public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "vmakarov at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/108552] Linux i386 kernel 5.14 memory corruption for pre_compound_page() when gcov is enabled
Date: Fri, 27 Jan 2023 22:30:07 +0000	[thread overview]
Message-ID: <bug-108552-4-AXHi4vK4WJ@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-108552-4@http.gcc.gnu.org/bugzilla/>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108552

--- Comment #35 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #34)
> Seems right now DECL_NONALIASED is only used on these coverage vars and on
> Fortran caf tokens, so perhaps a quick workaround would be on the LRA side
> never reread stuff from MEMs with VAR_P && DECL_NONALIASED MEM_EXPRs.  CCing
> Vlad on that.

The following patch can do this:

diff --git a/gcc/lra-constraints.cc b/gcc/lra-constraints.cc                    
index 7bffbc07ee2..d80a6a9f41d 100644                                           
--- a/gcc/lra-constraints.cc                                                    
+++ b/gcc/lra-constraints.cc                                                    
@@ -515,6 +515,7 @@ get_equiv (rtx x)                                           
 {                                                                              
   int regno;                                                                   
   rtx res;                                                                     
+  tree expr;                                                                   

   if (! REG_P (x) || (regno = REGNO (x)) < FIRST_PSEUDO_REGISTER               
       || ! ira_reg_equiv[regno].defined_p                                      
@@ -525,6 +526,10 @@ get_equiv (rtx x)                                          
     {                                                                          
       if (targetm.cannot_substitute_mem_equiv_p (res))                         
        return x;                                                               
+      if ((expr = MEM_EXPR (res)) != NULL                                      
+         && (expr = get_base_address (expr)) != NULL                           
+         && VAR_P (expr) && DECL_NONALIASED (expr))                            
+       return x;                                                               
       return res;                                                              
     }                                                                          
   if ((res = ira_reg_equiv[regno].constant) != NULL_RTX)

  parent reply	other threads:[~2023-01-27 22:30 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-26  8:00 [Bug c/108552] New: " feng.tang at intel dot com
2023-01-26  8:01 ` [Bug c/108552] " feng.tang at intel dot com
2023-01-26  8:02 ` [Bug target/108552] " pinskia at gcc dot gnu.org
2023-01-26  8:05 ` pinskia at gcc dot gnu.org
2023-01-26  8:13 ` feng.tang at intel dot com
2023-01-26  8:19 ` pinskia at gcc dot gnu.org
2023-01-26 11:35 ` feng.tang at intel dot com
2023-01-26 11:37 ` feng.tang at intel dot com
2023-01-26 11:39 ` feng.tang at intel dot com
2023-01-26 16:03 ` feng.tang at intel dot com
2023-01-26 16:07 ` feng.tang at intel dot com
2023-01-26 19:06 ` pinskia at gcc dot gnu.org
2023-01-26 19:22 ` torvalds@linux-foundation.org
2023-01-27  9:52 ` ubizjak at gmail dot com
2023-01-27 10:47 ` ubizjak at gmail dot com
2023-01-27 10:56 ` ubizjak at gmail dot com
2023-01-27 12:23 ` ubizjak at gmail dot com
2023-01-27 12:29 ` ubizjak at gmail dot com
2023-01-27 12:31 ` [Bug tree-optimization/108552] " ubizjak at gmail dot com
2023-01-27 12:51 ` ubizjak at gmail dot com
2023-01-27 12:52 ` ubizjak at gmail dot com
2023-01-27 13:17 ` jakub at gcc dot gnu.org
2023-01-27 13:40 ` ubizjak at gmail dot com
2023-01-27 14:14 ` jakub at gcc dot gnu.org
2023-01-27 14:59 ` rguenth at gcc dot gnu.org
2023-01-27 15:01 ` rguenth at gcc dot gnu.org
2023-01-27 15:13 ` rguenth at gcc dot gnu.org
2023-01-27 15:15 ` jakub at gcc dot gnu.org
2023-01-27 15:18 ` rguenth at gcc dot gnu.org
2023-01-27 15:20 ` jakub at gcc dot gnu.org
2023-01-27 17:00 ` torvalds@linux-foundation.org
2023-01-27 17:05 ` torvalds@linux-foundation.org
2023-01-27 17:15 ` torvalds@linux-foundation.org
2023-01-27 17:19 ` jakub at gcc dot gnu.org
2023-01-27 17:29 ` jakub at gcc dot gnu.org
2023-01-27 22:30 ` vmakarov at gcc dot gnu.org [this message]
2023-01-28 14:20 ` feng.tang at intel dot com
2023-01-28 14:27 ` feng.tang at intel dot com
2023-01-28 14:29 ` feng.tang at intel dot com
2023-01-28 23:40 ` hubicka at ucw dot cz
2023-01-29 10:08 ` jakub at gcc dot gnu.org
2023-01-30  7:05 ` rguenth at gcc dot gnu.org
2023-01-30  7:09 ` rguenth at gcc dot gnu.org
2023-01-30  8:06 ` torvalds@linux-foundation.org
2023-01-30  8:30 ` jakub at gcc dot gnu.org
2023-01-30  8:44 ` rguenth at gcc dot gnu.org
2023-01-30  8:46 ` rguenther at suse dot de
2023-01-30 18:54 ` torvalds@linux-foundation.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-108552-4-AXHi4vK4WJ@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).