News for hackers. Stuff that matters.
Re:Flavors of SIMD by gnufx on Tuesday February 19, 2019 @12:58PM (#19201124)
I can't comment on the GPU comments, but you may be better off leaving the vectorization to gcc than using the simd pragma. On haswell, it uses avx2, but not fma, so you loose a factor of two on GEMM, for instance. The GCC manual also gives an example for the ivdep pragma.
Re:Flavors of SIMD by pjmlp on Wednesday February 20, 2019 @09:58AM (#19208286)
Would not that be just a question of missing improvements?

If I recall correctly, OpenJDK can use FMA thanks to Intel contributions.

Re:Flavors of SIMD by gnufx on Wednesday February 20, 2019 @06:50PM (#19213253)
I don't see how openjdk is related to the openmp pragma. GCC has no problem using FMA if you just let it, avoiding the pragma which simply says "simd".
Re:Flavors of SIMD by pjmlp on Thursday February 21, 2019 @01:38AM (#19215008)
I understood that GCC auto-vectorization wouldn't do it currently, and hence gave an example where auto-vectorization does make use of it, assuming I remember Intel's session at CodeONE correctly.