Re: glibc 2.29 - Winter is coming...

[Florian Weimer <fweimer at redhat dot com>]

* Gabriel F. T. Gomes:

> On Wed, Jan 02 2019, Siddhesh Poyarekar wrote:
>> On 02/01/19 4:29 PM, Florian Weimer wrote:
>> > * Siddhesh Poyarekar:
>> > 
>> > > I have reviewed all original release blockers, including a couple of
>> > > additional ones that got added later but I couldn't get to all of
>> > > them. As of now, the pending blockers do not look suitable for
>> > > inclusion at this stage and so will be deferred to 2.30 development.
>> > 
>> > I support this decision, with the possible exception of “preserve
>> > sprintf(buf, "%s", buf) behaviour” and “posix: Clear close-on-exec for
>> > posix_spawn adddup2 (BZ#23640)”.
>> 
>> Both of those fall in the category of "I've reviewed them and expect them to
>> be merged in by this week with the necessary changes".
>
> For the sprintf case, do we have consensus to commit the original patch
> *with the test case*?  I think that the presence of the test case is
> important to avoid future, *unintended* changes in behavior (its
> presence does not imply that the behavior cannot change, only that it
> will make the change more visible for reviewers),

Agreed.

> and commiting the test
> case has an OK from Zack [1] and Joseph [2].  On the other hand, and as
> pointed out by Siddhesh [3], Paul could have a sustained objection.
>
> [1] https://sourceware.org/ml/libc-alpha/2018-12/msg00962.html
> [2] https://sourceware.org/ml/libc-alpha/2018-12/msg01074.html
> [3] https://sourceware.org/ml/libc-alpha/2018-12/msg01016.html

I think you can commit it now, considering that it has been reviewed
positively.

If the objection turns out to be of the sustained kind, we can still
back it out again before the release.

Thanks,
Florian