https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94871

--- Comment #3 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Roger Sayle <[email protected]>:

https://gcc.gnu.org/g:0895e430b5e59e3ee18c4763c493ee05a0c363e8

commit r17-1974-g0895e430b5e59e3ee18c4763c493ee05a0c363e8
Author: Roger Sayle <[email protected]>
Date:   Mon Jun 29 14:07:08 2026 +0100

    PR target/94871: Simplify (not (eq x y)) as (ne x y) for vectors.

    This patch is my proposed fix for PR target/94871, a missed optimization
    for simplifying vector comparisons in the RTL optimizers (combine) that
    includes both middle-end and i386 backend changes.

    The testcase from the original bugzilla PR is:

    typedef long long int64_t;
    typedef signed char int8_t;

    typedef double v2df __attribute__((vector_size(16)));
    typedef int64_t v2di __attribute__((vector_size(16)));
    typedef int8_t v16qi __attribute__((vector_size(16)));

    inline v2di set1_epi8(int8_t a)
    {
        return (v2di)(v16qi){a, a, a, a, a, a, a, a, a, a, a, a, a, a, a, a};
    }

    v2di foo(v2df a, v2df b)
    {
        return ((v2di)__builtin_ia32_cmpeqpd(a, b) ^ set1_epi8(0xFF));
    }

    where GCC with -O2 currently generates:

    foo:    cmpeqpd %xmm1, %xmm0
            pcmpeqd %xmm1, %xmm1
            pxor    %xmm1, %xmm0
            ret

    where with this pair of changes, we instead generate:

    foo:    cmpneqpd        %xmm1, %xmm0
            ret

    Basically, the usual optimizations (not (eq x y)) -> (ne x y) and
    (not (ne x y)) -> (eq x y), but applied to vector mode comparisons
    (and in this case floating point vector comparisons).

    The first (i386 backend) part is to observe that the middle-end
    RTL optimizers enjoy canonicalizing floating point vector comparisons
    to return integer vector results.  Hence -fdump-rtl-combine-all
    contains the lines:

    Failed to match this instruction:
    (set (reg:V2DI 105)
        (eq:V2DI (reg:V2DF 106 [ aD.2972 ])
            (reg:V2DF 107 [ bD.2973 ])))

    Currently sse.md contains patterns for matching vector comparisons
    for comparing V2DFs returning a V2DF, but not the equivalent patterns
    for comparing V2DFs but returning a V2DI.  Very easily handled by
    cloning the existing define_insn but substituting <sseintvecmode> for
    the comparison result mode.

    The second (simplify-rtx.cc) part is to tweak the current
    (not (eq X Y)) -> (ne X Y) transformation in
    simplify_unary_operation_1 to allow/handle vector modes.
    In fact, this is a (latent) bug fix, as the current logic
    checks STORE_FLAG_VALUE even for modes where this isn't
    relevant.  The target's STORE_FLAG_VALUE is applicable for
    SCALAR_INT_MODE_P comparisons; VECTOR_STORE_FLAG_VALUE
    should be used for VECTOR integer modes.

    For symmetry, I've also added the equivalent optimization for
    targets with STORE_FLAG_VALUE == 1 or VECTOR_STORE_FLAG_VALUE
    == const1_rtx (such as aarch64), such that (not (neg (eq x y)))
    -> (neg (ne x y)).

    2026-06-29  Roger Sayle  <[email protected]>

    gcc/ChangeLog
            PR target/94871
            * simplify-rtx.cc (simplify_unary_operation_1) <case NOT>:
            Update (not (eq X Y)) -> (ne X Y) to support vector comparions
            using VECTOR_STORE_FLAG_VALUE, and check SCALAR_INT_MODE_P
            when checking STORE_FLAG_VALUE.  Add equivalent optimization
            (not (neg (eq X Y))) -> (neg (ne X Y)) for targets with
            STORE_FLAG_VALUE == 1 or VECTOR_STORE_FLAG_VALUE is const1_rtx.

            * config/i386/sse.md (*<sse>_imaskcmp<mode>3_comm): Clone
            of *<sse>_maskcmp<mode>3_comm pattern with integer vector
            result.

    gcc/testsuite/ChangeLog
            PR target/94871
            * gcc.target/i386/pr94871.c: New test case.

Reply via email to