From 68cacd29167b1926d237bd1b153aa2a990201729 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 23 Mar 2011 16:03:06 +0100 Subject: perf_events: Fix stale ->cgrp pointer in update_cgrp_time_from_cpuctx() This patch solves a stale pointer problem in update_cgrp_time_from_cpuctx(). The cpuctx->cgrp was not cleared on all possible event exit paths, including: close() perf_release() perf_release_kernel() list_del_event() This patch fixes list_del_event() to clear cpuctx->cgrp when there are no cgroup events left in the context. [ This second version makes the code compile when CONFIG_CGROUP_PERF is not enabled. We unconditionally define perf_cpu_context->cgrp. ] Signed-off-by: Stephane Eranian Cc: peterz@infradead.org Cc: perfmon2-devel@lists.sf.net Cc: paulus@samba.org Cc: davem@davemloft.net LKML-Reference: <20110323150306.GA1580@quad> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'kernel/perf_event.c') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 3472bb1a070..0c714226ae0 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -941,6 +941,7 @@ static void perf_group_attach(struct perf_event *event) static void list_del_event(struct perf_event *event, struct perf_event_context *ctx) { + struct perf_cpu_context *cpuctx; /* * We can have double detach due to exit/hot-unplug + close. */ @@ -949,8 +950,17 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) event->attach_state &= ~PERF_ATTACH_CONTEXT; - if (is_cgroup_event(event)) + if (is_cgroup_event(event)) { ctx->nr_cgroups--; + cpuctx = __get_cpu_context(ctx); + /* + * if there are no more cgroup events + * then cler cgrp to avoid stale pointer + * in update_cgrp_time_from_cpuctx() + */ + if (!ctx->nr_cgroups) + cpuctx->cgrp = NULL; + } ctx->nr_events--; if (event->attr.inherit_stat) -- cgit v1.2.3-70-g09d2 From 880f57318450dbead6a03f9e31a1468924d6dd88 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 23 Mar 2011 19:29:39 +0100 Subject: perf: Better fit max unprivileged mlock pages for tools needs The maximum kilobytes of locked memory that an unprivileged user can reserve is of 512 kB = 128 pages by default, scaled to the number of onlined CPUs, which fits well with the tools that use 128 data pages by default. However tools actually use 129 pages, because they need one more for the user control page. Thus the default mlock threshold is not sufficient for the default tools needs and we always end up to evaluate the constant mlock rlimit policy, which doesn't have this scaling with the number of online CPUs. Hence, on systems that have more than 16 CPUs, we overlap the rlimit threshold and fail to mmap: $ perf record ls Error: failed to mmap with 1 (Operation not permitted) Just increase the max unprivileged mlock threshold by one page so that it supports well perf tools even after 16 CPUs. Reported-by: Han Pingtian Reported-by: Peter Zijlstra Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Frederic Weisbecker Acked-by: Arnaldo Carvalho de Melo Cc: Stephane Eranian Cc: Stable LKML-Reference: <1300904979-5508-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/perf_event.c') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 0c714226ae0..c75925c4d1e 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -145,7 +145,8 @@ static struct srcu_struct pmus_srcu; */ int sysctl_perf_event_paranoid __read_mostly = 1; -int sysctl_perf_event_mlock __read_mostly = 512; /* 'free' kb per user */ +/* Minimum for 128 pages + 1 for the user control page */ +int sysctl_perf_event_mlock __read_mostly = 516; /* 'free' kb per user */ /* * max perf event sample rate -- cgit v1.2.3-70-g09d2 From 20443384fe090c5f8aeb016e7e85659c5bbdd69f Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 31 Mar 2011 03:33:29 +0200 Subject: perf: Rebase max unprivileged mlock threshold on top of page size Ensure we allow 512 kiB + 1 page for user control without assuming a 4096 bytes page size. Reported-by: Peter Zijlstra Signed-off-by: Frederic Weisbecker Signed-off-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: Stephane Eranian Cc: LKML-Reference: <1301535209-9679-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/perf_event.c') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index c75925c4d1e..261690923ff 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -145,8 +145,8 @@ static struct srcu_struct pmus_srcu; */ int sysctl_perf_event_paranoid __read_mostly = 1; -/* Minimum for 128 pages + 1 for the user control page */ -int sysctl_perf_event_mlock __read_mostly = 516; /* 'free' kb per user */ +/* Minimum for 512 kiB + 1 user control page */ +int sysctl_perf_event_mlock __read_mostly = 512 + (PAGE_SIZE / 1024); /* 'free' kiB per user */ /* * max perf event sample rate -- cgit v1.2.3-70-g09d2 From fd1edb3aa2c1d92618d8f0c6d15d44ea41fcac6a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 28 Mar 2011 13:13:56 +0200 Subject: perf: Fix task_struct reference leak sys_perf_event_open() had an imbalance in the number of task refs it took causing memory leakage Cc: Jiri Olsa Cc: Oleg Nesterov Cc: stable@kernel.org # .37+ Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel/perf_event.c') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 261690923ff..27960f114ef 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -6531,6 +6531,11 @@ SYSCALL_DEFINE5(perf_event_open, goto err_alloc; } + if (task) { + put_task_struct(task); + task = NULL; + } + /* * Look up the group leader (we will attach this event to it): */ -- cgit v1.2.3-70-g09d2 From e566b76ed30768140df8f0023904aed5a41244f7 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 6 Apr 2011 02:54:54 +0200 Subject: perf_event: Fix cgrp event scheduling bug in perf_enable_on_exec() There is a bug in perf_event_enable_on_exec() when cgroup events are active on a CPU: the cgroup events may be scheduled twice causing event state corruptions which eventually may lead to kernel panics. The reason is that the function needs to first schedule out the cgroup events, just like for the per-thread events. The cgroup event are scheduled back in automatically from the perf_event_context_sched_in() function. The patch also adds a WARN_ON_ONCE() is perf_cgroup_switch() to catch any bogus state. Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110406005454.GA1062@quad Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'kernel/perf_event.c') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 27960f114ef..8e81a9860a0 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -364,6 +364,7 @@ void perf_cgroup_switch(struct task_struct *task, int mode) } if (mode & PERF_CGROUP_SWIN) { + WARN_ON_ONCE(cpuctx->cgrp); /* set cgrp before ctxsw in to * allow event_filter_match() to not * have to pass task around @@ -2423,6 +2424,14 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx) if (!ctx || !ctx->nr_events) goto out; + /* + * We must ctxsw out cgroup events to avoid conflict + * when invoking perf_task_event_sched_in() later on + * in this function. Otherwise we end up trying to + * ctxswin cgroup events which are already scheduled + * in. + */ + perf_cgroup_sched_out(current); task_ctx_sched_out(ctx, EVENT_ALL); raw_spin_lock(&ctx->lock); @@ -2447,6 +2456,9 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx) raw_spin_unlock(&ctx->lock); + /* + * Also calls ctxswin for cgroup events, if any: + */ perf_event_context_sched_in(ctx, ctx->task); out: local_irq_restore(flags); -- cgit v1.2.3-70-g09d2