[llvm] 0ee32c4 - [AMDGPU] Implement llvm.lrint intrinsic lowering (#98931)
via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 24 12:34:35 PDT 2024
Author: Sumanth Gundapaneni
Date: 2024-07-24T23:34:31+04:00
New Revision: 0ee32c45730c94be1b7d5fa60a0e8dff5751d014
URL: https://github.com/llvm/llvm-project/commit/0ee32c45730c94be1b7d5fa60a0e8dff5751d014
DIFF: https://github.com/llvm/llvm-project/commit/0ee32c45730c94be1b7d5fa60a0e8dff5751d014.diff
LOG: [AMDGPU] Implement llvm.lrint intrinsic lowering (#98931)
This patch enabled the target-independent lowering of llvm.lrint via
GlobalISel.
For SelectionDAG, the instrinsic is custom lowered for AMDGPU.
Added:
llvm/test/CodeGen/AMDGPU/lrint.ll
Modified:
llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
llvm/lib/CodeGen/TargetLoweringBase.cpp
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
Removed:
################################################################################
diff --git a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
index 6c7885c491f41..b490ab2fc7617 100644
--- a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
@@ -3900,6 +3900,17 @@ LegalizerHelper::lower(MachineInstr &MI, unsigned TypeIdx, LLT LowerHintTy) {
changeOpcode(MI, TargetOpcode::G_INTRINSIC_ROUNDEVEN);
return Legalized;
}
+ case TargetOpcode::G_INTRINSIC_LRINT:
+ case TargetOpcode::G_INTRINSIC_LLRINT: {
+ Register DstReg = MI.getOperand(0).getReg();
+ Register SrcReg = MI.getOperand(1).getReg();
+ LLT SrcTy = MRI.getType(SrcReg);
+ auto Round =
+ MIRBuilder.buildInstr(TargetOpcode::G_FRINT, {SrcTy}, {SrcReg});
+ MIRBuilder.buildFPTOSI(DstReg, Round);
+ MI.eraseFromParent();
+ return Legalized;
+ }
case TargetOpcode::G_ATOMIC_CMPXCHG_WITH_SUCCESS: {
auto [OldValRes, SuccessRes, Addr, CmpVal, NewVal] = MI.getFirst5Regs();
Register NewOldValRes = MRI.cloneVirtualRegister(OldValRes);
@@ -4755,6 +4766,8 @@ LegalizerHelper::fewerElementsVector(MachineInstr &MI, unsigned TypeIdx,
case G_FCEIL:
case G_FFLOOR:
case G_FRINT:
+ case G_INTRINSIC_LRINT:
+ case G_INTRINSIC_LLRINT:
case G_INTRINSIC_ROUND:
case G_INTRINSIC_ROUNDEVEN:
case G_INTRINSIC_TRUNC:
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
index 3a39f6a4d2b4a..bdb7917073020 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
@@ -4336,6 +4336,16 @@ bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
// targets where it is not needed.
Results.push_back(Node->getOperand(0));
break;
+ case ISD::LRINT:
+ case ISD::LLRINT: {
+ SDValue Arg = Node->getOperand(0);
+ EVT ArgVT = Arg.getValueType();
+ EVT ResVT = Node->getValueType(0);
+ SDLoc dl(Node);
+ SDValue RoundNode = DAG.getNode(ISD::FRINT, dl, ArgVT, Arg);
+ Results.push_back(DAG.getNode(ISD::FP_TO_SINT, dl, ResVT, RoundNode));
+ break;
+ }
case ISD::GLOBAL_OFFSET_TABLE:
case ISD::GlobalAddress:
case ISD::GlobalTLSAddress:
diff --git a/llvm/lib/CodeGen/TargetLoweringBase.cpp b/llvm/lib/CodeGen/TargetLoweringBase.cpp
index bc5fc9659469d..6ca9955993d24 100644
--- a/llvm/lib/CodeGen/TargetLoweringBase.cpp
+++ b/llvm/lib/CodeGen/TargetLoweringBase.cpp
@@ -824,15 +824,15 @@ void TargetLoweringBase::initActions() {
Expand);
// These library functions default to expand.
