public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "rearnsha at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/106187] armhf: Miscompilation at O2 level (O0 / O1 are working)
Date: Mon, 25 Jul 2022 13:04:51 +0000	[thread overview]
Message-ID: <bug-106187-4-Qu3PT4dBtu@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-106187-4@http.gcc.gnu.org/bugzilla/>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106187

--- Comment #45 from Richard Earnshaw <rearnsha at gcc dot gnu.org> ---
The problem with changing rtx_equal_for_cselib_1 is that it is essentially
commutative in its operands - it doesn't disambiguate with x substituting for y
or vice-versa, so we cannot tell if an operation is a load or a store.

A minimal fix, which just suppresses stores would be:

@@ -81,6 +81,10 @@ reload_cse_noop_set_p (rtx set)
   if (cselib_reg_set_mode (SET_DEST (set)) != GET_MODE (SET_DEST (set)))
     return 0;

+  /* Fixme: we need to check that removing a store doesn't change
+     the alias computations.  */
+  if (flag_strict_aliasing && MEM_P (SET_DEST (set)))
+    return 0;
   return rtx_equal_for_cselib_p (SET_DEST (set), SET_SRC (set));
 }

But we could no-doubt improve on that.

  parent reply	other threads:[~2022-07-25 13:04 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-04 16:23 [Bug c++/106187] New: armhf: Miscompilation with -O2 mathieu.malaterre at gmail dot com
2022-07-04 16:37 ` [Bug c++/106187] " mathieu.malaterre at gmail dot com
2022-07-04 16:37 ` mathieu.malaterre at gmail dot com
2022-07-04 16:42 ` [Bug c++/106187] armhf: Miscompilation at all optimization levels mathieu.malaterre at gmail dot com
2022-07-04 20:19 ` pinskia at gcc dot gnu.org
2022-07-04 20:22 ` pinskia at gcc dot gnu.org
2022-07-05  7:18 ` [Bug target/106187] " mathieu.malaterre at gmail dot com
2022-07-05  7:46 ` [Bug target/106187] armhf: Miscompilation at O2 level (O0 / O1 are working) jan.wassenberg at gmail dot com
2022-07-05  8:00 ` mathieu.malaterre at gmail dot com
2022-07-07  7:50 ` mathieu.malaterre at gmail dot com
2022-07-07  8:00 ` mathieu.malaterre at gmail dot com
2022-07-07  9:38 ` rearnsha at gcc dot gnu.org
2022-07-08  9:01 ` mathieu.malaterre at gmail dot com
2022-07-08  9:01 ` mathieu.malaterre at gmail dot com
2022-07-08  9:03 ` mathieu.malaterre at gmail dot com
2022-07-08 13:50 ` rearnsha at gcc dot gnu.org
2022-07-08 13:59 ` malat at debian dot org
2022-07-08 14:16 ` rearnsha at gcc dot gnu.org
2022-07-08 14:18 ` malat at debian dot org
2022-07-08 14:51 ` rearnsha at gcc dot gnu.org
2022-07-08 15:03 ` malat at debian dot org
2022-07-08 17:24 ` rearnsha at gcc dot gnu.org
2022-07-08 17:31 ` rearnsha at gcc dot gnu.org
2022-07-08 19:27 ` jan.wassenberg at gmail dot com
2022-07-14 13:03 ` rearnsha at gcc dot gnu.org
2022-07-14 13:18 ` rearnsha at gcc dot gnu.org
2022-07-14 16:09 ` pinskia at gcc dot gnu.org
2022-07-18 15:45 ` rearnsha at gcc dot gnu.org
2022-07-18 15:48 ` rearnsha at gcc dot gnu.org
2022-07-19  7:27 ` rguenth at gcc dot gnu.org
2022-07-19  9:00 ` rearnsha at gcc dot gnu.org
2022-07-19  9:13 ` rguenth at gcc dot gnu.org
2022-07-21  9:25 ` rearnsha at gcc dot gnu.org
2022-07-22 12:52 ` rearnsha at gcc dot gnu.org
2022-07-22 13:24 ` rearnsha at gcc dot gnu.org
2022-07-25  6:12 ` rguenth at gcc dot gnu.org
2022-07-25  9:44 ` rearnsha at gcc dot gnu.org
2022-07-25  9:50 ` rguenther at suse dot de
2022-07-25  9:59 ` rearnsha at gcc dot gnu.org
2022-07-25 10:24 ` rguenth at gcc dot gnu.org
2022-07-25 10:26 ` [Bug rtl-optimization/106187] " rguenth at gcc dot gnu.org
2022-07-25 10:33 ` rearnsha at gcc dot gnu.org
2022-07-25 10:42 ` rguenth at gcc dot gnu.org
2022-07-25 10:48 ` rearnsha at gcc dot gnu.org
2022-07-25 11:03 ` rguenther at suse dot de
2022-07-25 11:05 ` rguenth at gcc dot gnu.org
2022-07-25 13:04 ` rearnsha at gcc dot gnu.org [this message]
2022-07-25 14:45 ` rguenther at suse dot de
2022-07-27 13:35 ` rearnsha at gcc dot gnu.org
2022-07-28 16:51 ` rearnsha at gcc dot gnu.org
2022-08-03  9:07 ` cvs-commit at gcc dot gnu.org
2022-08-03  9:16 ` rearnsha at gcc dot gnu.org
2022-08-10  7:06 ` malat at debian dot org
2022-08-10  7:11 ` pinskia at gcc dot gnu.org
2022-09-02  9:28 ` malat at debian dot org
2022-09-02 12:30 ` cvs-commit at gcc dot gnu.org
2022-09-02 12:32 ` rearnsha at gcc dot gnu.org
2022-09-02 14:07 ` cvs-commit at gcc dot gnu.org
2022-09-27 15:28 ` malat at debian dot org
2022-09-27 15:54 ` rearnsha at gcc dot gnu.org
2024-01-20 17:20 ` 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-106187-4-Qu3PT4dBtu@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).