From e50298e5f4bb1942c7dc7b2722964a60e6434869 Mon Sep 17 00:00:00 2001 From: Martin Kronbichler Date: Mon, 24 Jun 2019 11:53:38 +0200 Subject: [PATCH] Changelog --- doc/news/changes/minor/20190624MartinKronbichler | 6 ++++++ 1 file changed, 6 insertions(+) create mode 100644 doc/news/changes/minor/20190624MartinKronbichler diff --git a/doc/news/changes/minor/20190624MartinKronbichler b/doc/news/changes/minor/20190624MartinKronbichler new file mode 100644 index 0000000000..7a6d9bdacd --- /dev/null +++ b/doc/news/changes/minor/20190624MartinKronbichler @@ -0,0 +1,6 @@ +Improved: The functions vectorized_load_and_transpose() and +vectorized_transpose_and_store() for AVX-512 were rewritten to better utilize +execution units with two load units and one shuffle/swizzle unit as common on +recent Intel CPUs. +
+(Martin Kronbichler, 2019/06/24) -- 2.39.5