This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: builtin fe[gs]etround
- From: Marc Glisse <marc dot glisse at inria dot fr>
- To: Richard Biener <richard dot guenther at gmail dot com>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Mon, 24 Feb 2014 14:47:49 +0100 (CET)
- Subject: Re: builtin fe[gs]etround
- Authentication-results: sourceware.org; auth=none
- References: <alpine dot DEB dot 2 dot 02 dot 1402231133510 dot 29647 at stedding dot saclay dot inria dot fr> <CAFiYyc04ejz8FVHqipc=k7VbevnWFZsejB5UGPxr7BiSX3PEJg at mail dot gmail dot com> <alpine dot DEB dot 2 dot 10 dot 1402241247280 dot 10108 at stedding dot saclay dot inria dot fr> <CAFiYyc0hRF2DVzjxgG8vsm1XAjS6dynEHwbf4Z4oQJ+XkAbCKQ at mail dot gmail dot com>
On Mon, 24 Feb 2014, Richard Biener wrote:
Well, your asm cannot modify it as you don't have a use or clobber for FP
state (but there isn't any...).
Ah, right, I was focusing on "volatile" and forgot there is no "memory"
clobber in those asms, cool, that may simplify things a lot.
Also I think the 'volatile' in the asms isn't needed.
Right, I had it because of PR 56027 where we don't use fesetround but
the SSE intrinsics. But according to your comment #3 there, volatile is
needed even with fesetround.
--
Marc Glisse