From ab6c6694be2a576f47ca6c4e8676d18af828e615 Mon Sep 17 00:00:00 2001 From: lkcl Date: Sun, 26 Sep 2021 12:08:37 +0100 Subject: [PATCH] --- 3d_gpu/architecture/dynamic_simd/assign.mdwn | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/3d_gpu/architecture/dynamic_simd/assign.mdwn b/3d_gpu/architecture/dynamic_simd/assign.mdwn index 447d92e86..364f5601c 100644 --- a/3d_gpu/architecture/dynamic_simd/assign.mdwn +++ b/3d_gpu/architecture/dynamic_simd/assign.mdwn @@ -54,6 +54,13 @@ partitions: a : AAAA AAAA AAAA AAAA (16 bits) b : B7B6 B5B4 B3B2 B1B0 (8 bits) +As the source is Scalar, it must be copied (broadcast) into each +partition of the output, B, starting at the beginning of each +partition. With each partition being smaller than A (except +in one case) truncation is guaranteed. The exception is when +all pattitions are open (1x) and the length of A and B are +the same. + The partition options are: | partition | o3 | o2 | o1 | o0 | -- 2.30.2