From: lkcl Date: Fri, 7 Apr 2023 20:49:07 +0000 (+0100) Subject: (no commit message) X-Git-Tag: opf_rfc_ls012_v1~87 X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=647635414d3c0c9ebe14091f0b4d63c3c41baafc;p=libreriscv.git --- diff --git a/openpower/sv/rfc/ls012.mdwn b/openpower/sv/rfc/ls012.mdwn index 1ee99d542..21266a75a 100644 --- a/openpower/sv/rfc/ls012.mdwn +++ b/openpower/sv/rfc/ls012.mdwn @@ -14,6 +14,21 @@ of a borderline Vectoriseable Defined Word is `mv.swizzle` which only really becomes high-priority for Vector GPU and HPC Workloads, but has less merit as a Scalar-only operation. +Power ISA Scalar (SFFS) has not been significantly advanced in 12 years. +With VSX bring 914 instructions and 128-bit it is far too much for any +new team to consider (10 years development effort) and far outside of +Embedded or Tablet/Desktop/Laptop power budgets. Thus bringing Power Scalar +up-to-date to modern standards is a reasonable goal, and the advantage is +that lessons can be learned from other ISAs. + +SVP64 Prefixing - also known by the terms "Zero-Overhead-Loop-Prefixing" +as well as "True-Scalable Vector Prefixing" - also literally brings new +dimensions to the Power ISA. Thus when adding new Scalar "Defined Words" +it has to unavoidably be taken into consideration their value when +Vectorised. + + + Instruction count guide and approximate priority order: * 6 - SVP64 Management [[ls008]] [[ls009]] [[ls010]]