https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91568
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |marxin at gcc dot gnu.org
Known to work| |10.0
Target Milestone|--- |9.3
Known to fail| |9.2.0
--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
Reduced test-case:
$ cat pr91568.f90
subroutine h3dall(z,hvec,hder)
complex *16 hvec(1),hder(0:1)
complex *16 z,zinv,ztmp/1.0/
if (abs0ifder.eq.1) then
zinv=1.0/z
do i=1,nterms
dtmp=i
ztmp=zinv*dtmp
hder(i)=hvec(i-1)-ztmp*hvec(i)
enddo
endif
end
$ gcc-9 pr91568.f90 -Ofast
pr91568.f90:9:18:
6 | do i=1,nterms
| 2
......
9 | hder(i)=hvec(i-1)-ztmp*hvec(i)
| 1
Warning: Array reference at (1) out of bounds (0 < 1) in loop beginning at (2)
during GIMPLE pass: vect
pr91568.f90:1:0:
1 | subroutine h3dall(z,hvec,hder)
|
internal compiler error: in vect_schedule_slp_instance, at tree-vect-slp.c:3922
0x7fa337825bca __libc_start_main
../csu/libc-start.c:308
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <https://bugs.opensuse.org/> for instructions.
I can reproduce that only on 9-branch.
