https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91213
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |NEW
Resolution|FIXED |---
--- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
trunk GCC riscv:
li a5,31
subw a0,a5,a0
vs trunk clang riscv:
xori a0, a0, 31
ret
(In reply to Robin Dapp from comment #8)
> Hacked something together, inspired by the other cases that try two
> different sequences. Does this go into the right direction? Works for me
> on s390. I see some regressions related to predictive commoning that I will
> look into.
The only different to this patch is now you can use the ranger instead to get
the tree_nonzero_bits to get more information.
