[Bug regression/48262] New: Subversion id 171341 breaks various vectorization files on powerpc

[Bug regression/48262] New: Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

           Summary: Subversion id 171341 breaks various vectorization
                    files on powerpc
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: regression
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: meissner@gcc.gnu.org
              Host: powerpc64-linux
            Target: powerpc64-linux
             Build: powerpc64-linux


Created attachment 23763
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23763
-save-temps of a file that shows the bug (compile with -O3 -m64 -mvsx)

I updated GCC to the head release and tried to build spec on power7, and I had
a lot of failures.  I also got failures in one of my vectorization tests.

The failures are in:
oo2.c: In function ‘trunkp7vsx_align_amove_0_0_sign_long’:
foo2.c:120:36: internal compiler error: in expand_insn, at optabs.c:7171
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

GNU gdb (GDB) SUSE (7.0-0.4.16)
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "ppc64-suse-linux".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from
/home/meissner/fsf-build-ppc64/power7-meissner/gcc/cc1...done.
Breakpoint 1 at 0x1020460c: file
/home/meissner/fsf-src/power7-meissner/gcc/diagnostic.c, line 893.
Breakpoint 2 at 0x102052c4: file
/home/meissner/fsf-src/power7-meissner/gcc/diagnostic.c, line 838.
Breakpoint 3 at 0x109e7f94
Breakpoint 4 at 0x109e7884
(gdb) r -O3 -m64 -mvsx foo2.i
Starting program: /home/meissner/fsf-build-ppc64/power7-meissner/gcc/cc1 -O3
-m64 -mvsx foo2.i
Missing separate debuginfo for /usr/lib64/libstdc++.so.6
Try: zypper install -C
"debuginfo(build-id)=ed25a9207e96ab232788f89131590142a443bca0"
Missing separate debuginfo for /lib64/libgcc_s.so.1
Try: zypper install -C
"debuginfo(build-id)=3a099294fce09d9fce7bf2a37e64588381d3c56d"
 atof atoi atol atoll gnu_dev_major gnu_dev_minor gnu_dev_makedev vprintf
getchar fgetc_unlocked getc_unlocked getchar_unlocked putchar fputc_unlocked
putc_unlocked putchar_unlocked feof_unlocked ferror_unlocked
trunkp7vsx_align_amove_0_0_sign_long
Analyzing compilation unit
Performing interprocedural optimizations
 <*free_lang_data> <visibility> <early_local_cleanups> <increase_alignment>
<whole-program> <ipa-profile> <cp> <inline> <pure-const> <static-var>Assembling
functions:
 trunkp7vsx_align_amove_0_0_sign_long
Breakpoint 1, fancy_abort (file=0x10a30f18
"/home/meissner/fsf-src/power7-meissner/gcc/optabs.c", line=7171,
function=0x10a30e60 "expand_insn") at
/home/meissner/fsf-src/power7-meissner/gcc/diagnostic.c:893
893       internal_error ("in %s, at %s:%d", function, trim_filename (file),
line);
(gdb) up
#1  0x00000000103c2e6c in expand_insn (icode=<value optimized out>, nops=<value
optimized out>, ops=<value optimized out>) at
/home/meissner/fsf-src/power7-meissner/gcc/optabs.c:7171
7171        gcc_unreachable ();
(gdb) up
#2  expand_insn (icode=<value optimized out>, nops=<value optimized out>,
ops=<value optimized out>) at
/home/meissner/fsf-src/power7-meissner/gcc/optabs.c:7167
7167    expand_insn (enum insn_code icode, unsigned int nops,
(gdb) up
#3  0x00000000102864f4 in expand_assignment (nontemporal=<value optimized out>,
from=Unhandled dwarf expression opcode 0xf3
) at /home/meissner/fsf-src/power7-meissner/gcc/expr.c:4184
4184          expand_insn (icode, 2, ops);
(gdb) up
#4  expand_assignment (nontemporal=<value optimized out>, from=Unhandled dwarf
expression opcode 0xf3
) at /home/meissner/fsf-src/power7-meissner/gcc/expr.c:4108
4108    expand_assignment (tree to, tree from, bool nontemporal)
(gdb) print to
$1 = (union tree_node *) 0xfffb63bc5d0
(gdb) pt
 <target_mem_ref 0xfffb63bc5d0
    type <vector_type 0xfffb6409c30
        type <integer_type 0xfffb62405e8 long int public DI
            size <integer_cst 0xfffb6150780 constant 64>
            unit size <integer_cst 0xfffb61507a8 constant 8>
            align 64 symtab 0 alias set 2 canonical type 0xfffb62405e8
precision 64 min <integer_cst 0xfffb6150708 -9223372036854775808> max
<integer_cst 0xfffb6150730 9223372036854775807> context <translation_unit_decl
0xfffb61eaea8 D.2915>
            pointer_to_this <pointer_type 0xfffb62440f8>>
        V2DI
        size <integer_cst 0xfffb6150910 constant 128>
        unit size <integer_cst 0xfffb6150938 constant 16>
        align 64 symtab 0 alias set 2 canonical type 0xfffb6409648 nunits 2
        pointer_to_this <pointer_type 0xfffb640afe0>>

