* [tree-ssa] bootstrap failure
@ 2003-04-25 12:32 Diego Novillo
2003-04-25 12:36 ` Diego Novillo
0 siblings, 1 reply; 13+ messages in thread
From: Diego Novillo @ 2003-04-25 12:32 UTC (permalink / raw)
To: Jason Merrill, Andrew Macleod, Daniel Berlin; +Cc: gcc
Folks,
I got a bootstrap failure last night while running SPEC95.
One of these patches introduced this regression.
I haven't looked very hard, but I *think* it might be your patch
Jason. I had done a bootstrap yesterday that included Andrew's
and Daniel's patch (although that bootstrap didn't include
libjava). It can't be Daniel's because his change is not in the
tree-ssa path.
Thanks. Diego.
+2003-04-24 Jason Merrill <jason@redhat.com>
+
+ * gimplify.c (simplify_cond_expr): Avoid redundant gimplification.
+
+ Add TREE_VEC of case labels to the SWITCH_EXPR during gimplification.
+ * tree.def (CASE_LABEL_EXPR): Add an operand for the LABEL_DECL.
+ * tree.h (SWITCH_LABELS, CASE_LABEL): New macros.
+ * c-simplify.c (c_simplify_stmt) <CASE_LABEL>: Create LABEL_DECL here.
+ * expr.c (expand_expr) <CASE_LABEL_EXPR>: Not here.
+ * gimplify.c (gimplify_ctx): Add case_labels field.
+ (gimplify_switch_expr, gimple_add_case_label): New fns.
+ (simplify_expr): Use them.
+
+2003-04-24 Andrew MacLeod <amacleod@redhat.com>
+
+ * tree-flow.h (processed_out_of_ssa): Rename to out_of_ssa_tag.
+ * tree-ssa-live.c (register_ssa_partition): Add variable to partition.
+ (change_partition_var): Use out_of_ssa_tag.
+ (create_ssa_var_map): Add all uses, defs and PHI elements to partition.
+ (new_tree_live_info): Create a live range info structure.
+ (delete_tree_live_info): Free storage.
+ (live_worklist): Fill in the live range info for a variable for the
+ blocks between the def and all the blocks containing uses.
+ (set_if_valid): Set partition bit if variable is in a partition.
+ (add_livein_if_notdef): Set live on entry bit for a var's partition
+ if a definition has not been seen.
+ (calculate_live_on_entry): Create partition live on entry bitmaps for
+ all basic blocks .
+ (calculate_live_on_exit): Calculate live on exit information for each
+ basic block.
+ (init_root_var): Initialize and fill in a root_var structure.
+ (remove_root_var_partition): remove a partition from a root_var list.
+ (delete_root_var): Free storage.
+ (dump_root_var): Display root_var summary.
+ (dump_var_map): Show extra info for ssa name versions.
+ * tree-ssa-live.h (NO_PARTITION): Define.
+ (register_ssa_partition): Remove.
+ (partition_to_var): Use partition_find after decompressing.
+ (var_to_partition): Return NO_PARTITION if var is not in a partition.
+ (struct tree_live_info_d): Define live range info structure.
+ (partition_is_global): Return 1 if used outside a basic block.
+ (live_entry_blocks): Return bitmap over blocks that partition is live
+ on entry to.
+ (live_on_exit): Return bitmap of partitions live on exit from a block.
+ (struct root_var_d): Define a root_var structure.
+ (ROOT_VAR_NONE): Define.
+ (num_root_vars): Number of variables in root_var object.
+ (root_var): Return variable for a root_var index.
+ (first_root_var_partition): Return first partition for a root_var.
+ (next_root_var_partition): Get next partition for a root_var.
+ (find_root_var): Find root_var index for a specific partition.
+ * tree-ssa.c (eliminate_extraneous_phis): Remove.
+ (set_if_valid): Set partition bit if variable is in a partition.
+ (add_conflicts_if_valid): Add conflict between variable and all
+ related partitions set in a bitvector.
