[Bug driver/48832] New: -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[Bug driver/48832] New: -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[marcus at jet dot franken.de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48832

           Summary: -O2 does not imply -fomit-frame-pointer, contrary to
                    --help=optimizers -v -Q
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: driver
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: marcus@jet.franken.de
              Host: x86_64-suse-linux
            Target: i586-suse-linux
             Build: i586-suse-linux


$ gcc -v -O2 -m32 -Q --help=optimizers 2>&1|grep omit-frame
  -fomit-frame-pointer                  [enabled]
$ cat >xx.c
int f() { return 1; } 
$ gcc -O2 -m32 -c xx.c ; objdump -d xx.o
...
00000000 <f>:
   0:   55                      push   %ebp
   1:   b8 01 00 00 00          mov    $0x1,%eax
   6:   89 e5                   mov    %esp,%ebp
   8:   5d                      pop    %ebp
   9:   c3                      ret    
$


seems gcc is lying in --help=optimizers

I cannot see why in the sources though.

[Bug driver/48832] -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48832

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2011.05.02 10:24:33
     Ever Confirmed|0                           |1

--- Comment #1 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-05-02 10:24:33 UTC ---
Works for me:

> gcc-4.6 -S t.c -m32 -O2
> cat t.s
        .file   "t.c"
        .text
        .p2align 4,,15
        .globl  f
        .type   f, @function
f:
.LFB0:
        movl    $1, %eax
        ret
...

please show us that output from -v

[Bug driver/48832] -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[marcus at jet dot franken.de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48832

marcus at jet dot franken.de changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Version|4.6.0                       |4.7.0

--- Comment #2 from marcus at jet dot franken.de 2011-05-02 19:48:25 UTC ---
Ok, I tested 4.5 and TRUNK.

both 4.5 and TRUNK have the ebp frame setup.

gcc-4.6 does not.

[Bug driver/48832] -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48832

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-05-03 11:15:30 UTC ---
On the 4.5 branch this is expected.  On trunk I get as well

> ./xgcc -B. -S t.c -m32 -O2
> cat t.s
        .file   "t.c"
        .text
        .p2align 4,,15
        .globl  f
        .type   f, @function
f:
.LFB0:
        movl    $1, %eax
        ret

how did you configure?  What is the output of the gcc command when appending
-v?

[Bug driver/48832] -O2 does not imply -fomit-frame-pointer, contrary to --help=optimizers -v -Q

[marcus at jet dot franken.de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48832

marcus at jet dot franken.de changed:

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

--- Comment #4 from marcus at jet dot franken.de 2011-05-03 16:46:31 UTC ---
configured by /mp3/gcc.trunk/configure, generated by GNU Autoconf 2.64,
  with options \" '--prefix=/mp3/buildgcc/BIN' '--enable-lto'
'--enable-checking' '--enable-frame-pointer' '--enable-languages=c,lto'\"


oops. not sure where this came from, but I suspect from an earlier debugging
attempt.

so please disregard, my faul.t