Bug 1798970 Comment 0 Edit History

Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.

Original comment by
on 
FMA3 is added to Intel SIMD at the same time as AVX2. We currently don't test this CPU feature in the SM. To get expected performance FMA lowering expected to be (as described at https://github.com/WebAssembly/relaxed-simd/issues/27):

   - relaxed f32x4.fma = VFMADD231PS
   - relaxed f32x4.fnma = VFNMADD231PS
   - relaxed f64x2.fma = VFMADD231PS
   - relaxed f64x2.fnma = VFNMADD231PS
Revision 1 by
on 
FMA3 is added to Intel SIMD at the same time as AVX2. We currently don't test this CPU feature in the SM. To get expected performance FMA lowering expected to be (as described at https://github.com/WebAssembly/relaxed-simd/issues/27):

   - relaxed f32x4.fma = VFMADD231PS
   - relaxed f32x4.fnma = VFNMADD231PS
   - relaxed f64x2.fma = VFMADD231PD
   - relaxed f64x2.fnma = VFNMADD231PD

Back to Bug 1798970 Comment 0