public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
@ 2012-10-11 15:08 dominiq at lps dot ens.fr
  2012-10-11 15:17 ` [Bug tree-optimization/54901] " rguenth at gcc dot gnu.org
                   ` (4 more replies)
  0 siblings, 5 replies; 6+ messages in thread
From: dominiq at lps dot ens.fr @ 2012-10-11 15:08 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

             Bug #: 54901
           Summary: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are
                    miscompiled with '-m64 -O3 -funroll-loops
                    -fwhole-program' after revision 192213
    Classification: Unclassified
           Product: gcc
           Version: 4.8.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: dominiq@lps.ens.fr
                CC: steven@gcc.gnu.org


The tests air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3
-funroll-loops -fwhole-program' after revision 192213. The backtraces are:

(1) air

...
  ITERATION#  TIME             FINAL          MASS RESIDUAL

Program received signal SIGSEGV, Segmentation fault.
0x000000010000269b in points (d=..., np=<value optimized out>, nd=<value
optimized out>) at air.f90:1110
1110                Dr(k,j) = d1(N-k,N-j)
(gdb) bt
#0  0x000000010000269b in points (d=..., np=<value optimized out>, nd=<value
optimized out>) at air.f90:1110
#1  0x0000000100004b52 in MAIN__ () at air.f90:160

(2) aermod

...
 First hour of profile data
 YR MO DY HR HEIGHT F  WDIR    WSPD AMB_TMP sigmaA  sigmaW  sigmaV
 99 01 01 01    7.0 1    1.    2.60   262.6   99.0  -99.00  -99.00

 F indicates top of profile (=1) or below (=0)
+Now Processing Data For Day No.    1 of 1999

Program received signal SIGSEGV, Segmentation fault.
wake_sig (x=<value optimized out>, xd=104.343887, xold=44.0807877,
turbz=0.221128643, turby=0.300897658, szold=Cannot access memory at address 0x8
) at aermod.f90:38204
38204             SZ = SZOLD + DELX*TURBZ
(gdb) bt
#0  wake_sig (x=<value optimized out>, xd=104.343887, xold=44.0807877,
turbz=0.221128643, turby=0.300897658, szold=Cannot access memory at address 0x8
) at aermod.f90:38204
#1  0x0000000100071ec9 in wake_dfsn (ldbhr=.FALSE., xi=40.076088,
szi=0.596251428, syi=0.618943393, z=<value optimized out>, z=<value optimized
out>)
    at aermod.f90:36900
#2  0x0000000100084eee in numrise (ldbhr=<value optimized out>, h=<value
optimized out>, reff=<value optimized out>, texit=<value optimized out>, 
    wexit=<value optimized out>, ntr=<value optimized out>, xtr=<value
optimized out>, ytr=<value optimized out>, ztr=<value optimized out>, 
    rtr=<value optimized out>, linwake=.TRUE., numwake=<value optimized out>,
numwake=<value optimized out>, rtr=<value optimized out>, 
    ztr=<value optimized out>, ytr=<value optimized out>, xtr=<value optimized
out>, ntr=<value optimized out>, wexit=<value optimized out>, 
    texit=<value optimized out>, reff=<value optimized out>, h=<value optimized
out>, ldbhr=<value optimized out>) at aermod.f90:35360
#3  0x0000000100095856 in prmdelh (xarg=30.6145649, l_inwake=.TRUE.) at
aermod.f90:39784
#4  0x00000001000a143f in prmcalc (xbrec=<value optimized out>, ybrec=<value
optimized out>, ybrec=<value optimized out>, 
    xbrec=<value optimized out>) at aermod.f90:39767
#5  0x00000001000c237f in pcalc () at aermod.f90:6282
#6  0x00000001000c6117 in hrloop () at aermod.f90:5761
#7  0x00000001000d66f6 in aermod () at aermod.f90:2194
#8  0x00000001000f9e9f in main (argc=<value optimized out>, argv=<value
optimized out>) at aermod.f90:2038
#9  0x00000001000009f4 in start ()

(3) mdbx

...
  STEP LP  KIN.E   POT.E   TOT.E   DIFFUS     PX       PY       PZ   
  ---- -- ------- ------- ------- -------- -------- -------- --------

Program received signal SIGSEGV, Segmentation fault.
0x0000000100005a97 in cbuild (ransq=<value optimized out>, icode=1,
ransq=<value optimized out>) at mdbx.f90:1510
1510             IF ( npc(icell).GT.npcusd ) npcusd = npc(icell)
(gdb) bt
#0  0x0000000100005a97 in cbuild (ransq=<value optimized out>, icode=1,
ransq=<value optimized out>) at mdbx.f90:1510
#1  0x0000000100007bc7 in mlist (lmethd=<value optimized out>) at mdbx.f90:1240
#2  0x000000010000deb0 in master (nsteps=100, nlist=<value optimized out>,
method=<value optimized out>, skin=<value optimized out>, 
    ncorr=<value optimized out>, nprint=10) at mdbx.f90:1058
#3  0x000000010000ec17 in mdbnch () at mdbx.f90:120
#4  0x000000010000fa5f in main (argc=<value optimized out>, argv=<value
optimized out>) at mdbx.f90:171
#5  0x00000001000014b4 in start ()


