mm/page_counter: mark intentional data races
Subsystems (2)
PATCH 1/1 mm/page_counter: mark intentional data races
The commit 3e32cb2e0a12 ("mm: memcontrol: lockless page counters") had memcg->memsw->failcnt and ->watermark could be accessed concurrently as reported by KCSAN,
Reported by Kernel Concurrency Sanitizer on:
BUG: KCSAN: data-race in page_counter_try_charge / page_counter_try_charge
read to 0xffff8fb18c4cd190 of 8 bytes by task 1081 on cpu 59:
page_counter_try_charge+0x4d/0x150 mm/page_counter.c:138 try_charge+0x131/0xd50 __memcg_kmem_charge_memcg+0x58/0x140 __memcg_kmem_charge+0xcc/0x280 __alloc_pages_nodemask+0x1e1/0x450 alloc_pages_current+0xa6/0x120 pte_alloc_one+0x17/0xd0 __pte_alloc+0x3a/0x1f0 copy_p4d_range+0xc36/0x1990 copy_page_range+0x21d/0x360 dup_mmap+0x5f5/0x7a0 dup_mm+0xa2/0x240 copy_process+0x1b3f/0x3460 _do_fork+0xaa/0xa20 __x64_sys_clone+0x13b/0x170 do_syscall_64+0x91/0xb47 entry_SYSCALL_64_after_hwframe+0x49/0xbe
write to 0xffff8fb18c4cd190 of 8 bytes by task 1153 on cpu 120:
page_counter_try_charge+0x5b/0x150 mm/page_counter.c:139 try_charge+0x131/0xd50 mem_cgroup_try_charge+0x159/0x460 mem_cgroup_try_charge_delay+0x3d/0xa0 wp_page_copy+0x14d/0x930 do_wp_page+0x107/0x7b0 __handle_mm_fault+0xce6/0xd40 handle_mm_fault+0xfc/0x2f0 do_page_fault+0x263/0x6f9 page_fault+0x34/0x40
Since the failcnt and watermark are tolerant of some inaccuracy, a data race will not be harmful, thus mark them as intentional data races with the data_race() macro.
--- mm/page_counter.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
mm/page_counter.c +5 -5
@@ -82,8 +82,8 @@ void page_counter_charge(struct page_counter *counter, unsigned long nr_pages)* This is indeed racy, but we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new); } }
@@ -126,7 +126,7 @@ bool page_counter_try_charge(struct page_counter *counter,* This is racy, but we can live with some * inaccuracy in the failcnt. */ - c->failcnt++; +data_race (c->failcnt++); *fail = c; goto failed; }
@@ -135,8 +135,8 @@ bool page_counter_try_charge(struct page_counter *counter,* Just like with failcnt, we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new); }return true;
Signature · 1 line
2.21.0 (Apple Git-122.2)
Michal Hocko
maintainerQuoted 38 lines · Qian Cai
The commit
3e32cb2e0a12("mm: memcontrol: lockless page counters") hadmemcg->memsw->failcntand ->watermark could be accessed concurrently as reported by KCSAN,Reported by Kernel Concurrency Sanitizer on:
BUG: KCSAN: data-race in page_counter_try_charge / page_counter_try_chargeread to
0xffff8fb18c4cd190of 8 bytes by task 1081 on cpu 59:page_counter_try_charge+0x4d/0x150 mm/page_counter.c:138 try_charge+0x131/0xd50 __memcg_kmem_charge_memcg+0x58/0x140 __memcg_kmem_charge+0xcc/0x280 __alloc_pages_nodemask+0x1e1/0x450 alloc_pages_current+0xa6/0x120 pte_alloc_one+0x17/0xd0 __pte_alloc+0x3a/0x1f0 copy_p4d_range+0xc36/0x1990 copy_page_range+0x21d/0x360 dup_mmap+0x5f5/0x7a0 dup_mm+0xa2/0x240 copy_process+0x1b3f/0x3460 _do_fork+0xaa/0xa20 __x64_sys_clone+0x13b/0x170 do_syscall_64+0x91/0xb47 entry_SYSCALL_64_after_hwframe+0x49/0xbewrite to
0xffff8fb18c4cd190of 8 bytes by task 1153 on cpu 120:page_counter_try_charge+0x5b/0x150 mm/page_counter.c:139 try_charge+0x131/0xd50 mem_cgroup_try_charge+0x159/0x460 mem_cgroup_try_charge_delay+0x3d/0xa0 wp_page_copy+0x14d/0x930 do_wp_page+0x107/0x7b0 __handle_mm_fault+0xce6/0xd40 handle_mm_fault+0xfc/0x2f0 do_page_fault+0x263/0x6f9 page_fault+0x34/0x40
Quoted 3 lines
Since the failcnt and watermark are tolerant of some inaccuracy, a data race will not be harmful, thus mark them as intentional data races with the
data_race()macro.
