public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c/28839] New: ICE in tree-vectorizer.c
@ 2006-08-24 22:55 aldot at gcc dot gnu dot org
2006-08-24 22:59 ` [Bug c/28839] " aldot at gcc dot gnu dot org
` (9 more replies)
0 siblings, 10 replies; 11+ messages in thread
From: aldot at gcc dot gnu dot org @ 2006-08-24 22:55 UTC (permalink / raw)
To: gcc-bugs
/root/src/buildroot/build_x86_64/staging_dir/libexec/gcc/x86_64-linux-uclibc/4.2.0/cc1
-fpreprocessed gettext_test-lock.i -quiet -dumpbase gettext_test-lock.i
-mtune=generic -auxbase gettext_test-lock -O2 -version -ftree-vectorize
-funswitch-loops -fvariable-expansion-in-unroller -o /tmp/ccflUJcQ.s
GNU C version 4.2.0 20060824 (experimental) (x86_64-linux-uclibc)
compiled by GNU C version 4.1.2 20060613 (prerelease) (Debian 4.1.1-5).
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 3bea3173e03ea3601524f7ad8491537d
test-lock.c: In function 'test_once':
test-lock.c:566: internal compiler error: in
slpeel_update_phi_nodes_for_guard1, at tree-vectorizer.c:595
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
Full backtrace with original args:
(gdb) b slpeel_update_phi_nodes_for_guard1
Breakpoint 1 at 0x4fa9c0: file
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vectorizer.c,
line 508.
(gdb) set args -fpreprocessed gettext_test-lock.i -quiet -dumpbase
gettext_test-lock.i -mtune=nocona -march=nocona -msse -msse2 -msse3
-mfpmath=sse -auxbase gettext_test-lock -O2 -version -fPIC -ftree-loop-optimize
-ftree-vectorize -ftree-vect-loop-version -funroll-loops -freorder-blocks
-freorder-blocks-and-partition -fprefetch-loop-arrays -funswitch-loops
-fbranch-target-load-optimize -ftree-loop-linear -ftree-loop-im
-ftree-loop-ivcanon -fgcse-after-reload -fvariable-expansion-in-unroller
-fivopts -o /tmp/ccInilaw.s
(gdb) run
Starting program:
/root/src/buildroot/build_x86_64/staging_dir/libexec/gcc/x86_64-linux-uclibc/4.2.0/cc1
-fpreprocessed gettext_test-lock.i -quiet -dumpbase gettext_test-lock.i
-mtune=nocona -march=nocona -msse -msse2 -msse3 -mfpmath=sse -auxbase
gettext_test-lock -O2 -version -fPIC -ftree-loop-optimize -ftree-vectorize
-ftree-vect-loop-version -funroll-loops -freorder-blocks
-freorder-blocks-and-partition -fprefetch-loop-arrays -funswitch-loops
-fbranch-target-load-optimize -ftree-loop-linear -ftree-loop-im
-ftree-loop-ivcanon -fgcse-after-reload -fvariable-expansion-in-unroller
-fivopts -o /tmp/ccInilaw.s
GNU C version 4.2.0 20060824 (experimental) (x86_64-linux-uclibc)
compiled by GNU C version 4.1.2 20060613 (prerelease) (Debian 4.1.1-5).
