[PATCH v3] linux: Use rseq area unconditionally in sched_getcpu (bug 31479)

Michael Jeanson mjeanson@efficios.com
Wed Mar 13 20:52:49 GMT 2024


On 2024-03-13 16:51, Michael Jeanson wrote:
> On 2024-03-13 15:28, Florian Weimer wrote:
>> Originally, nptl/descr.h included <sys/rseq.h>, but we removed that
>> in commit 2c6b4b272e6b4d07303af25709051c3e96288f2d ("nptl:
>> Unconditionally use a 32-byte rseq area").  After that, it was
>> not ensured that the RSEQ_SIG macro was defined during sched_getcpu.c
>> compilation that provided a definition.  This commit always checks
>> the rseq area for CPU number information before using the other
>> approaches.
>>
>> This adds an unnecessary (but well-predictable) branch on
>> architectures which do not define RSEQ_SIG, but its cost is small
>> compared to the system call.  Most architectures that have vDSO
>> acceleration for getcpu also have rseq support.
>>
>> Fixes: 2c6b4b272e6b4d07303af25709051c3e96288f2d
>> Fixes: 1d350aa06091211863e41169729cee1bca39f72f
> 
> Would adding back the '#include <sys/rseq.h>' in nptl/descr.h have any adverse 
> consequences?
> 
> It would restore the original behavior without any effect on architectures 
> which do not define RSEQ_SIG.

Just saw the V2 discussion, please ignore this.


More information about the Libc-alpha mailing list