[llvm] r274650 - [SystemZ] Remove AND mask of bottom 6 bits when result is used for shift/rotate
Elliot Colp via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 6 11:13:11 PDT 2016
Author: colpell
Date: Wed Jul 6 13:13:11 2016
New Revision: 274650
URL: http://llvm.org/viewvc/llvm-project?rev=274650&view=rev
Log:
[SystemZ] Remove AND mask of bottom 6 bits when result is used for shift/rotate
On SystemZ, shift and rotate instructions only use the bottom 6 bits of the shift/rotate amount.
Therefore, if the amount is ANDed with an immediate mask that has all of the bottom 6 bits set, we
can remove the AND operation entirely.
Differential Revision: http://reviews.llvm.org/D21854
Added:
llvm/trunk/test/CodeGen/SystemZ/rot-02.ll
llvm/trunk/test/CodeGen/SystemZ/shift-12.ll
Modified:
llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.cpp
llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.h
llvm/trunk/test/CodeGen/SystemZ/rot-01.ll
Modified: llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.cpp?rev=274650&r1=274649&r2=274650&view=diff
==============================================================================
--- llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.cpp Wed Jul 6 13:13:11 2016
@@ -440,6 +440,10 @@ SystemZTargetLowering::SystemZTargetLowe
setTargetDAGCombine(ISD::EXTRACT_VECTOR_ELT);
setTargetDAGCombine(ISD::FP_ROUND);
setTargetDAGCombine(ISD::BSWAP);
+ setTargetDAGCombine(ISD::SHL);
+ setTargetDAGCombine(ISD::SRA);
+ setTargetDAGCombine(ISD::SRL);
+ setTargetDAGCombine(ISD::ROTL);
// Handle intrinsics.
setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::Other, Custom);
@@ -2874,7 +2878,7 @@ lowerDYNAMIC_STACKALLOC(SDValue Op, Sele
// Add extra space for alignment if needed.
if (ExtraAlignSpace)
NeededSpace = DAG.getNode(ISD::ADD, DL, MVT::i64, NeededSpace,
- DAG.getConstant(ExtraAlignSpace, DL, MVT::i64));
+ DAG.getConstant(ExtraAlignSpace, DL, MVT::i64));
// Get the new stack pointer value.
SDValue NewSP = DAG.getNode(ISD::SUB, DL, MVT::i64, OldSP, NeededSpace);
@@ -5069,6 +5073,50 @@ SDValue SystemZTargetLowering::combineBS
return SDValue();
}
+SDValue SystemZTargetLowering::combineSHIFTROT(
+ SDNode *N, DAGCombinerInfo &DCI) const {
+
+ SelectionDAG &DAG = DCI.DAG;
+
+ // Shift/rotate instructions only use the last 6 bits of the second operand
+ // register. If the second operand is the result of an AND with an immediate
+ // value that has its last 6 bits set, we can safely remove the AND operation.
+ SDValue N1 = N->getOperand(1);
+ if (N1.getOpcode() == ISD::AND) {
+ auto *AndMask = dyn_cast<ConstantSDNode>(N1.getOperand(1));
+
+ // The AND mask is constant
+ if (AndMask) {
+ auto AmtVal = AndMask->getZExtValue();
+
+ // Bottom 6 bits are set
+ if ((AmtVal & 0x3f) == 0x3f) {
+ SDValue AndOp = N1->getOperand(0);
+
+ // This is the only use, so remove the node
+ if (N1.hasOneUse()) {
+ // Combine the AND away
+ DCI.CombineTo(N1.getNode(), AndOp);
+
+ // Return N so it isn't rechecked
+ return SDValue(N, 0);
+
+ // The node will be reused, so create a new node for this one use
+ } else {
+ SDValue Replace = DAG.getNode(N->getOpcode(), SDLoc(N),
+ N->getValueType(0), N->getOperand(0),
+ AndOp);
+ DCI.AddToWorklist(Replace.getNode());
+
+ return Replace;
+ }
+ }
+ }
+ }
+
+ return SDValue();
+}
+
SDValue SystemZTargetLowering::PerformDAGCombine(SDNode *N,
DAGCombinerInfo &DCI) const {
switch(N->getOpcode()) {
@@ -5081,7 +5129,12 @@ SDValue SystemZTargetLowering::PerformDA
case SystemZISD::JOIN_DWORDS: return combineJOIN_DWORDS(N, DCI);
case ISD::FP_ROUND: return combineFP_ROUND(N, DCI);
case ISD::BSWAP: return combineBSWAP(N, DCI);
+ case ISD::SHL:
+ case ISD::SRA:
+ case ISD::SRL:
+ case ISD::ROTL: return combineSHIFTROT(N, DCI);
}
+
return SDValue();
}
Modified: llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.h?rev=274650&r1=274649&r2=274650&view=diff
==============================================================================
--- llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.h (original)
+++ llvm/trunk/lib/Target/SystemZ/SystemZISelLowering.h Wed Jul 6 13:13:11 2016
@@ -552,6 +552,7 @@ private:
SDValue combineJOIN_DWORDS(SDNode *N, DAGCombinerInfo &DCI) const;
SDValue combineFP_ROUND(SDNode *N, DAGCombinerInfo &DCI) const;
SDValue combineBSWAP(SDNode *N, DAGCombinerInfo &DCI) const;
+ SDValue combineSHIFTROT(SDNode *N, DAGCombinerInfo &DCI) const;
// If the last instruction before MBBI in MBB was some form of COMPARE,
// try to replace it with a COMPARE AND BRANCH just before MBBI.
