[Bug middle-end/126410] [14/15/16 Regression] vectorizer aligns 16-byte array to 64byte

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126410

--- Comment #10 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-16 branch has been updated by Richard Biener
<rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:f1c1322dca42d88edef68762ab52a4c6aacd218a

commit r16-9460-gf1c1322dca42d88edef68762ab52a4c6aacd218a
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Sun Jul 26 19:21:55 2026 +0800

    vectorizer: Don't enlarge base alignment during costing

    We errorneously commit the recorded base alignment during costing
    since the costing refactoring.  This causes alignment to be
    computed based on a possibly larger vector type than used in the
    end.

    gcc/

            PR middle-end/126410
            * tree-vect-stmts.cc (vectorizable_store): Only commit to
            recorded base alignment when transforming.
            (vectorizable_load): Likewise.

    gcc/testsuite/

            PR middle-end/126410
            * gcc.target/i386/pr126410-1a.c: New test.
            * gcc.target/i386/pr126410-1b.c: Likewise.
            * gcc.target/i386/pr126410-1c.c: Likewise.
            * gcc.target/i386/pr126410-2a.c: Likewise.
            * gcc.target/i386/pr126410-2b.c: Likewise.
            * gcc.target/i386/pr126410-2c.c: Likewise.

    Co-Authored-By: Richard Biener  <rguenther@suse.de>
    Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
    (cherry picked from commit 7c7c6626410c261d188a96d9b2a15d3781d0dcfb)