This is the mail archive of the mailing list for the GCC 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: [4.0 PATCH] Fix PR c++/29106: A wrong-code regression on the 4.0 branch

On Sat, 27 Jan 2007, Volker Reichelt wrote:

| The patch addresses a wrong-code regression on the 4.0 branch
| that appeared in GCC 4.0.3.
| It's a backport of two patches from the 4.1 branch to the 4.0 branch.
| They have been on the 4.1 branch for quite some time now, and didn't
| cause any trouble there.

GCC-4.0.x is what it is -- we all know it is the source of many
wrong-code generations.  The purpose of GCC-4.0.4 is not to fix
them.  Most of them have been fixed in GCC-4.1.x.  My job is
to get GCC-4.0.4 out, with all its imperfections.

| The first patch from Mark for PR 29106 fixes the wrong-code bug.
| It is documented in
| The second patch from Simon for PR 28284 fixes a minor diagnostic
| regression that is introduced by the former patch. It is documented in
| Bootstrapped and regtested on i686-pc-linux-gnu.
| Ok for the 4.0 branch?

Yes; but please to consider that GCC-4.0.4 is frozen unless we've a
major blocker.  We are not going to fix more PRs in GCC-4.0.4 even
though we know of patches on GCC-4.1.x.  GCC-4.0.4 is near to
permanent hibernation.


-- Gaby

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