From a2153222677327be43251012f462d19a7e98ce14 Mon Sep 17 00:00:00 2001 From: =?utf8?q?S=C3=B8ren=20Sandmann=20Pedersen?= Date: Sun, 3 Apr 2011 20:32:30 -0400 Subject: [PATCH] ARM: Tiny improvement in over_n_8888_8888_ca_process_pixblock_head Instead of two mvn d24, d24 mvn d25, d25 use just one mvn q12, q12 Also move another vmvn instruction into the created pipeline bubble, as pointed out by Siarhei. --- pixman/pixman-arm-neon-asm.S | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/pixman/pixman-arm-neon-asm.S b/pixman/pixman-arm-neon-asm.S index 71b30ac..1d3e64e 100644 --- a/pixman/pixman-arm-neon-asm.S +++ b/pixman/pixman-arm-neon-asm.S @@ -1358,11 +1358,10 @@ generate_composite_function \ * * output: updated dest in {d28, d29, d30, d31} */ - vmvn.8 d24, d24 - vmvn.8 d25, d25 + vmvn.8 q12, q12 + vmvn.8 d26, d26 vmull.u8 q8, d24, d4 vmull.u8 q9, d25, d5 - vmvn.8 d26, d26 vmvn.8 d27, d3 vmull.u8 q10, d26, d6 vmull.u8 q11, d27, d7 -- 2.7.4