From 763c87309da149ec7984d7d75c98749b6e05d060 Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Fri, 27 Mar 2020 17:29:15 +0000
Subject: [PATCH] [X86][SSE] Add some additional v8i16 'truncation' style
 shuffle tests

---
 llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll | 164 +++++++++++++++++++++++++
 1 file changed, 164 insertions(+)

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll b/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
index eb4e2ed..80345a5 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
@@ -2512,6 +2512,170 @@ define <8 x i16> @shuffle_v8i16_12305674(<8 x i16> %a) {
   ret <8 x i16> %shuffle
 }
 
+define <8 x i16> @shuffle_v8i16_02460246(<8 x i16> %a) {
+; SSE2-LABEL: shuffle_v8i16_02460246:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,4]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,3,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,4,5]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: shuffle_v8i16_02460246:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: shuffle_v8i16_02460246:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
+; SSE41-NEXT:    retq
+;
+; AVX-LABEL: shuffle_v8i16_02460246:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
+; AVX-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 0, i32 2, i32 4, i32 6>
+  ret <8 x i16> %shuffle
+}
+
+define <8 x i16> @shuffle_v8i16_04040404(<8 x i16> %a) {
+; SSE2-LABEL: shuffle_v8i16_04040404:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,6,4]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: shuffle_v8i16_04040404:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: shuffle_v8i16_04040404:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
+; SSE41-NEXT:    retq
+;
+; AVX-LABEL: shuffle_v8i16_04040404:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
+; AVX-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 0, i32 4, i32 0, i32 4, i32 0, i32 4, i32 0, i32 4>
+  ret <8 x i16> %shuffle
+}
+
+define <8 x i16> @shuffle_v8i16_02468ACE(<8 x i16> %a, <8 x i16> %b) {
+; SSE2-LABEL: shuffle_v8i16_02468ACE:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,6,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: shuffle_v8i16_02468ACE:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; SSSE3-NEXT:    pshufb %xmm2, %xmm1
+; SSSE3-NEXT:    pshufb %xmm2, %xmm0
+; SSSE3-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: shuffle_v8i16_02468ACE:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; SSE41-NEXT:    pshufb %xmm2, %xmm1
+; SSE41-NEXT:    pshufb %xmm2, %xmm0
+; SSE41-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: shuffle_v8i16_02468ACE:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v8i16_02468ACE:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    retq
+;
+; AVX512VL-LABEL: shuffle_v8i16_02468ACE:
+; AVX512VL:       # %bb.0:
+; AVX512VL-NEXT:    # kill: def $xmm0 killed $xmm0 def $ymm0
+; AVX512VL-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
+; AVX512VL-NEXT:    vzeroupper
+; AVX512VL-NEXT:    retq
+;
+; XOP-LABEL: shuffle_v8i16_02468ACE:
+; XOP:       # %bb.0:
+; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13],xmm1[0,1,4,5,8,9,12,13]
+; XOP-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
+  ret <8 x i16> %shuffle
+}
+
+define <8 x i16> @shuffle_v8i16_048C048C(<8 x i16> %a, <8 x i16> %b) {
+; SSE-LABEL: shuffle_v8i16_048C048C:
+; SSE:       # %bb.0:
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; SSE-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,0,2,4,5,6,7]
+; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
+; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: shuffle_v8i16_048C048C:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm1 = xmm1[0,2,0,2,4,5,6,7]
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
+; AVX1-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX1-NEXT:    retq
+;
+; AVX2-SLOW-LABEL: shuffle_v8i16_048C048C:
+; AVX2-SLOW:       # %bb.0:
+; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm1 = xmm1[0,2,0,2,4,5,6,7]
+; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
+; AVX2-SLOW-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX2-SLOW-NEXT:    retq
+;
+; AVX2-FAST-LABEL: shuffle_v8i16_048C048C:
+; AVX2-FAST:       # %bb.0:
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,1,8,9,0,1,8,9,8,9,10,11,12,13,14,15]
+; AVX2-FAST-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-FAST-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX2-FAST-NEXT:    retq
+;
+; AVX512VL-LABEL: shuffle_v8i16_048C048C:
+; AVX512VL:       # %bb.0:
+; AVX512VL-NEXT:    vpbroadcastq {{.*#+}} xmm2 = [3377734080528384,3377734080528384]
+; AVX512VL-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
+; AVX512VL-NEXT:    retq
+;
+; XOP-LABEL: shuffle_v8i16_048C048C:
+; XOP:       # %bb.0:
+; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,8,9],xmm1[0,1,8,9],xmm0[0,1,8,9],xmm1[0,1,8,9]
+; XOP-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 8, i32 12, i32 0, i32 4, i32 8, i32 12>
+  ret <8 x i16> %shuffle
+}
+
 ;
 ; Shuffle to logical bit shifts
 ;
-- 
2.7.4