public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu
@ 2014-07-08  3:49 su at cs dot ucdavis.edu
  2014-07-08  5:18 ` [Bug lto/61741] " andi-gcc at firstfloor dot org
                   ` (8 more replies)
  0 siblings, 9 replies; 10+ messages in thread
From: su at cs dot ucdavis.edu @ 2014-07-08  3:49 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

            Bug ID: 61741
           Summary: wrong code by LTO on x86_64-linux-gnu
           Product: gcc
           Version: 4.10.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: lto
          Assignee: unassigned at gcc dot gnu.org
          Reporter: su at cs dot ucdavis.edu

The current gcc trunk miscompiles the following code when using LTO on
x86_64-linux-gnu in both 32-bit and 64-bit modes. 

This also seems to affect 4.6.x -- 4.9.x: 
(1) the generated code hangs for 4.6.x -- 4.8.x; and
(2) the generated code produces wrong result for 4.9.x.

$ gcc-trunk -v
Using built-in specs.
COLLECT_GCC=gcc-trunk
COLLECT_LTO_WRAPPER=/usr/local/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/4.10.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../gcc-trunk/configure --prefix=/usr/local/gcc-trunk
--enable-languages=c,c++ --disable-werror --enable-multilib
Thread model: posix
gcc version 4.10.0 20140707 (experimental) [trunk revision 212319] (GCC) 
$ 
$ gcc-trunk -O0 -c foo.c          
$ gcc-trunk -O0 -c main.c
$ gcc-trunk -Os main.o foo.o 
$ a.out
0
$ 
$ gcc-trunk -flto -O0 -c foo.c
$ gcc-trunk -flto -O0 -c main.c
$ gcc-trunk -flto -Os main.o foo.o
$ a.out
1
$ 
$ cat foo.c
extern int a, b; 

void
foo (void)
{
  char c = 0;
  for (; a; a--)
    for (; c >= 0; c++);
  if (!c)
    b = 1;
}
$ cat main.c
int printf (const char *, ...);

extern void foo (void); 

int a = 1, b;

int
main ()
{
  foo ();
  printf ("%d\n", b);
  return 0;
}
$


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug lto/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
@ 2014-07-08  5:18 ` andi-gcc at firstfloor dot org
  2014-07-08  5:36 ` su at cs dot ucdavis.edu
                   ` (7 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: andi-gcc at firstfloor dot org @ 2014-07-08  5:18 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

Andi Kleen <andi-gcc at firstfloor dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andi-gcc at firstfloor dot org

--- Comment #1 from Andi Kleen <andi-gcc at firstfloor dot org> ---
 char c = 0;
  for (; a; a--)
    for (; c >= 0; c++);

Don't you rely on signed overflow which is undefined?

-fsanitize=signed-integer-overflow
-fwrapv


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug lto/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
  2014-07-08  5:18 ` [Bug lto/61741] " andi-gcc at firstfloor dot org
@ 2014-07-08  5:36 ` su at cs dot ucdavis.edu
  2014-07-08  9:19 ` rguenth at gcc dot gnu.org
                   ` (6 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: su at cs dot ucdavis.edu @ 2014-07-08  5:36 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

--- Comment #2 from Zhendong Su <su at cs dot ucdavis.edu> ---
(In reply to Andi Kleen from comment #1)
>  char c = 0;
>   for (; a; a--)
>     for (; c >= 0; c++);
> 
> Don't you rely on signed overflow which is undefined?

No, there is no signed overflow since the variable c is a char so the
increments are performed on int first and then cast back to char: (char) ((int)
c + 1).   

The same also holds for short.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug lto/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
  2014-07-08  5:18 ` [Bug lto/61741] " andi-gcc at firstfloor dot org
  2014-07-08  5:36 ` su at cs dot ucdavis.edu
@ 2014-07-08  9:19 ` rguenth at gcc dot gnu.org
  2014-07-09 12:06 ` rguenth at gcc dot gnu.org
                   ` (5 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-08  9:19 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2014-07-08
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
I believe I fixed sth similar for 4.10 already (it's mismatching
-fstrict-overflow
on compile-time vs. link-time).

I will have a look again.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug lto/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (2 preceding siblings ...)
  2014-07-08  9:19 ` rguenth at gcc dot gnu.org
@ 2014-07-09 12:06 ` rguenth at gcc dot gnu.org
  2014-07-09 17:15 ` [Bug c/61741] " rguenth at gcc dot gnu.org
                   ` (4 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-09 12:06 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
The testcase fails without LTO as well if you use -Os -fno-strict-overflow,
fixed by -fno-tree-vrp (not necessarily caused by it though).  Also
fails with -O2 -fno-strict-overflow.

Smells similar to PR61184.  The patch originally posted for that bug is
thus still required.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug c/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (3 preceding siblings ...)
  2014-07-09 12:06 ` rguenth at gcc dot gnu.org
@ 2014-07-09 17:15 ` rguenth at gcc dot gnu.org
  2014-07-09 17:15 ` rguenth at gcc dot gnu.org
                   ` (3 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-09 17:15 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|lto                         |c
      Known to work|                            |4.10.0
      Known to fail|                            |4.8.3, 4.9.1

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Wed Jul  9 17:14:11 2014
New Revision: 212400

URL: https://gcc.gnu.org/viewcvs?rev=212400&root=gcc&view=rev
Log:
2014-07-09  Richard Biener  <rguenther@suse.de>

    PR c-family/61741
    * c-gimplify.c (c_gimplify_expr): Gimplify self-modify expressions
    using unsigned arithmetic if overflow does not wrap instead of
    if overflow is undefined.