+ (coalesce_ssa_name): Create a conflict graph and coalesce all
+ partitions which don't conflict and are related by the same root_var.
+ (assign_vars): Use a root_var object, and assign different real
+ variables to all partitions.
+ (rewrite_out_of_ssa): Call compact_var_map() once, and don't call
+ eliminate_extraneous_phis.
+
+2003-04-23 Daniel Berlin <dberlin@dberlin.org>
+
+ * tree-flow.h (create_global_var): Add prototype.
+
+ * tree-dfa.c (create_global_var): Externalize.
+
+ * tree-alias-common.c: Set DECL_CONTEXT on our temp alias vars.
+ (call_may_clobber): Make a copy of this function, since our version
+ will be slightly different soon.
+ (create_alias_vars): We need global_var, so create it if necessary.
+
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-04-25 12:32 [tree-ssa] bootstrap failure Diego Novillo
@ 2003-04-25 12:36 ` Diego Novillo
2003-04-25 12:53 ` Andrew MacLeod
0 siblings, 1 reply; 13+ messages in thread
From: Diego Novillo @ 2003-04-25 12:36 UTC (permalink / raw)
To: Jason Merrill; +Cc: Andrew Macleod, Daniel Berlin, gcc
On Fri, 2003-04-25 at 07:11, Diego Novillo wrote:
> Folks,
>
> I got a bootstrap failure last night while running SPEC95.
> One of these patches introduced this regression.
>
> I haven't looked very hard, but I *think* it might be your patch
> Jason. I had done a bootstrap yesterday that included Andrew's
> and Daniel's patch (although that bootstrap didn't include
> libjava). It can't be Daniel's because his change is not in the
> tree-ssa path.
>
Maybe it would help if I included the error as well.
/notnfs/dnovillo/sbox/tree-ssa-branch/bld/gcc/xgcc -shared-libgcc -B/notnfs/dnovillo/sbox/tree-ssa-branch/bld/gcc/ -nostdinc++ -L/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libstdc++-v3/src -L/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libstdc++-v3/src/.libs -B/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/bin/ -B/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/lib/ -isystem /home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/include -isystem /home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/sys-include -DHAVE_CONFIG_H -I. -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava -I./include -I./gcj -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava -Iinclude -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/include -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../boehm-gc/include -DGC_LINUX_THREADS=1 -D_REENTRANT=1 -DTHREAD_LOCAL_ALLOC=1 -DSILENT=1 -DNO_SIGNALS=1 -DNO_EXECUTE_PERMISSION=1 -DALL_INTERIOR_POINTERS=1 -DJAVA_FINALIZATION=1 -DGC_GCJ_SUPPORT=1 -DATOMIC_UNCOLLECTABLE=1 -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/libltdl -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/libltdl -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/.././libjava/../gcc -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../zlib -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../libffi/include -I../libffi/include -O2 -g -O2 -fno-rtti -fnon-call-exceptions -fdollars-in-identifiers -Wswitch-enum -D__NO_MATH_INLINES -ffloat-store -W -Wall -D_GNU_SOURCE -DPREFIX=\"/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst\" -DLIBDIR=\"/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/lib\" -g -O2 -D_GNU_SOURCE -Wp,-MD,.deps/verify.pp -c /home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/verify.cc -fPIC -DPIC -o .libs/verify.o
/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/include/jvm.h: In
function `void debug_print(const char*, ...)':
/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/verify.cc:40: warning: unused
parameter `const char*fmt'
cc1plus: out of memory allocating 758112432 bytes after a total of 7397644 bytes
make[3]: *** [verify.lo] Error 1
make[3]: Leaving directory `/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libjava'
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-04-25 12:36 ` Diego Novillo
@ 2003-04-25 12:53 ` Andrew MacLeod
2003-04-25 17:03 ` Diego Novillo
0 siblings, 1 reply; 13+ messages in thread
From: Andrew MacLeod @ 2003-04-25 12:53 UTC (permalink / raw)
To: Diego Novillo; +Cc: Jason Merrill, Daniel Berlin, gcc
On Fri, 2003-04-25 at 07:16, Diego Novillo wrote:
> On Fri, 2003-04-25 at 07:11, Diego Novillo wrote:
> > Folks,
> >
> > I got a bootstrap failure last night while running SPEC95.
