; } *ppos += cnt; diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 49b3ef54ec46..e03cbf1e38f3 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -215,7 +215,7 @@ static int __filter_add_pred(struct ftrace_event_call *call, } } - return -ENOMEM; + return -ENOSPC; } static int is_string_field(const char *type) @@ -319,7 +319,7 @@ int filter_add_subsystem_pred(struct event_subsystem *system, } if (i == MAX_FILTER_PRED) - return -EINVAL; + return -ENOSPC; events_for_each(call) { int err; -- cgit v1.2.3 From 9eeba6138cefc0435695463ddadb0d95e0a6bcd2 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 11 Apr 2009 03:17:17 +0200 Subject: lockdep: warn about lockdep disabling after kernel taint Impact: provide useful missing info for developers Kernel taint can occur in several situations such as warnings, load of prorietary or staging modules, bad page, etc... But when such taint happens, a developer might still be working on the kernel, expecting that lockdep is still enabled. But a taint disables lockdep without ever warning about it. Such a kernel behaviour doesn't really help for kernel development. This patch adds this missing warning. Since the taint is done most of the time after the main message that explain the real source issue, it seems safe to warn about it inside add_taint() so that it appears at last, without hurting the main information. v2: Use a generic helper to disable lockdep instead of an open coded xchg(). Signed-off-by: Frederic Weisbecker Cc: Peter Zijlstra LKML-Reference: <1239412638-6739-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/panic.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index 3fd8c5bf8b39..940ca14f6dbf 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -213,8 +213,14 @@ unsigned long get_taint(void) void add_taint(unsigned flag) { - /* can't trust the integrity of the kernel anymore: */ - debug_locks = 0; + /* + * Can't trust the integrity of the kernel anymore. + * We don't call directly debug_locks_off() because the issue + * is not necessarily serious enough to set oops_in_progress to 1 + */ + if (__debug_locks_off()) + printk(KERN_WARNING "Disabling lockdep due to kernel taint\n"); + set_bit(flag, &tainted_mask); } EXPORT_SYMBOL(add_taint); -- cgit v1.2.3 From 574bbe782057fdf0490dc7dec906a2dc26363e20 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 11 Apr 2009 03:17:18 +0200 Subject: lockdep: continue lock debugging despite some taints Impact: broaden lockdep checks Lockdep is disabled after any kernel taints. This might be convenient to ignore bad locking issues which sources come from outside the kernel tree. Nevertheless, it might be a frustrating experience for the staging developers or those who experience a warning but are focused on another things that require lockdep. The v2 of this patch simply don't disable anymore lockdep in case of TAINT_CRAP and TAINT_WARN events. Signed-off-by: Frederic Weisbecker Cc: LTP Cc: Peter Zijlstra Cc: Greg KH LKML-Reference: <1239412638-6739-2-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/panic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index 940ca14f6dbf..934fb377f4b3 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -217,8 +217,10 @@ void add_taint(unsigned flag) * Can't trust the integrity of the kernel anymore. * We don't call directly debug_locks_off() because the issue * is not necessarily serious enough to set oops_in_progress to 1 + * Also we want to keep up lockdep for staging development and + * post-warning case. */ - if (__debug_locks_off()) + if (flag != TAINT_CRAP && flag != TAINT_WARN && __debug_locks_off()) printk(KERN_WARNING "Disabling lockdep due to kernel taint\n"); set_bit(flag, &tainted_mask); -- cgit v1.2.3 From c751085943362143f84346d274e0011419c84202 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 12 Apr 2009 20:06:56 +0200 Subject: PM/Hibernate: Wait for SCSI devices scan to complete during resume There is a race between resume from hibernation and the asynchronous scanning of SCSI devices and to prevent it from happening we need to call scsi_complete_async_scans() during resume from hibernation. In addition, if the resume from hibernation is userland-driven, it's better to wait for all device probes in the kernel to complete before attempting to open the resume device. Signed-off-by: Rafael J. Wysocki Acked-by: Arjan van de Ven Signed-off-by: Linus Torvalds --- kernel/power/disk.c | 8 ++++++++ kernel/power/user.c | 9 +++++++++ 2 files changed, 17 insertions(+) (limited to 'kernel') diff --git a/kernel/power/disk.c b/kernel/power/disk.c index 5f21ab2bbcdf..0854770b63b9 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include "power.h" @@ -644,6 +645,13 @@ static int software_resume(void) if (noresume) return 0; + /* + * We can't depend on SCSI devices being available after loading one of + * their modules if scsi_complete_async_scans() is not called and the + * resume device usually is a SCSI one. + */ + scsi_complete_async_scans(); + /* * name_to_dev_t() below takes a sysfs buffer mutex when sysfs * is configured into the kernel. Since the regular hibernate diff --git a/kernel/power/user.c b/kernel/power/user.c index 6c85359364f2..ed97375daae9 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -24,6 +24,7 @@ #include #include #include +#include #include @@ -92,6 +93,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) filp->private_data = data; memset(&data->handle, 0, sizeof(struct snapshot_handle)); if ((filp->f_flags & O_ACCMODE) == O_RDONLY) { + /* Hibernating. The image device should be accessible. */ data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device, 0, NULL) : -1; data->mode = O_RDONLY; @@ -99,6 +101,13 @@ static int snapshot_open(struct inode *inode, struct file *filp) if (error) pm_notifier_call_chain(PM_POST_HIBERNATION); } else { + /* + * Resuming. We may need to wait for the image device to + * appear. + */ + wait_for_device_probe(); + scsi_complete_async_scans(); + data->swap = -1; data->mode = O_WRONLY; error = pm_notifier_call_chain(PM_RESTORE_PREPARE); -- cgit v1.2.3