[llvm] ddb82fc - [ARM] Handle any extend whilst lowering mull
David Green via llvm-commits
llvm-commits at lists.llvm.org
Wed Jan 6 02:51:34 PST 2021
Author: David Green
Date: 2021-01-06T10:51:12Z
New Revision: ddb82fc76ceb92e6f361d35f1ee8dedaee756854
URL: https://github.com/llvm/llvm-project/commit/ddb82fc76ceb92e6f361d35f1ee8dedaee756854
DIFF: https://github.com/llvm/llvm-project/commit/ddb82fc76ceb92e6f361d35f1ee8dedaee756854.diff
LOG: [ARM] Handle any extend whilst lowering mull
Similar to 78d8a821e23e but for ARM, this handles any_extend whilst
creating MULL nodes, treating them as zextends.
Differential Revision: https://reviews.llvm.org/D93834
Added:
Modified:
llvm/lib/Target/ARM/ARMISelLowering.cpp
llvm/test/CodeGen/ARM/vmla.ll
llvm/test/CodeGen/ARM/vmls.ll
llvm/test/CodeGen/ARM/vmul.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/ARM/ARMISelLowering.cpp b/llvm/lib/Target/ARM/ARMISelLowering.cpp
index 6a8355f0c3e8a..efe2efe91bcf8 100644
--- a/llvm/lib/Target/ARM/ARMISelLowering.cpp
+++ b/llvm/lib/Target/ARM/ARMISelLowering.cpp
@@ -8724,10 +8724,11 @@ static bool isSignExtended(SDNode *N, SelectionDAG &DAG) {
return false;
}
-/// isZeroExtended - Check if a node is a vector value that is zero-extended
-/// or a constant BUILD_VECTOR with zero-extended elements.
+/// isZeroExtended - Check if a node is a vector value that is zero-extended (or
+/// any-extended) or a constant BUILD_VECTOR with zero-extended elements.
static bool isZeroExtended(SDNode *N, SelectionDAG &DAG) {
- if (N->getOpcode() == ISD::ZERO_EXTEND || ISD::isZEXTLoad(N))
+ if (N->getOpcode() == ISD::ZERO_EXTEND || N->getOpcode() == ISD::ANY_EXTEND ||
+ ISD::isZEXTLoad(N))
return true;
if (isExtendedBUILD_VECTOR(N, DAG, false))
return true;
@@ -8795,13 +8796,14 @@ static SDValue SkipLoadExtensionForVMULL(LoadSDNode *LD, SelectionDAG& DAG) {
}
/// SkipExtensionForVMULL - For a node that is a SIGN_EXTEND, ZERO_EXTEND,
-/// extending load, or BUILD_VECTOR with extended elements, return the
-/// unextended value. The unextended vector should be 64 bits so that it can
+/// ANY_EXTEND, extending load, or BUILD_VECTOR with extended elements, return
+/// the unextended value. The unextended vector should be 64 bits so that it can
/// be used as an operand to a VMULL instruction. If the original vector size
/// before extension is less than 64 bits we add a an extension to resize
/// the vector to 64 bits.