> > One of these patches introduced this regression.
> >
> > I haven't looked very hard, but I *think* it might be your patch
> > Jason. I had done a bootstrap yesterday that included Andrew's
> > and Daniel's patch (although that bootstrap didn't include
> > libjava). It can't be Daniel's because his change is not in the
> > tree-ssa path.
> >
> Maybe it would help if I included the error as well.
>
> /notnfs/dnovillo/sbox/tree-ssa-branch/bld/gcc/xgcc -shared-libgcc -B/notnfs/dnovillo/sbox/tree-ssa-branch/bld/gcc/ -nostdinc++ -L/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libstdc++-v3/src -L/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libstdc++-v3/src/.libs -B/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/bin/ -B/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/lib/ -isystem /home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/include -isystem /home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/i686-pc-linux-gnu/sys-include -DHAVE_CONFIG_H -I. -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava -I./include -I./gcj -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava -Iinclude -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/include -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../boehm-gc/include -DGC_LINUX_THREADS=1 -D_REENTRANT=1 -DTHREAD_LOCAL_ALLOC=1 -DSILENT=1 -DNO_SIGNALS=1 -DNO_EXECUTE_PERMISSION=1 -DALL_INTERIOR_POINTERS=1 -DJAVA_FINALIZATION=1 -DGC_GCJ_SUPPORT=1 -DATOMIC_UNCOLLECTABLE=1 -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/libltdl -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/libltdl -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/.././libjava/../gcc -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../zlib -I/home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/../libffi/include -I../libffi/include -O2 -g -O2 -fno-rtti -fnon-call-exceptions -fdollars-in-identifiers -Wswitch-enum -D__NO_MATH_INLINES -ffloat-store -W -Wall -D_GNU_SOURCE -DPREFIX=\"/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst\" -DLIBDIR=\"/home/dnovillo/perf/sbox/tree-ssa-branch/local/inst/lib\" -g -O2 -D_GNU_SOURCE -Wp,-MD,.deps/verify.pp -c /home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/verify.cc -fPIC -DPIC -o .libs/verify.o
> /home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/include/jvm.h: In
> function `void debug_print(const char*, ...)':
> /home/dnovillo/perf/sbox/tree-ssa-branch/local/src/libjava/verify.cc:40: warning: unused
> parameter `const char*fmt'
>
> cc1plus: out of memory allocating 758112432 bytes after a total of 7397644 bytes
> make[3]: *** [verify.lo] Error 1
> make[3]: Leaving directory `/notnfs/dnovillo/sbox/tree-ssa-branch/bld/i686-pc-linux-gnu/libjava'
>
Then it could be mine... Im likely to be chewing up a lot of memory in
places :-) I'll try to have a look today.
Andrew
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-04-25 12:53 ` Andrew MacLeod
@ 2003-04-25 17:03 ` Diego Novillo
0 siblings, 0 replies; 13+ messages in thread
From: Diego Novillo @ 2003-04-25 17:03 UTC (permalink / raw)
To: Andrew MacLeod; +Cc: Jason Merrill, Daniel Berlin, gcc
On Fri, Apr 25, 2003 at 08:29:42AM -0400, Andrew MacLeod wrote:
> Then it could be mine... Im likely to be chewing up a lot of memory in
> places :-) I'll try to have a look today.
