This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: basic asm and memory clobbers
- From: Andrew Haley <aph at redhat dot com>
- To: David Wohlferd <dw at LimeGreenSocks dot com>, Richard Henderson <rth at redhat dot com>, Jakub Jelinek <jakub at redhat dot com>
- Cc: Segher Boessenkool <segher at kernel dot crashing dot org>, Jeff Law <law at redhat dot com>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>, rth at gcc dot gnu dot org, pinskia at gcc dot gnu dot org, Sandra Loosemore <sandra at codesourcery dot com>
- Date: Tue, 24 Nov 2015 09:48:23 +0000
- Subject: Re: basic asm and memory clobbers
- Authentication-results: sourceware.org; auth=none
- References: <564AC155 dot 4040601 at LimeGreenSocks dot com> <564B9CB1 dot 1060001 at redhat dot com> <564E762B dot 6070705 at LimeGreenSocks dot com> <564EF338 dot 4030703 at redhat dot com> <564EF7FF dot 1070107 at LimeGreenSocks dot com> <564F008B dot 8040703 at redhat dot com> <564F1436 dot 2060005 at LimeGreenSocks dot com> <564F1A7D dot 9050607 at redhat dot com> <20151120152039 dot GA15922 at gate dot crashing dot org> <564F3C6E dot 5010908 at redhat dot com> <20151120153431 dot GJ5675 at tucnak dot redhat dot com> <564F46CF dot 2030403 at redhat dot com> <564FB2ED dot 70803 at LimeGreenSocks dot com> <56506A03 dot 7030609 at LimeGreenSocks dot com> <5652E4A6 dot 9090905 at redhat dot com> <56537EFF dot 2040206 at LimeGreenSocks dot com>
On 23/11/15 21:02, David Wohlferd wrote:
>> On 11/23/2015 2:04 AM, Andrew Haley wrote:
>> > My warning still holds: there are modes of compilation on some
>> > machines where you can't clobber all registers without causing reload
>> > failures. This is why Jeff didn't fix this in 1999. So, if we really
>> > do want to clobber "all" registers in basic asm it'll take a lot of
>> > work.
>
> I was always reluctant to see this change made. In addition to the
> issues you mention, I had questions about the impact on the surrounding
> code. I like Richard's direction much better. We can start with a
> disabled warning, then upgrade as seems warranted.
I don't see any problem with unconditionally clobbering memory.
Andrew.