    arg 0 <ssa_name 0xfffb6393700
        type <pointer_type 0xfffb6409b88 type <vector_type 0xfffb6409648>
            public unsigned restrict DI size <integer_cst 0xfffb6150780 64>
unit size <integer_cst 0xfffb61507a8 8>
            align 64 symtab 0 alias set -1 canonical type 0xfffb6409b88>
        visited var <var_decl 0xfffb63e32a0 vect_p.29>def_stmt vect_p.29_94 =
vec1_15 + base_off.22_85;

        version 94
        ptr-info 0xfffb61db160>
    arg 1 <integer_cst 0xfffb63f7aa8 type <pointer_type 0xfffb6408730> constant
0>
    arg 2 <ssa_name 0xfffb6392520
        type <integer_type 0xfffb6240000 long unsigned int public unsigned
sizetype DI size <integer_cst 0xfffb6150780 64> unit size <integer_cst
0xfffb61507a8 8>
            align 64 symtab 0 alias set -1 canonical type 0xfffb6240000
precision 64 min <integer_cst 0xfffb61507d0 0> max <integer_cst 0xfffb6150aa0
-1>>
        visited var <var_decl 0xfffb63e8660 ivtmp.122>def_stmt ivtmp.122_61 =
PHI <ivtmp.122_62(22), 0(31)>
        version 61>>
(gdb) print from
Unhandled dwarf expression opcode 0xf3
(gdb) pt
 <target_mem_ref 0xfffb63bc5d0
    type <vector_type 0xfffb6409c30
        type <integer_type 0xfffb62405e8 long int public DI
            size <integer_cst 0xfffb6150780 constant 64>
            unit size <integer_cst 0xfffb61507a8 constant 8>
            align 64 symtab 0 alias set 2 canonical type 0xfffb62405e8
precision 64 min <integer_cst 0xfffb6150708 -9223372036854775808> max
<integer_cst 0xfffb6150730 9223372036854775807> context <translation_unit_decl
0xfffb61eaea8 D.2915>
            pointer_to_this <pointer_type 0xfffb62440f8>>
        V2DI
        size <integer_cst 0xfffb6150910 constant 128>
        unit size <integer_cst 0xfffb6150938 constant 16>
        align 64 symtab 0 alias set 2 canonical type 0xfffb6409648 nunits 2
        pointer_to_this <pointer_type 0xfffb640afe0>>

    arg 0 <ssa_name 0xfffb6393700
        type <pointer_type 0xfffb6409b88 type <vector_type 0xfffb6409648>
            public unsigned restrict DI size <integer_cst 0xfffb6150780 64>
unit size <integer_cst 0xfffb61507a8 8>
            align 64 symtab 0 alias set -1 canonical type 0xfffb6409b88>
        visited var <var_decl 0xfffb63e32a0 vect_p.29>def_stmt vect_p.29_94 =
vec1_15 + base_off.22_85;

        version 94
        ptr-info 0xfffb61db160>
    arg 1 <integer_cst 0xfffb63f7aa8 type <pointer_type 0xfffb6408730> constant
0>
    arg 2 <ssa_name 0xfffb6392520
        type <integer_type 0xfffb6240000 long unsigned int public unsigned
sizetype DI size <integer_cst 0xfffb6150780 64> unit size <integer_cst
0xfffb61507a8 8>
            align 64 symtab 0 alias set -1 canonical type 0xfffb6240000
precision 64 min <integer_cst 0xfffb61507d0 0> max <integer_cst 0xfffb6150aa0
-1>>
        visited var <var_decl 0xfffb63e8660 ivtmp.122>def_stmt ivtmp.122_61 =
PHI <ivtmp.122_62(22), 0(31)>

        version 61>>
(gdb)

