From bbe1a59b3a3d4af3869647d294618a23f6c9c6a4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 22 Jan 2007 20:40:33 -0800 Subject: [PATCH] fix "kvm: add vm exit profiling" export profile_hits() on !SMP too. Cc: Ingo Molnar Cc: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/profile.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/profile.c b/kernel/profile.c index a6574a18514e..d6579d511069 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -331,7 +331,6 @@ out: local_irq_restore(flags); put_cpu(); } -EXPORT_SYMBOL_GPL(profile_hits); static int __devinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) @@ -401,6 +400,8 @@ void profile_hits(int type, void *__pc, unsigned int nr_hits) } #endif /* !CONFIG_SMP */ +EXPORT_SYMBOL_GPL(profile_hits); + void profile_tick(int type) { struct pt_regs *regs = get_irq_regs(); -- cgit v1.2.3 From 1b5180b65122666a36a1a232b7b9b38b21a9dcdd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 23 Jan 2007 10:45:50 +0100 Subject: [PATCH] notifiers: fix blocking_notifier_call_chain() scalability while lock-profiling the -rt kernel i noticed weird contention during mmap-intense workloads, and the tracer showed the following gem, in one of our MM hotpaths: threaded-2771 1.... 65us : sys_munmap (sysenter_do_call) threaded-2771 1.... 66us : profile_munmap (sys_munmap) threaded-2771 1.... 66us : blocking_notifier_call_chain (profile_munmap) threaded-2771 1.... 66us : rt_down_read (blocking_notifier_call_chain) ouch! a global rw-semaphore taken in one of the most performance- sensitive codepaths of the kernel. And i dont even have oprofile enabled! All distro kernels have CONFIG_PROFILING enabled, so this scalability problem affects the majority of Linux users. The fix is to enhance blocking_notifier_call_chain() to only take the lock if there appears to be work on the call-chain. With this patch applied i get nicely saturated system, and much higher munmap performance, on SMP systems. And as a bonus this also fixes a similar scalability bottleneck in the thread-exit codepath: profile_task_exit() ... Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Acked-by: Nick Piggin Signed-off-by: Linus Torvalds --- kernel/sys.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index c7675c1bfdf2..6e2101dec0fc 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -323,11 +323,18 @@ EXPORT_SYMBOL_GPL(blocking_notifier_chain_unregister); int blocking_notifier_call_chain(struct blocking_notifier_head *nh, unsigned long val, void *v) { - int ret; + int ret = NOTIFY_DONE; - down_read(&nh->rwsem); - ret = notifier_call_chain(&nh->head, val, v); - up_read(&nh->rwsem); + /* + * We check the head outside the lock, but if this access is + * racy then it does not matter what the result of the test + * is, we re-check the list after having taken the lock anyway: + */ + if (rcu_dereference(nh->head)) { + down_read(&nh->rwsem); + ret = notifier_call_chain(&nh->head, val, v); + up_read(&nh->rwsem); + } return ret; } -- cgit v1.2.3 From 8528b0f1de1101c6002036fd53638fb21111d0ea Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 23 Jan 2007 14:16:31 -0800 Subject: Clear spurious irq stat information when adding irq handler Any newly added irq handler may obviously make any old spurious irq status invalid, since the new handler may well be the thing that is supposed to handle any interrupts that came in. So just clear the statistics when adding handlers. Pointed-out-by: Alan Cox Acked-by: Thomas Gleixner Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- kernel/irq/manage.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index b385878c6e80..8b961adc3bd2 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -315,6 +315,9 @@ int setup_irq(unsigned int irq, struct irqaction *new) /* Undo nested disables: */ desc->depth = 1; } + /* Reset broken irq detection when installing new handler */ + desc->irq_count = 0; + desc->irqs_unhandled = 0; spin_unlock_irqrestore(&desc->lock, flags); new->irq = irq; -- cgit v1.2.3