Re: [PATCH 2/2] add masked-epilogue tuning

2025-07-07 Thread Richard Biener
On Mon, 7 Jul 2025, Hongtao Liu wrote: > On Mon, Jul 7, 2025 at 3:18 PM Hongtao Liu wrote: > > > > On Fri, Jul 4, 2025 at 5:45 PM Richard Biener wrote: > > > > > > The following adds a x86 tuning to enable the use of AVX512 masked > > > epilogues in cases we heuristically determine it to be not

Re: [PATCH 2/2] add masked-epilogue tuning

2025-07-07 Thread Hongtao Liu
On Mon, Jul 7, 2025 at 3:18 PM Hongtao Liu wrote: > > On Fri, Jul 4, 2025 at 5:45 PM Richard Biener wrote: > > > > The following adds a x86 tuning to enable the use of AVX512 masked > > epilogues in cases we heuristically determine it to be not detrimental > > by high chance. Basically problemat

Re: [PATCH 2/2] add masked-epilogue tuning

2025-07-07 Thread Hongtao Liu
On Fri, Jul 4, 2025 at 5:45 PM Richard Biener wrote: > > The following adds a x86 tuning to enable the use of AVX512 masked > epilogues in cases we heuristically determine it to be not detrimental > by high chance. Basically problematic cases are when there are > data streams that are both stored

Re: [PATCH 2/2] add masked-epilogue tuning

2025-07-04 Thread Jan Hubicka
> The following adds a x86 tuning to enable the use of AVX512 masked > epilogues in cases we heuristically determine it to be not detrimental > by high chance. Basically problematic cases are when there are > data streams that are both stored and loaded from and an outer loop > could end up execut

[PATCH 2/2] add masked-epilogue tuning

2025-07-04 Thread Richard Biener
The following adds a x86 tuning to enable the use of AVX512 masked epilogues in cases we heuristically determine it to be not detrimental by high chance. Basically problematic cases are when there are data streams that are both stored and loaded from and an outer loop could end up executing only t