Modified: llvm/trunk/test/CodeGen/SystemZ/rot-01.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/SystemZ/rot-01.ll?rev=274650&r1=274649&r2=274650&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/SystemZ/rot-01.ll (original)
+++ llvm/trunk/test/CodeGen/SystemZ/rot-01.ll Wed Jul 6 13:13:11 2016
@@ -21,9 +21,9 @@ define i32 @f1(i32 %val, i32 %amt) {
; Test 64-bit rotate.
define i64 @f2(i64 %val, i64 %amt) {
; CHECK-LABEL: f2:
-; CHECK: nill %r3, 63
+; CHECK: nill %r3, 31
; CHECK: rllg %r2, %r2, 0(%r3)
- %mod = urem i64 %amt, 64
+ %mod = urem i64 %amt, 32
%inv = sub i64 64, %mod
%parta = shl i64 %val, %mod
Added: llvm/trunk/test/CodeGen/SystemZ/rot-02.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/SystemZ/rot-02.ll?rev=274650&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/SystemZ/rot-02.ll (added)
+++ llvm/trunk/test/CodeGen/SystemZ/rot-02.ll Wed Jul 6 13:13:11 2016
@@ -0,0 +1,86 @@
+; Test removal of AND operations that don't affect last 6 bits of rotate amount
+; operand.
+;
+; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
+
+; Test that AND is not removed when some lower 6 bits are not set.
+define i32 @f1(i32 %val, i32 %amt) {
+; CHECK-LABEL: f1:
+; CHECK: nil{{[lf]}} %r3, 31
+; CHECK: rll %r2, %r2, 0(%r3)
+ %and = and i32 %amt, 31
+
+ %inv = sub i32 32, %and
+ %parta = shl i32 %val, %and
+ %partb = lshr i32 %val, %inv
+
+ %rotl = or i32 %parta, %partb
+
+ ret i32 %rotl
+}
+
+; Test removal of AND mask with only bottom 6 bits set.
+define i32 @f2(i32 %val, i32 %amt) {
+; CHECK-LABEL: f2:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: rll %r2, %r2, 0(%r3)
+ %and = and i32 %amt, 63
+
+ %inv = sub i32 32, %and
+ %parta = shl i32 %val, %and
+ %partb = lshr i32 %val, %inv
+
+ %rotl = or i32 %parta, %partb
+
+ ret i32 %rotl
+}
+
+; Test removal of AND mask including but not limited to bottom 6 bits.
+define i32 @f3(i32 %val, i32 %amt) {
+; CHECK-LABEL: f3:
+; CHECK-NOT: nil{{[lf]}} %r3, 255
+; CHECK: rll %r2, %r2, 0(%r3)
+ %and = and i32 %amt, 255
+
+ %inv = sub i32 32, %and
+ %parta = shl i32 %val, %and
+ %partb = lshr i32 %val, %inv
+
+ %rotl = or i32 %parta, %partb
+
+ ret i32 %rotl
+}
+
+; Test removal of AND mask from RLLG.
+define i64 @f4(i64 %val, i64 %amt) {
+; CHECK-LABEL: f4:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: rllg %r2, %r2, 0(%r3)
+ %and = and i64 %amt, 63
+
+ %inv = sub i64 64, %and
+ %parta = shl i64 %val, %and
+ %partb = lshr i64 %val, %inv
+
+ %rotl = or i64 %parta, %partb
+
+ ret i64 %rotl
+}
+
+; Test that AND is not entirely removed if the result is reused.