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [Bug tree-optimization/54901] [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
  2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
@ 2012-10-11 15:17 ` rguenth at gcc dot gnu.org
  2012-10-16  8:55 ` vbyakovl23 at gmail dot com
                   ` (3 subsequent siblings)
  4 siblings, 0 replies; 6+ messages in thread
From: rguenth at gcc dot gnu.org @ 2012-10-11 15:17 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.8.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [Bug tree-optimization/54901] [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
  2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
  2012-10-11 15:17 ` [Bug tree-optimization/54901] " rguenth at gcc dot gnu.org
@ 2012-10-16  8:55 ` vbyakovl23 at gmail dot com
  2012-10-16  9:13 ` steven at gcc dot gnu.org
                   ` (2 subsequent siblings)
  4 siblings, 0 replies; 6+ messages in thread
From: vbyakovl23 at gmail dot com @ 2012-10-16  8:55 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

Vladimir Yakovlev <vbyakovl23 at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |vbyakovl23 at gmail dot com

--- Comment #1 from Vladimir Yakovlev <vbyakovl23 at gmail dot com> 2012-10-16 08:55:16 UTC ---
Dominique, could you attach the tests.


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [Bug tree-optimization/54901] [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
  2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
  2012-10-11 15:17 ` [Bug tree-optimization/54901] " rguenth at gcc dot gnu.org
  2012-10-16  8:55 ` vbyakovl23 at gmail dot com
@ 2012-10-16  9:13 ` steven at gcc dot gnu.org
  2012-10-16  9:51 ` dominiq at lps dot ens.fr
  2012-10-21 11:55 ` dominiq at lps dot ens.fr
  4 siblings, 0 replies; 6+ messages in thread
From: steven at gcc dot gnu.org @ 2012-10-16  9:13 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

Steven Bosscher <steven at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|steven at gcc dot gnu.org   |

--- Comment #2 from Steven Bosscher <steven at gcc dot gnu.org> 2012-10-16 09:12:39 UTC ---
Probably a dup of the discussion going on here:
http://gcc.gnu.org/ml/gcc-patches/2012-10/msg01414.html


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [Bug tree-optimization/54901] [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
  2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
                   ` (2 preceding siblings ...)
  2012-10-16  9:13 ` steven at gcc dot gnu.org
@ 2012-10-16  9:51 ` dominiq at lps dot ens.fr
  2012-10-21 11:55 ` dominiq at lps dot ens.fr
  4 siblings, 0 replies; 6+ messages in thread
From: dominiq at lps dot ens.fr @ 2012-10-16  9:51 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

--- Comment #3 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2012-10-16 09:51:21 UTC ---
> Dominique, could you attach the tests.

http://www.polyhedron.com/polyhedron_benchmark_suite0html

> Probably a dup of the discussion going on here:
> http://gcc.gnu.org/ml/gcc-patches/2012-10/msg01414.html

Probably, as it seems fixed by some revision before r192440 (included). I'ld
like to investigate which revision caused the miscompilations and which one
fixed them. Please let me close this PR as fixed myself.


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [Bug tree-optimization/54901] [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213
  2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
                   ` (3 preceding siblings ...)
  2012-10-16  9:51 ` dominiq at lps dot ens.fr
@ 2012-10-21 11:55 ` dominiq at lps dot ens.fr
  4 siblings, 0 replies; 6+ messages in thread
From: dominiq at lps dot ens.fr @ 2012-10-21 11:55 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54901

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

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

--- Comment #4 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2012-10-21 11:54:47 UTC ---
Fixed by revision 192413:

Author:    hubicka
Date:    Fri Oct 12 20:36:09 2012 UTC (8 days, 15 hours ago)
Changed paths:    4
Log Message:    

    * web.c (web_main): Do not set DF_RD_PRUNE_DEAD_DEFS flag.

> Probably a dup of the discussion going on here:
> http://gcc.gnu.org/ml/gcc-patches/2012-10/msg01414.html

So this looks true. Closing.


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2012-10-21 11:55 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-10-11 15:08 [Bug tree-optimization/54901] New: [4.8 Regression] air.f90, aermod.f90, and mdbx.f90 are miscompiled with '-m64 -O3 -funroll-loops -fwhole-program' after revision 192213 dominiq at lps dot ens.fr
2012-10-11 15:17 ` [Bug tree-optimization/54901] " rguenth at gcc dot gnu.org
2012-10-16  8:55 ` vbyakovl23 at gmail dot com
2012-10-16  9:13 ` steven at gcc dot gnu.org
2012-10-16  9:51 ` dominiq at lps dot ens.fr
2012-10-21 11:55 ` dominiq at lps dot ens.fr

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).