This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Protect _dl_profile_fixup data-dependency order [BZ #23690]


Florian Weimer <fweimer@redhat.com> writes:

> * Tulio Magno Quites Machado Filho:
>
>> Yes.  There is already something attached to the bug report, but it does
>> require dozens of threads to reproduce here.
>> If you're fine with that, I can adapt it.
>
> That's okay (depending on the exact value of “dozens” 8-).

I'm using 70, but I believe we should limit this according to the machine.
I don't think we need to run this test on processors with sequential con

> I assume it's burning CPU all the time?  Then you should put it into
> nptl/ because those tests run serialized.  If you think it should run
> longer than five second, we perhaps should make it an xtest.

Yes, it does burn CPU for ~30s on a POWER8.

-- 
Tulio Magno


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]