On 11/07/2017 06:15 PM, Andreas Schwab wrote:
This breaks g++.dg/torture/pr64669.C with -O3 on ia64:

$ gcc/xg++ -Bgcc/ ../../gcc/gcc/testsuite/g++.dg/torture/pr64669.C -nostdinc++ 
-Iia64-suse-linux/libstdc++-v3/include/ia64-suse-linux 
-Iia64-suse-linux/libstdc++-v3/include -I../libstdc++-v3/libsupc++ 
-I../libstdc++-v3/include/backward -I../libstdc++-v3/testsuite/util -O3 -S -o 
pr64669.s
../../gcc/gcc/testsuite/g++.dg/torture/pr64669.C: In member function ‘const 
char* Lex::advance_one_char(const char*, bool, unsigned int*, bool*)’:
../../gcc/gcc/testsuite/g++.dg/torture/pr64669.C:65:1: error: qsort comparator 
non-negative on sorted output: 1
  }
  ^
during RTL pass: mach
../../gcc/gcc/testsuite/g++.dg/torture/pr64669.C:65:1: internal compiler error: 
qsort checking failed
0x40000000001cc80f qsort_chk_error
         ../../gcc/vec.c:222
0x40000000025d20cf qsort_chk(void*, unsigned long, unsigned long, int (*)(void 
const*, void const*))
         ../../gcc/vec.c:274
0x40000000014f680f vec<_expr*, va_heap, vl_embed>::qsort(int (*)(void const*, 
void const*))
         ../../gcc/vec.h:973
0x40000000014f680f vec<_expr*, va_heap, vl_ptr>::qsort(int (*)(void const*, 
void const*))
         ../../gcc/vec.h:1735
0x40000000014f680f fill_vec_av_set
         ../../gcc/sel-sched.c:3725
0x40000000014fc59f fill_ready_list
         ../../gcc/sel-sched.c:4022
0x40000000014fc59f find_best_expr
         ../../gcc/sel-sched.c:4382
0x40000000014fc59f fill_insns
         ../../gcc/sel-sched.c:5539
0x40000000014fc59f schedule_on_fences
         ../../gcc/sel-sched.c:7356
0x40000000014fc59f sel_sched_region_2
         ../../gcc/sel-sched.c:7494
0x4000000001503acf sel_sched_region_1
         ../../gcc/sel-sched.c:7536
0x4000000001503acf sel_sched_region(int)
         ../../gcc/sel-sched.c:7637
0x4000000001504e6f run_selective_scheduling()
         ../../gcc/sel-sched.c:7713
0x4000000001df9bdf ia64_reorg
         ../../gcc/config/ia64/ia64.c:9854
0x400000000146d40f execute
         ../../gcc/reorg.c:3947

Andreas.


Hi Andreas.

That will be very probably dup of PR82398.

Martin

Reply via email to