This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: glibc-2.20 status - 2014-07-30
- From: Allan McRae <allan at archlinux dot org>
- To: Siddhesh Poyarekar <siddhesh dot poyarekar at gmail dot com>, "Joseph S. Myers" <joseph at codesourcery dot com>
- Cc: libc-alpha <libc-alpha at sourceware dot org>
- Date: Sat, 09 Aug 2014 09:30:23 +1000
- Subject: Re: glibc-2.20 status - 2014-07-30
- Authentication-results: sourceware.org; auth=none
- References: <53D87188 dot 2020204 at archlinux dot org> <Pine dot LNX dot 4 dot 64 dot 1408081506260 dot 26654 at digraph dot polyomino dot org dot uk> <CAAHN_R3hDVUejBWyus-QN-cR2Gmfhae90i0nwDPbGDvkzGhB5g at mail dot gmail dot com>
On 09/08/14 01:23, Siddhesh Poyarekar wrote:
> On 8 August 2014 20:41, Joseph S. Myers <joseph@codesourcery.com> wrote:
>> On Wed, 30 Jul 2014, Allan McRae wrote:
>>
>>> * blockers
>>> - -Wundef (is there place to track the status of this?)
>>
>> I see no sign of any activity on this in the past week. We're more than a
>> month behind schedule - can we please just freeze? Freeze delays should
>> only be for something user-visible - not anything purely internal about
>> build warnings - and only where there is significant active work (if it's
>> freeze time and there are unresolved problems with a recent change and no
>> active work on fixing them, the default should be to revert the change and
>> work getting it back for the following release). -Wundef should not have
>> occasioned a freeze delay unless patches posted *by the end of June* were
>> substantially complete and satisfactory for resolving all the warnings,
>> with only a few revisions needed following review.
>
> I had taken up responsibility for this and I've now narrowed it to two
> changes: one to the IS_IN_* scheme and the other for the _POSIX_*
> macros. I've been working on them on and off last week in between a
> lot of NMIs, but I don't have anything useful to show yet.
>
> Roland and I had agreed at the Cauldron that we ought to get this
> closed in 2.20 lest it remains languishing forever, but if there's
> disagreement on that then maybe it needs to be discussed. Either way,
> I can commit to not letting them languish too long after 2.20 even if
> 2.20 freezes right now if that helps.
>
I had kept this as a blocker based on Roland and Carlos pushing for it.
There needs to be input from them on this before I remove it.
Note that the s/390 ABI blocker was only fixed a week ago so we are not
delay much specifically by this yet. Also, any real freeze was not going
to happen until after Cauldron. I'm cutting a tarball today to get the
translation in progress.
Siddhesh: am I right in assuming that these patches would be
non-intrusive (i.e. code unchanged) so that I would be happy including
them post freeze?
Allan