From: Segher Boessenkool <segher@kernel.crashing.org>
To: "Liu, Hongtao" <hongtao.liu@intel.com>
Cc: Richard Biener <richard.guenther@gmail.com>,
GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] Canonicalize (vec_duplicate (not A)) to (not (vec_duplicate A)).
Date: Thu, 3 Jun 2021 14:59:33 -0500 [thread overview]
Message-ID: <20210603195933.GW18427@gate.crashing.org> (raw)
In-Reply-To: <CY4PR11MB2007083223D73249A34A3777E53C9@CY4PR11MB2007.namprd11.prod.outlook.com>
On Thu, Jun 03, 2021 at 11:03:43AM +0000, Liu, Hongtao wrote:
> >A very typical example is how UMIN is optimised:
> >
> > case UMIN:
> > if (trueop1 == CONST0_RTX (mode) && ! side_effects_p (op0))
> > return op1;
> > if (rtx_equal_p (trueop0, trueop1) && ! side_effects_p (op0))
> > return op0;
> > tem = simplify_associative_operation (code, mode, op0, op1);
> > if (tem)
> > return tem;
> > break;
> >
> >(the stuff using "tem").
> >
> >Hongtao, can we do something similar here? Does that work well? Please try
> >it out :-)
>
> In simplify_rtx, no simplication occurs, there is just the difference between
> (vec_duplicate (not REG)) and (not (vec_duplicate (REG)). So here tem will only be 0.
simplify-rtx is used by combine. When you do and+not+splat for example
my suggestion should kick in. Try it out, don't just dismiss it?
> Basically we don't know it's a simplication until combine successfully split the
> 3->2 instructions (not + broadcast + and to andnot + broadcast), but it's pretty awkward
> to do this in combine.
But you need to do this *before* it is split. That is the whole point.
> Consider andnot is existed for many backends, I think a canonicalization is needed here.
Please do note that that is not as easy as yoou may think: you need to
make sure nothing ever creates non-canonical code.
> Maybe we can add insn canonicalization for transforming (and (vect_duplicate (not A)) B) to
> (and (not (duplicate (not A)) B) instead of (vec_duplicate (not A)) to (not (vec_duplicate A))?
I don't understand what this means?
Segher
next prev parent reply other threads:[~2021-06-03 20:00 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-25 5:49 [PATCH][i386] Split not+broadcast+pand to broadcast+pandn Hongtao Liu
2021-05-25 6:11 ` Andrew Pinski
2021-05-25 6:23 ` Hongtao Liu
2021-05-25 6:29 ` Andrew Pinski
2021-05-25 6:34 ` Hongtao Liu
2021-05-26 1:21 ` Hongtao Liu
2021-05-26 4:12 ` Andrew Pinski
2021-05-26 5:17 ` Hongtao Liu
2021-06-01 8:32 ` Hongtao Liu
2021-06-01 13:54 ` Segher Boessenkool
2021-06-01 14:02 ` Segher Boessenkool
2021-06-02 5:39 ` liuhongt
2021-06-02 5:39 ` [PATCH] Canonicalize (vec_duplicate (not A)) to (not (vec_duplicate A)) liuhongt
2021-06-02 7:07 ` Richard Biener
2021-06-02 20:46 ` Segher Boessenkool
2021-06-03 11:03 ` Liu, Hongtao
2021-06-03 11:06 ` Jakub Jelinek
2021-06-03 19:59 ` Segher Boessenkool [this message]
2021-06-04 2:48 ` Liu, Hongtao
2021-06-02 5:49 ` Hongtao Liu
2021-06-02 5:41 ` [PATCH] Canonicalize (vec_duplicate (not A)) to (not (vec_duplicate A)) liuhongt
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=20210603195933.GW18427@gate.crashing.org \
--to=segher@kernel.crashing.org \
--cc=gcc-patches@gcc.gnu.org \
--cc=hongtao.liu@intel.com \
--cc=richard.guenther@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).