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

Michael Meissner <meissner at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |richard.sandiford at linaro
                   |                            |dot org

--- Comment #1 from Michael Meissner <meissner at gcc dot gnu.org> 2011-03-23 22:34:45 UTC ---
I tracked down the bug, and it is occurring at subversion id 171341 submitted
by richard.sandiford@linaro.org.

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-23 23:30:15 UTC ---
gcc.c-torture/compile/900116-1.c fails at -O3 -maltivec but with a different
ICE but revision which caused it is the same.
Some debugging from that file:
(gdb) p code
$2 = VEC_EXTRACT_EVEN_EXPR
(gdb) p debug_generic_expr (type)
vector(16) char
$3 = void
(gdb) p this_optab
$4 = (optab) 0x1778538
(gdb) p *this_optab
$5 = {code = UNKNOWN, libcall_suffix = 0 '\000', libcall_basename = 0x0,
libcall_gen = 0, handlers = {{insn_code = 0} <repeats 48 times>, 
    {insn_code = -26}, {insn_code = -27}, {insn_code = -29}, {insn_code = 0},
{insn_code = 0}, {insn_code = -28}, {insn_code = 0}}}
(gdb) p ops
$6 = (sepops) 0x7fffffffb870
(gdb) p *ops
$7 = {code = VEC_EXTRACT_EVEN_EXPR, location = 1236, type = 0x7ffff6d7a930, op0
= 0x7ffff6df5c60, op1 = 0x7ffff6d98688, op2 = 0x0}
(gdb) p debug_generic_expr(ops.op1)
vect_var_.59_10
$8 = void
(gdb) p debug_generic_expr(ops.op0)
vect_var_.57_116
$9 = void
------ CUT -----
So it looks like it broke VEC_EXTRACT_EVEN_EXPR also.

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-23 23:33:57 UTC ---
CODE_FOR_movmisalignv2di

(mem:V2DI (plus:DI (reg/f:DI 203 [ vect_p.29 ])
        (reg:DI 188 [ ivtmp.122 ])) [3 MEM[base: vect_p.29_94, index:
ivtmp.122_61, offset: 0B]+0 S16 A64])

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-23 23:40:18 UTC ---
;; Under VSX, vectors of 4/8 byte alignments do not need to be aligned
;; since the load already handles it.
(define_expand "movmisalign<mode>"
 [(set (match_operand:VEC_N 0 "vfloat_operand" "")
       (match_operand:VEC_N 1 "vfloat_operand" ""))]
 "VECTOR_MEM_VSX_P (<MODE>mode) && TARGET_ALLOW_MOVMISALIGN"
 "")

Only accepts registers while expr.c thinks the operands should allow memory
too.

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-23 23:43:15 UTC ---
Here is a patch for the movmisalign issue:
Index: vector.md
===================================================================
--- vector.md    (revision 171367)
+++ vector.md    (working copy)
@@ -871,8 +871,8 @@ (define_expand "vec_realign_load_<mode>"
 ;; Under VSX, vectors of 4/8 byte alignments do not need to be aligned
 ;; since the load already handles it.
 (define_expand "movmisalign<mode>"
- [(set (match_operand:VEC_N 0 "vfloat_operand" "")
-       (match_operand:VEC_N 1 "vfloat_operand" ""))]
+ [(set (match_operand:VEC_N 0 "nonimmediate_operand" "")
+       (match_operand:VEC_N 1 "any_operand" ""))]
  "VECTOR_MEM_VSX_P (<MODE>mode) && TARGET_ALLOW_MOVMISALIGN"
  "")

--- CUT ---
Still looking into the other issue.

