This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: libstdc++ and race detectors
- From: Paolo Carlini <paolo dot carlini at oracle dot com>
- To: Kostya Serebryany <kcc at google dot com>
- Cc: Jonathan Wakely <jwakely dot gcc at gmail dot com>, libstdc++ at gcc dot gnu dot org, Julian Seward <jseward at acm dot org>, Bart Van Assche <bvanassche at acm dot org>, "Frank, Matthew I" <matthew dot i dot frank at intel dot com>
- Date: Thu, 12 Aug 2010 13:59:08 +0200
- Subject: Re: libstdc++ and race detectors
- References: <AANLkTim68MweWz6dJkpl4t2Ub6EIlhpgPYas_Gc1WurR@mail.gmail.com> <4C3C5BF8.2050805@oracle.com> <AANLkTikNmQ7jpetsnO5KOgD2LQmglLU8X67Gwr8twW1w@mail.gmail.com> <4C3C79B0.4070703@oracle.com> <AANLkTinWYzsa0wZMoAvXGgFt6emEtVhooRONYbKCx_C9@mail.gmail.com> <AANLkTinguZkC2Itigx3CpG2tuRhkoRSPvBZ9GeYWOGKD@mail.gmail.com> <AANLkTimKWPH1F9TpauoYYAYq-KtVRn565uNkF3p9InCh@mail.gmail.com> <AANLkTikO34M61MUuOziCfuzAGUYvEwNeybq9ulCvNOxB@mail.gmail.com> <4C3D8A35.60302@oracle.com> <AANLkTilTEp-YkjwTsuvj9UXY9cx168_OJAHQBe0i7n49@mail.gmail.com> <4C3D9A7B.2010107@oracle.com> <AANLkTimyphRPCwv1UPngm1JcAH7CKlXoi7N5Wpix6zVF@mail.gmail.com> <AANLkTimaZI6NhN2ymB3fdkZI8YLDsJ8BUWD8k-RFjQLk@mail.gmail.com> <AANLkTil1nRjJzryXCRhyJfX8jTEHH948app2csV5wqLd@mail.gmail.com> <AANLkTikJ_V3ggetht4TrdP46BW402i3pL6PW98kXuRKH@mail.gmail.com> <AANLkTikTd0BER3rqaYPoJ6pz47GE5rSDrH7s9Bw=jvmT@mail.gmail.com> <4C627D32.9020503@oracle.com> <AANLkTi=oAHjYj4v6XDSb=RnMb0jeJiXCa-WtZMaLp7Bt@mail.gmail.com>
Hi,
> I changed the patch:
> - removed the docs section (but leaved small docs section in c++config)
> - annotated all other cases of refcount decrement
> - tried hard to follow the coding style.
>
Great. The patch looks good to me. Barring comments from the other
maintainers, I will apply it later today. Thanks! For the future, please
remember that patches always come with a ChangeLog entry, I'll do it
this time. And of course, you should always clearly say how you tested
the patch, on which "triple" (machine, OS): you may always find that
minor adjustments are needed for the testsuite, like changed dg-error
line numbers. I'll do that too, on x86_64-linux.
>> Well, it's true that without a Copyright assignment
>> (right? Or you are covered by Google?) we are on risky grounds, but if
>> you can keep the patch otherwise as simple as possible, we can still
>> make it this time.
>>
> I was assured that I am 'covered by Google'
>
Excellent.
>
>> And yes, maybe better taking out from the comment the
>> documentation bits and opening a PR as a reminder to ourselves.
>>
> How do I do it?
>
Just open a Bugzilla PR.
Thanks,
Paolo.