[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-30 Thread rguenth at gcc dot gnu dot org 


--- Comment #7 from rguenth at gcc dot gnu dot org  2010-07-30 09:22 ---
It can't be the same bug, please open a new one.  Thx.


-- 

rguenth at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|REOPENED|RESOLVED
 Resolution||FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-29 Thread howarth at nitro dot med dot uc dot edu 


--- Comment #6 from howarth at nitro dot med dot uc dot edu  2010-07-30 
02:48 ---
At r162696, this bug is back on x86_64-apple-darwin10...


gfortran -r -nostdlib -o cns_solve-1007222053.exe array_reduced.o cns_reduced.o
genic_reduced.o -DINTEGER='long int' -DCNS_ARCH_TYPE_LINUX -fdefault-integer-8
-O2 -fwhopr -fwhole-program -w
In file included from genic_reduced.f:602:0,
 from genic_reduced.f:602,
 from :3:
array_reduced.f: In function ‘find52_.constprop.0’:
array_reduced.f:34:0: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.

however it doesn't seem to be backtracing now in gdb.


-- 

howarth at nitro dot med dot uc dot edu changed:

   What|Removed |Added

 Status|RESOLVED|REOPENED
 Resolution|FIXED   |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-23 Thread rguenth at gcc dot gnu dot org 


--- Comment #5 from rguenth at gcc dot gnu dot org  2010-07-23 10:15 ---
Subject: Bug 45037

Author: rguenth
Date: Fri Jul 23 10:15:27 2010
New Revision: 162451

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=162451
Log:
2010-07-23  Richard Guenther  

PR tree-optimization/45037
* tree-ssa-loop-ivopts.c (copy_ref_info): Handle NULL base.

Modified:
trunk/gcc/ChangeLog
trunk/gcc/tree-ssa-loop-ivopts.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-23 Thread rguenth at gcc dot gnu dot org 


--- Comment #4 from rguenth at gcc dot gnu dot org  2010-07-23 10:15 ---
Fixed.


-- 

rguenth at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution||FIXED
   Target Milestone|--- |4.6.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-23 Thread rguenth at gcc dot gnu dot org 


--- Comment #3 from rguenth at gcc dot gnu dot org  2010-07-23 08:49 ---
Mine.  IPA-CP makes us produce

  MEM[([0:] * restrict)&C.3242][__result_find52_49]

but nobody folds this (C.3242 is constant zero) because the access is likely
undefined.


-- 

rguenth at gcc dot gnu dot org changed:

   What|Removed |Added

 AssignedTo|unassigned at gcc dot gnu   |rguenth at gcc dot gnu dot
   |dot org |org
 Status|UNCONFIRMED |ASSIGNED
 Ever Confirmed|0   |1
   Last reconfirmed|-00-00 00:00:00 |2010-07-23 08:49:13
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-22 Thread howarth at nitro dot med dot uc dot edu 


--- Comment #2 from howarth at nitro dot med dot uc dot edu  2010-07-23 
01:59 ---
The ICE in lto1 backtraces as...

 gdb /sw/lib/gcc4.6/libexec/gcc/x86_64-apple-darwin10.4.0/4.6.0/lto1

(gdb) r -fPIC -quiet -dumpdir ./ -dumpbase cns_solve-1007222053.exe.ltrans2
-mmacosx-version-min=10.6.4 -mtune=generic -auxbase-strip
cns_solve-1007222053.exe.ltrans2.ltrans.o -O2 -w -version -fdefault-integer-8
-fwhole-program -fltrans cns_solve-1007222053.exe.ltrans2.o -o
cns_solve-1007222053.exe.ltrans2.s
Starting program:
/sw/lib/gcc4.6/libexec/gcc/x86_64-apple-darwin10.4.0/4.6.0/lto1 -fPIC -quiet
-dumpdir ./ -dumpbase cns_solve-1007222053.exe.ltrans2
-mmacosx-version-min=10.6.4 -mtune=generic -auxbase-strip
cns_solve-1007222053.exe.ltrans2.ltrans.o -O2 -w -version -fdefault-integer-8
-fwhole-program -fltrans cns_solve-1007222053.exe.ltrans2.o -o
cns_solve-1007222053.exe.ltrans2.s
Reading symbols for shared libraries .. done
GNU GIMPLE (GCC) version 4.6.0 20100722 (experimental)
(x86_64-apple-darwin10.4.0)
compiled by GNU C version 4.6.0 20100722 (experimental), GMP version
4.3.2, MPFR version 2.4.2-p3, MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
GNU GIMPLE (GCC) version 4.6.0 20100722 (experimental)
(x86_64-apple-darwin10.4.0)
compiled by GNU C version 4.6.0 20100722 (experimental), GMP version
4.3.2, MPFR version 2.4.2-p3, MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072

Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x
rewrite_use_address (use=0x14151e260, cand=) at ../../gcc-4.6-20100722/gcc/tree-ssa-loop-ivopts.c:5607
5607  if ((INDIRECT_REF_P (base)
(gdb) bt
#0  rewrite_use_address (use=0x14151e260, cand=) at
../../gcc-4.6-20100722/gcc/tree-ssa-loop-ivopts.c:5607
#1  0x000100593e24 in tree_ssa_iv_optimize () at
../../gcc-4.6-20100722/gcc/tree-ssa-loop-ivopts.c:5717
#2  0x0001005a0d2a in tree_ssa_loop_ivopts () at
../../gcc-4.6-20100722/gcc/tree-ssa-loop.c:634
#3  0x00010044c438 in execute_one_pass (pass=0x1008f73c0) at
../../gcc-4.6-20100722/gcc/passes.c:1564
#4  0x00010044c6dd in execute_pass_list (pass=0x1008f73c0) at
../../gcc-4.6-20100722/gcc/passes.c:1619
#5  0x00010044c6ef in execute_pass_list (pass=0x1008f79c0) at
../../gcc-4.6-20100722/gcc/passes.c:1620
#6  0x00010044c6ef in execute_pass_list (pass=0x1008f6be0) at
../../gcc-4.6-20100722/gcc/passes.c:1620
#7  0x000100540825 in tree_rest_of_compilation (fndecl=0x141718c00) at
../../gcc-4.6-20100722/gcc/tree-optimize.c:452
#8  0x00010069974d in cgraph_expand_function (node=0x141720150) at
../../gcc-4.6-20100722/gcc/cgraphunit.c:1643
#9  0x00010069ba69 in cgraph_optimize () at
../../gcc-4.6-20100722/gcc/cgraphunit.c:1722
#10 0x000100017a49 in lto_main (debug_p=) at ../../gcc-4.6-20100722/gcc/lto/lto.c:1852
#11 0x0001004f9729 in toplev_main (argc=20, argv=0x7fff5fbfeea8) at
../../gcc-4.6-20100722/gcc/toplev.c:978
#12 0x00010f44 in start ()
(gdb) 


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037

[Bug lto/45037] lto1 ICEs when linking cns_solve

2010-07-22 Thread howarth at nitro dot med dot uc dot edu 


--- Comment #1 from howarth at nitro dot med dot uc dot edu  2010-07-23 
01:57 ---
Created an attachment (id=21291)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21291&action=view)
reduced testcase of lto1 crash when linking cns_solve.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45037