From 36b3db03b4741b8935b68fffc7e69951d8d70a89 Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Fri, 15 Nov 2019 18:08:18 +0200 Subject: perf/core: Fix the mlock accounting, again Commit: 5e6c3c7b1ec2 ("perf/aux: Fix tracking of auxiliary trace buffer allocation") tried to guess the correct combination of arithmetic operations that would undo the AUX buffer's mlock accounting, and failed, leaking the bottom part when an allocation needs to be charged partially to both user->locked_vm and mm->pinned_vm, eventually leaving the user with no locked bonus: $ perf record -e intel_pt//u -m1,128 uname [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.061 MB perf.data ] $ perf record -e intel_pt//u -m1,128 uname Permission error mapping pages. Consider increasing /proc/sys/kernel/perf_event_mlock_kb, or try again with a smaller value of -m/--mmap_pages. (current value: 1,128) Fix this by subtracting both locked and pinned counts when AUX buffer is unmapped. Reported-by: Thomas Richter Tested-by: Thomas Richter Signed-off-by: Alexander Shishkin Acked-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mark Rutland Cc: Namhyung Kim Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Signed-off-by: Ingo Molnar --- kernel/events/core.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 00a014670ed0..8f66a4833ded 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5607,10 +5607,8 @@ static void perf_mmap_close(struct vm_area_struct *vma) perf_pmu_output_stop(event); /* now it's safe to free the pages */ - if (!rb->aux_mmap_locked) - atomic_long_sub(rb->aux_nr_pages, &mmap_user->locked_vm); - else - atomic64_sub(rb->aux_mmap_locked, &vma->vm_mm->pinned_vm); + atomic_long_sub(rb->aux_nr_pages - rb->aux_mmap_locked, &mmap_user->locked_vm); + atomic64_sub(rb->aux_mmap_locked, &vma->vm_mm->pinned_vm); /* this has to be the last one */ rb_free_aux(rb); -- cgit v1.2.3-58-ga151 From c4b75479741c9c3a4f0abff5baa5013d27640ac1 Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Wed, 20 Nov 2019 19:06:40 +0200 Subject: perf/core: Make the mlock accounting simple again Commit: d44248a41337 ("perf/core: Rework memory accounting in perf_mmap()") does a lot of things to the mlock accounting arithmetics, while the only thing that actually needed to happen is subtracting the part that is charged to the mm from the part that is charged to the user, so that the former isn't charged twice. Signed-off-by: Alexander Shishkin Acked-by: Song Liu Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Peter Zijlstra Cc: Wanpeng Li Cc: Yauheni Kaliuta Cc: songliubraving@fb.com Link: https://lkml.kernel.org/r/20191120170640.54123-1-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'kernel/events') diff --git a/kernel/events/core.c b/kernel/events/core.c index 8f66a4833ded..7e8980d0b997 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5825,13 +5825,7 @@ accounting: user_locked = atomic_long_read(&user->locked_vm) + user_extra; - if (user_locked <= user_lock_limit) { - /* charge all to locked_vm */ - } else if (atomic_long_read(&user->locked_vm) >= user_lock_limit) { - /* charge all to pinned_vm */ - extra = user_extra; - user_extra = 0; - } else { + if (user_locked > user_lock_limit) { /* * charge locked_vm until it hits user_lock_limit; * charge the rest from pinned_vm -- cgit v1.2.3-58-ga151