public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/58320] New: [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0
@ 2013-09-05 10:06 doko at gcc dot gnu.org
  2013-09-05 10:36 ` [Bug target/58320] " rguenth at gcc dot gnu.org
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: doko at gcc dot gnu.org @ 2013-09-05 10:06 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58320

            Bug ID: 58320
           Summary: [4.7 / 4.8 / 4.9 Regression] code used in autoconf
                    test fails with -O2, works with -O0
           Product: gcc
           Version: 4.7.4
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: doko at gcc dot gnu.org

seen on arm-linux-gnueabihf, as part of an autoconf test in librep:

$ cat stack.c 
#include <stdlib.h>

void inner (char *foo) {
  char bar;
  exit (!(foo >= &bar));
}

void main () {
  char foo;
  inner (&foo);
}

$ gcc-4.6 -o conftest -O2 stack.c && ./conftest ; echo $?
0
$ gcc-4.7 -o conftest -O2 stack.c && ./conftest ; echo $?
1
$ gcc-4.8 -o conftest -O2 stack.c && ./conftest ; echo $?
1
$ gcc-4.8 -o conftest -O0 stack.c && ./conftest ; echo $?
0


^ permalink raw reply	[flat|nested] 4+ messages in thread
* [Bug target/58320] [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0
  2013-09-05 10:06 [Bug target/58320] New: [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0 doko at gcc dot gnu.org
@ 2013-09-05 10:36 ` rguenth at gcc dot gnu.org
  2013-09-05 12:12 ` schwab@linux-m68k.org
  2013-09-05 13:01 ` rguenth at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: rguenth at gcc dot gnu.org @ 2013-09-05 10:36 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58320

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |INVALID

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Well, it's clearly a bogus test - you are comparing addresses of two distinct
objects which invokes undefined behavior.


^ permalink raw reply	[flat|nested] 4+ messages in thread
* [Bug target/58320] [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0
  2013-09-05 10:06 [Bug target/58320] New: [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0 doko at gcc dot gnu.org
  2013-09-05 10:36 ` [Bug target/58320] " rguenth at gcc dot gnu.org
@ 2013-09-05 12:12 ` schwab@linux-m68k.org
  2013-09-05 13:01 ` rguenth at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: schwab@linux-m68k.org @ 2013-09-05 12:12 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58320

--- Comment #2 from Andreas Schwab <schwab@linux-m68k.org> ---
Casting to uintptr_t should probably make this work as intented.


^ permalink raw reply	[flat|nested] 4+ messages in thread
* [Bug target/58320] [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0
  2013-09-05 10:06 [Bug target/58320] New: [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0 doko at gcc dot gnu.org
  2013-09-05 10:36 ` [Bug target/58320] " rguenth at gcc dot gnu.org
  2013-09-05 12:12 ` schwab@linux-m68k.org
@ 2013-09-05 13:01 ` rguenth at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: rguenth at gcc dot gnu.org @ 2013-09-05 13:01 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58320

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Well, until the compiler comes along and inlines 'inner'.

This is really a GIGO test.


^ permalink raw reply	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2013-09-05 13:01 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-09-05 10:06 [Bug target/58320] New: [4.7 / 4.8 / 4.9 Regression] code used in autoconf test fails with -O2, works with -O0 doko at gcc dot gnu.org
2013-09-05 10:36 ` [Bug target/58320] " rguenth at gcc dot gnu.org
2013-09-05 12:12 ` schwab@linux-m68k.org
2013-09-05 13:01 ` 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).