    * c-c++-common/torture/pr61741.c: New testcase.

Added:
    trunk/gcc/testsuite/c-c++-common/torture/pr61741.c
Modified:
    trunk/gcc/c-family/ChangeLog
    trunk/gcc/c-family/c-gimplify.c
    trunk/gcc/testsuite/ChangeLog

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk (sofar).


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug c/61741] wrong code by LTO on x86_64-linux-gnu
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (4 preceding siblings ...)
  2014-07-09 17:15 ` [Bug c/61741] " rguenth at gcc dot gnu.org
@ 2014-07-09 17:15 ` rguenth at gcc dot gnu.org
  2014-07-10 12:21 ` [Bug c/61741] wrong code with -fno-strict-overflow rguenth at gcc dot gnu.org
                   ` (2 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-09 17:15 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|lto                         |c
      Known to work|                            |4.10.0
      Known to fail|                            |4.8.3, 4.9.1

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Wed Jul  9 17:14:11 2014
New Revision: 212400

URL: https://gcc.gnu.org/viewcvs?rev=212400&root=gcc&view=rev
Log:
2014-07-09  Richard Biener  <rguenther@suse.de>

    PR c-family/61741
    * c-gimplify.c (c_gimplify_expr): Gimplify self-modify expressions
    using unsigned arithmetic if overflow does not wrap instead of
    if overflow is undefined.

    * c-c++-common/torture/pr61741.c: New testcase.

Added:
    trunk/gcc/testsuite/c-c++-common/torture/pr61741.c
Modified:
    trunk/gcc/c-family/ChangeLog
    trunk/gcc/c-family/c-gimplify.c
    trunk/gcc/testsuite/ChangeLog

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk (sofar).


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug c/61741] wrong code with -fno-strict-overflow
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (5 preceding siblings ...)
  2014-07-09 17:15 ` rguenth at gcc dot gnu.org
@ 2014-07-10 12:21 ` rguenth at gcc dot gnu.org
  2014-07-17  7:58 ` rguenth at gcc dot gnu.org
  2014-07-17  7:58 ` rguenth at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-10 12:21 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Thu Jul 10 12:20:32 2014
New Revision: 212430

URL: https://gcc.gnu.org/viewcvs?rev=212430&root=gcc&view=rev
Log:
2014-07-10  Richard Biener  <rguenther@suse.de>

    PR c-family/61741
    * c-c++-common/torture/pr61741.c: Use signed char.

Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/c-c++-common/torture/pr61741.c


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug c/61741] wrong code with -fno-strict-overflow
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (6 preceding siblings ...)
  2014-07-10 12:21 ` [Bug c/61741] wrong code with -fno-strict-overflow rguenth at gcc dot gnu.org
@ 2014-07-17  7:58 ` rguenth at gcc dot gnu.org
  2014-07-17  7:58 ` rguenth at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-17  7:58 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
      Known to work|                            |4.9.2
         Resolution|---                         |FIXED
   Target Milestone|---                         |4.9.2

--- Comment #9 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk and 4.9 branch.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug c/61741] wrong code with -fno-strict-overflow
  2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
                   ` (7 preceding siblings ...)
  2014-07-17  7:58 ` rguenth at gcc dot gnu.org
@ 2014-07-17  7:58 ` rguenth at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2014-07-17  7:58 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61741

--- Comment #8 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Thu Jul 17 07:57:30 2014
New Revision: 212742

URL: https://gcc.gnu.org/viewcvs?rev=212742&root=gcc&view=rev
Log:
2014-07-17  Richard Biener  <rguenther@suse.de>

    Backport from mainline
    2014-07-10  Richard Biener  <rguenther@suse.de>

    PR c-family/61741
    * c-c++-common/torture/pr61741.c: Use signed char.

    2014-07-09  Richard Biener  <rguenther@suse.de>

    PR c-family/61741
    * c-gimplify.c (c_gimplify_expr): Gimplify self-modify expressions
    using unsigned arithmetic if overflow does not wrap instead of
    if overflow is undefined.

    * c-c++-common/torture/pr61741.c: New testcase.

Added:
    branches/gcc-4_9-branch/gcc/testsuite/c-c++-common/pr61741.c
Modified:
    branches/gcc-4_9-branch/gcc/c-family/ChangeLog
    branches/gcc-4_9-branch/gcc/c-family/c-gimplify.c
    branches/gcc-4_9-branch/gcc/testsuite/ChangeLog


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2014-07-17  7:58 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-07-08  3:49 [Bug lto/61741] New: wrong code by LTO on x86_64-linux-gnu su at cs dot ucdavis.edu
2014-07-08  5:18 ` [Bug lto/61741] " andi-gcc at firstfloor dot org
2014-07-08  5:36 ` su at cs dot ucdavis.edu
2014-07-08  9:19 ` rguenth at gcc dot gnu.org
2014-07-09 12:06 ` rguenth at gcc dot gnu.org
2014-07-09 17:15 ` [Bug c/61741] " rguenth at gcc dot gnu.org
2014-07-09 17:15 ` rguenth at gcc dot gnu.org
2014-07-10 12:21 ` [Bug c/61741] wrong code with -fno-strict-overflow rguenth at gcc dot gnu.org
2014-07-17  7:58 ` rguenth at gcc dot gnu.org
2014-07-17  7:58 ` rguenth at gcc dot gnu.org

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).