From 9065b3468478818e9c5918380af665f2d499a322 Mon Sep 17 00:00:00 2001 From: cnugteren Date: Sun, 15 May 2016 14:04:34 +0200 Subject: Added support for staggered/shuffled offsets for GEMM to improve performance for large power-of-2 kernels on AMD GPUs --- CHANGELOG | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'CHANGELOG') diff --git a/CHANGELOG b/CHANGELOG index 92c0c5ad..187fca73 100644 --- a/CHANGELOG +++ b/CHANGELOG @@ -1,6 +1,6 @@ Development version (next release) -- +- Improved performance of large power-of-2 xGEMM kernels for AMD GPUs Version 0.7.0 - Added exports to be able to create a DLL on Windows (thanks to Marco Hutter) -- cgit v1.2.3