[Bug target/98670] [11 Regression] ICE in final_scan_insn_1, at final.c:3091 since r11-6628-gb668a06e37f72fd9

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98670

--- Comment #2 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:0efdc7b31c2aeb3b0414a838e90014172b87302f

commit r11-6664-g0efdc7b31c2aeb3b0414a838e90014172b87302f
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Thu Jan 14 12:55:19 2021 +0100

    i386: Fix the pmovzx SSE4.1 define_insn_and_split patterns [PR98670]

    I've made two mistakes in the *sse4_1_zero_extend* define_insn_and_split
    patterns.  One is that when it uses vector_operand, it should use Bm rather
    than m constraint, and the other one is that because it is a post-reload
    splitter it needs isa attribute to select which alternatives are valid for
    which ISAs.  Sorry for messing this up.

    2021-01-14  Jakub Jelinek  <jakub@redhat.com>

            PR target/98670
            * config/i386/sse.md (*sse4_1_zero_extendv8qiv8hi2_3,
            *sse4_1_zero_extendv4hiv4si2_3, *sse4_1_zero_extendv2siv2di2_3):
            Use Bm instead of m for non-avx.  Add isa attribute.

            * gcc.target/i386/pr98670.c: New test.