From: Luke Kenneth Casson Leighton Date: Thu, 27 Apr 2023 09:42:04 +0000 (+0100) Subject: whitespace X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=dc3aa04071262dabc0b5376cc0e58163ab09ad99;p=libreriscv.git whitespace --- diff --git a/openpower/sv/cookbook/chacha20.mdwn b/openpower/sv/cookbook/chacha20.mdwn index 44ae30d2a..cee371553 100644 --- a/openpower/sv/cookbook/chacha20.mdwn +++ b/openpower/sv/cookbook/chacha20.mdwn @@ -122,17 +122,22 @@ Let's list the additions only: ## Introduction to Vertical-First Mode -We're going to use Vertical-First mode (VF) to implement this, so we will -first do a short introduction to VF. In normal Horizontal Vector mode, or even in traditional SIMD mode, -the instruction is executed across a (Horizontal) Vector. So, if we have, for example `VL=8`, then the instruction +We're going to use Vertical-First mode (VF) to implement this, so we +will first do a short introduction to VF. In normal Horizontal Vector +mode, or even in traditional SIMD mode, the instruction is executed +across a (Horizontal) Vector. So, if we have, for example `VL=8`, then +the instruction sv.add *RT, *RA, *RB # RT = RA + RB -will be executed on all elements of the vector, **before** moving to the next assembly instruction. -This behaviour changes in Vertical-First mode. In VF mode, the instruction is executed on a **single** -element of the vector and then moves to the next assembly instruction. It continues to do so until execution -reaches the `svstep` instruction where processing will be moved to the next element/register in the vector. Branching -to the beginning of the loop does not occur automatically though, a branch instruction will have to be added manually. +will be executed on all elements of the vector, **before** moving to +the next assembly instruction. This behaviour changes in Vertical-First +mode. In VF mode, the instruction is executed on a **single** element of +the vector and then moves to the next assembly instruction. It continues +to do so until execution reaches the `svstep` instruction where processing +will be moved to the next element/register in the vector. Branching to +the beginning of the loop does not occur automatically though, a branch +instruction will have to be added manually. ## Application of VF mode in the Xchacha20 loop