Skip to contentSkip to conversation

perf: Fix the mlock accounting, again

PATCH In review Alexander Shishkin 2 participants 3 messages Nov 15, 2019 – Nov 18, 2019 perf events

lore
Share & tools
lore

Alexander Shishkin

author reviewer
#0 lore raw

PATCH 1/1 perf: 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 changed 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.

---
 kernel/events/core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
kernel/events/core.c +2 -4
index 16d80ad8d6d7..522438887206 100644
@@ -5664,10 +5664,8 @@ 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);
Signature · 1 line
2.24.0
author reviewer
#1 lore raw
Quoted 7 lines · Alexander Shishkin

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 changed partially to both user->locked_vm

s/changed/charged/

#2 lore raw
In reply to Alexander Shishkin
Quoted 34 lines · Alexander Shishkin

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 changed 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.

---
 kernel/events/core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
kernel/events/core.c +0 -0
index 16d80ad8d6d7..522438887206 100644
@@ -5664,10 +5664,8 @@ static void perf_mmap_close(struct vm_area_struct *vma)
 		perf_pmu_output_stop(event);
 

kernel/events/core.c static void perf_mmap_close(struct vm_area_struct *vma) · line 5671

Quoted 10 lines
kernel/events/core.c +2 -4
 		/* 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);

I have tested your patch on our test suite which discovered the original issue. Your patch works nicely, thanks for fixing this.

Signature · 5 lines
Thomas Richter, Dept 3252, IBM s390 Linux Development, Boeblingen, Germany
--
Vorsitzender des Aufsichtsrats: Matthias Hartmann
Geschäftsführung: Dirk Wittkopp
Sitz der Gesellschaft: Böblingen / Registergericht: Amtsgericht Stuttgart, HRB 243294