https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113204
--- Comment #9 from Sam James <sjames at gcc dot gnu.org> ---
(In reply to Xi Ruoyao from comment #5)
> I've hit a similar ICE testing libbacktrace with LTO bootstrapped GCC on
> LoongArch:
I hit this today too.
Unfortunately, it seems that libbacktrace gets relinked (and maybe sometimes
rebuilt?) repeatedly and I couldn't convert all the files to preprocessed
variants in time before the build finished, and I was too lazy to properly
suspend it.
But I can reproduce it with:
```
$ gcc btest-btest.i btest_lto-testlib.i fileline.i dwarf.o elf.o -O2 -flto
/var/tmp/portage/sys-devel/gcc-15.0.9999/work/gcc-15.0.9999/libbacktrace/btest.c:52:1:
warning: ‘optnone’ attribute directive ignored [-Wattributes]
/var/tmp/portage/sys-devel/gcc-15.0.9999/work/gcc-15.0.9999/libbacktrace/btest.c:166:1:
warning: ‘optnone’ attribute directive ignored [-Wattributes]
lto-wrapper: warning: Extra option to ‘-Xassembler’: -O2, dropping all
‘-Xassembler’ and ‘-Wa’ options.
lto1: error: qsort comparator non-negative on sorted output: 204
during IPA pass: static-var
lto1: internal compiler error: qsort checking failed
0x558d86f3778f internal_error(char const*, ...)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/diagnostic-global-context.cc:491
0x558d86f3784f qsort_chk_error
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/vec.cc:204
0x558d858bc2f0 qsort_chk(void*, unsigned long, unsigned long, int (*)(void
const*, void const*, void*), void*)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/vec.cc:246
0x558d873b50f2 gcc_sort_r(void*, unsigned long, unsigned long, int (*)(void
const*, void const*, void*), void*)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/sort.cc:296
0x558d871d971b vec<symtab_node*, va_heap, vl_embed>::sort(int (*)(void const*,
void const*, void*), void*)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/vec.h:1232
0x558d871d971b vec<symtab_node*, va_heap, vl_ptr>::sort(int (*)(void const*,
void const*, void*), void*)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/vec.h:2252
0x558d871d971b lto_output()
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto-streamer-out.cc:2792
0x558d878d6805 write_lto
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/passes.cc:2780
0x558d878cf8f7 ipa_write_optimization_summaries(lto_symtab_encoder_d*)
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/passes.cc:2970
0x558d878cf1e4 stream_out
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:179
0x558d878cf1e4 stream_out_partitions_1
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:231
0x558d8788ed99 stream_out_partitions
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:290
0x558d8788ed99 lto_wpa_write_files
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:419
0x558d8788ed99 do_whole_program_analysis
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:596
0x558d8788ed99 lto_main()
/usr/src/debug/sys-devel/gcc-15.0.9999/gcc-15.0.9999/gcc/lto/lto.cc:670
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <https://bugs.gentoo.org/> for instructions.
```
and at least dwarf.o and elf.o are from C this time so Go is out of the
equation.
This keeps irritating me so at some point I'll try playing with it like richi
suggested I guess.