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 3bea3173e03ea3601524f7ad8491537d
Breakpoint 1, slpeel_update_phi_nodes_for_guard1 (guard_edge=0x2b5d60da1800,
loop=0xcc8b50, is_new_loop=0 '\0', new_exit_bb=0x7fffffd46a08,
defs=0x7fffffd46a10)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vectorizer.c:508
508 {
(gdb) bt
#0 slpeel_update_phi_nodes_for_guard1 (guard_edge=0x2b5d60da1800,
loop=0xcc8b50, is_new_loop=0 '\0', new_exit_bb=0x7fffffd46a08,
defs=0x7fffffd46a10)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vectorizer.c:508
#1 0x00000000004fbd8b in slpeel_tree_peel_loop_to_edge (
loop=<value optimized out>, loops=<value optimized out>,
e=<value optimized out>, first_niters=0x2b5d60dac900,
niters=0x2b5d60b26e70, update_first_loop_count=0 '\0')
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vectorizer.c:1168
#2 0x0000000000501e17 in vect_do_peeling_for_loop_bound (loop_vinfo=0xcb6ef0,
ratio=0x7fffffd46bc8, loops=0xcbdae0)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vect-transform.c:2609
#3 0x00000000005068f5 in vect_transform_loop (loop_vinfo=0xcb6ef0,
loops=0xcbdae0)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vect-transform.c:3060
#4 0x00000000004f9336 in vectorize_loops (loops=0xcbdae0)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-vectorizer.c:2066
#5 0x00000000004ee310 in tree_vectorize ()
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-ssa-loop.c:193
#6 0x00000000007c6c6c in execute_one_pass (pass=0xb8e600)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/passes.c:864
#7 0x00000000007c6ddc in execute_pass_list (pass=0xb8e600)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/passes.c:911
#8 0x00000000007c6dee in execute_pass_list (pass=0xb8e480)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/passes.c:912
#9 0x00000000007c6dee in execute_pass_list (pass=0xb8dac0)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/passes.c:912
#10 0x000000000048e89e in tree_rest_of_compilation (fndecl=0x2b5d60b3e9a0)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/tree-optimize.c:417
#11 0x000000000040c665 in c_expand_body (fndecl=0x2b5d60da1800)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/c-decl.c:6757
#12 0x000000000081d5f6 in cgraph_expand_function (node=0x2b5d60b760c0)
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/cgraphunit.c:1230
#13 0x000000000082010d in cgraph_optimize ()
at
/root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/cgraphunit.c:1295
#14 0x0000000000416a8b in c_write_global_declarations ()
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/c-decl.c:7872
#15 0x000000000079864f in toplev_main (argc=<value optimized out>,
argv=<value optimized out>)
at /root/src/buildroot/toolchain_build_x86_64/gcc-4.2.0/gcc/toplev.c:1012
#16 0x00002b5d606784ca in __libc_start_main () from /lib/libc.so.6
#17 0x000000000040250a in _start () at ../sysdeps/x86_64/elf/start.S:113
(gdb) c
Continuing.
test-lock.c: In function 'test_once':
test-lock.c:566: internal compiler error: in
slpeel_update_phi_nodes_for_guard1, at tree-vectorizer.c:595
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
Program exited with code 04.
--
Summary: ICE in tree-vectorizer.c
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: aldot at gcc dot gnu dot org
GCC build triplet: x86_64-linux-gnu
GCC host triplet: x86_64-linux-gnu
GCC target triplet: x86_64-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug c/28839] ICE in tree-vectorizer.c
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
@ 2006-08-24 22:59 ` aldot at gcc dot gnu dot org
2006-08-25 5:16 ` [Bug tree-optimization/28839] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops pinskia at gcc dot gnu dot org
` (8 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: aldot at gcc dot gnu dot org @ 2006-08-24 22:59 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from aldot at gcc dot gnu dot org 2006-08-24 22:59 -------
Created an attachment (id=12132)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12132&action=view)
preprocessed from a testcase of plain gettext-0.15
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
2006-08-24 22:59 ` [Bug c/28839] " aldot at gcc dot gnu dot org
@ 2006-08-25 5:16 ` pinskia at gcc dot gnu dot org
2006-08-25 5:17 ` [Bug tree-optimization/28839] [4.2 Regression] " pinskia at gcc dot gnu dot org
` (7 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2006-08-25 5:16 UTC (permalink / raw)
To: gcc-bugs
------- Comment #2 from pinskia at gcc dot gnu dot org 2006-08-25 05:16 -------
Testcase:
static int ready[10];
void test_once (int t,int t1)
{
int i, repeat;
for (i = 0; i < 10; i++)
{
ready[i] = 0;
if (t1)
if (b())
abort ();
}
if (t)
abort ();
}
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
2006-08-24 22:59 ` [Bug c/28839] " aldot at gcc dot gnu dot org
2006-08-25 5:16 ` [Bug tree-optimization/28839] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops pinskia at gcc dot gnu dot org
@ 2006-08-25 5:17 ` pinskia at gcc dot gnu dot org
2006-08-26 18:06 ` rguenth at gcc dot gnu dot org
` (6 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2006-08-25 5:17 UTC (permalink / raw)
To: gcc-bugs
------- Comment #3 from pinskia at gcc dot gnu dot org 2006-08-25 05:17 -------
Confirmed, a regression from 4.1.2.
