Re: GCC-3.3.3 release status (2003-12-01)

Re: GCC-3.3.3 release status (2003-12-01)

[Nathanael Nerode]

> If some bugs have already
>been fixed on mainline and the correponding patches could be safely 
>back-ported to 3_3-branch, please let me know.

This was never reported to Bugzilla, but it was in fact a regression from
3.2.  The fix should be backported to 3_3-branch, and is quite safe.

http://gcc.gnu.org/ml/gcc-patches/2003-11/msg01543.html

Please, if you approve it, just go ahead and commit it for me.  :-)

-- 
Nathanael Nerode  <neroden at gcc.gnu.org>
http://home.twcny.rr.com/nerode/neroden/fdl.html

Re: GCC-3.3.3 release status (2003-12-01)

[Hans-Peter Nilsson]

On 2 Dec 2003, Gabriel Dos Reis wrote:
> If some bugs have already
> been fixed on mainline and the correponding patches could be safely
> back-ported to 3_3-branch, please let me know.  I'm trying hard to
> follow all patch submissions but I may miss some, please drop me a
> note in case.

PR12598 Note the question in the patch: "Can I also
 import gcc.dg/torture/dg-torture.exp on the 3.3 branch?"

PR13256 (not resolved yet; seemingly trivial patch pending)

brgds, H-P

GCC-3.3.3 release status (2003-12-01)

[Gabriel Dos Reis]

Hi,

  I've been planning to send this sort of report for some time.  It
was delayed mainly due to my moving to a new country/continent.  I
just needed to get an apartment and a connection :-)  Anyway, thanks
to Gerald (who kept his promise) I can now send this note you're
reading now.

  As stated in message

    http://gcc.gnu.org/ml/gcc/2003-10/msg01390.html

GCC-3.3.3 is scheduled for Januray 31, 2004.  

  The current status is that gcc-3_3-branch counts 68 PRs marked as
regressions.  Many of them are marked NEW (although I believe there
are some GNATS -> Bugzilla concersion artefacts here), very few are
assigned to people.  Most the regressions seem to be middle end and/or
back end problems.  That is a high number.

  The gcc-3_3-branch is primarily open for regression fixes.  But
non-regression fixes are also welcome as much as they don't
destabilize the branch.  I would encourage maintainers and people
submitting patches for 3.4 to state whether they think particular
patches might worth back-porting to 3_3-branch. Zack had helped today
with such analyzis.  Many thanks to him. If some bugs have already
been fixed on mainline and the correponding patches could be safely 
back-ported to 3_3-branch, please let me know.  I'm trying hard to
follow all patch submissions but I may miss some, please drop me a
note in case.

  Starting from January 1, 2004, only regression-fixes may be
committed to the branch.  Such patches will require my explicit
approval. Documentation patches are exempted.

Next report on December 8, 2003.

Thanks,

-- Gaby