public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jakub Jelinek <jakub@redhat.com>
To: Uros Bizjak <ubizjak@gmail.com>
Cc: "H.J. Lu" <hjl.tools@gmail.com>,
	       "gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] i386: Use EXT_REX_SSE_REG_P in *movoi_internal_avx/movti_internal
Date: Mon, 11 Feb 2019 16:24:00 -0000	[thread overview]
Message-ID: <20190211162424.GG2135@tucnak> (raw)
In-Reply-To: <CAFULd4Ydin5mS3Ovmw5gJHfrWFm=Qxfw4aadCSwTqvMTfjyGLA@mail.gmail.com>

On Mon, Feb 11, 2019 at 04:56:45PM +0100, Uros Bizjak wrote:
> > Let's first define what MODE_XI means in standard_sse_constant_opcode
> > as well as in all these mov patterns for with and without AVX512VL.   Without
> > a clear definition, we can't get out of this mess.
> 
> INT_MODE (OI, 32);
> INT_MODE (XI, 64);
> 
> So, XI_MODE represents 64 INTEGER bytes = 64 * 8 = 512 bit operation,
> in case of const_1, all 512 bits set.
> 
> We can load zeros with narrower instruction, (e.g. 256 bit by inherent
> zeroing of highpart in case of 128 bit xor), so TImode in this case.
> 
> Some targets prefer V4SF mode, so they will emit float xorps for zeroing
> 
> Then the introduction of AVX512F fubared everything by overloading the
> meaning of insn mode.

I don't see much changes in AVX512F here, most of the behavior has been
there already in AVX.
Most of the SSE/AVX/AVX512 instructions affect the whole register,
usually there is DEST[MAX_VL-1:VL] <- 0 at the end of each instruction.
But, using the MAX_VL to determine get_attr_mode doesn't seem really useful,
because that changes dynamically at runtime based on the actual hw, not on
what we've been compiled for.
So, I believe we want to use that VL value to determine the bitsize of the
mode corresponding to get_attr_mode.  And in that case, for
*movoi_internal_avx and *movti_internal, I believe the right mode is MODE_OI
resp. MODE_TI for AVX512VL, because e.g.
vmovdqa32 %ymm12, %ymm23
is a VL = 256 instruction, not VL = 512.  Similarly, if we want to set
%ymm25 to all ones, i.e. movoi_internal_avx, we use
vpternlogd	$0xFF, %ymm25, %ymm25, %ymm25
which is again VL = 256 instruction, so should use MODE_OI.
We'd need to use
vmovdqa32 %zmm12, %zmm23
or
vpternlogd	$0xFF, %zmm25, %zmm25, %zmm25
instructions for AVX512F without AVX512VL, but as has been discussed, this
won't really happen, because hard_regno_mode_ok refuses to allocate 256-bit
or 128-bit modes in ext sse registers.

	Jakub

  parent reply	other threads:[~2019-02-11 16:24 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-07 21:11 [PATCH] i386: Use OI/TImode in *mov[ot]i_internal_avx with AVX512VL H.J. Lu
2019-02-08  9:51 ` Uros Bizjak
2019-02-08 11:29   ` H.J. Lu
2019-02-09  0:31     ` [PATCH] i386: Use EXT_REX_SSE_REG_P in *movoi_internal_avx/movti_internal H.J. Lu
2019-02-09  9:50       ` Uros Bizjak
2019-02-09  9:56         ` Jakub Jelinek
2019-02-09 10:40           ` Jakub Jelinek
2019-02-09 10:51             ` Jakub Jelinek
2019-02-09 12:12               ` H.J. Lu
2019-02-09 12:22                 ` Jakub Jelinek
2019-02-09 13:39                   ` Jakub Jelinek
2019-02-11 13:11                     ` H.J. Lu
2019-02-11 13:15                       ` Uros Bizjak
2019-02-11 13:29                         ` H.J. Lu
2019-02-11 13:51                           ` Uros Bizjak
2019-02-11 14:32                             ` H.J. Lu
2019-02-11 15:57                               ` Uros Bizjak
2019-02-11 16:03                                 ` H.J. Lu
2019-02-11 16:24                                 ` Jakub Jelinek [this message]
2019-02-14  2:34                                   ` [PATCH] i386: Properly encode xmm16-xmm31/ymm16-ymm31 for vector move H.J. Lu
2019-02-11 13:47                         ` [PATCH] i386: Use EXT_REX_SSE_REG_P in *movoi_internal_avx/movti_internal Jakub Jelinek
2019-02-12 18:03     ` [PATCH] i386: Use OI/TImode in *mov[ot]i_internal_avx with AVX512VL Uros Bizjak
2019-02-12 19:01       ` H.J. Lu
2019-02-11  2:35   ` Alan Modra
2019-02-11  7:23     ` Uros Bizjak

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=20190211162424.GG2135@tucnak \
    --to=jakub@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hjl.tools@gmail.com \
    --cc=ubizjak@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).