This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Import GCC 4.2.0 PRs
- From: Mark Mitchell <mark at codesourcery dot com>
- To: gcc at gcc dot gnu dot org, "gcc-patches >> GCC Patches" <gcc-patches at gcc dot gnu dot org>
- Cc: Jason Merrill <jason at redhat dot com>, Paul Brook <paul at codesourcery dot com>, Ulrich Weigand <uweigand at de dot ibm dot com>, Ian Lance Taylor <iant at google dot com>, Daniel Berlin <dberlin at dberlin dot org>, "Hubicha, Jan" <jh at suse dot cz>, Richard Henderson <rth at redhat dot com>, Alexandre Oliva <aoliva at redhat dot com>, Andrew Pinski <pinskia at gmail dot com>, Nathan Sidwell <nathan at codesourcery dot com>, Richard Guenther <richard dot guenther at gmail dot com>, David Edelsohn <dje at watson dot ibm dot com>
- Date: Mon, 12 Mar 2007 11:47:21 -0700
- Subject: Import GCC 4.2.0 PRs
Here are some GCC 4.2.0 P1s which I think it would be good for GCC to
have resolved before the release, together with names of people I'd like
to volunteer to help. (Naturally, I have no command authority, and I'd
encourage anyone else who wants to help to pitch in, but I'm trying to
tap a few likely suspects.)
* PR 27945 (Merill) -- layout of structures with packed fields. This is
an ABI change. We've had the current behavior in 4.0 and 4.1, so it's
not essential that this be fixed in 4.2.0, but from Jason's comments
last November in the PR it doesn't sound that hard to fix.
* PR 28544 (Brook, Weigand) -- this is an ARM ICE which Paul has tracked
to reload, but he's not sure what to do there. Perhaps Ulrich can help.
* PR 29286 (Berlin, Taylor) -- This is the problem with our alias
analysis not understanding placement new. Ignoring the full complexity
of C/C++ object lifetimes, new vs. malloc vs. casts, etc., it's clear
that the test case is being mishandled. Is there anything we can do?
This is an unfortunate problem, and all our efforts to make the
optimizers smarter will exacerbate this problem.
* PR 29585 (Berlin) -- This is a crash in alias analysis with anonymous
namespaces. I can't imagine that anonymous namespaces have any deep
implications for aliasing, so I would hope this is an easy fix.
* PR 29841 (Hubicka, Henderson) -- This is an oft-reported crash on
__builtin_trap with a fix inside flow/scheduler. A quick review by
someone familiar with the code should close this out.
* PR 29906 (Oliva) -- This is a crash during DWARF generation for a
small C++ test case.
* PR 30132 (Pinksi, Berlin) -- This is an aliasing crash on complex
numbers. Andrew, you mentioned you might have a patch.
* PR 30221 (Sidwell, Mitchell) -- An ICE-on-valid for initializing
arrays of pointer-to-member functions in C++.
* PR 30590 (Guenther, Merill) -- Wrong code generated by NRV.
* PR 30700 (Hubicka) -- Cgraph causes undefined references.
* PR 30704 (Edelsohn) -- Bad code generation for long long on big-endian
targets.
--
Mark Mitchell
CodeSourcery
mark@codesourcery.com
(650) 331-3385 x713