I am not familiar with KCSAN and git grep for data_race on the current linux-next doesn't really show any users of this macro. Is there a general consensus that data_race is going to be used to silence all KCSAN false positives?
Quoted 42 lines
Signed-off-by Qian Cai <[email protected]>--- mm/page_counter.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)mm/page_counter.c +5
-5@@ -82,8 +82,8 @@ void page_counter_charge(struct page_counter *counter, unsigned long nr_pages)* This is indeed racy, but we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new); } }@@ -126,7 +126,7 @@ bool page_counter_try_charge(struct page_counter *counter,* This is racy, but we can live with some * inaccuracy in the failcnt. */ - c->failcnt++; +data_race (c->failcnt++); *fail = c; goto failed; }@@ -135,8 +135,8 @@ bool page_counter_try_charge(struct page_counter *counter,* Just like with failcnt, we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new); }return true;Signature · 1 line
2.21.0 (Apple Git-122.2)
Signature · 2 lines
Michal Hocko SUSE Labs
Quoted 6 lines · Michal Hocko
Quoted 4 lines · Qian Cai
The commit
3e32cb2e0a12("mm: memcontrol: lockless page counters") hadmemcg->memsw->failcntand ->watermark could be accessed concurrently as reported by KCSAN,
Quoted 6 lines
Quoted 6 lines
Reported by Kernel Concurrency Sanitizer on:
BUG: KCSAN: data-race in page_counter_try_charge / page_counter_try_chargeread to
0xffff8fb18c4cd190of 8 bytes by task 1081 on cpu 59:page_counter_try_charge+0x4d/0x150 mm/page_counter.c:138 try_charge+0x131/0xd50
Why are the line numbers for the remaining symbols missing? Doesn't scripts/decode_stacktrace.sh give you all line numbers?
[ As an aside: if you want to use what syzbot uses to put line numbers on symbols, which is a bit faster:
https://github.com/google/syzkaller/tree/master/tools/syz-symbolize
https://github.com/google/syzkaller/blob/master/docs/linux/setup.md
then 'go build tools/syz-symbolize'. ]
Quoted 28 lines
Quoted 28 lines
__memcg_kmem_charge_memcg+0x58/0x140 __memcg_kmem_charge+0xcc/0x280 __alloc_pages_nodemask+0x1e1/0x450 alloc_pages_current+0xa6/0x120 pte_alloc_one+0x17/0xd0 __pte_alloc+0x3a/0x1f0 copy_p4d_range+0xc36/0x1990 copy_page_range+0x21d/0x360 dup_mmap+0x5f5/0x7a0 dup_mm+0xa2/0x240 copy_process+0x1b3f/0x3460 _do_fork+0xaa/0xa20 __x64_sys_clone+0x13b/0x170 do_syscall_64+0x91/0xb47 entry_SYSCALL_64_after_hwframe+0x49/0xbewrite to
0xffff8fb18c4cd190of 8 bytes by task 1153 on cpu 120:page_counter_try_charge+0x5b/0x150 mm/page_counter.c:139 try_charge+0x131/0xd50 mem_cgroup_try_charge+0x159/0x460 mem_cgroup_try_charge_delay+0x3d/0xa0 wp_page_copy+0x14d/0x930 do_wp_page+0x107/0x7b0 __handle_mm_fault+0xce6/0xd40 handle_mm_fault+0xfc/0x2f0 do_page_fault+0x263/0x6f9 page_fault+0x34/0x40
Quoted 8 lines
Quoted 3 lines
Since the failcnt and watermark are tolerant of some inaccuracy, a data race will not be harmful, thus mark them as intentional data races with the
data_race()macro.I am not familiar with KCSAN and git grep for
data_raceon the current linux-next doesn't really show any users of this macro. Is there a general consensus thatdata_raceis going to be used to silence all KCSAN false positives?
