This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH, i386] Add prefixes avoidance tuning for silvermont target
- From: Ilya Enkovich <enkovich dot gnu at gmail dot com>
- To: Uros Bizjak <ubizjak at gmail dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Mon, 14 Jul 2014 18:08:29 +0400
- Subject: Re: [PATCH, i386] Add prefixes avoidance tuning for silvermont target
- Authentication-results: sourceware.org; auth=none
- References: <CAFULd4ZGWB72AH44rh3KHKtjJo+uFqFUk6i0T-pX4MaCDNu4eA at mail dot gmail dot com> <CAMbmDYZKq6CJa3tGtAseXzQZceNBSQBQevbgKbHHsyy6vnD8VQ at mail dot gmail dot com> <CAFULd4ZMK9nUwjD670DweBtrY-Ro8D1YomnLkddm7vuz=sm+cg at mail dot gmail dot com> <CAMbmDYYKv+CYsaYH+_gABQomB3o5rfZ+dbO=iiUbKmxjqmKiDg at mail dot gmail dot com> <CAFULd4aL+MYcYH4+bgkCi7=0U9c6GOyzyP8=25X_PcZa4dHWWw at mail dot gmail dot com> <CAMbmDYb167cYiOx5qVuOy=2hxBV6oU_TJWkKzusqTuugyBpHUw at mail dot gmail dot com>
2014-07-03 17:38 GMT+04:00 Ilya Enkovich <firstname.lastname@example.org>:
> 2014-07-03 16:07 GMT+04:00 Uros Bizjak <email@example.com>:
>> On Thu, Jul 3, 2014 at 1:50 PM, Ilya Enkovich <firstname.lastname@example.org> wrote:
>>> I didn't find a nice way to fix peephole2 patterns to take register
>>> constraints into account. Is there any way to do it?
>> Use REX_SSE_REGNO_P (REGNO (operands[...])) in the insn C constraint.
> Peephole doesn't know whether it works with tuned instruction or not,
> right? I would need to mark all instructions I modify with some
> attribute and then check for it in peephole.
>>> Also fully restrict xmm8-15 does not seem right. It is just costly
>>> but not fully disallowed.
>> As said earlier, you can try "Ya*x" as a constraint.
> I tried it. It does not seem to affect allocation much. I do not see
> any gain on targeted tests.