[X86][SSE] Added tests for pre-truncating arithmetic operations that have already...

As discussed on D28219 - it is profitable to combine trunc(binop (s/zext(x), s/zext(y)) to binop(trunc(s/zext(x)), trunc(s/zext(y))) assuming the trunc(ext()) will simplify further

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@292487 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/test/CodeGen/X86/vector-trunc-math.ll b/test/CodeGen/X86/vector-trunc-math.ll
index f828ed0ba6e7387a20f0dc3aa179ec541ad215ca..f7e9fdff81b31a7fe31e5d11403910959b84a02f 100644 (file)
--- a/test/CodeGen/X86/vector-trunc-math.ll
+++ b/test/CodeGen/X86/vector-trunc-math.ll
@@ -412,6 +412,65 @@ define <16 x i8> @trunc_add_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
   ret <16 x i8> %2
 }
 
+define <8 x i16> @trunc_add_v8i32_v8i16_sext_8i8(<16 x i8> %a0, <8 x i32> %a1) {
+; SSE-LABEL: trunc_add_v8i32_v8i16_sext_8i8:
+; SSE:       # BB#0:
+; SSE-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1],xmm3[2],xmm0[2],xmm3[3],xmm0[3],xmm3[4],xmm0[4],xmm3[5],xmm0[5],xmm3[6],xmm0[6],xmm3[7],xmm0[7]
+; SSE-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0,0,1,1,2,2,3,3]
+; SSE-NEXT:    psrad $24, %xmm3
+; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,1,2,3]
+; SSE-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
+; SSE-NEXT:    psrad $24, %xmm0
+; SSE-NEXT:    paddd %xmm2, %xmm0
+; SSE-NEXT:    paddd %xmm1, %xmm3
+; SSE-NEXT:    pslld $16, %xmm3
+; SSE-NEXT:    psrad $16, %xmm3
+; SSE-NEXT:    pslld $16, %xmm0
+; SSE-NEXT:    psrad $16, %xmm0
+; SSE-NEXT:    packssdw %xmm0, %xmm3
+; SSE-NEXT:    movdqa %xmm3, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: trunc_add_v8i32_v8i16_sext_8i8:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vpmovsxbd %xmm0, %xmm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovsxbd %xmm0, %xmm0
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm3
+; AVX1-NEXT:    vpaddd %xmm3, %xmm0, %xmm0
+; AVX1-NEXT:    vpaddd %xmm1, %xmm2, %xmm1
+; 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 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: trunc_add_v8i32_v8i16_sext_8i8:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpmovsxbd %xmm0, %ymm0
+; AVX2-NEXT:    vpaddd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13],zero,zero,zero,zero,zero,zero,zero,zero,ymm0[16,17,20,21,24,25,28,29],zero,zero,zero,zero,zero,zero,zero,zero
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT:    # kill: %XMM0<def> %XMM0<kill> %YMM0<kill>
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: trunc_add_v8i32_v8i16_sext_8i8:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vpmovsxbd %xmm0, %ymm0
+; AVX512-NEXT:    vpaddd %ymm1, %ymm0, %ymm0
+; AVX512-NEXT:    vpmovdw %zmm0, %ymm0
+; AVX512-NEXT:    # kill: %XMM0<def> %XMM0<kill> %YMM0<kill>
+; AVX512-NEXT:    retq
+  %1 = shufflevector <16 x i8> %a0, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %2 = sext <8 x i8> %1 to <8 x i32>
+  %3 = add <8 x i32> %2, %a1
+  %4 = trunc <8 x i32> %3 to <8 x i16>
+  ret <8 x i16> %4
+}
+
 ;
 ; add to constant
 ;
@@ -2244,6 +2303,74 @@ define <16 x i8> @trunc_mul_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
   ret <16 x i8> %2
 }
 
+define <8 x i16> @trunc_mul_v8i32_v8i16_zext_8i8(<16 x i8> %a0, <8 x i32> %a1) {
+; SSE-LABEL: trunc_mul_v8i32_v8i16_zext_8i8:
+; SSE:       # BB#0:
+; SSE-NEXT:    pxor %xmm3, %xmm3
+; SSE-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3],xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7]
+; SSE-NEXT:    movdqa %xmm0, %xmm4
+; SSE-NEXT:    punpckhwd {{.*#+}} xmm4 = xmm4[4],xmm3[4],xmm4[5],xmm3[5],xmm4[6],xmm3[6],xmm4[7],xmm3[7]
+; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3]
+; SSE-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3]
+; SSE-NEXT:    pmuludq %xmm1, %xmm0
+; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
+; SSE-NEXT:    pmuludq %xmm3, %xmm1
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm4[1,1,3,3]
+; SSE-NEXT:    pmuludq %xmm2, %xmm4
+; SSE-NEXT:    pshufd {{.*#+}} xmm3 = xmm4[0,2,2,3]
+; SSE-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[1,1,3,3]
+; SSE-NEXT:    pmuludq %xmm1, %xmm2
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm2[0,2,2,3]
+; SSE-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm1[0],xmm3[1],xmm1[1]
+; SSE-NEXT:    pslld $16, %xmm3
+; SSE-NEXT:    psrad $16, %xmm3
+; SSE-NEXT:    pslld $16, %xmm0
+; SSE-NEXT:    psrad $16, %xmm0
+; SSE-NEXT:    packssdw %xmm3, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: trunc_mul_v8i32_v8i16_zext_8i8:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm2 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm3
+; AVX1-NEXT:    vpmulld %xmm3, %xmm0, %xmm0
+; AVX1-NEXT:    vpmulld %xmm1, %xmm2, %xmm1
+; 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 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: trunc_mul_v8i32_v8i16_zext_8i8:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX2-NEXT:    vpmulld %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13],zero,zero,zero,zero,zero,zero,zero,zero,ymm0[16,17,20,21,24,25,28,29],zero,zero,zero,zero,zero,zero,zero,zero
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT:    # kill: %XMM0<def> %XMM0<kill> %YMM0<kill>
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: trunc_mul_v8i32_v8i16_zext_8i8:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512-NEXT:    vpmulld %ymm1, %ymm0, %ymm0
+; AVX512-NEXT:    vpmovdw %zmm0, %ymm0
+; AVX512-NEXT:    # kill: %XMM0<def> %XMM0<kill> %YMM0<kill>
+; AVX512-NEXT:    retq
+  %1 = shufflevector <16 x i8> %a0, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %2 = zext <8 x i8> %1 to <8 x i32>
+  %3 = mul <8 x i32> %2, %a1
+  %4 = trunc <8 x i32> %3 to <8 x i16>
+  ret <8 x i16> %4
+}
+
 ;
 ; mul to constant
 ;