RFR: 8289552: Make intrinsic conversions between bit representations of half precision values and floats [v6]

Sandhya Viswanathan sviswanathan at openjdk.org
Thu Sep 1 23:48:54 UTC 2022


On Thu, 1 Sep 2022 18:31:07 GMT, Smita Kamath <svkamath at openjdk.org> wrote:

>> 8289552: Make intrinsic conversions between bit representations of half precision values and floats
>
> Smita Kamath has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Addressed review comments, updated microbenchmark

src/hotspot/cpu/x86/assembler_x86.cpp line 1931:

> 1929: }
> 1930: 
> 1931: void Assembler::vcvtps2ph(Address dst, KRegister mask, XMMRegister src, int imm8, int vector_len) {

This could be named as evcvtps2ph as the instruction with mask register is evex only.

-------------

PR: https://git.openjdk.org/jdk/pull/9781


More information about the core-libs-dev mailing list