From 0271583d2276c93c20dd837a77736b2c6d499c5a Mon Sep 17 00:00:00 2001 From: lkcl Date: Thu, 28 Oct 2021 18:00:31 +0100 Subject: [PATCH] --- 3d_gpu/architecture/dynamic_simd/shape.mdwn | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/3d_gpu/architecture/dynamic_simd/shape.mdwn b/3d_gpu/architecture/dynamic_simd/shape.mdwn index 1773e2220..40974b7c5 100644 --- a/3d_gpu/architecture/dynamic_simd/shape.mdwn +++ b/3d_gpu/architecture/dynamic_simd/shape.mdwn @@ -424,7 +424,7 @@ Summary: * **Fixed_width Priority** - all operations (add, mul, sub, shift, div) - expected to work (caveat: layout() should check rounding) + expected to work (caveat: layout() should check partition alignment) * **Element-width Priority** - all operations expected to work (no caveats) * **Fixed-and-Elwidth (no priority)** @@ -452,3 +452,12 @@ In this case, the result will be an elwidths priority SimdShape, where the layout() function is already capable of computing the required overall width based on the (newly-computed) vec_el_widths. +Note that, interestingly, when fixed_width priority SimdShapes are +used on both LHS and RHS, it is effectively an identical case to +when the RHS is an integer, because the fixed_width of the RHS is +itself an integer. + +Summary: + +* **Both LHS and RHS Fixed-width Priority** + - Exactly the same as for when RHS is an Integer -- 2.30.2