static SDValue SkipExtensionForVMULL(SDNode *N, SelectionDAG &DAG) {
- if (N->getOpcode() == ISD::SIGN_EXTEND || N->getOpcode() == ISD::ZERO_EXTEND)
+ if (N->getOpcode() == ISD::SIGN_EXTEND ||
+ N->getOpcode() == ISD::ZERO_EXTEND || N->getOpcode() == ISD::ANY_EXTEND)
return AddRequiredExtensionForVMULL(N->getOperand(0), DAG,
N->getOperand(0)->getValueType(0),
N->getValueType(0),
diff --git a/llvm/test/CodeGen/ARM/vmla.ll b/llvm/test/CodeGen/ARM/vmla.ll
index 14d425da2df48..43474efdf86b1 100644
--- a/llvm/test/CodeGen/ARM/vmla.ll
+++ b/llvm/test/CodeGen/ARM/vmla.ll
@@ -156,9 +156,7 @@ define <2 x i64> @vmlalu32(<2 x i64> %A, <2 x i32> %B, <2 x i32> %C) nounwind {
define <8 x i16> @vmlala8(<8 x i16> %A, <8 x i8> %B, <8 x i8> %C) nounwind {
; CHECK-LABEL: vmlala8:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u8 q8, d3
-; CHECK-NEXT: vmovl.u8 q9, d2
-; CHECK-NEXT: vmla.i16 q0, q9, q8
+; CHECK-NEXT: vmlal.u8 q0, d2, d3
; CHECK-NEXT: vbic.i16 q0, #0xff00
; CHECK-NEXT: bx lr
%tmp4 = zext <8 x i8> %B to <8 x i16>
@@ -172,9 +170,7 @@ define <8 x i16> @vmlala8(<8 x i16> %A, <8 x i8> %B, <8 x i8> %C) nounwind {
define <4 x i32> @vmlala16(<4 x i32> %A, <4 x i16> %B, <4 x i16> %C) nounwind {
; CHECK-LABEL: vmlala16:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u16 q8, d3
-; CHECK-NEXT: vmovl.u16 q9, d2
-; CHECK-NEXT: vmla.i32 q0, q9, q8
+; CHECK-NEXT: vmlal.u16 q0, d2, d3
; CHECK-NEXT: vmov.i32 q8, #0xffff
; CHECK-NEXT: vand q0, q0, q8
; CHECK-NEXT: bx lr
@@ -189,32 +185,10 @@ define <4 x i32> @vmlala16(<4 x i32> %A, <4 x i16> %B, <4 x i16> %C) nounwind {
define <2 x i64> @vmlala32(<2 x i64> %A, <2 x i32> %B, <2 x i32> %C) nounwind {
; CHECK-LABEL: vmlala32:
; CHECK: @ %bb.0:
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vmovl.u32 q8, d3
-; CHECK-NEXT: vmovl.u32 q9, d2
-; CHECK-NEXT: vmov.32 r0, d16[0]
-; CHECK-NEXT: vmov.32 r1, d18[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r3, d17[0]
-; CHECK-NEXT: vmov.32 r2, d19[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r1, r0
-; CHECK-NEXT: mla r1, r1, r12, r5
-; CHECK-NEXT: umull r5, r4, r2, r3
-; CHECK-NEXT: mla r2, r2, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
-; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r3, r2
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r0, r1
-; CHECK-NEXT: vmov.32 d16[1], r0
-; CHECK-NEXT: vadd.i64 q8, q0, q8
-; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: vmlal.u32 q0, d2, d3
+; CHECK-NEXT: vmov.i64 q8, #0xffffffff
+; CHECK-NEXT: vand q0, q0, q8
+; CHECK-NEXT: bx lr
%tmp4 = zext <2 x i32> %B to <2 x i64>
%tmp5 = zext <2 x i32> %C to <2 x i64>
%tmp6 = mul <2 x i64> %tmp4, %tmp5
@@ -282,10 +256,7 @@ entry:
define arm_aapcs_vfpcc <4 x i32> @test_vmlal_lanea16(<4 x i32> %arg0_uint32x4_t, <4 x i16> %arg1_uint16x4_t, <4 x i16> %arg2_uint16x4_t) nounwind readnone {
; CHECK-LABEL: test_vmlal_lanea16:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: vdup.16 d16, d3[1]
-; CHECK-NEXT: vmovl.u16 q9, d2
-; CHECK-NEXT: vmovl.u16 q8, d16
-; CHECK-NEXT: vmla.i32 q0, q9, q8
+; CHECK-NEXT: vmlal.u16 q0, d2, d3[1]
; CHECK-NEXT: vmov.i32 q8, #0xffff
