From 0e76de7c653eb0caf148d732e6269a4445cdd09f Mon Sep 17 00:00:00 2001 From: Nigel Tao Date: Mon, 25 Jan 2021 10:12:32 +1100 Subject: [PATCH] Add std/png bench comment for the roll back --- std/png/decode_filter_sse42.wuffs | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/std/png/decode_filter_sse42.wuffs b/std/png/decode_filter_sse42.wuffs index e96a3048..7c19f8d6 100644 --- a/std/png/decode_filter_sse42.wuffs +++ b/std/png/decode_filter_sse42.wuffs @@ -37,6 +37,10 @@ // c.poke_u24le!(a: x128.truncate_u32()) // } // } +// +// Note that "more SIMD" doesn't always mean faster compute. See +// https://github.com/google/wuffs/commit/1660f9268621ed4415b3b363f0a0e1026d4aa83d +// "Have std/png filter_1_distance_? use more SIMD" for a pessimizing example. pri func decoder.filter_1_distance_4_sse42!(curr: slice base.u8), choose cpu_arch >= x86_sse42,