>
Yes. It's your patch. It worked on the 1Gb machine, but not on
this one. Jason, you're off the hook ;)
Diego.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-29 6:41 ` Diego Novillo
@ 2003-11-29 11:53 ` Andrew MacLeod
0 siblings, 0 replies; 13+ messages in thread
From: Andrew MacLeod @ 2003-11-29 11:53 UTC (permalink / raw)
To: Diego Novillo; +Cc: Steven Bosscher, gcc mailing list
On Fri, 2003-11-28 at 23:26, Diego Novillo wrote:
> On Fri, 2003-11-28 at 19:28, Steven Bosscher wrote:
>
> > I also just updated my tree, and I _still_ have a comparison failure for
> > intl.o. Honza said on IRC earlier today that he is seeing comparison
> > failures as well. What's going on here....
> >
> No idea. I've just finished bootstraps on ia64, amd64 and x86. All
> without a glitch (though all had local changes). We'll see what the
> testers say tomorrow.
My checkout from this morning with a scratch build and bootstrap on x86
worked fine too.
Andrew
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-29 2:30 ` Steven Bosscher
@ 2003-11-29 6:41 ` Diego Novillo
2003-11-29 11:53 ` Andrew MacLeod
0 siblings, 1 reply; 13+ messages in thread
From: Diego Novillo @ 2003-11-29 6:41 UTC (permalink / raw)
To: Steven Bosscher; +Cc: gcc
On Fri, 2003-11-28 at 19:28, Steven Bosscher wrote:
> I also just updated my tree, and I _still_ have a comparison failure for
> intl.o. Honza said on IRC earlier today that he is seeing comparison
> failures as well. What's going on here....
>
No idea. I've just finished bootstraps on ia64, amd64 and x86. All
without a glitch (though all had local changes). We'll see what the
testers say tomorrow.
Diego.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-28 20:38 ` Diego Novillo
@ 2003-11-29 2:30 ` Steven Bosscher
2003-11-29 6:41 ` Diego Novillo
0 siblings, 1 reply; 13+ messages in thread
From: Steven Bosscher @ 2003-11-29 2:30 UTC (permalink / raw)
To: Diego Novillo; +Cc: gcc
On Friday 28 November 2003 19:10, Diego Novillo wrote:
> On Fri, 2003-11-28 at 12:11, Steven Bosscher wrote:
> > on i686-pc-linux-gnu
>
> I just updated my local tree and it bootstraps fine on x86.
I also just updated my tree, and I _still_ have a comparison failure for
intl.o. Honza said on IRC earlier today that he is seeing comparison
failures as well. What's going on here....
Gr.
Steven
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-28 18:10 ` Steven Bosscher
@ 2003-11-28 20:38 ` Diego Novillo
2003-11-29 2:30 ` Steven Bosscher
0 siblings, 1 reply; 13+ messages in thread
From: Diego Novillo @ 2003-11-28 20:38 UTC (permalink / raw)
To: Steven Bosscher; +Cc: gcc
On Fri, 2003-11-28 at 12:11, Steven Bosscher wrote:
> on i686-pc-linux-gnu
>
I just updated my local tree and it bootstraps fine on x86.
Diego.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-28 18:06 [tree-ssa] bootstrap failure Steven Bosscher
2003-11-28 18:10 ` Steven Bosscher
@ 2003-11-28 18:12 ` Diego Novillo
1 sibling, 0 replies; 13+ messages in thread
From: Diego Novillo @ 2003-11-28 18:12 UTC (permalink / raw)
To: Steven Bosscher; +Cc: gcc
On Fri, 2003-11-28 at 12:08, Steven Bosscher wrote:
> Bootstrap comparison failure!
> intl.o differs
> make[1]: *** [gnucompare] Error 1
> make[1]: Leaving directory `/home/steven/devel/ssabuild/gcc'
> make: *** [bootstrap] Error 2
>
> configured with "--enable-languages=c,c++,objc,f95"
> last updated Fri Nov 28 13:39:34 UTC 2003.
>
> Does anyone else see this?
>
Nope. All the testers worked fine last night. Host? OS?
