From 932c6db16ee5f108a5eb42df79481452c71108b8 Mon Sep 17 00:00:00 2001 From: Luke Kenneth Casson Leighton Date: Thu, 30 Jun 2022 12:46:47 +0100 Subject: [PATCH] clarify intro --- openpower/sv.mdwn | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/openpower/sv.mdwn b/openpower/sv.mdwn index 1255e4d52..869f50c09 100644 --- a/openpower/sv.mdwn +++ b/openpower/sv.mdwn @@ -190,14 +190,15 @@ AVX-512 and SVE2 truly "Scalable".* # Major opcodes summary -Simple-V itself only requires four instructions with 6-bit Minor XO +Simple-V itself only requires five instructions with 6-bit Minor XO (bits 26-31), and the SVP64 Prefix Encoding requires 25% space of the EXT001 Major Opcode. There are **no** Vector Instructions and consequently **no further opcode space is required**. That said: for the target workloads for which Scalable Vectors are typically -used, the Scalar ISA on which SV critically relies is somewhat anaemic. +used, the Scalar ISA on which those workloads critically rely +is somewhat anaemic. The Libre-SOC Team has therefore been addressing that by developing a number of Scalar instructions in specialist areas (Big Integer, Cryptography, 3D, Audio/Video, DSP) and it is these which require @@ -245,10 +246,12 @@ is under severe design pressure as it is insufficient to hold the full extent of the instruction additions required to create a Hybrid 3D CPU-VPU-GPU. -**Whilst SVP64 is only 4 instructions +**Whilst SVP64 is only 5 instructions the heavy focus on VSX for the past 12 years has left the SFFS Level -anaemic and out-of-date compared to ARM and x86. Approximately -100 additional Scalar Instructions are up for proposal** +anaemic and out-of-date compared to ARM and x86. This is partially +a blessing as the Scalar ISA has remained clean. Approximately +100 additional (optional) Scalar Instructions are up for proposal to bring SFFS +up-to-date** # Other -- 2.30.2