[Bug target/93172] with AVX512 masked mov assigning zero can use {z}

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93172 --- Comment #3 from Segher Boessenkool --- Why should it be limited that way? simplify_rtx does not / should not care about peculiarities of a certain architecture or microarchitecture. A transform like this would be a good idea I think, and wo

[Bug target/93172] with AVX512 masked mov assigning zero can use {z}

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93172 --- Comment #2 from Jakub Jelinek --- This isn't a regression, so I don't think we can handle it for GCC 10. Anyway, with the simpler ~k instead of _knot_mask16(k) we don't optimize that in combine either: (set (reg:V16SF 88) (vec_merge:V16S

[Bug target/93172] with AVX512 masked mov assigning zero can use {z}

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93172 Martin Liška changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|