From 09a167e325b879b07d9e622af7612e99e0053d99 Mon Sep 17 00:00:00 2001 From: lkcl Date: Sat, 2 Oct 2021 20:08:11 +0100 Subject: [PATCH] --- 3d_gpu/architecture/dynamic_simd.mdwn | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/3d_gpu/architecture/dynamic_simd.mdwn b/3d_gpu/architecture/dynamic_simd.mdwn index ac0760f1f..0d1e9d805 100644 --- a/3d_gpu/architecture/dynamic_simd.mdwn +++ b/3d_gpu/architecture/dynamic_simd.mdwn @@ -1,3 +1,13 @@ +# Links + +* top level SIMD +* m.If/Switch +* Limited Cat +* RFC for nmigen integration +* Formal proof of PartitionedSignal +* Formal proof of PartitionedSignal nmigen interaction +* Partition-context-sensitive length adjustment + # Dynamic Partitioned SIMD The Dynamic Partitioned SIMD Signal is effectively a parallelisation @@ -5,6 +15,8 @@ of nmigen's Signal. It is expected to work transparently as if it was a nmigen Signal, in every way, as a full peer of a nmigen Signal, with no requirement on the part of the developer to even know that it is performing parallel dynamically-partitioned operations. +All standard nmigen language constructs are both expected and required +to behave exactly as they do for scalar Signals. nmigen 32-bit Signal: @@ -112,17 +124,7 @@ basic HDL is literally an order of magnitude larger: .... - - -# Links - -* top level SIMD -* m.If/Switch -* Limited Cat -* RFC for nmigen integration -* Formal proof of PartitionedSignal -* Formal proof of PartitionedSignal nmigen interaction -* Partition-context-sensitive length adjustment +# Alternative implementation concepts # Rationale / Introduction -- 2.30.2