--
pinskia at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |pinskia at gcc dot gnu dot
| |org
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Known to fail| |4.2.0
Known to work| |4.1.1 4.1.2
Last reconfirmed|0000-00-00 00:00:00 |2006-08-25 05:17:05
date| |
Summary|ICE in tree-vectorizer.c |[4.2 Regression] ICE in
|with -O2 -ftree-vectorize - |tree-vectorizer.c with -O2 -
|funswitch-loops |ftree-vectorize -funswitch-
| |loops
Target Milestone|--- |4.2.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (2 preceding siblings ...)
2006-08-25 5:17 ` [Bug tree-optimization/28839] [4.2 Regression] " pinskia at gcc dot gnu dot org
@ 2006-08-26 18:06 ` rguenth at gcc dot gnu dot org
2006-08-26 20:48 ` pinskia at gcc dot gnu dot org
` (5 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: rguenth at gcc dot gnu dot org @ 2006-08-26 18:06 UTC (permalink / raw)
To: gcc-bugs
------- Comment #4 from rguenth at gcc dot gnu dot org 2006-08-26 18:06 -------
Also fails on i686 with -O2 -msse2 -mfpmath=sse -ftree-vectorize
-funswitch-loops
and is a bug in unswitching which creates
<bb 2>:
if (t1_7 != 0) goto <L17>; else goto <L18>;
<L18>:;
# ready_20 = PHI <ready_4(3), ready_12(5)>;
# i_1 = PHI <0(3), i_17(5)>;
# ready_2 = PHI <ready_4(3), ready_12(5)>;
<L15>:;
# ready_12 = V_MAY_DEF <ready_2>;
ready[i_1] = 0;
note ready_20 and ready_2 which are both life on entry to block L15 which
is invalid for virtual operands. Updating of the SSA form is broken here.
Slightly smaller testcase:
tatic int ready[10];
void test_once (int t1)
{
int i;
for (i = 0; i < 10; i++)
{
ready[i] = 0;
if (t1)
if (f())
return;
}
abort ();
}
--
rguenth at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |rakdver at gcc dot gnu dot
| |org
GCC target triplet|x86_64-linux-gnu |x86_64-linux-gnu, i?86-*-*
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (3 preceding siblings ...)
2006-08-26 18:06 ` rguenth at gcc dot gnu dot org
@ 2006-08-26 20:48 ` pinskia at gcc dot gnu dot org
2006-08-30 11:10 ` tbm at cyrius dot com
` (4 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2006-08-26 20:48 UTC (permalink / raw)
To: gcc-bugs
------- Comment #5 from pinskia at gcc dot gnu dot org 2006-08-26 20:48 -------
(In reply to comment #4)
> Also fails on i686 with -O2 -msse2 -mfpmath=sse -ftree-vectorize
> -funswitch-loops
> and is a bug in unswitching which creates
Seems like we should have something in verrifying part of tree-ssa to catch
this case.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (4 preceding siblings ...)
2006-08-26 20:48 ` pinskia at gcc dot gnu dot org
@ 2006-08-30 11:10 ` tbm at cyrius dot com
2006-08-30 13:19 ` rakdver at gcc dot gnu dot org
` (3 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: tbm at cyrius dot com @ 2006-08-30 11:10 UTC (permalink / raw)
To: gcc-bugs
------- Comment #6 from tbm at cyrius dot com 2006-08-30 11:10 -------
This bug happens *a lot* compiling the Debian archie with -ftree-vectorize -O3.
Additional testcases available on request.
--
tbm at cyrius dot com changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |tbm at cyrius dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (5 preceding siblings ...)
