From: Uros Bizjak <ubizjak@gmail.com>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: gcc-patches@gcc.gnu.org
Subject: Re: [PATCH 27/42] i386: Make _mm_empty () as NOP when MMX is disabled
Date: Sat, 16 Feb 2019 08:58:00 -0000 [thread overview]
Message-ID: <CAFULd4YReyjBbwZhwWJiyGTtwOrNnv1C1myXnxb-JvjJ=4uErw@mail.gmail.com> (raw)
In-Reply-To: <20190216003408.23761-28-hjl.tools@gmail.com>
On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> With SSE emulation of MMX intrinsics, we should make _mm_empty () as NOP
> when MMX is disabled.
>
> PR target/89021
> * config/i386/mmx.md (mmx_<emms>): Renamed to ...
> (mmx_<emms>_1): This.
> (mmx_<emms>): New expander.
> ---
> gcc/config/i386/mmx.md | 29 ++++++++++++++++++++++++++++-
> 1 file changed, 28 insertions(+), 1 deletion(-)
>
> diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
> index 9cf0251293a..0f925c0b1ea 100644
> --- a/gcc/config/i386/mmx.md
> +++ b/gcc/config/i386/mmx.md
> @@ -1848,7 +1848,34 @@
> [(UNSPECV_EMMS "emms")
> (UNSPECV_FEMMS "femms")])
>
> -(define_insn "mmx_<emms>"
> +(define_expand "mmx_<emms>"
> + [(unspec_volatile [(const_int 0)] EMMS)
> + (clobber (reg:XF ST0_REG))
> + (clobber (reg:XF ST1_REG))
> + (clobber (reg:XF ST2_REG))
> + (clobber (reg:XF ST3_REG))
> + (clobber (reg:XF ST4_REG))
> + (clobber (reg:XF ST5_REG))
> + (clobber (reg:XF ST6_REG))
> + (clobber (reg:XF ST7_REG))
> + (clobber (reg:DI MM0_REG))
> + (clobber (reg:DI MM1_REG))
> + (clobber (reg:DI MM2_REG))
> + (clobber (reg:DI MM3_REG))
> + (clobber (reg:DI MM4_REG))
> + (clobber (reg:DI MM5_REG))
> + (clobber (reg:DI MM6_REG))
> + (clobber (reg:DI MM7_REG))]
> + "TARGET_MMX || TARGET_MMX_WITH_SSE"
> +{
> + if (TARGET_MMX)
> + emit_insn (gen_mmx_<emms>_1 ());
> + else
> + emit_insn (gen_nop ());
> + DONE;
The above should be written as:
if (!TARGET_MMX)
{
emit_insn (gen_nop ()));
DONE;
}
> +})
> +
> +(define_insn "mmx_<emms>_1"
The old insn should be renamed to "*mmx_<emms>".
Uros.
> [(unspec_volatile [(const_int 0)] EMMS)
> (clobber (reg:XF ST0_REG))
> (clobber (reg:XF ST1_REG))
> --
> 2.20.1
>
>
next prev parent reply other threads:[~2019-02-16 8:58 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-16 0:34 [PATCH 00/42] V7: Emulate MMX intrinsics with SSE H.J. Lu
2019-02-16 0:34 ` [PATCH 01/42] i386: Allow MMX register modes in SSE registers H.J. Lu
2019-02-16 0:34 ` [PATCH 33/42] i386: Emulate MMX ssse3_psign<mode>3 with SSE H.J. Lu
2019-02-16 0:34 ` [PATCH 14/42] i386: Emulate MMX sse_cvtps2pi/sse_cvttps2pi " H.J. Lu
2019-02-16 0:34 ` [PATCH 11/42] i386: Emulate MMX mmx_eq/mmx_gt<mode>3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 10/42] i386: Emulate MMX mmx_andnot<mode>3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 02/42] i386: Emulate MMX packsswb/packssdw/packuswb with SSE2 H.J. Lu
2019-02-16 0:34 ` [PATCH 20/42] i386: Emulate MMX mmx_umulv4hi3_highpart with SSE H.J. Lu
2019-02-16 0:34 ` [PATCH 03/42] i386: Emulate MMX punpcklXX/punpckhXX with SSE punpcklXX H.J. Lu
2019-02-16 0:34 ` [PATCH 34/42] i386: Emulate MMX ssse3_palignrdi with SSE H.J. Lu
2019-02-16 0:34 ` [PATCH 16/42] i386: Emulate MMX mmx_pextrw " H.J. Lu
2019-02-16 0:34 ` [PATCH 18/42] i386: Emulate MMX V4HI smaxmin/V8QI umaxmin " H.J. Lu
