Open chriselrod opened 2 years ago
If anyone wants to give it a go, see here for an example: https://github.com/JuliaSIMD/LoopVectorization.jl/blob/ed466fb1ca7e92b70b98d6ee50eb5544b64678e1/test/shuffleloadstores.jl#L143-L157
I'd be happy to answer any questions.
This requires the FMA instruction set, meaning x86_64 only. Thus, using these should be behind a check for this.
If anyone wants to give it a go, see here for an example: https://github.com/JuliaSIMD/LoopVectorization.jl/blob/ed466fb1ca7e92b70b98d6ee50eb5544b64678e1/test/shuffleloadstores.jl#L143-L157
I'd be happy to answer any questions.
This requires the FMA instruction set, meaning x86_64 only. Thus, using these should be behind a check for this.