Re: [PATCH][x86] Fix regression from x86 multi-epilogue tuning

2025-05-14 Thread Jan Hubicka
> With the avx512_two_epilogues tuning enabled for zen4 and zen5 > the gcc.target/i386/vect-epilogues-5.c testcase below regresses > and ends up using AVX2 sized vectors for the masked epilogue > rather than AVX512 sized vectors. The following patch rectifies > this and adds coverage for the inten

[PATCH][x86] Fix regression from x86 multi-epilogue tuning

2025-05-14 Thread Richard Biener
With the avx512_two_epilogues tuning enabled for zen4 and zen5 the gcc.target/i386/vect-epilogues-5.c testcase below regresses and ends up using AVX2 sized vectors for the masked epilogue rather than AVX512 sized vectors. The following patch rectifies this and adds coverage for the intended behavi