- setOperationAction(
- {ISD::FCBRT, ISD::FLOG, ISD::FLOG2, ISD::FLOG10, ISD::FEXP,
- ISD::FEXP2, ISD::FEXP10, ISD::FFLOOR, ISD::FNEARBYINT, ISD::FCEIL,
- ISD::FRINT, ISD::FTRUNC, ISD::LRINT, ISD::LLRINT, ISD::FROUNDEVEN,
- ISD::FTAN, ISD::FACOS, ISD::FASIN, ISD::FATAN, ISD::FCOSH,
- ISD::FSINH, ISD::FTANH},
- {MVT::f32, MVT::f64, MVT::f128}, Expand);
-
- setOperationAction({ISD::LROUND, ISD::LLROUND},
+ setOperationAction({ISD::FCBRT, ISD::FLOG, ISD::FLOG2, ISD::FLOG10,
+ ISD::FEXP, ISD::FEXP2, ISD::FEXP10, ISD::FFLOOR,
+ ISD::FNEARBYINT, ISD::FCEIL, ISD::FRINT, ISD::FTRUNC,
+ ISD::FROUNDEVEN, ISD::FTAN, ISD::FACOS, ISD::FASIN,
+ ISD::FATAN, ISD::FCOSH, ISD::FSINH, ISD::FTANH},
+ {MVT::f32, MVT::f64, MVT::f128}, Expand);
+
+ // FIXME: Query RuntimeLibCalls to make the decision.
+ setOperationAction({ISD::LRINT, ISD::LLRINT, ISD::LROUND, ISD::LLROUND},
{MVT::f32, MVT::f64, MVT::f128}, LibCall);
setOperationAction({ISD::FTAN, ISD::FACOS, ISD::FASIN, ISD::FATAN, ISD::FCOSH,
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index af0ffc8fc7d07..ec386000302f1 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -408,6 +408,9 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(const TargetMachine &TM,
setOperationAction(ISD::FRINT, {MVT::f16, MVT::f32, MVT::f64}, Custom);
+ setOperationAction({ISD::LRINT, ISD::LLRINT}, {MVT::f16, MVT::f32, MVT::f64},
+ Expand);
+
setOperationAction(ISD::FREM, {MVT::f16, MVT::f32, MVT::f64}, Custom);
if (Subtarget->has16BitInsts())
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
index bdec2c54b4450..c6c4b8f930647 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
@@ -1146,6 +1146,11 @@ AMDGPULegalizerInfo::AMDGPULegalizerInfo(const GCNSubtarget &ST_,
.scalarize(0)
.lower();
+ getActionDefinitionsBuilder({G_INTRINSIC_LRINT, G_INTRINSIC_LLRINT})
+ .clampScalar(0, S16, S64)
+ .scalarize(0)
+ .lower();
+
if (ST.has16BitInsts()) {
getActionDefinitionsBuilder(
{G_INTRINSIC_TRUNC, G_FCEIL, G_INTRINSIC_ROUNDEVEN})
diff --git a/llvm/test/CodeGen/AMDGPU/lrint.ll b/llvm/test/CodeGen/AMDGPU/lrint.ll
new file mode 100644
index 0000000000000..31e6cf6ea645c
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/lrint.ll
@@ -0,0 +1,771 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --extra_scrub --version 5
+
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9,GFX9-SDAG %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9,GFX9-GISEL %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GCN,GFX10,GFX10-SDAG %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GCN,GFX10,GFX10-GISEL %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck -check-prefixes=GCN,GFX11,GFX11-SDAG %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck -check-prefixes=GCN,GFX11,GFX11-GISEL %s
+
+declare float @llvm.rint.f32(float)
+declare i32 @llvm.lrint.i32.f32(float)
+declare i32 @llvm.lrint.i32.f64(double)
+declare i64 @llvm.lrint.i64.f32(float)
+declare i64 @llvm.lrint.i64.f64(double)
+declare i64 @llvm.llrint.i64.f32(float)
+declare half @llvm.rint.f16(half)
+declare i32 @llvm.lrint.i32.f16(half %arg)
+declare <2 x float> @llvm.rint.v2f32.v2f32(<2 x float> %arg)
+declare <2 x i32> @llvm.lrint.v2i32.v2f32(<2 x float> %arg)
+declare <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> %arg)
+
+define float @intrinsic_frint(float %arg) {
+; GCN-LABEL: intrinsic_frint:
+; GCN: ; %bb.0: ; %entry
+; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT: v_rndne_f32_e32 v0, v0
+; GCN-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call float @llvm.rint.f32(float %arg)
+ ret float %res
+}
+
+define i32 @intrinsic_lrint_i32_f32(float %arg) {
+; GFX9-LABEL: intrinsic_lrint_i32_f32:
+; GFX9: ; %bb.0: ; %entry
+; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX9-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: intrinsic_lrint_i32_f32:
+; GFX10: ; %bb.0: ; %entry
+; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX10-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-LABEL: intrinsic_lrint_i32_f32:
+; GFX11: ; %bb.0: ; %entry
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX11-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i32 @llvm.lrint.i32.f32(float %arg)
+ ret i32 %res
+}
+
+define i32 @intrinsic_lrint_i32_f64(double %arg) {
+; GFX9-LABEL: intrinsic_lrint_i32_f64:
+; GFX9: ; %bb.0: ; %entry
+; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX9-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
+; GFX9-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: intrinsic_lrint_i32_f64:
+; GFX10: ; %bb.0: ; %entry
+; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX10-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
+; GFX10-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-LABEL: intrinsic_lrint_i32_f64:
+; GFX11: ; %bb.0: ; %entry
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
+; GFX11-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i32 @llvm.lrint.i32.f64(double %arg)
+ ret i32 %res
+}
+
+define i64 @intrinsic_lrint_i64_f32(float %arg) {
+; GFX9-SDAG-LABEL: intrinsic_lrint_i64_f32:
+; GFX9-SDAG: ; %bb.0: ; %entry
+; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0x2f800000
+; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v0|, s4
+; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0xcf800000
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v1
+; GFX9-SDAG-NEXT: v_fma_f32 v1, v1, s4, |v0|
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v2, v2, v3
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v1, v3
+; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
+; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v2, v3, vcc
+; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: intrinsic_lrint_i64_f32:
+; GFX9-GISEL: ; %bb.0: ; %entry
+; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0x2f800000
+; GFX9-GISEL-NEXT: v_mul_f32_e64 v2, |v1|, v2
+; GFX9-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0xcf800000
+; GFX9-GISEL-NEXT: v_fma_f32 v1, v2, v3, |v1|
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v1, v3
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v1, v2, v3
+; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
+; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
+; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-SDAG-LABEL: intrinsic_lrint_i64_f32:
+; GFX10-SDAG: ; %bb.0: ; %entry
+; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
+; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX10-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX10-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-GISEL-LABEL: intrinsic_lrint_i64_f32:
+; GFX10-GISEL: ; %bb.0: ; %entry
+; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX10-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
+; GFX10-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX10-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-SDAG-LABEL: intrinsic_lrint_i64_f32:
+; GFX11-SDAG: ; %bb.0: ; %entry
+; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
+; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX11-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
+; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-GISEL-LABEL: intrinsic_lrint_i64_f32:
+; GFX11-GISEL: ; %bb.0: ; %entry
+; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX11-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i64 @llvm.lrint.i64.f32(float %arg)
+ ret i64 %res
+}
+
+define i64 @intrinsic_lrint_i64_f64(double %arg) {
+; GFX9-SDAG-LABEL: intrinsic_lrint_i64_f64:
+; GFX9-SDAG: ; %bb.0: ; %entry
+; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX9-SDAG-NEXT: s_movk_i32 s4, 0xffe0
+; GFX9-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], s4
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0
+; GFX9-SDAG-NEXT: s_mov_b32 s5, 0xc1f00000
+; GFX9-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX9-SDAG-NEXT: v_fma_f64 v[0:1], v[2:3], s[4:5], v[0:1]
+; GFX9-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX9-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: intrinsic_lrint_i64_f64:
+; GFX9-GISEL: ; %bb.0: ; %entry
+; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3df00000
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v4, 0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v5, 0xc1f00000
+; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX9-GISEL-NEXT: v_mul_f64 v[2:3], v[0:1], v[2:3]
+; GFX9-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX9-GISEL-NEXT: v_fma_f64 v[0:1], v[2:3], v[4:5], v[0:1]
+; GFX9-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX9-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-SDAG-LABEL: intrinsic_lrint_i64_f64:
+; GFX10-SDAG: ; %bb.0: ; %entry
+; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX10-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
+; GFX10-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX10-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX10-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX10-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-GISEL-LABEL: intrinsic_lrint_i64_f64:
+; GFX10-GISEL: ; %bb.0: ; %entry
+; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX10-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
+; GFX10-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX10-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX10-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX10-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-SDAG-LABEL: intrinsic_lrint_i64_f64:
+; GFX11-SDAG: ; %bb.0: ; %entry
+; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
+; GFX11-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX11-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX11-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-GISEL-LABEL: intrinsic_lrint_i64_f64:
+; GFX11-GISEL: ; %bb.0: ; %entry
+; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX11-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX11-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX11-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i64 @llvm.lrint.i64.f64(double %arg)
+ ret i64 %res
+}
+
+define i64 @intrinsic_llrint_i64_f32(float %arg) {
+; GFX9-SDAG-LABEL: intrinsic_llrint_i64_f32:
+; GFX9-SDAG: ; %bb.0: ; %entry
+; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0x2f800000
+; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v0|, s4
+; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0xcf800000
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v1
+; GFX9-SDAG-NEXT: v_fma_f32 v1, v1, s4, |v0|
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v2, v2, v3
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v1, v3
+; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
+; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v2, v3, vcc
+; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: intrinsic_llrint_i64_f32:
+; GFX9-GISEL: ; %bb.0: ; %entry
+; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0x2f800000
+; GFX9-GISEL-NEXT: v_mul_f32_e64 v2, |v1|, v2
+; GFX9-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0xcf800000
+; GFX9-GISEL-NEXT: v_fma_f32 v1, v2, v3, |v1|
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v1, v3
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v1, v2, v3
+; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
+; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
+; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-SDAG-LABEL: intrinsic_llrint_i64_f32:
+; GFX10-SDAG: ; %bb.0: ; %entry
+; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
+; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX10-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX10-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-GISEL-LABEL: intrinsic_llrint_i64_f32:
+; GFX10-GISEL: ; %bb.0: ; %entry
+; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX10-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
+; GFX10-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX10-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-SDAG-LABEL: intrinsic_llrint_i64_f32:
+; GFX11-SDAG: ; %bb.0: ; %entry
+; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
+; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX11-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
+; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-GISEL-LABEL: intrinsic_llrint_i64_f32:
+; GFX11-GISEL: ; %bb.0: ; %entry
+; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
+; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX11-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
+; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i64 @llvm.llrint.i64.f32(float %arg)
+ ret i64 %res
+}
+
+define i64 @intrinsic_llrint_i64_f64(double %arg) {
+; GFX9-SDAG-LABEL: intrinsic_llrint_i64_f64:
+; GFX9-SDAG: ; %bb.0: ; %entry
+; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX9-SDAG-NEXT: s_movk_i32 s4, 0xffe0
+; GFX9-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], s4
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0
+; GFX9-SDAG-NEXT: s_mov_b32 s5, 0xc1f00000
+; GFX9-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX9-SDAG-NEXT: v_fma_f64 v[0:1], v[2:3], s[4:5], v[0:1]
+; GFX9-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX9-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: intrinsic_llrint_i64_f64:
+; GFX9-GISEL: ; %bb.0: ; %entry
+; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3df00000
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v4, 0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v5, 0xc1f00000
+; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX9-GISEL-NEXT: v_mul_f64 v[2:3], v[0:1], v[2:3]
