[llvm] AMDGPU: Convert vector 64-bit shl to 32-bit if shift amt >= 32 (PR #132964)
via llvm-commits
llvm-commits at lists.llvm.org
Fri Mar 28 06:04:37 PDT 2025
https://github.com/LU-JOHN updated https://github.com/llvm/llvm-project/pull/132964
>From 7b967f1023ee4558ad58c6dd4801162e5dc99e3c Mon Sep 17 00:00:00 2001
From: John Lu <John.Lu at amd.com>
Date: Tue, 25 Mar 2025 12:40:21 -0500
Subject: [PATCH 1/2] Convert 64-bit shl to 32-bit if shift amt >= 32
Signed-off-by: John Lu <John.Lu at amd.com>
---
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 44 ++++++++++++++-----
llvm/test/CodeGen/AMDGPU/shl64_reduce.ll | 34 ++++++++------
2 files changed, 52 insertions(+), 26 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index dbbd67cea27e5..22621bea2a40f 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -4084,7 +4084,7 @@ SDValue AMDGPUTargetLowering::performShlCombine(SDNode *N,
}
}
- if (VT != MVT::i64)
+ if (VT.getScalarType() != MVT::i64)
return SDValue();
// i64 (shl x, C) -> (build_pair 0, (shl x, C -32))
@@ -4092,21 +4092,24 @@ SDValue AMDGPUTargetLowering::performShlCombine(SDNode *N,
// On some subtargets, 64-bit shift is a quarter rate instruction. In the
// common case, splitting this into a move and a 32-bit shift is faster and
// the same code size.
- EVT TargetType = VT.getHalfSizedIntegerVT(*DAG.getContext());
- EVT TargetVecPairType = EVT::getVectorVT(*DAG.getContext(), TargetType, 2);
KnownBits Known = DAG.computeKnownBits(RHS);
- if (Known.getMinValue().getZExtValue() < TargetType.getSizeInBits())
+ EVT ElementType = VT.getScalarType();
+ EVT TargetScalarType = ElementType.getHalfSizedIntegerVT(*DAG.getContext());
+ EVT TargetType = (VT.isVector() ? VT.changeVectorElementType(TargetScalarType)
+ : TargetScalarType);
+
+ if (Known.getMinValue().getZExtValue() < TargetScalarType.getSizeInBits())
return SDValue();
SDValue ShiftAmt;
if (CRHS) {
- ShiftAmt =
- DAG.getConstant(RHSVal - TargetType.getSizeInBits(), SL, TargetType);
+ ShiftAmt = DAG.getConstant(RHSVal - TargetScalarType.getSizeInBits(), SL,
+ TargetType);
} else {
SDValue truncShiftAmt = DAG.getNode(ISD::TRUNCATE, SL, TargetType, RHS);
const SDValue ShiftMask =
- DAG.getConstant(TargetType.getSizeInBits() - 1, SL, TargetType);
+ DAG.getConstant(TargetScalarType.getSizeInBits() - 1, SL, TargetType);
// This AND instruction will clamp out of bounds shift values.
// It will also be removed during later instruction selection.
ShiftAmt = DAG.getNode(ISD::AND, SL, TargetType, truncShiftAmt, ShiftMask);
@@ -4116,9 +4119,21 @@ SDValue AMDGPUTargetLowering::performShlCombine(SDNode *N,
SDValue NewShift =
DAG.getNode(ISD::SHL, SL, TargetType, Lo, ShiftAmt, N->getFlags());
- const SDValue Zero = DAG.getConstant(0, SL, TargetType);
-
- SDValue Vec = DAG.getBuildVector(TargetVecPairType, SL, {Zero, NewShift});
+ const SDValue Zero = DAG.getConstant(0, SL, TargetScalarType);
+ SDValue Vec;
+
+ if (VT.isVector()) {
+ EVT ConcatType = TargetType.getDoubleNumVectorElementsVT(*DAG.getContext());
+ unsigned NElts = TargetType.getVectorNumElements();
+ SmallVector<SDValue, 8> Ops;
+ DAG.ExtractVectorElements(NewShift, Ops, 0, NElts);
+ for (unsigned I = 0; I != NElts; ++I)
+ Ops.insert(Ops.begin() + 2 * I, Zero);
+ Vec = DAG.getNode(ISD::BUILD_VECTOR, SL, ConcatType, Ops);
+ } else {
+ EVT ConcatType = EVT::getVectorVT(*DAG.getContext(), TargetType, 2);
+ Vec = DAG.getBuildVector(ConcatType, SL, {Zero, NewShift});
+ }
return DAG.getNode(ISD::BITCAST, SL, VT, Vec);
}
@@ -5182,9 +5197,14 @@ SDValue AMDGPUTargetLowering::PerformDAGCombine(SDNode *N,
break;
}
case ISD::SHL: {
- if (DCI.getDAGCombineLevel() < AfterLegalizeDAG)
+ // Range metadata can be invalidated when loads are converted to legal types
+ // (e.g. v2i64 -> v4i32).
