Bug 50181 - insn does not satisfy constraints for 481.wrf when generating profile data
Summary: insn does not satisfy constraints for 481.wrf when generating profile data
Status: RESOLVED WORKSFORME
Alias: None
Product: gcc
Classification: Unclassified
Component: target (show other bugs)
Version: 4.6.2
: P3 normal
Target Milestone: ---
Assignee: Not yet assigned to anyone
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-08-24 21:11 UTC by Bill Schmidt
Modified: 2014-01-03 23:13 UTC (History)
3 users (show)

See Also:
Host: powerpc64-linux
Target: powerpc64-linux
Build: powerpc64-linux
Known to work: 4.4.7, 4.7.0
Known to fail: 4.5.4, 4.6.3
Last reconfirmed:


Attachments
Reduced test case (1.46 KB, text/plain)
2011-08-24 21:11 UTC, Bill Schmidt
Details
Smaller C test case (234 bytes, text/plain)
2012-02-28 14:53 UTC, Peter Bergner
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Bill Schmidt 2011-08-24 21:11:55 UTC
Created attachment 25094 [details]
Reduced test case

The following occurs on powerpc64-linux in gcc-4_6-branch and
ibm/gcc-4_6-branch, but not on trunk.  The options listed are the minimum
required to exhibit the bug.  The reduced test case is attached.

> /home/wschmidt/gcc/install/gcc-4_6-branch/libexec/gcc/powerpc64-linux/4.6.2/f951 -mcpu=power7 -O3 -fprofile-generate wrf_io-min.f90  
reorder transpose fieldio netcdf_err uppercase lowercase extorderstr extorder getindices getdim gettimeindex getname datecheck getdh allochandle ext_ncd_read_field
Analyzing compilation unit
Performing interprocedural optimizations
 <*free_lang_data> <visibility> <early_local_cleanups> <tree_profile_ipa> <increase_alignment> <whole-program> <ipa-profile> <cp> <inline> <pure-const> <static-var>Assembling functions:
 allochandle getdh datecheck getname gettimeindex getdim getindices extorder extorderstr lowercase uppercase netcdf_err fieldio transpose reorder ext_ncd_read_field
wrf_io-min.f90: In function ‘ext_ncd_read_field’:
wrf_io-min.f90:184:0: error: insn does not satisfy its constraints:
(insn 451 320 450 8 (set (reg:V4SI 7 7)
        (const_vector:V4SI [
                (const_int 1 [0x1])
                (const_int 1 [0x1])
                (const_int 1 [0x1])
                (const_int 1 [0x1])
            ])) wrf_io-min.f90:176 793 {*vsx_movv4si}
     (nil))
wrf_io-min.f90:184:0: internal compiler error: in reload_cse_simplify_operands, at postreload.c:403
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
Comment 1 Peter Bergner 2012-02-28 14:53:57 UTC
Created attachment 26769 [details]
Smaller C test case

Here is a smaller C test case that doesn't require -fprofile-generate to ICE:

bergner@igoo:~/gcc/BUGS/pr50181> /home/bergner/gcc/build/gcc-fsf-4_6-base/gcc/xgcc -B/home/bergner/gcc/build/gcc-fsf-4_6-base/gcc -O3 -mcpu=power7 -S pr50181.c 
pr50181.c: In function ‘testcase’:
pr50181.c:29:1: error: insn does not satisfy its constraints:
(insn 69 7 68 2 (set (reg:V4SI 10 10)
        (const_vector:V4SI [
                (const_int 1 [0x1])
                (const_int 1 [0x1])
                (const_int 1 [0x1])
                (const_int 1 [0x1])
            ])) pr50181.c:17 789 {*vsx_movv4si}
     (nil))
pr50181.c:29:1: internal compiler error: in reload_cse_simplify_operands, at postreload.c:403
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
Comment 2 Steven Bosscher 2013-12-26 23:35:23 UTC
gcc 4.6 is no longer maintained. Is there still a bug here to fix?
(The testcase of comment #1 works for me with r206195 on ppc64-linux.)
Comment 3 wschmidt 2014-01-02 14:06:50 UTC
Thanks, Steven!  I have been on an extended vacation but will have a
look at this shortly.  Very likely this was fixed long ago and the
bugzilla not cleaned up.

Bill

On Thu, 2013-12-26 at 23:35 +0000, steven at gcc dot gnu.org wrote:
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50181
> 
> --- Comment #2 from Steven Bosscher <steven at gcc dot gnu.org> ---
> gcc 4.6 is no longer maintained. Is there still a bug here to fix?
> (The testcase of comment #1 works for me with r206195 on ppc64-linux.)
>
Comment 4 Bill Schmidt 2014-01-03 23:13:41 UTC
This appears to have been resolved some time ago.  Closing.