+; GFX9-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX9-GISEL-NEXT: v_fma_f64 v[0:1], v[2:3], v[4:5], v[0:1]
+; GFX9-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX9-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-SDAG-LABEL: intrinsic_llrint_i64_f64:
+; GFX10-SDAG: ; %bb.0: ; %entry
+; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX10-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
+; GFX10-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX10-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX10-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX10-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-GISEL-LABEL: intrinsic_llrint_i64_f64:
+; GFX10-GISEL: ; %bb.0: ; %entry
+; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX10-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
+; GFX10-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX10-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX10-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX10-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-SDAG-LABEL: intrinsic_llrint_i64_f64:
+; GFX11-SDAG: ; %bb.0: ; %entry
+; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-SDAG-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
+; GFX11-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX11-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX11-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-GISEL-LABEL: intrinsic_llrint_i64_f64:
+; GFX11-GISEL: ; %bb.0: ; %entry
+; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-GISEL-NEXT: v_rndne_f64_e32 v[0:1], v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
+; GFX11-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
+; GFX11-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
+; GFX11-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
+; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i64 @llvm.llrint.i64.f64(double %arg)
+ ret i64 %res
+}
+
+define half @intrinsic_frint_half(half %arg) {
+; GCN-LABEL: intrinsic_frint_half:
+; GCN: ; %bb.0: ; %entry
+; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT: v_rndne_f16_e32 v0, v0
+; GCN-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call half @llvm.rint.f16(half %arg)
+ ret half %res
+}
+
+define i32 @intrinsic_lrint_i32_f16(half %arg) {
+; GFX9-LABEL: intrinsic_lrint_i32_f16:
+; GFX9: ; %bb.0: ; %entry
+; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-NEXT: v_rndne_f16_e32 v0, v0
+; GFX9-NEXT: v_cvt_f32_f16_e32 v0, v0
+; GFX9-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX9-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: intrinsic_lrint_i32_f16:
+; GFX10: ; %bb.0: ; %entry
+; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT: v_rndne_f16_e32 v0, v0
+; GFX10-NEXT: v_cvt_f32_f16_e32 v0, v0
+; GFX10-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX10-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-LABEL: intrinsic_lrint_i32_f16:
+; GFX11: ; %bb.0: ; %entry
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_rndne_f16_e32 v0, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_cvt_f32_f16_e32 v0, v0
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX11-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call i32 @llvm.lrint.i32.f16(half %arg)
+ ret i32 %res
+}
+
+define <2 x float> @intrinsic_frint_v2f32_v2f32(<2 x float> %arg) {
+; GCN-LABEL: intrinsic_frint_v2f32_v2f32:
+; GCN: ; %bb.0: ; %entry
+; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT: v_rndne_f32_e32 v0, v0
+; GCN-NEXT: v_rndne_f32_e32 v1, v1
+; GCN-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call <2 x float> @llvm.rint.v2f32.v2f32(<2 x float> %arg)
+ ret <2 x float> %res
+}
+
+define <2 x i32> @intrinsic_lrint_v2i32_v2f32(<2 x float> %arg) {
+; GFX9-LABEL: intrinsic_lrint_v2i32_v2f32:
+; GFX9: ; %bb.0: ; %entry
+; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-NEXT: v_rndne_f32_e32 v1, v1
+; GFX9-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX9-NEXT: v_cvt_i32_f32_e32 v1, v1
+; GFX9-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: intrinsic_lrint_v2i32_v2f32:
+; GFX10: ; %bb.0: ; %entry
+; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-NEXT: v_rndne_f32_e32 v1, v1
+; GFX10-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX10-NEXT: v_cvt_i32_f32_e32 v1, v1
+; GFX10-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-LABEL: intrinsic_lrint_v2i32_v2f32:
+; GFX11: ; %bb.0: ; %entry
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-NEXT: v_rndne_f32_e32 v1, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, v0
