From: Sanjay Patel <spatel@rotateright.com>
Date: Mon, 25 Mar 2019 15:54:34 +0000 (+0000)
Subject: [x86] add tests for vector zext; NFC
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=2b324fde545d4f8365658621d2d7d946f01d47b1;p=llvm

[x86] add tests for vector zext; NFC

The AVX1 lowering is poor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@356914 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/test/CodeGen/X86/vector-zext.ll b/test/CodeGen/X86/vector-zext.ll
index 4e436f61e83..c21fc6a73a8 100644
--- a/test/CodeGen/X86/vector-zext.ll
+++ b/test/CodeGen/X86/vector-zext.ll
@@ -2563,3 +2563,160 @@ entry:
   %e = zext <8 x i6> %d to <8 x i64>
   ret <8 x i64> %e
 }
+
+define <4 x i64> @splatshuf_zext_v4i64(<4 x i32> %x) {
+; SSE2-LABEL: splatshuf_zext_v4i64:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
+; SSE2-NEXT:    pxor %xmm1, %xmm1
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: splatshuf_zext_v4i64:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
+; SSSE3-NEXT:    pxor %xmm1, %xmm1
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: splatshuf_zext_v4i64:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[0,0,0,0]
+; SSE41-NEXT:    pxor %xmm2, %xmm2
+; SSE41-NEXT:    pmovzxdq {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero
+; SSE41-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm2[2],xmm1[3],xmm2[3]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: splatshuf_zext_v4i64:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm1 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
+; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splatshuf_zext_v4i64:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpbroadcastd %xmm0, %xmm0
+; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: splatshuf_zext_v4i64:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpbroadcastd %xmm0, %xmm0
+; AVX512-NEXT:    vpmovzxdq {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX512-NEXT:    retq
+  %shuf = shufflevector <4 x i32> %x, <4 x i32> undef, <4 x i32> zeroinitializer
+  %ext = zext <4 x i32> %shuf to <4 x i64>
+  ret <4 x i64> %ext
+}
+
+define <8 x i32> @splatshuf_zext_v8i32(<8 x i16> %x) {
+; SSE2-LABEL: splatshuf_zext_v8i32:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,1,3]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,5,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,0]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm0[0,1,2,3,6,5,5,4]
+; SSE2-NEXT:    pxor %xmm2, %xmm2
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: splatshuf_zext_v8i32:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[0,1,2,3,6,7,14,15,0,1,6,7,6,7,14,15]
+; SSSE3-NEXT:    pxor %xmm2, %xmm2
+; SSSE3-NEXT:    movdqa %xmm1, %xmm0
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: splatshuf_zext_v8i32:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[0,1,2,3,6,7,14,15,0,1,6,7,6,7,14,15]
+; SSE41-NEXT:    pxor %xmm2, %xmm2
+; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
+; SSE41-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: splatshuf_zext_v8i32:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,14,15,0,1,6,7,6,7,14,15]
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splatshuf_zext_v8i32:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,14,15,0,1,6,7,6,7,14,15]
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: splatshuf_zext_v8i32:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,14,15,0,1,6,7,6,7,14,15]
+; AVX512-NEXT:    vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX512-NEXT:    retq
+  %shuf = shufflevector <8 x i16> %x, <8 x i16> undef, <8 x i32> <i32 0, i32 1, i32 3, i32 7, i32 0, i32 undef, i32 3, i32 7>
+  %ext = zext <8 x i16> %shuf to <8 x i32>
+  ret <8 x i32> %ext
+}
+
+define <16 x i16> @splatshuf_zext_v16i16(<16 x i8> %x) {
+; SSE2-LABEL: splatshuf_zext_v16i16:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm0 = xmm0[8,8,9,9,10,10,11,11,12,12,13,13,14,14,15,15]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,6,6,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,2,2,3]
+; SSE2-NEXT:    pxor %xmm1, %xmm1
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: splatshuf_zext_v16i16:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[14],zero,xmm0[14],zero,xmm0[14],zero,xmm0[14],zero,xmm0[14],zero,xmm0[14],zero,xmm0[14],zero,xmm0[14],zero
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: splatshuf_zext_v16i16:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14]
+; SSE41-NEXT:    pxor %xmm2, %xmm2
+; SSE41-NEXT:    pmovzxbw {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
+; SSE41-NEXT:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: splatshuf_zext_v16i16:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14]
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm1 = xmm0[8],xmm1[8],xmm0[9],xmm1[9],xmm0[10],xmm1[10],xmm0[11],xmm1[11],xmm0[12],xmm1[12],xmm0[13],xmm1[13],xmm0[14],xmm1[14],xmm0[15],xmm1[15]
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splatshuf_zext_v16i16:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14]
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: splatshuf_zext_v16i16:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14]
+; AVX512-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX512-NEXT:    retq
+  %shuf = shufflevector <16 x i8> %x, <16 x i8> undef, <16 x i32> <i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14, i32 14>
+  %ext = zext <16 x i8> %shuf to <16 x i16>
+  ret <16 x i16> %ext
+}