[Bug regression/48262] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-24 00:07:49 UTC ---
(In reply to comment #2)
> gcc.c-torture/compile/900116-1.c fails at -O3 -maltivec but with a different
> ICE but revision which caused it is the same.

The problem here is the wrong mode is being used to create the output reg.
Here is the patch for the other issue:
Index: altivec.md
===================================================================
--- altivec.md    (revision 171367)
+++ altivec.md    (working copy)
@@ -2422,7 +2422,7 @@ (define_insn "altivec_stvrxl"

 (define_expand "vec_extract_evenv4si"
  [(set (match_operand:V4SI 0 "register_operand" "")
-        (unspec:V8HI [(match_operand:V4SI 1 "register_operand" "")
+        (unspec:V4SI [(match_operand:V4SI 1 "register_operand" "")
                       (match_operand:V4SI 2 "register_operand" "")]
               UNSPEC_EXTEVEN_V4SI))]
   "TARGET_ALTIVEC"
@@ -2455,7 +2455,7 @@ (define_expand "vec_extract_evenv4si"

 (define_expand "vec_extract_evenv4sf"
  [(set (match_operand:V4SF 0 "register_operand" "")
-        (unspec:V8HI [(match_operand:V4SF 1 "register_operand" "")
+        (unspec:V4SF [(match_operand:V4SF 1 "register_operand" "")
                       (match_operand:V4SF 2 "register_operand" "")]
                       UNSPEC_EXTEVEN_V4SF))]
   "TARGET_ALTIVEC"
@@ -2487,7 +2487,7 @@ (define_expand "vec_extract_evenv4sf"
 }")

 (define_expand "vec_extract_evenv8hi"
- [(set (match_operand:V4SI 0 "register_operand" "")
+ [(set (match_operand:V8HI 0 "register_operand" "")
         (unspec:V8HI [(match_operand:V8HI 1 "register_operand" "")
                       (match_operand:V8HI 2 "register_operand" "")]
                       UNSPEC_EXTEVEN_V8HI))]
@@ -2520,9 +2520,9 @@ (define_expand "vec_extract_evenv8hi"
 }")

 (define_expand "vec_extract_evenv16qi"
- [(set (match_operand:V4SI 0 "register_operand" "")
-        (unspec:V8HI [(match_operand:V16QI 1 "register_operand" "")
-                      (match_operand:V16QI 2 "register_operand" "")]
+ [(set (match_operand:V16QI 0 "register_operand" "")
+        (unspec:V16QI [(match_operand:V16QI 1 "register_operand" "")
+                       (match_operand:V16QI 2 "register_operand" "")]
                       UNSPEC_EXTEVEN_V16QI))]
   "TARGET_ALTIVEC"
   "
@@ -2554,7 +2554,7 @@ (define_expand "vec_extract_evenv16qi"

 (define_expand "vec_extract_oddv4si"
  [(set (match_operand:V4SI 0 "register_operand" "")
-        (unspec:V8HI [(match_operand:V4SI 1 "register_operand" "")
+        (unspec:V4SI [(match_operand:V4SI 1 "register_operand" "")
                       (match_operand:V4SI 2 "register_operand" "")]
                       UNSPEC_EXTODD_V4SI))]
   "TARGET_ALTIVEC"
@@ -2587,7 +2587,7 @@ (define_expand "vec_extract_oddv4si"

 (define_expand "vec_extract_oddv4sf"
  [(set (match_operand:V4SF 0 "register_operand" "")
-        (unspec:V8HI [(match_operand:V4SF 1 "register_operand" "")
+        (unspec:V4SF [(match_operand:V4SF 1 "register_operand" "")
                       (match_operand:V4SF 2 "register_operand" "")]
                       UNSPEC_EXTODD_V4SF))]
   "TARGET_ALTIVEC"

