On Thu, Jul 27, 2017 at 12:48 PM, Andrew Morton
<akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Thu, 27 Jul 2017 09:46:08 -0700 Dima Zavin <dmitriyz@xxxxxxxxx> wrote:
>
>> In codepaths that use the begin/retry interface for reading
>> mems_allowed_seq with irqs disabled, there exists a race condition that
>> stalls the patch process after only modifying a subset of the
>> static_branch call sites.
>>
>> This problem manifested itself as a dead lock in the slub
>> allocator, inside get_any_partial. The loop reads
>> mems_allowed_seq value (via read_mems_allowed_begin),
>> performs the defrag operation, and then verifies the consistency
>> of mem_allowed via the read_mems_allowed_retry and the cookie
>> returned by xxx_begin. The issue here is that both begin and retry
>> first check if cpusets are enabled via cpusets_enabled() static branch.
>> This branch can be rewritted dynamically (via cpuset_inc) if a new
>> cpuset is created. The x86 jump label code fully synchronizes across
>> all CPUs for every entry it rewrites. If it rewrites only one of the
>> callsites (specifically the one in read_mems_allowed_retry) and then
>> waits for the smp_call_function(do_sync_core) to complete while a CPU is
>> inside the begin/retry section with IRQs off and the mems_allowed value
>> is changed, we can hang. This is because begin() will always return 0
>> (since it wasn't patched yet) while retry() will test the 0 against
>> the actual value of the seq counter.
>>
>> The fix is to cache the value that's returned by cpusets_enabled() at the
>> top of the loop, and only operate on the seqcount (both begin and retry) if
>> it was true.
>
> Tricky. Hence we should have a nice code comment somewhere describing
> all of this.
>
>> --- a/include/linux/cpuset.h
>> +++ b/include/linux/cpuset.h
>> @@ -16,6 +16,11 @@
>> #include <linux/mm.h>
>> #include <linux/jump_label.h>
>>
>> +struct cpuset_mems_cookie {
>> + unsigned int seq;
>> + bool was_enabled;
>> +};
>
> At cpuset_mems_cookie would be a good site - why it exists, what it
> does, when it is used and how.
Will do. I actually had a comment here but removed it in lieu of
commit message :) Will put it back.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
