https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126923

--- Comment #19 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by H.J. Lu <[email protected]>:

https://gcc.gnu.org/g:e7a2a4825d90ddaeed7eb57c3e63d6cad7ccf78f

commit r17-3462-ge7a2a4825d90ddaeed7eb57c3e63d6cad7ccf78f
Author: H.J. Lu <[email protected]>
Date:   Thu Aug 20 15:24:16 2026 +0800

    ext-floating14.C: Compile with -mno-avx512bf16 -mno-avxneconvert

    Compile ext-floating14.C with -mno-avx512bf16 -mno-avxneconvert for x86
    to avoid

    FAIL: g++.dg/cpp23/ext-floating14.C  -std=gnu++23 (test for excess errors)
    FAIL: g++.dg/cpp23/ext-floating14.C  -std=gnu++26 (test for excess errors)
    FAIL: g++.dg/cpp23/ext-floating14.C  -std=gnu++29 (test for excess errors)

    with warnings like

    ext-floating14.C:112:111: warning: â__bfloat16â is redefined from
typedef âshortâ to real â__bf16â since GCC 13.1, be careful of implicit
conversion between â__bf16â and âshortâ; an explicit bitcast may be
needed here

    when AVXNECONVERT or AVX512BF16 are enabled by default.

            PR testsuite/126923
            * g++.dg/cpp23/ext-floating14.C (dg-additional-options): Add
            -mno-avx512bf16 -mno-avxneconvert for x86.

    Signed-off-by: H.J. Lu <[email protected]>

Reply via email to