public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/103771] [12 Regression] Missed vectorization under -mavx512f -mavx512vl after r12-5489
Date: Wed, 19 Jan 2022 08:48:44 +0000 [thread overview]
Message-ID: <bug-103771-4-ofMOkwcYJp@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-103771-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103771
--- Comment #30 from rguenther at suse dot de <rguenther at suse dot de> ---
On Wed, 19 Jan 2022, crazylht at gmail dot com wrote:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103771
>
> --- Comment #29 from Hongtao.liu <crazylht at gmail dot com> ---
> (In reply to Richard Biener from comment #28)
> > (In reply to Hongtao.liu from comment #25)
> > > in fold_unary_loc
> > > ---cut from fold-const.cc-----
> > > 9276 else if (TREE_CODE (arg0) == COND_EXPR)
> > > 9277 {
> > > 9278 tree arg01 = TREE_OPERAND (arg0, 1);
> > > 9279 tree arg02 = TREE_OPERAND (arg0, 2);
> > > 9280 if (! VOID_TYPE_P (TREE_TYPE (arg01)))
> > > 9281 arg01 = fold_build1_loc (loc, code, type,
> > > 9282 fold_convert_loc (loc,
> > > 9283 TREE_TYPE (op0),
> > > arg01));
> > > 9284 if (! VOID_TYPE_P (TREE_TYPE (arg02)))
> > > 9285 arg02 = fold_build1_loc (loc, code, type,
> > > 9286 fold_convert_loc (loc,
> > > 9287 TREE_TYPE (op0),
> > > arg02));
> > > 9288=> tem = fold_build3_loc (loc, COND_EXPR, type, TREE_OPERAND
> > > (arg0, 0),
> > > 9289 arg01, arg02);
> > >
> > > -----------end---------------
> > >
> > > gcc always tries to simplify (convert (cond (cmp a b) c d) ---- > (cond (cmp
> > > a b) (convert c) (convert d)), exactly the opposite of what this case wants.
> >
> > It also then undos this if the result didn't simplify and plays trick to
> > avoid
> > recursions.
> >
> > I think this particular transform ought to be specialized, maybe to
> > (T)p?(T')a:(T')b or maybe done during gimplification or RTL expansion only.
> >
> > The "cheap" way of avoiding a conflict is to wrap the match.pd pattern
> > with opposite logic in
> >
> > #if GIMPLE
> > #endif
> >
> It doesn't work,
> > (with a comment explaining this)
> >
> > Note that we can move a conversion out only if the sources of the conversions
> > have compatible types but we always can move a conversion in.
> >
> > Alternatively this transform can also be done in a vectorizer pattern based
> > on vector compatibility of the ?: predicate with the data.
> yes, I'm thinking of doing this in fold_build_cond_expr which is only used by
> pass_ifcvt to generate cond_expr.
That's also a reasonable place but the vectorizer pattern recog phase
might have more contextual information to determine the best types
to use. fold_build_cond_expr is probably easiest to adjust though.
next prev parent reply other threads:[~2022-01-19 8:48 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-20 8:59 [Bug target/103771] New: " wwwhhhyyy333 at gmail dot com
2021-12-20 11:45 ` [Bug target/103771] [12 Regression] " pinskia at gcc dot gnu.org
2022-01-04 15:31 ` tnfchris at gcc dot gnu.org
2022-01-05 9:00 ` crazylht at gmail dot com
2022-01-13 8:43 ` crazylht at gmail dot com
2022-01-13 9:02 ` crazylht at gmail dot com
2022-01-13 9:04 ` crazylht at gmail dot com
2022-01-13 9:28 ` rguenth at gcc dot gnu.org
2022-01-13 9:50 ` rguenth at gcc dot gnu.org
2022-01-13 10:04 ` crazylht at gmail dot com
2022-01-13 10:11 ` rsandifo at gcc dot gnu.org
2022-01-13 10:22 ` crazylht at gmail dot com
2022-01-13 10:43 ` crazylht at gmail dot com
2022-01-13 10:44 ` rguenther at suse dot de
2022-01-13 10:49 ` rguenther at suse dot de
2022-01-13 10:50 ` crazylht at gmail dot com
2022-01-13 12:40 ` crazylht at gmail dot com
2022-01-13 13:42 ` rguenther at suse dot de
2022-01-13 14:42 ` crazylht at gmail dot com
2022-01-14 9:40 ` rsandifo at gcc dot gnu.org
2022-01-14 10:18 ` rguenth at gcc dot gnu.org
2022-01-17 8:22 ` crazylht at gmail dot com
2022-01-18 8:31 ` crazylht at gmail dot com
2022-01-18 8:36 ` crazylht at gmail dot com
2022-01-18 10:44 ` rguenther at suse dot de
2022-01-18 10:49 ` pinskia at gcc dot gnu.org
2022-01-19 6:03 ` crazylht at gmail dot com
2022-01-19 6:10 ` pinskia at gcc dot gnu.org
2022-01-19 6:25 ` crazylht at gmail dot com
2022-01-19 7:44 ` rguenth at gcc dot gnu.org
2022-01-19 8:33 ` crazylht at gmail dot com
2022-01-19 8:48 ` rguenther at suse dot de [this message]
2022-01-20 8:52 ` cvs-commit at gcc dot gnu.org
2022-01-27 6:03 ` crazylht at gmail dot com
2022-02-01 5:21 ` pinskia at gcc dot gnu.org
2022-02-01 16:13 ` law at gcc dot gnu.org
2022-02-01 16:18 ` pinskia at gcc dot gnu.org
2022-02-13 9:58 ` cvs-commit at gcc dot gnu.org
2022-02-14 7:36 ` rguenth at gcc dot gnu.org
2022-02-14 7:45 ` crazylht at gmail dot com
2022-02-14 7:53 ` rguenth at gcc dot gnu.org
2022-02-17 11:01 ` cvs-commit at gcc dot gnu.org
2023-04-26 6:55 ` [Bug target/103771] [12/13/14 " rguenth at gcc dot gnu.org
2023-05-06 23:21 ` pinskia at gcc dot gnu.org
2023-05-08 7:38 ` cvs-commit at gcc dot gnu.org
2023-05-08 7:39 ` pinskia at gcc dot gnu.org
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=bug-103771-4-ofMOkwcYJp@http.gcc.gnu.org/bugzilla/ \
--to=gcc-bugzilla@gcc.gnu.org \
--cc=gcc-bugs@gcc.gnu.org \
/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).