From 90a237bf83f5a583bb6bb329d17cf1e614044c5d Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Tue, 6 Feb 2018 12:16:10 +0000
Subject: [X86][SSE] Add PACKSS support for truncation of clamped values

Followup to D42544 that matches PACKSSWB cases for non-AVX512, SSE and PACKSSDW cases will have to wait until we can add support for general SMIN/SMAX matching.

llvm-svn: 324339
---
 llvm/test/CodeGen/X86/vector-trunc-ssat.ll | 54 ++++--------------------------
 1 file changed, 7 insertions(+), 47 deletions(-)

(limited to 'llvm/test')

diff --git a/llvm/test/CodeGen/X86/vector-trunc-ssat.ll b/llvm/test/CodeGen/X86/vector-trunc-ssat.ll
index d4c22be94d4..2dbfe76a6e8 100644
--- a/llvm/test/CodeGen/X86/vector-trunc-ssat.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-ssat.ll
@@ -3274,28 +3274,14 @@ define <16 x i8> @trunc_ssat_v16i16_v16i8(<16 x i16> %a0) {
 ; AVX1-LABEL: trunc_ssat_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [127,127,127,127,127,127,127,127]
-; AVX1-NEXT:    vpminsw %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpminsw %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [65408,65408,65408,65408,65408,65408,65408,65408]
-; AVX1-NEXT:    vpmaxsw %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpmaxsw %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpacksswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_ssat_v16i16_v16i8:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vpminsw {{.*}}(%rip), %ymm0, %ymm0
-; AVX2-NEXT:    vpmaxsw {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpacksswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3409,44 +3395,18 @@ define <32 x i8> @trunc_ssat_v32i16_v32i8(<32 x i16> %a0) {
 ; AVX1-LABEL: trunc_ssat_v32i16_v32i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [127,127,127,127,127,127,127,127]
-; AVX1-NEXT:    vpminsw %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpminsw %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm4
-; AVX1-NEXT:    vpminsw %xmm3, %xmm4, %xmm4
-; AVX1-NEXT:    vpminsw %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [65408,65408,65408,65408,65408,65408,65408,65408]
-; AVX1-NEXT:    vpmaxsw %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpmaxsw %xmm3, %xmm4, %xmm4
-; AVX1-NEXT:    vpmaxsw %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpmaxsw %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
-; AVX1-NEXT:    vpshufb %xmm3, %xmm4, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX1-NEXT:    vpacksswb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
+; AVX1-NEXT:    vpacksswb %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_ssat_v32i16_v32i8:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX2-NEXT:    vpminsw %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vpminsw %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408,65408]
-; AVX2-NEXT:    vpmaxsw %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vpmaxsw %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
+; AVX2-NEXT:    vpacksswb %xmm2, %xmm1, %xmm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpacksswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
 ;
-- 
cgit v1.2.3