https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100711
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |segher at gcc dot gnu.org
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
7: r82:SI=~r89:SI
REG_DEAD r89:SI
8: r88:V4SI=vec_duplicate(r82:SI)
REG_DEAD r82:SI
9: r87:V4SI=r88:V4SI&r90:V4SI
REG_DEAD r90:V4SI
REG_DEAD r88:V4SI
I suppose we're confused about the vec_duplicate. Would generally swapping
the duplicate and the bit_not be profitable? Eventually it's a simplification
combine could try - I belive it has some cases where it tries variants of the
original instructions when combining. Adding a combine helper pattern
looks like putting too much burden on the backend IMHO.
We don't have a generic nand optab so handling this in ISEL on gimple
isn't straight-forward.
But combine and/or forwprop could do this.