; CHECK-NEXT: vand q0, q0, q8
; CHECK-NEXT: bx lr
@@ -302,33 +273,10 @@ entry:
define arm_aapcs_vfpcc <2 x i64> @test_vmlal_lanea32(<2 x i64> %arg0_uint64x2_t, <2 x i32> %arg1_uint32x2_t, <2 x i32> %arg2_uint32x2_t) nounwind readnone {
; CHECK-LABEL: test_vmlal_lanea32:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vdup.32 d16, d3[1]
-; CHECK-NEXT: vmovl.u32 q9, d2
-; CHECK-NEXT: vmovl.u32 q8, d16
-; CHECK-NEXT: vmov.32 r0, d18[0]
-; CHECK-NEXT: vmov.32 r3, d19[0]
-; CHECK-NEXT: vmov.32 r1, d16[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r2, d17[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r0, r1
-; CHECK-NEXT: mla r0, r0, r12, r5
-; CHECK-NEXT: umull r5, r4, r3, r2
-; CHECK-NEXT: mla r3, r3, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
-; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r2, r3
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r1, r0
-; CHECK-NEXT: vmov.32 d16[1], r0
-; CHECK-NEXT: vadd.i64 q8, q0, q8
-; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: vmlal.u32 q0, d2, d3[1]
+; CHECK-NEXT: vmov.i64 q8, #0xffffffff
+; CHECK-NEXT: vand q0, q0, q8
+; CHECK-NEXT: bx lr
entry:
%0 = shufflevector <2 x i32> %arg2_uint32x2_t, <2 x i32> undef, <2 x i32> <i32 1, i32 1> ; <<2 x i32>> [#uses=1]
%1 = zext <2 x i32> %arg1_uint32x2_t to <2 x i64>
diff --git a/llvm/test/CodeGen/ARM/vmls.ll b/llvm/test/CodeGen/ARM/vmls.ll
index eef90041b96fc..21b7ee841c14e 100644
--- a/llvm/test/CodeGen/ARM/vmls.ll
+++ b/llvm/test/CodeGen/ARM/vmls.ll
@@ -156,9 +156,7 @@ define <2 x i64> @vmlslu32(<2 x i64> %A, <2 x i32> %B, <2 x i32> %C) nounwind {
define <8 x i16> @vmlsla8(<8 x i16> %A, <8 x i8> %B, <8 x i8> %C) nounwind {
; CHECK-LABEL: vmlsla8:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u8 q8, d3
-; CHECK-NEXT: vmovl.u8 q9, d2
-; CHECK-NEXT: vmls.i16 q0, q9, q8
+; CHECK-NEXT: vmlsl.u8 q0, d2, d3
; CHECK-NEXT: vbic.i16 q0, #0xff00
; CHECK-NEXT: bx lr
%tmp4 = zext <8 x i8> %B to <8 x i16>
@@ -172,9 +170,7 @@ define <8 x i16> @vmlsla8(<8 x i16> %A, <8 x i8> %B, <8 x i8> %C) nounwind {
define <4 x i32> @vmlsla16(<4 x i32> %A, <4 x i16> %B, <4 x i16> %C) nounwind {
; CHECK-LABEL: vmlsla16:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u16 q8, d3
-; CHECK-NEXT: vmovl.u16 q9, d2
-; CHECK-NEXT: vmls.i32 q0, q9, q8
+; CHECK-NEXT: vmlsl.u16 q0, d2, d3
; CHECK-NEXT: vmov.i32 q8, #0xffff
; CHECK-NEXT: vand q0, q0, q8
; CHECK-NEXT: bx lr
@@ -189,32 +185,10 @@ define <4 x i32> @vmlsla16(<4 x i32> %A, <4 x i16> %B, <4 x i16> %C) nounwind {
define <2 x i64> @vmlsla32(<2 x i64> %A, <2 x i32> %B, <2 x i32> %C) nounwind {
; CHECK-LABEL: vmlsla32:
; CHECK: @ %bb.0:
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vmovl.u32 q8, d3
-; CHECK-NEXT: vmovl.u32 q9, d2
-; CHECK-NEXT: vmov.32 r0, d16[0]
-; CHECK-NEXT: vmov.32 r1, d18[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r3, d17[0]
-; CHECK-NEXT: vmov.32 r2, d19[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r1, r0
-; CHECK-NEXT: mla r1, r1, r12, r5
-; CHECK-NEXT: umull r5, r4, r2, r3
-; CHECK-NEXT: mla r2, r2, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
-; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r3, r2
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r0, r1
-; CHECK-NEXT: vmov.32 d16[1], r0
-; CHECK-NEXT: vsub.i64 q8, q0, q8
-; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: vmlsl.u32 q0, d2, d3
+; CHECK-NEXT: vmov.i64 q8, #0xffffffff
+; CHECK-NEXT: vand q0, q0, q8
+; CHECK-NEXT: bx lr
%tmp4 = zext <2 x i32> %B to <2 x i64>
%tmp5 = zext <2 x i32> %C to <2 x i64>
%tmp6 = mul <2 x i64> %tmp4, %tmp5
@@ -282,10 +256,7 @@ entry:
define arm_aapcs_vfpcc <4 x i32> @test_vmlsl_lanea16(<4 x i32> %arg0_uint32x4_t, <4 x i16> %arg1_uint16x4_t, <4 x i16> %arg2_uint16x4_t) nounwind readnone {
; CHECK-LABEL: test_vmlsl_lanea16:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: vdup.16 d16, d3[1]
-; CHECK-NEXT: vmovl.u16 q9, d2
-; CHECK-NEXT: vmovl.u16 q8, d16
-; CHECK-NEXT: vmls.i32 q0, q9, q8
+; CHECK-NEXT: vmlsl.u16 q0, d2, d3[1]
; CHECK-NEXT: vmov.i32 q8, #0xffff
; CHECK-NEXT: vand q0, q0, q8
; CHECK-NEXT: bx lr
@@ -302,33 +273,10 @@ entry:
define arm_aapcs_vfpcc <2 x i64> @test_vmlsl_lanea32(<2 x i64> %arg0_uint64x2_t, <2 x i32> %arg1_uint32x2_t, <2 x i32> %arg2_uint32x2_t) nounwind readnone {
; CHECK-LABEL: test_vmlsl_lanea32:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vdup.32 d16, d3[1]
-; CHECK-NEXT: vmovl.u32 q9, d2
-; CHECK-NEXT: vmovl.u32 q8, d16
-; CHECK-NEXT: vmov.32 r0, d18[0]
-; CHECK-NEXT: vmov.32 r3, d19[0]
-; CHECK-NEXT: vmov.32 r1, d16[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r2, d17[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r0, r1
-; CHECK-NEXT: mla r0, r0, r12, r5
-; CHECK-NEXT: umull r5, r4, r3, r2
-; CHECK-NEXT: mla r3, r3, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
-; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r2, r3
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r1, r0
-; CHECK-NEXT: vmov.32 d16[1], r0
-; CHECK-NEXT: vsub.i64 q8, q0, q8
-; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: vmlsl.u32 q0, d2, d3[1]
+; CHECK-NEXT: vmov.i64 q8, #0xffffffff
+; CHECK-NEXT: vand q0, q0, q8
+; CHECK-NEXT: bx lr
entry:
%0 = shufflevector <2 x i32> %arg2_uint32x2_t, <2 x i32> undef, <2 x i32> <i32 1, i32 1> ; <<2 x i32>> [#uses=1]
%1 = zext <2 x i32> %arg1_uint32x2_t to <2 x i64>
diff --git a/llvm/test/CodeGen/ARM/vmul.ll b/llvm/test/CodeGen/ARM/vmul.ll
index c57ace2a6c673..4f7464c7e202d 100644
--- a/llvm/test/CodeGen/ARM/vmul.ll
+++ b/llvm/test/CodeGen/ARM/vmul.ll
@@ -286,9 +286,7 @@ define <2 x i64> @vmullu32_int(<2 x i32> %A, <2 x i32> %B) nounwind {
define <8 x i16> @vmulla8(<8 x i8> %A, <8 x i8> %B) nounwind {
; CHECK-LABEL: vmulla8:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u8 q8, d1
-; CHECK-NEXT: vmovl.u8 q9, d0
-; CHECK-NEXT: vmul.i16 q0, q9, q8
+; CHECK-NEXT: vmull.u8 q0, d0, d1
; CHECK-NEXT: vbic.i16 q0, #0xff00
; CHECK-NEXT: bx lr
%tmp3 = zext <8 x i8> %A to <8 x i16>
@@ -301,9 +299,7 @@ define <8 x i16> @vmulla8(<8 x i8> %A, <8 x i8> %B) nounwind {
define <4 x i32> @vmulla16(<4 x i16> %A, <4 x i16> %B) nounwind {
; CHECK-LABEL: vmulla16:
; CHECK: @ %bb.0:
-; CHECK-NEXT: vmovl.u16 q8, d1
-; CHECK-NEXT: vmovl.u16 q9, d0