--- CUT ---
I have not looked to see if this fixes all the issues reported by the testsuite
and I have not bootstrapped/tested it.  I also don't have access to a PPC
machine any more.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|                            |2011.03.24 00:13:45
          Component|regression                  |target
               Host|powerpc64-linux             |
     Ever Confirmed|0                           |1
            Summary|Subversion id 171341 breaks |[4.7 Regression] Subversion
                   |various vectorization files |id 171341 breaks various
                   |on powerpc                  |vectorization files on
                   |                            |powerpc
   Target Milestone|---                         |4.7.0
              Build|powerpc64-linux             |

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-03-24 00:13:45 UTC ---
Confirmed.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[rsandifo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

rsandifo@gcc.gnu.org <rsandifo at gcc dot gnu.org> changed:

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

--- Comment #8 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> 2011-03-24 08:52:21 UTC ---
Unfortunately, this kind of target.md fallout was to
some extent expected.  Thanks Andrew for fixing it.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #9 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-03-24 14:16:54 UTC ---
On powerpc-apple-darwin9, all the tests ran by vect.exp pass with the patches
in comments #5 and #6 on top of revision 171367.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #10 from Michael Meissner <meissner at gcc dot gnu.org> 2011-03-24 15:02:47 UTC ---
The vec_extract patches look good.

However, I think the movmisalign patch is a band-aid.  The exander should call
force_reg if the predicates don't match for movmisalign (probably ternary ops)
like it does for every other RTL expansion.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #11 from Michael Meissner <meissner at gcc dot gnu.org> 2011-03-24 19:44:12 UTC ---
I was wrong with my previous comment about movmisalign, and both patches look
good.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #12 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-03-24 20:00:17 UTC ---
What about the patch in http://gcc.gnu.org/ml/gcc-patches/2011-03/msg01595.html
? Does it come on top of the patches in comments #5 and #6 or does it replace
them?

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #13 from Michael Meissner <meissner at gcc dot gnu.org> 2011-03-24 20:09:06 UTC ---
My reading of the patches, is the rs6000 port was broken, and Andrew's two
patches fixes this.

Richard's patch would have the effect that we wouldn't notice if the two
patches aren't installed.  But it is always better to have correct RTL, so I
would say even if Richard's patch goes in, we should apply Andrew's two patches
as well.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[rsandifo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #14 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> 2011-03-25 09:20:35 UTC ---
FWIW, my patch just fixed the expansion of commutative operations.
movmisalign and vector extractions are handled by two separate
pieces of code, so I don't think my patch would affect the two
problems that Andrew identified.

I agree with Mike that both patches look good and are still needed.

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #15 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-03-30 21:42:39 UTC ---
Test results with the patches in comment #5 and #6 posted at
http://gcc.gnu.org/ml/gcc-testresults/2011-03/msg02627.html and
http://gcc.gnu.org/ml/gcc-testresults/2011-03/msg02983.html with no new
regression. Could the patches be committed?

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #16 from Michael Meissner <meissner at gcc dot gnu.org> 2011-03-31 18:38:41 UTC ---
Created attachment 23841
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23841
Patch to fix the problem, originally from Andrew Pinski

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at linux dot vnet.ibm.com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

--- Comment #17 from Michael Meissner <meissner at linux dot vnet.ibm.com> 2011-03-31 18:43:44 UTC ---
On Wed, Mar 30, 2011 at 09:43:05PM +0000, dominiq at lps dot ens.fr wrote:
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262
> 
> --- Comment #15 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-03-30 21:42:39 UTC ---
> Test results with the patches in comment #5 and #6 posted at
> http://gcc.gnu.org/ml/gcc-testresults/2011-03/msg02627.html and
> http://gcc.gnu.org/ml/gcc-testresults/2011-03/msg02983.html with no new
> regression. Could the patches be committed?
> 
> -- 
> Configure bugmail: http://gcc.gnu.org/bugzilla/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.

I was hoping Andrew would submit the patches.  I just sent them out, for
review:

http://gcc.gnu.org/ml/gcc-patches/2011-03/msg02249.html

[Bug target/48262] [4.7 Regression] Subversion id 171341 breaks various vectorization files on powerpc

[meissner at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48262

Michael Meissner <meissner at gcc dot gnu.org> changed:

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

--- Comment #18 from Michael Meissner <meissner at gcc dot gnu.org> 2011-04-01 18:46:13 UTC ---
Fixed in subversion id 171847