+ // Try to convert vector shl before type legalization so that range metadata
+ // can be utilized.
+ if (!(N->getValueType(0).isVector() &&
+ DCI.getDAGCombineLevel() == BeforeLegalizeTypes) &&
+ DCI.getDAGCombineLevel() < AfterLegalizeDAG)
break;
-
return performShlCombine(N, DCI);
}
case ISD::SRL: {
diff --git a/llvm/test/CodeGen/AMDGPU/shl64_reduce.ll b/llvm/test/CodeGen/AMDGPU/shl64_reduce.ll
index 69242f4e44840..21b7ed4d6b779 100644
--- a/llvm/test/CodeGen/AMDGPU/shl64_reduce.ll
+++ b/llvm/test/CodeGen/AMDGPU/shl64_reduce.ll
@@ -72,39 +72,41 @@ define i64 @shl_metadata_cant_be_narrowed_to_i32(i64 %arg0, ptr %arg1.ptr) {
ret i64 %shl
}
-; FIXME: This case should be reduced
define <2 x i64> @shl_v2_metadata(<2 x i64> %arg0, ptr %arg1.ptr) {
; CHECK-LABEL: shl_v2_metadata:
; CHECK: ; %bb.0:
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; CHECK-NEXT: flat_load_dwordx4 v[4:7], v[4:5]
+; CHECK-NEXT: flat_load_dwordx4 v[3:6], v[4:5]
; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
-; CHECK-NEXT: v_lshlrev_b64 v[0:1], v4, v[0:1]
-; CHECK-NEXT: v_lshlrev_b64 v[2:3], v6, v[2:3]
+; CHECK-NEXT: v_lshlrev_b32_e32 v1, v3, v0
+; CHECK-NEXT: v_lshlrev_b32_e32 v3, v5, v2
+; CHECK-NEXT: v_mov_b32_e32 v0, 0
+; CHECK-NEXT: v_mov_b32_e32 v2, 0
; CHECK-NEXT: s_setpc_b64 s[30:31]
%shift.amt = load <2 x i64>, ptr %arg1.ptr, !range !0, !noundef !{}
%shl = shl <2 x i64> %arg0, %shift.amt
ret <2 x i64> %shl
}
-; FIXME: This case should be reduced
define <3 x i64> @shl_v3_metadata(<3 x i64> %arg0, ptr %arg1.ptr) {
; CHECK-LABEL: shl_v3_metadata:
; CHECK: ; %bb.0:
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; CHECK-NEXT: flat_load_dword v12, v[6:7] offset:16
+; CHECK-NEXT: flat_load_dword v1, v[6:7] offset:16
; CHECK-NEXT: flat_load_dwordx4 v[8:11], v[6:7]
; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
-; CHECK-NEXT: v_lshlrev_b64 v[4:5], v12, v[4:5]
-; CHECK-NEXT: v_lshlrev_b64 v[0:1], v8, v[0:1]
-; CHECK-NEXT: v_lshlrev_b64 v[2:3], v10, v[2:3]
+; CHECK-NEXT: v_lshlrev_b32_e32 v5, v1, v4
+; CHECK-NEXT: v_lshlrev_b32_e32 v1, v8, v0
+; CHECK-NEXT: v_lshlrev_b32_e32 v3, v10, v2
+; CHECK-NEXT: v_mov_b32_e32 v0, 0
+; CHECK-NEXT: v_mov_b32_e32 v2, 0
+; CHECK-NEXT: v_mov_b32_e32 v4, 0
; CHECK-NEXT: s_setpc_b64 s[30:31]
%shift.amt = load <3 x i64>, ptr %arg1.ptr, !range !0, !noundef !{}
%shl = shl <3 x i64> %arg0, %shift.amt
ret <3 x i64> %shl
}
-; FIXME: This case should be reduced
define <4 x i64> @shl_v4_metadata(<4 x i64> %arg0, ptr %arg1.ptr) {
; CHECK-LABEL: shl_v4_metadata:
; CHECK: ; %bb.0:
@@ -113,11 +115,15 @@ define <4 x i64> @shl_v4_metadata(<4 x i64> %arg0, ptr %arg1.ptr) {
; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
; CHECK-NEXT: flat_load_dwordx4 v[13:16], v[8:9] offset:16
; CHECK-NEXT: ; kill: killed $vgpr8 killed $vgpr9
-; CHECK-NEXT: v_lshlrev_b64 v[0:1], v10, v[0:1]
-; CHECK-NEXT: v_lshlrev_b64 v[2:3], v12, v[2:3]
+; CHECK-NEXT: v_lshlrev_b32_e32 v1, v10, v0
+; CHECK-NEXT: v_lshlrev_b32_e32 v3, v12, v2
; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
-; CHECK-NEXT: v_lshlrev_b64 v[4:5], v13, v[4:5]
-; CHECK-NEXT: v_lshlrev_b64 v[6:7], v15, v[6:7]
+; CHECK-NEXT: v_lshlrev_b32_e32 v5, v13, v4
+; CHECK-NEXT: v_lshlrev_b32_e32 v7, v15, v6
+; CHECK-NEXT: v_mov_b32_e32 v0, 0
+; CHECK-NEXT: v_mov_b32_e32 v2, 0
+; CHECK-NEXT: v_mov_b32_e32 v4, 0
+; CHECK-NEXT: v_mov_b32_e32 v6, 0
; CHECK-NEXT: s_setpc_b64 s[30:31]
%shift.amt = load <4 x i64>, ptr %arg1.ptr, !range !0, !noundef !{}
%shl = shl <4 x i64> %arg0, %shift.amt
>From 636238885c41b20bdbc605b23de1e3741b41286f Mon Sep 17 00:00:00 2001
From: John Lu <John.Lu at amd.com>
Date: Fri, 28 Mar 2025 08:04:13 -0500
Subject: [PATCH 2/2] Recode with array indexing
Signed-off-by: John Lu <John.Lu at amd.com>
---
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index 22621bea2a40f..d36355860a33b 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -4096,8 +4096,8 @@ SDValue AMDGPUTargetLowering::performShlCombine(SDNode *N,
EVT ElementType = VT.getScalarType();
EVT TargetScalarType = ElementType.getHalfSizedIntegerVT(*DAG.getContext());
- EVT TargetType = (VT.isVector() ? VT.changeVectorElementType(TargetScalarType)
- : TargetScalarType);
+ EVT TargetType = VT.isVector() ? VT.changeVectorElementType(TargetScalarType)
+ : TargetScalarType;
if (Known.getMinValue().getZExtValue() < TargetScalarType.getSizeInBits())
return SDValue();
@@ -4125,11 +4125,13 @@ SDValue AMDGPUTargetLowering::performShlCombine(SDNode *N,
if (VT.isVector()) {
EVT ConcatType = TargetType.getDoubleNumVectorElementsVT(*DAG.getContext());
unsigned NElts = TargetType.getVectorNumElements();
- SmallVector<SDValue, 8> Ops;
- DAG.ExtractVectorElements(NewShift, Ops, 0, NElts);
+ SmallVector<SDValue, 8> HiOps;
+ SmallVector<SDValue, 16> HiAndLoOps(NElts * 2, Zero);
+
+ DAG.ExtractVectorElements(NewShift, HiOps, 0, NElts);
for (unsigned I = 0; I != NElts; ++I)
- Ops.insert(Ops.begin() + 2 * I, Zero);
- Vec = DAG.getNode(ISD::BUILD_VECTOR, SL, ConcatType, Ops);
+ HiAndLoOps[2 * I + 1] = HiOps[I];
+ Vec = DAG.getNode(ISD::BUILD_VECTOR, SL, ConcatType, HiAndLoOps);
} else {
EVT ConcatType = EVT::getVectorVT(*DAG.getContext(), TargetType, 2);
Vec = DAG.getBuildVector(ConcatType, SL, {Zero, NewShift});
More information about the llvm-commits
mailing list