maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[X86][AVX] Only combine loads to broadcasts for legal types 

Thanks to @echristo for spotting this.

llvm-svn: 354961
This commit is contained in:
Simon Pilgrim 2019-02-27 11:17:25 +00:00
parent 5b27402bee
commit 71bb6850cf
3 changed files with 79 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
llvm/lib/Target/X86/X86ISelLowering.cpp
View File

@ -7448,15 +7448,17 @@ static SDValue EltsFromConsecutiveLoads(EVT VT, ArrayRef<SDValue> Elts,
if (RepeatSize > ScalarSize)
RepeatVT = EVT::getVectorVT(*DAG.getContext(), RepeatVT,
RepeatSize / ScalarSize);
if (SDValue RepeatLoad = EltsFromConsecutiveLoads(
RepeatVT, RepeatedLoads, DL, DAG, Subtarget, isAfterLegalize)) {
EVT BroadcastVT =
EVT::getVectorVT(*DAG.getContext(), RepeatVT.getScalarType(),
VT.getSizeInBits() / ScalarSize);
unsigned Opcode = RepeatSize > ScalarSize ? X86ISD::SUBV_BROADCAST
: X86ISD::VBROADCAST;
SDValue Broadcast = DAG.getNode(Opcode, DL, BroadcastVT, RepeatLoad);
return DAG.getBitcast(VT, Broadcast);
EVT BroadcastVT =
EVT::getVectorVT(*DAG.getContext(), RepeatVT.getScalarType(),
VT.getSizeInBits() / ScalarSize);
if (TLI.isTypeLegal(BroadcastVT)) {
if (SDValue RepeatLoad = EltsFromConsecutiveLoads(
RepeatVT, RepeatedLoads, DL, DAG, Subtarget, isAfterLegalize)) {
unsigned Opcode = RepeatSize > ScalarSize ? X86ISD::SUBV_BROADCAST
: X86ISD::VBROADCAST;
SDValue Broadcast = DAG.getNode(Opcode, DL, BroadcastVT, RepeatLoad);
return DAG.getBitcast(VT, Broadcast);
}
}
}
}

26
llvm/test/CodeGen/X86/avx-vbroadcast.ll
View File

@ -856,6 +856,32 @@ define <4 x double> @broadcast_shuffle1032(double* %p) {
ret <4 x double> %4
}
define void @broadcast_v16i32(i32* %a, <16 x i32>* %b) {
; X32-LABEL: broadcast_v16i32:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: vbroadcastss (%ecx), %ymm0
; X32-NEXT: vmovups %ymm0, 32(%eax)
; X32-NEXT: vmovups %ymm0, (%eax)
; X32-NEXT: vzeroupper
; X32-NEXT: retl
;
; X64-LABEL: broadcast_v16i32:
; X64: ## %bb.0:
; X64-NEXT: vbroadcastss (%rdi), %ymm0
; X64-NEXT: vmovups %ymm0, 32(%rsi)
; X64-NEXT: vmovups %ymm0, (%rsi)
; X64-NEXT: vzeroupper
; X64-NEXT: retq
%1 = load i32, i32* %a, align 4
%2 = insertelement <8 x i32> undef, i32 %1, i32 0
%3 = shufflevector <8 x i32> %2, <8 x i32> undef, <8 x i32> zeroinitializer
%4 = shufflevector <8 x i32> undef, <8 x i32> %3, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
store <16 x i32> %4, <16 x i32>* %b, align 4
ret void
}
;
; When VBROADCAST replaces an existing load, ensure it still respects lifetime dependencies.
;

42
llvm/test/CodeGen/X86/avx2-vbroadcast.ll
View File

@ -1037,6 +1037,48 @@ define <4 x double> @splat_concat4(double %d) {
ret <4 x double> %5
}
define void @broadcast_v16i32(i32* %a, <16 x i32>* %b) {
; X32-AVX2-LABEL: broadcast_v16i32:
; X32-AVX2: ## %bb.0:
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-AVX2-NEXT: vbroadcastss (%ecx), %ymm0
; X32-AVX2-NEXT: vmovups %ymm0, 32(%eax)
; X32-AVX2-NEXT: vmovups %ymm0, (%eax)
; X32-AVX2-NEXT: vzeroupper
; X32-AVX2-NEXT: retl
;
; X64-AVX2-LABEL: broadcast_v16i32:
; X64-AVX2: ## %bb.0:
; X64-AVX2-NEXT: vbroadcastss (%rdi), %ymm0
; X64-AVX2-NEXT: vmovups %ymm0, 32(%rsi)
; X64-AVX2-NEXT: vmovups %ymm0, (%rsi)
; X64-AVX2-NEXT: vzeroupper
; X64-AVX2-NEXT: retq
;
; X32-AVX512VL-LABEL: broadcast_v16i32:
; X32-AVX512VL: ## %bb.0:
; X32-AVX512VL-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-AVX512VL-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-AVX512VL-NEXT: vbroadcastss (%ecx), %zmm0
; X32-AVX512VL-NEXT: vmovups %zmm0, (%eax)
; X32-AVX512VL-NEXT: vzeroupper
; X32-AVX512VL-NEXT: retl
;
; X64-AVX512VL-LABEL: broadcast_v16i32:
; X64-AVX512VL: ## %bb.0:
; X64-AVX512VL-NEXT: vbroadcastss (%rdi), %zmm0
; X64-AVX512VL-NEXT: vmovups %zmm0, (%rsi)
; X64-AVX512VL-NEXT: vzeroupper
; X64-AVX512VL-NEXT: retq
%1 = load i32, i32* %a, align 4
%2 = insertelement <8 x i32> undef, i32 %1, i32 0
%3 = shufflevector <8 x i32> %2, <8 x i32> undef, <8 x i32> zeroinitializer
%4 = shufflevector <8 x i32> undef, <8 x i32> %3, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
store <16 x i32> %4, <16 x i32>* %b, align 4
ret void
}
; Test cases for <rdar://problem/16074331>.
; Instruction selection for broacast instruction fails if
; the load cannot be folded into the broadcast.