This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix for PR 61561
- From: Ramana Radhakrishnan <ramana dot gcc at googlemail dot com>
- To: Yuri Gribov <tetra2005 at gmail dot com>
- Cc: Richard Earnshaw <rearnsha at arm dot com>, Marat Zakirov <m dot zakirov at samsung dot com>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, "ktkachov at gcc dot gnu dot org" <ktkachov at gcc dot gnu dot org>, Gribov Yury <y dot gribov at samsung dot com>, Slava Garbuzov <v dot garbuzov at samsung dot com>, Marat Zakirov <marat61 at gmail dot com>
- Date: Thu, 10 Jul 2014 14:07:51 +0100
- Subject: Re: [PATCH] Fix for PR 61561
- Authentication-results: sourceware.org; auth=none
- References: <000b01cf8bcf$f6e83d20$e4b8b760$ at samsung dot com> <53A30620 dot 8020009 at arm dot com> <CAJOtW+5f1RSBEDii-S7Mv34oHNJ6AjTKwvJDoZeLGLwJaQbFvw at mail dot gmail dot com>
- Reply-to: ramrad01 at arm dot com
On Thu, Jun 19, 2014 at 9:19 PM, Yuri Gribov <firstname.lastname@example.org> wrote:
>> Thirdly, we also need to fix movhi_bytes (for pre-v4) thumb2_movhi_insn
>> (for thumb2) and, quite possibly, thumb1_movhi_insn (for thumb1). There
>> may well be additional changes for movqi variants as well.
> A general question: how should one test ARM backend patches? Is it
> enough to regtest ARM and Thumb2 on some modern Cortex? If not - what
> other variants should be tested?
We don't expect everyone to test every single variant as one wouldn't
be able to make forward progress.
1. Cross testing armv7-a / Thumb2 on qemu is acceptable.
2. If you have the resources (now with A15's) bootstrapping and
regression testing on a modern machine would be so much better.
3. Cross test on qemu for older variants that may fail- you could
consider cross testing on just the older variants. Testing on qemu
comes with it's pitfalls, the threaded tests are a joke / the asan
tests continue to randomly fail but in general it gives you a good
enough indication about how badly things maybe broken.
4. Watch out for any regressions / bug reports from folks on bugzilla
/ the mailing lists and deal with any fallout that comes promptly.