+; GFX11-NEXT: v_cvt_i32_f32_e32 v1, v1
+; GFX11-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call <2 x i32> @llvm.lrint.v2i32.v2f32(<2 x float> %arg)
+ ret <2 x i32> %res
+}
+
+define <2 x i64> @intrinsic_lrint_v2i64_v2f32(<2 x float> %arg) {
+; GFX9-SDAG-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX9-SDAG: ; %bb.0: ; %entry
+; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-SDAG-NEXT: s_mov_b32 s4, 0x2f800000
+; GFX9-SDAG-NEXT: v_mul_f32_e64 v2, |v0|, s4
+; GFX9-SDAG-NEXT: v_floor_f32_e32 v2, v2
+; GFX9-SDAG-NEXT: s_mov_b32 s5, 0xcf800000
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v2
+; GFX9-SDAG-NEXT: v_fma_f32 v2, v2, s5, |v0|
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v4, 31, v0
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v3, v3, v4
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v2, v4
+; GFX9-SDAG-NEXT: v_rndne_f32_e32 v2, v1
+; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v2|, s4
+; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v5, v1
+; GFX9-SDAG-NEXT: v_fma_f32 v1, v1, s5, |v2|
+; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v6, v1
+; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v4
+; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v3, v4, vcc
+; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v2
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v2, v6, v3
+; GFX9-SDAG-NEXT: v_xor_b32_e32 v4, v5, v3
+; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v2, vcc, v2, v3
+; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v3, vcc, v4, v3, vcc
+; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX9-GISEL: ; %bb.0: ; %entry
+; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX9-GISEL-NEXT: v_trunc_f32_e32 v2, v0
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x2f800000
+; GFX9-GISEL-NEXT: v_mul_f32_e64 v4, |v2|, v3
+; GFX9-GISEL-NEXT: v_floor_f32_e32 v4, v4
+; GFX9-GISEL-NEXT: v_mov_b32_e32 v5, 0xcf800000
+; GFX9-GISEL-NEXT: v_fma_f32 v2, v4, v5, |v2|
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v4, v4
+; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v6, 31, v0
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v2, v6
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v2, v4, v6
+; GFX9-GISEL-NEXT: v_rndne_f32_e32 v4, v1
+; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v4
+; GFX9-GISEL-NEXT: v_mul_f32_e64 v3, |v1|, v3
+; GFX9-GISEL-NEXT: v_floor_f32_e32 v3, v3
+; GFX9-GISEL-NEXT: v_fma_f32 v1, v3, v5, |v1|
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v5, v1
+; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v3, v3
+; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v6
+; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v4, 31, v4
+; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v2, v6, vcc
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v2, v5, v4
+; GFX9-GISEL-NEXT: v_xor_b32_e32 v3, v3, v4
+; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v2, vcc, v2, v4
+; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v3, vcc, v3, v4, vcc
+; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-SDAG-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX10-SDAG: ; %bb.0: ; %entry
+; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-SDAG-NEXT: v_rndne_f32_e32 v1, v1
+; GFX10-SDAG-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v0|
+; GFX10-SDAG-NEXT: v_mul_f32_e64 v3, 0x2f800000, |v1|
+; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v5, 31, v0
+; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v6, 31, v1
+; GFX10-SDAG-NEXT: v_floor_f32_e32 v2, v2
+; GFX10-SDAG-NEXT: v_floor_f32_e32 v3, v3
+; GFX10-SDAG-NEXT: v_fma_f32 v4, 0xcf800000, v2, |v0|
+; GFX10-SDAG-NEXT: v_fma_f32 v0, 0xcf800000, v3, |v1|
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v3
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v4
+; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v2, v2, v5
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v3, v3, v6
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v5
+; GFX10-SDAG-NEXT: v_xor_b32_e32 v4, v0, v6
+; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v5
+; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v5, vcc_lo
+; GFX10-SDAG-NEXT: v_sub_co_u32 v2, vcc_lo, v4, v6
+; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v3, v6, vcc_lo
+; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX10-GISEL-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX10-GISEL: ; %bb.0: ; %entry
+; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX10-GISEL-NEXT: v_rndne_f32_e32 v1, v1
+; GFX10-GISEL-NEXT: v_trunc_f32_e32 v2, v0
+; GFX10-GISEL-NEXT: v_trunc_f32_e32 v3, v1
+; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v6, 31, v0
+; GFX10-GISEL-NEXT: v_mul_f32_e64 v4, 0x2f800000, |v2|
+; GFX10-GISEL-NEXT: v_mul_f32_e64 v5, 0x2f800000, |v3|
+; GFX10-GISEL-NEXT: v_floor_f32_e32 v4, v4
+; GFX10-GISEL-NEXT: v_floor_f32_e32 v5, v5
+; GFX10-GISEL-NEXT: v_fma_f32 v2, 0xcf800000, v4, |v2|
+; GFX10-GISEL-NEXT: v_fma_f32 v0, 0xcf800000, v5, |v3|
+; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v1
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v4
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v4, v5
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v6
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v2, v2, v6
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v5, v0, v3
+; GFX10-GISEL-NEXT: v_xor_b32_e32 v4, v4, v3
+; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v6
+; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v6, vcc_lo
+; GFX10-GISEL-NEXT: v_sub_co_u32 v2, vcc_lo, v5, v3
+; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v4, v3, vcc_lo
+; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-SDAG-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX11-SDAG: ; %bb.0: ; %entry
+; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-SDAG-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-SDAG-NEXT: v_rndne_f32_e32 v1, v1
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v0|
+; GFX11-SDAG-NEXT: v_mul_f32_e64 v3, 0x2f800000, |v1|
+; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v5, 31, v0
+; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v6, 31, v1
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-SDAG-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-SDAG-NEXT: v_floor_f32_e32 v3, v3
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-SDAG-NEXT: v_fma_f32 v4, 0xcf800000, v2, |v0|
+; GFX11-SDAG-NEXT: v_fma_f32 v0, 0xcf800000, v3, |v1|
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v3
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v4
+; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v2, v2, v5
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v3, v3, v6
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v5
+; GFX11-SDAG-NEXT: v_xor_b32_e32 v4, v0, v6
+; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3)
+; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v5
+; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v5, vcc_lo
+; GFX11-SDAG-NEXT: v_sub_co_u32 v2, vcc_lo, v4, v6
+; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v3, v6, vcc_lo
+; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX11-GISEL-LABEL: intrinsic_lrint_v2i64_v2f32:
+; GFX11-GISEL: ; %bb.0: ; %entry
+; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-GISEL-NEXT: v_rndne_f32_e32 v0, v0
+; GFX11-GISEL-NEXT: v_rndne_f32_e32 v1, v1
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_trunc_f32_e32 v2, v0
+; GFX11-GISEL-NEXT: v_trunc_f32_e32 v3, v1
+; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v6, 31, v0
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
+; GFX11-GISEL-NEXT: v_mul_f32_e64 v4, 0x2f800000, |v2|
+; GFX11-GISEL-NEXT: v_mul_f32_e64 v5, 0x2f800000, |v3|
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_floor_f32_e32 v4, v4
+; GFX11-GISEL-NEXT: v_floor_f32_e32 v5, v5
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-GISEL-NEXT: v_fma_f32 v2, 0xcf800000, v4, |v2|
+; GFX11-GISEL-NEXT: v_fma_f32 v0, 0xcf800000, v5, |v3|
+; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v1
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_4)
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v4
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v4, v5
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v6
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v2, v2, v6
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v5, v0, v3
+; GFX11-GISEL-NEXT: v_xor_b32_e32 v4, v4, v3
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v6
+; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v6, vcc_lo
+; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-GISEL-NEXT: v_sub_co_u32 v2, vcc_lo, v5, v3
+; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v4, v3, vcc_lo
+; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
+entry:
+ %res = tail call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> %arg)
+ ret <2 x i64> %res
+}
+
More information about the llvm-commits
mailing list