Diego.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] bootstrap failure
2003-11-28 18:06 [tree-ssa] bootstrap failure Steven Bosscher
@ 2003-11-28 18:10 ` Steven Bosscher
2003-11-28 20:38 ` Diego Novillo
2003-11-28 18:12 ` Diego Novillo
1 sibling, 1 reply; 13+ messages in thread
From: Steven Bosscher @ 2003-11-28 18:10 UTC (permalink / raw)
To: gcc, dnovillo
On Friday 28 November 2003 18:08, Steven Bosscher wrote:
> Hi,
>
> I have a comparison failure when I try to bootstrap the tree-ssa branch:
grrrrrr....
on i686-pc-linux-gnu
^ permalink raw reply [flat|nested] 13+ messages in thread
* [tree-ssa] bootstrap failure
@ 2003-11-28 18:06 Steven Bosscher
2003-11-28 18:10 ` Steven Bosscher
2003-11-28 18:12 ` Diego Novillo
0 siblings, 2 replies; 13+ messages in thread
From: Steven Bosscher @ 2003-11-28 18:06 UTC (permalink / raw)
To: gcc, dnovillo
Hi,
I have a comparison failure when I try to bootstrap the tree-ssa branch:
case "gnucompare" in *compare | *compare-lean ) stage=2 ;; * ) stage=`echo gnucompare | sed -e 's,^[a-z]*compare\([0-9][0-9]*\).*,\1,'` ;; esac; \
if [ -f .bad_compare ]; then \
echo "Bootstrap comparison failure!"; \
cat .bad_compare; \
exit 1; \
else \
case "gnucompare" in \
*-lean ) rm -rf stage$stage ;; \
*) ;; \
esac; true; \
fi
Bootstrap comparison failure!
intl.o differs
make[1]: *** [gnucompare] Error 1
make[1]: Leaving directory `/home/steven/devel/ssabuild/gcc'
make: *** [bootstrap] Error 2
configured with "--enable-languages=c,c++,objc,f95"
last updated Fri Nov 28 13:39:34 UTC 2003.
Does anyone else see this?
Gr.
Steven
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [tree-ssa] Bootstrap failure
2003-07-29 11:37 [tree-ssa] Bootstrap failure Steven Bosscher
@ 2003-07-29 18:07 ` Phil Edwards
0 siblings, 0 replies; 13+ messages in thread
From: Phil Edwards @ 2003-07-29 18:07 UTC (permalink / raw)
To: Steven Bosscher; +Cc: gcc, dnovillo
On Tue, Jul 29, 2003 at 09:08:27AM +0200, Steven Bosscher wrote:
> -c ../../tree-ssa/gcc/crtstuff.c -DCRT_BEGIN \
> -o crtbegin.o
> /usr/include/bits/string2.h: In function `__strcspn_c1':
> /usr/include/bits/string2.h:926: internal compiler error: in var_ann, at tree-flow-inline.h:36
Same error here.
I have no range-of-changelogs to show, unfortunately.