It was discussed here:
https://lore.kernel.org/linux-fsdevel/CAHk-=wg5CkOEF8DTez1Qu0XTEFw_oHhxN98bDnFqbY7HL5AB2g@mail.gmail.com/
If they're intentional data races that should remain, data_race() is one option. There are 4 options (other than address the data race) to deal with 'false positives':
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/dev-tools/kcsan.rst#n101
That being said, every use of data_race() needs to be justified, and not just applied without understanding the issue. See below.
Quoted 5 lines
Quoted 5 lines
Signed-off-by Qian Cai <[email protected]>--- mm/page_counter.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
Quoted 12 lines
mm/page_counter.c void page_counter_charge(struct page_counter *counter, unsigned long nr_pages) · line 86
Quoted 12 lines
mm/page_counter.c +2
-2@@ -82,8 +82,8 @@ void page_counter_charge(struct page_counter *counter, unsigned long nr_pages)* This is indeed racy, but we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new);
These should be using 'READ_ONCE' and 'WRITE_ONCE' for c->watermark. Store or load tearing would change the logic here, since the comparison might see garbage.
Quoted 9 lines
mm/page_counter.c bool page_counter_try_charge(struct page_counter *counter, · line 129
Quoted 9 lines
mm/page_counter.c +1
-1} }@@ -126,7 +126,7 @@ bool page_counter_try_charge(struct page_counter *counter,* This is racy, but we can live with some * inaccuracy in the failcnt. */ - c->failcnt++; +data_race (c->failcnt++);
This is probably fine.
Quoted 11 lines
mm/page_counter.c bool page_counter_try_charge(struct page_counter *counter, · line 139
Quoted 11 lines
mm/page_counter.c +2
-2*fail = c; goto failed; }@@ -135,8 +135,8 @@ bool page_counter_try_charge(struct page_counter *counter,* Just like with failcnt, we can live with some * inaccuracy in the watermark. */ -if (new > c->watermark) - c->watermark = new; +if (data_race (new > c->watermark)) +data_race (c->watermark = new);
This should be READ_ONCE / WRITE_ONCE.
Quoted 9 lines
Quoted 5 lines
mm/page_counter.c +0
-1}return true; --2.21.0 (Apple Git-122.2)
--
Michal Hocko
SUSE Labs
Quoted 6 lines · Marco Elver
Why are the line numbers for the remaining symbols missing? Doesn't
scripts/decode_stacktrace.shgive you all line numbers?
I used scripts/faddr2line and never used decode_stacktrace.sh before. I did not insert other line numbers because it can be easily determined by looking at the function itself through an code browser.
Quoted 6 lines
[ As an aside: if you want to use what syzbot uses to put line numbers on symbols, which is a bit faster:
https://github.com/google/syzkaller/tree/master/tools/syz-symbolize
https://github.com/google/syzkaller/blob/master/docs/linux/setup.md
then 'go buildtools/syz-symbolize'. ]
That is good to know.
Quoted 7 lines · Marco Elver
These should be using '
READ_ONCE' and 'WRITE_ONCE' forc->watermark. Store or load tearing would change the logic here, since the comparison might see garbage.
I originally thought that it probably does not matter because it is racy there by doing lockless access anyway. Another thread could change the value at anytime.
Now, I agree set it to a garbage due to a data race could be quite unpleasant there.