Re: [x86 PATCH] PR tree-optimization/105668: Provide vcond_mask_v1tiv1ti pattern.

2022-05-23 Thread Richard Biener via Gcc-patches
On Mon, May 23, 2022 at 10:00 AM Uros Bizjak wrote: > > On Mon, May 23, 2022 at 9:16 AM Roger Sayle > wrote: > > > > > > This patch is an alternate/supplementary fix to PR tree-optimization/105668 > > that provides a vcond_mask_v1titi optab/define_expand to the i386 backend. > > An undocumented

Re: [x86 PATCH] PR tree-optimization/105668: Provide vcond_mask_v1tiv1ti pattern.

2022-05-23 Thread Uros Bizjak via Gcc-patches
On Mon, May 23, 2022 at 10:00 AM Uros Bizjak wrote: > > On Mon, May 23, 2022 at 9:16 AM Roger Sayle > wrote: > > > > > > This patch is an alternate/supplementary fix to PR tree-optimization/105668 > > that provides a vcond_mask_v1titi optab/define_expand to the i386 backend. > > An undocumented

Re: [x86 PATCH] PR tree-optimization/105668: Provide vcond_mask_v1tiv1ti pattern.

2022-05-23 Thread Uros Bizjak via Gcc-patches
On Mon, May 23, 2022 at 9:16 AM Roger Sayle wrote: > > > This patch is an alternate/supplementary fix to PR tree-optimization/105668 > that provides a vcond_mask_v1titi optab/define_expand to the i386 backend. > An undocumented feature/bug of GCC's vectorization is that any target that > provides

[x86 PATCH] PR tree-optimization/105668: Provide vcond_mask_v1tiv1ti pattern.

2022-05-23 Thread Roger Sayle
This patch is an alternate/supplementary fix to PR tree-optimization/105668 that provides a vcond_mask_v1titi optab/define_expand to the i386 backend. An undocumented feature/bug of GCC's vectorization is that any target that provides a vec_cmpeq has to also provide a matching vcond_mask. This ba