-; CHECK-NEXT: vmul.i32 q8, q9, q8
+; CHECK-NEXT: vmull.u16 q8, d0, d1
; CHECK-NEXT: vmov.i32 q9, #0xffff
; CHECK-NEXT: vand q0, q8, q9
; CHECK-NEXT: bx lr
@@ -317,31 +313,10 @@ define <4 x i32> @vmulla16(<4 x i16> %A, <4 x i16> %B) nounwind {
define <2 x i64> @vmulla32(<2 x i32> %A, <2 x i32> %B) nounwind {
; CHECK-LABEL: vmulla32:
; CHECK: @ %bb.0:
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vmovl.u32 q8, d1
-; CHECK-NEXT: vmovl.u32 q9, d0
-; CHECK-NEXT: vmov.32 r0, d16[0]
-; CHECK-NEXT: vmov.32 r1, d18[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r3, d17[0]
-; CHECK-NEXT: vmov.32 r2, d19[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r1, r0
-; CHECK-NEXT: mla r1, r1, r12, r5
-; CHECK-NEXT: umull r5, r4, r2, r3
-; CHECK-NEXT: mla r2, r2, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
+; CHECK-NEXT: vmull.u32 q8, d0, d1
; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r3, r2
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r0, r1
-; CHECK-NEXT: vmov.32 d16[1], r0
; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: bx lr
%tmp3 = zext <2 x i32> %A to <2 x i64>
%tmp4 = zext <2 x i32> %B to <2 x i64>
%tmp5 = mul <2 x i64> %tmp3, %tmp4
@@ -457,10 +432,7 @@ entry:
define arm_aapcs_vfpcc <4 x i32> @test_vmull_lanea16(<4 x i16> %arg0_uint16x4_t, <4 x i16> %arg1_uint16x4_t) nounwind readnone {
; CHECK-LABEL: test_vmull_lanea16:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: vdup.16 d16, d1[1]
-; CHECK-NEXT: vmovl.u16 q9, d0
-; CHECK-NEXT: vmovl.u16 q8, d16
-; CHECK-NEXT: vmul.i32 q8, q9, q8
+; CHECK-NEXT: vmull.u16 q8, d0, d1[1]
; CHECK-NEXT: vmov.i32 q9, #0xffff
; CHECK-NEXT: vand q0, q8, q9
; CHECK-NEXT: bx lr
@@ -476,32 +448,10 @@ entry:
define arm_aapcs_vfpcc <2 x i64> @test_vmull_lanea32(<2 x i32> %arg0_uint32x2_t, <2 x i32> %arg1_uint32x2_t) nounwind readnone {
; CHECK-LABEL: test_vmull_lanea32:
; CHECK: @ %bb.0: @ %entry
-; CHECK-NEXT: .save {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: push {r4, r5, r6, r7, r11, lr}
-; CHECK-NEXT: vdup.32 d16, d1[1]
-; CHECK-NEXT: vmovl.u32 q9, d0
-; CHECK-NEXT: vmovl.u32 q8, d16
-; CHECK-NEXT: vmov.32 r0, d18[0]
-; CHECK-NEXT: vmov.32 r3, d19[0]
-; CHECK-NEXT: vmov.32 r1, d16[0]
-; CHECK-NEXT: vmov.32 r12, d16[1]
-; CHECK-NEXT: vmov.32 r2, d17[0]
-; CHECK-NEXT: vmov.32 lr, d17[1]
-; CHECK-NEXT: vmov.32 r6, d19[1]
-; CHECK-NEXT: umull r7, r5, r0, r1
-; CHECK-NEXT: mla r0, r0, r12, r5
-; CHECK-NEXT: umull r5, r4, r3, r2
-; CHECK-NEXT: mla r3, r3, lr, r4
-; CHECK-NEXT: vmov.32 r4, d18[1]
+; CHECK-NEXT: vmull.u32 q8, d0, d1[1]
; CHECK-NEXT: vmov.i64 q9, #0xffffffff
-; CHECK-NEXT: mla r2, r6, r2, r3
-; CHECK-NEXT: vmov.32 d17[0], r5
-; CHECK-NEXT: vmov.32 d16[0], r7
-; CHECK-NEXT: vmov.32 d17[1], r2
-; CHECK-NEXT: mla r0, r4, r1, r0
-; CHECK-NEXT: vmov.32 d16[1], r0
; CHECK-NEXT: vand q0, q8, q9
-; CHECK-NEXT: pop {r4, r5, r6, r7, r11, pc}
+; CHECK-NEXT: bx lr
entry:
%0 = shufflevector <2 x i32> %arg1_uint32x2_t, <2 x i32> undef, <2 x i32> <i32 1, i32 1> ; <<2 x i32>> [#uses=1]
%1 = zext <2 x i32> %arg0_uint32x2_t to <2 x i64>
More information about the llvm-commits
mailing list