From 1b3e795fc908a4b804a0af6beed15843a4dc8a29 Mon Sep 17 00:00:00 2001 From: Matthew Barr Date: Tue, 7 Jun 2016 15:44:39 +1000 Subject: [PATCH] teddy: we only need the upper lane Just use an extract, no need to shuffle first. --- src/fdr/teddy_avx2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/fdr/teddy_avx2.c b/src/fdr/teddy_avx2.c index f282c505..ef06813c 100644 --- a/src/fdr/teddy_avx2.c +++ b/src/fdr/teddy_avx2.c @@ -204,7 +204,7 @@ do { \ if (unlikely(isnonzero256(var))) { \ u32 arrCnt = 0; \ m128 lo = cast256to128(var); \ - m128 hi = cast256to128(swap128in256(var)); \ + m128 hi = movdq_hi(var); \ bit_array_fast_teddy(lo, bitArr, &arrCnt, offset); \ bit_array_fast_teddy(hi, bitArr, &arrCnt, offset + 2); \ for (u32 i = 0; i < arrCnt; i++) { \