2006-08-30 11:10 ` tbm at cyrius dot com
@ 2006-08-30 13:19 ` rakdver at gcc dot gnu dot org
2006-08-31 18:16 ` rakdver at gcc dot gnu dot org
` (2 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: rakdver at gcc dot gnu dot org @ 2006-08-30 13:19 UTC (permalink / raw)
To: gcc-bugs
--
rakdver at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |rakdver at gcc dot gnu dot
|dot org |org
Status|NEW |ASSIGNED
Last reconfirmed|2006-08-25 05:17:05 |2006-08-30 13:19:20
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (6 preceding siblings ...)
2006-08-30 13:19 ` rakdver at gcc dot gnu dot org
@ 2006-08-31 18:16 ` rakdver at gcc dot gnu dot org
2006-08-31 19:34 ` rakdver at gcc dot gnu dot org
2006-09-01 4:02 ` pinskia at gcc dot gnu dot org
9 siblings, 0 replies; 11+ messages in thread
From: rakdver at gcc dot gnu dot org @ 2006-08-31 18:16 UTC (permalink / raw)
To: gcc-bugs
------- Comment #7 from rakdver at gcc dot gnu dot org 2006-08-31 18:16 -------
Patch: http://gcc.gnu.org/ml/gcc-patches/2006-08/msg01171.html
--
rakdver at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
URL| |http://gcc.gnu.org/ml/gcc-
| |patches/2006-
| |08/msg01171.html
Keywords| |patch
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (7 preceding siblings ...)
2006-08-31 18:16 ` rakdver at gcc dot gnu dot org
@ 2006-08-31 19:34 ` rakdver at gcc dot gnu dot org
2006-09-01 4:02 ` pinskia at gcc dot gnu dot org
9 siblings, 0 replies; 11+ messages in thread
From: rakdver at gcc dot gnu dot org @ 2006-08-31 19:34 UTC (permalink / raw)
To: gcc-bugs
------- Comment #8 from rakdver at gcc dot gnu dot org 2006-08-31 19:34 -------
Subject: Bug 28839
Author: rakdver
Date: Thu Aug 31 19:33:56 2006
New Revision: 116605
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=116605
Log:
PR tree-optimization/28839
* tree-into-ssa.c (prune_unused_phi_nodes): Take into account kills in
blocks in that phi arguments appear.
* gcc.dg/pr28839.c: New test.
Added:
trunk/gcc/testsuite/gcc.dg/pr28839.c
Modified:
trunk/gcc/ChangeLog
trunk/gcc/testsuite/ChangeLog
trunk/gcc/tree-into-ssa.c
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Bug tree-optimization/28839] [4.2 Regression] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
` (8 preceding siblings ...)
2006-08-31 19:34 ` rakdver at gcc dot gnu dot org
@ 2006-09-01 4:02 ` pinskia at gcc dot gnu dot org
9 siblings, 0 replies; 11+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2006-09-01 4:02 UTC (permalink / raw)
To: gcc-bugs
------- Comment #9 from pinskia at gcc dot gnu dot org 2006-09-01 04:02 -------
Fixed.
--
pinskia at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28839
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2006-09-01 4:02 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-08-24 22:55 [Bug c/28839] New: ICE in tree-vectorizer.c aldot at gcc dot gnu dot org
2006-08-24 22:59 ` [Bug c/28839] " aldot at gcc dot gnu dot org
2006-08-25 5:16 ` [Bug tree-optimization/28839] ICE in tree-vectorizer.c with -O2 -ftree-vectorize -funswitch-loops pinskia at gcc dot gnu dot org
2006-08-25 5:17 ` [Bug tree-optimization/28839] [4.2 Regression] " pinskia at gcc dot gnu dot org
2006-08-26 18:06 ` rguenth at gcc dot gnu dot org
2006-08-26 20:48 ` pinskia at gcc dot gnu dot org
2006-08-30 11:10 ` tbm at cyrius dot com
2006-08-30 13:19 ` rakdver at gcc dot gnu dot org
2006-08-31 18:16 ` rakdver at gcc dot gnu dot org
2006-08-31 19:34 ` rakdver at gcc dot gnu dot org
2006-09-01 4:02 ` pinskia at gcc dot gnu dot org
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).