Phil
--
If ye love wealth greater than liberty, the tranquility of servitude greater
than the animating contest for freedom, go home and leave us in peace. We seek
not your counsel, nor your arms. Crouch down and lick the hand that feeds you;
and may posterity forget that ye were our countrymen. - Samuel Adams
^ permalink raw reply [flat|nested] 13+ messages in thread
* [tree-ssa] Bootstrap failure
@ 2003-07-29 11:37 Steven Bosscher
2003-07-29 18:07 ` Phil Edwards
0 siblings, 1 reply; 13+ messages in thread
From: Steven Bosscher @ 2003-07-29 11:37 UTC (permalink / raw)
To: gcc, dnovillo
Something in the last 12 hours broke bootstrap with checking enabled for
the tree-ssa branch (last updated 06:22:22 UTC 2003) on i686-pc-linux-gnu:
/home/steven/devel/ssabuild/gcc/xgcc -B/home/steven/devel/ssabuild/gcc/ -B/opt/experimental//i686-pc-linux-gnu/bin/ -B/opt/experimental//i686-pc-linux-gnu/lib/ -isystem /opt/experimental//i686-pc-linux-gnu/include -isystem /opt/experimental//i686-pc-linux-gnu/sys-include -O2 -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -isystem ./include -I. -I. -I../../tree-ssa/gcc -I../../tree-ssa/gcc/. -I../../tree-ssa/gcc/config -I../../tree-ssa/gcc/../include -I../../tree-ssa/gcc/../libbanshee/libcompat -I../../tree-ssa/gcc/../libbanshee -I../../tree-ssa/gcc/../libbanshee/points-to -g0 -finhibit-size-directive -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss -fno-unit-at-a-time -fno-omit-frame-pointer \
-c ../../tree-ssa/gcc/crtstuff.c -DCRT_BEGIN \
-o crtbegin.o
/usr/include/bits/string2.h: In function `__strcspn_c1':
/usr/include/bits/string2.h:926: internal compiler error: in var_ann, at tree-flow-inline.h:36
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
make[2]: *** [crtbegin.o] Error 1
make[2]: Leaving directory `/home/steven/devel/ssabuild/gcc'
make[1]: *** [stage1_build] Error 2
make[1]: Leaving directory `/home/steven/devel/ssabuild/gcc'
make: *** [bootstrap] Error 2
#1 0x08104a0d in var_ann (t=0x0) at tree-flow-inline.h:36
36 abort ();
(gdb) l
31 #if defined ENABLE_CHECKING
32 if (t == NULL_TREE
33 || !DECL_P (t)
34 || (t->common.ann
35 && t->common.ann->common.type != VAR_ANN))
36 abort ();
37 #endif
38
39 return (var_ann_t) t->common.ann;
40 }
(gdb) p t
$2 = 0x0
(gdb) where
#0 fancy_abort (file=0x8512ba0 "../../tree-ssa/gcc/tree-flow-inline.h",
line=36, function=0x8512b86 "var_ann")
at ../../tree-ssa/gcc/diagnostic.c:1380
#1 0x08104a0d in var_ann (t=0x0) at tree-flow-inline.h:36
#2 0x0810400a in set_is_used (t=0x0) at ../../tree-ssa/gcc/tree-ssa.c:2124
#3 0x08100053 in insert_copy_on_edge (e=0x87628dc, dest=0x40217f50,
src=0x40218f6c) at ../../tree-ssa/gcc/tree-ssa.c:881
#4 0x08101123 in eliminate_phi (e=0x87628dc, i=0, g=0x875f980)
at ../../tree-ssa/gcc/tree-ssa.c:1173
#5 0x081025cf in rewrite_out_of_ssa (fndecl=0x40217c40)
at ../../tree-ssa/gcc/tree-ssa.c:1711
We lose the tree pointer in tree_simple:get_base_symbol (tree t) when it is
called from set_is_used (). get_base_symbol() expects a decl, a ref or an ssa
name, but we call it with an INTEGER_CST.
Probably fallout from the "constant propatation into PHIs"-patches??
Gr.
Steven
^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2003-11-29 6:44 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-04-25 12:32 [tree-ssa] bootstrap failure Diego Novillo
2003-04-25 12:36 ` Diego Novillo
2003-04-25 12:53 ` Andrew MacLeod
2003-04-25 17:03 ` Diego Novillo
2003-07-29 11:37 [tree-ssa] Bootstrap failure Steven Bosscher
2003-07-29 18:07 ` Phil Edwards
2003-11-28 18:06 [tree-ssa] bootstrap failure Steven Bosscher
2003-11-28 18:10 ` Steven Bosscher
2003-11-28 20:38 ` Diego Novillo
2003-11-29 2:30 ` Steven Bosscher
2003-11-29 6:41 ` Diego Novillo
2003-11-29 11:53 ` Andrew MacLeod
2003-11-28 18:12 ` Diego Novillo
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).