[X86][AVX] Insert zeros byte elements into 256/512-bit vectors using shuffle/and

Avoid extracting/inserting subvectors which makes it more difficult for shuffle combining to merge them together.

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 8d78571..808665f 100644 (file)
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -18854,7 +18854,7 @@ SDValue X86TargetLowering::LowerINSERT_VECTOR_ELT(SDValue Op,
   // a blend shuffle with a rematerializable vector than a costly integer
   // insertion.
   if ((IsZeroElt || IsAllOnesElt) && Subtarget.hasSSE41() &&
-      16 <= EltSizeInBits) {
+      (16 <= EltSizeInBits || (IsZeroElt && !VT.is128BitVector()))) {
     SmallVector<int, 8> BlendMask;
     for (unsigned i = 0; i != NumElts; ++i)
       BlendMask.push_back(i == IdxVal ? i + NumElts : i);

diff --git a/llvm/test/CodeGen/X86/clear_upper_vector_element_bits.ll b/llvm/test/CodeGen/X86/clear_upper_vector_element_bits.ll
index 16a9933..d81b3b1 100644 (file)
--- a/llvm/test/CodeGen/X86/clear_upper_vector_element_bits.ll
+++ b/llvm/test/CodeGen/X86/clear_upper_vector_element_bits.ll
@@ -665,11 +665,7 @@ define <16 x i16> @_clearupper16xi16b(<16 x i16>) nounwind {
 ;
 ; AVX-LABEL: _clearupper16xi16b:
 ; AVX:       # %bb.0:
-; AVX-NEXT:    vmovaps {{.*#+}} xmm1 = [255,0,255,0,255,0,255,0,255,0,255,0,255,0,255,0]
-; AVX-NEXT:    vandps %xmm1, %xmm0, %xmm2
-; AVX-NEXT:    vextractf128 $1, %ymm0, %xmm0
-; AVX-NEXT:    vandps %xmm1, %xmm0, %xmm0
-; AVX-NEXT:    vinsertf128 $1, %xmm0, %ymm2, %ymm0
+; AVX-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX-NEXT:    retq
   %x8 = bitcast <16 x i16> %0 to <32 x i8>
   %r0  = insertelement <32 x i8> %x8,  i8 zeroinitializer, i32 1

diff --git a/llvm/test/CodeGen/X86/insertelement-zero.ll b/llvm/test/CodeGen/X86/insertelement-zero.ll
index 72a02d2..6cd512a 100644 (file)
--- a/llvm/test/CodeGen/X86/insertelement-zero.ll
+++ b/llvm/test/CodeGen/X86/insertelement-zero.ll
@@ -463,36 +463,10 @@ define <32 x i8> @insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz(<32 x i8> %a) {
 ; SSE41-NEXT:    pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6],xmm2[7]
 ; SSE41-NEXT:    retq
 ;
-; AVX1-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
-; AVX1:       # %bb.0:
-; AVX1-NEXT:    xorl %eax, %eax
-; AVX1-NEXT:    vpinsrb $0, %eax, %xmm0, %xmm1
-; AVX1-NEXT:    vpinsrb $15, %eax, %xmm1, %xmm1
-; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
-; AVX1-NEXT:    vpxor %xmm2, %xmm2, %xmm2
-; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
-; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
-; AVX1-NEXT:    retq
-;
-; AVX2-SLOW-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
-; AVX2-SLOW:       # %bb.0:
-; AVX2-SLOW-NEXT:    xorl %eax, %eax
-; AVX2-SLOW-NEXT:    vpinsrb $0, %eax, %xmm0, %xmm1
-; AVX2-SLOW-NEXT:    vpinsrb $15, %eax, %xmm1, %xmm1
-; AVX2-SLOW-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX2-SLOW-NEXT:    vpxor %xmm2, %xmm2, %xmm2
-; AVX2-SLOW-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
-; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
-; AVX2-SLOW-NEXT:    retq
-;
-; AVX2-FAST-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
-; AVX2-FAST:       # %bb.0:
-; AVX2-FAST-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm1
-; AVX2-FAST-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX2-FAST-NEXT:    vpxor %xmm2, %xmm2, %xmm2
-; AVX2-FAST-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
-; AVX2-FAST-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT:    retq
+; AVX-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
+; AVX-NEXT:    retq
   %1 = insertelement <32 x i8> %a, i8 0, i32 0
   %2 = insertelement <32 x i8> %1, i8 0, i32 15
   %3 = insertelement <32 x i8> %2, i8 0, i32 30