2019-02-16 0:34 ` [PATCH 04/42] i386: Emulate MMX plusminus/sat_plusminus " H.J. Lu
2019-02-16 0:34 ` [PATCH 13/42] i386: Emulate MMX pshufw " H.J. Lu
2019-02-16 0:34 ` [PATCH 36/42] i386: Correct <ssse3_avx2>_pmulhrsw<mode>3[_mask] H.J. Lu
2019-02-16 9:28 ` Uros Bizjak
2019-02-16 0:34 ` [PATCH 06/42] i386: Emulate MMX smulv4hi3_highpart with SSE H.J. Lu
2019-02-16 0:34 ` [PATCH 08/42] i386: Emulate MMX ashr<mode>3/<shift_insn><mode>3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 23/42] i386: Emulate MMX mmx_uavgv4hi3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 17/42] i386: Emulate MMX mmx_pinsrw " H.J. Lu
2019-02-16 9:08 ` Uros Bizjak
2019-02-16 14:57 ` H.J. Lu
2019-02-16 0:34 ` [PATCH 29/42] i386: Emulate MMX ssse3_ph<plusminus_mnemonic>dv2si3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 09/42] i386: Emulate MMX <any_logic><mode>3 " H.J. Lu
2019-02-16 0:34 ` [PATCH 15/42] i386: Emulate MMX sse_cvtpi2ps " H.J. Lu
2019-02-16 0:35 ` [PATCH 30/42] i386: Emulate MMX ssse3_pmaddubsw " H.J. Lu
2019-02-16 0:35 ` [PATCH 07/42] i386: Emulate MMX mmx_pmaddwd " H.J. Lu
2019-02-16 0:35 ` [PATCH 42/42] i386: Add tests for MMX intrinsic emulations " H.J. Lu
2019-02-16 0:35 ` [PATCH 05/42] i386: Emulate MMX mulv4hi3 " H.J. Lu
2019-02-16 0:35 ` [PATCH 19/42] i386: Emulate MMX mmx_pmovmskb " H.J. Lu
2019-02-16 0:43 ` [PATCH 26/42] i386: Emulate MMX umulv1siv1di3 with SSE2 H.J. Lu
2019-02-16 0:43 ` [PATCH 28/42] i386: Emulate MMX ssse3_ph<plusminus_mnemonic>wv4hi3 with SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 40/42] i386: Allow MMX intrinsic emulation " H.J. Lu
2019-02-16 0:43 ` [PATCH 39/42] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE H.J. Lu
2019-02-16 9:50 ` Uros Bizjak
2019-02-16 0:43 ` [PATCH 35/42] i386: Emulate MMX abs<mode>2 with SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 21/42] i386: Emulate MMX maskmovq with SSE2 maskmovdqu H.J. Lu
2019-02-16 0:43 ` [PATCH 12/42] i386: Emulate MMX vec_dupv2si with SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 41/42] i386: Enable TM MMX intrinsics with SSE2 H.J. Lu
2019-02-16 0:43 ` [PATCH 25/42] i386: Emulate MMX movntq with SSE2 movntidi H.J. Lu
2019-02-16 0:43 ` [PATCH 22/42] i386: Emulate MMX mmx_uavgv8qi3 with SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 31/42] i386: Emulate MMX ssse3_pmulhrswv4hi3 " H.J. Lu
2019-02-16 0:43 ` [PATCH 37/42] Prevent allocation of MMX registers with TARGET_MMX_WITH_SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 24/42] i386: Emulate MMX mmx_psadbw with SSE H.J. Lu
2019-02-16 0:43 ` [PATCH 27/42] i386: Make _mm_empty () as NOP when MMX is disabled H.J. Lu
2019-02-16 8:58 ` Uros Bizjak [this message]
2019-02-16 14:56 ` H.J. Lu
2019-02-16 19:02 ` Uros Bizjak
2019-02-16 0:43 ` [PATCH 32/42] i386: Emulate MMX pshufb with SSE version H.J. Lu
2019-02-16 0:43 ` [PATCH 38/42] i386: Allow MMXMODE moves with TARGET_MMX_WITH_SSE H.J. Lu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAFULd4YReyjBbwZhwWJiyGTtwOrNnv1C1myXnxb-JvjJ=4uErw@mail.gmail.com' \
--to=ubizjak@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=hjl.tools@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).