+define i32 @f5(i32 %val, i32 %amt) {
+; CHECK-LABEL: f5:
+; CHECK: rll %r2, %r2, 0(%r3)
+; CHECK: nil{{[lf]}} %r3, 63
+; CHECK: ar %r2, %r3
+ %and = and i32 %amt, 63
+
+ %inv = sub i32 32, %and
+ %parta = shl i32 %val, %and
+ %partb = lshr i32 %val, %inv
+
+ %rotl = or i32 %parta, %partb
+
+ %reuse = add i32 %and, %rotl
+ ret i32 %reuse
+}
Added: llvm/trunk/test/CodeGen/SystemZ/shift-12.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/SystemZ/shift-12.ll?rev=274650&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/SystemZ/shift-12.ll (added)
+++ llvm/trunk/test/CodeGen/SystemZ/shift-12.ll Wed Jul 6 13:13:11 2016
@@ -0,0 +1,106 @@
+; Test removal of AND operations that don't affect last 6 bits of shift amount
+; operand.
+;
+; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
+
+; Test that AND is not removed when some lower 6 bits are not set.
+define i32 @f1(i32 %a, i32 %sh) {
+; CHECK-LABEL: f1:
+; CHECK: nil{{[lf]}} %r3, 31
+; CHECK: sll %r2, 0(%r3)
+ %and = and i32 %sh, 31
+ %shift = shl i32 %a, %and
+ ret i32 %shift
+}
+
+; Test removal of AND mask with only bottom 6 bits set.
+define i32 @f2(i32 %a, i32 %sh) {
+; CHECK-LABEL: f2:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: sll %r2, 0(%r3)
+ %and = and i32 %sh, 63
+ %shift = shl i32 %a, %and
+ ret i32 %shift
+}
+
+; Test removal of AND mask including but not limited to bottom 6 bits.
+define i32 @f3(i32 %a, i32 %sh) {
+; CHECK-LABEL: f3:
+; CHECK-NOT: nil{{[lf]}} %r3, 255
+; CHECK: sll %r2, 0(%r3)
+ %and = and i32 %sh, 255
+ %shift = shl i32 %a, %and
+ ret i32 %shift
+}
+
+; Test removal of AND mask from SRA.
+define i32 @f4(i32 %a, i32 %sh) {
+; CHECK-LABEL: f4:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: sra %r2, 0(%r3)
+ %and = and i32 %sh, 63
+ %shift = ashr i32 %a, %and
+ ret i32 %shift
+}
+
+; Test removal of AND mask from SRL.
+define i32 @f5(i32 %a, i32 %sh) {
+; CHECK-LABEL: f5:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: srl %r2, 0(%r3)
+ %and = and i32 %sh, 63
+ %shift = lshr i32 %a, %and
+ ret i32 %shift
+}
+
+; Test removal of AND mask from SLLG.
+define i64 @f6(i64 %a, i64 %sh) {
+; CHECK-LABEL: f6:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: sllg %r2, %r2, 0(%r3)
+ %and = and i64 %sh, 63
+ %shift = shl i64 %a, %and
+ ret i64 %shift
+}
+
+; Test removal of AND mask from SRAG.
+define i64 @f7(i64 %a, i64 %sh) {
+; CHECK-LABEL: f7:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: srag %r2, %r2, 0(%r3)
+ %and = and i64 %sh, 63
+ %shift = ashr i64 %a, %and
+ ret i64 %shift
+}
+
+; Test removal of AND mask from SRLG.
+define i64 @f8(i64 %a, i64 %sh) {
+; CHECK-LABEL: f8:
+; CHECK-NOT: nil{{[lf]}} %r3, 63
+; CHECK: srlg %r2, %r2, 0(%r3)
+ %and = and i64 %sh, 63
+ %shift = lshr i64 %a, %and
+ ret i64 %shift
+}
+
+; Test that AND with two register operands is not affected.
+define i32 @f9(i32 %a, i32 %b, i32 %sh) {
+; CHECK-LABEL: f9:
+; CHECK: nr %r3, %r4
+; CHECK: sll %r2, 0(%r3)
+ %and = and i32 %sh, %b
+ %shift = shl i32 %a, %and
+ ret i32 %shift
+}
+
+; Test that AND is not entirely removed if the result is reused.
+define i32 @f10(i32 %a, i32 %sh) {
+; CHECK-LABEL: f10:
+; CHECK: sll %r2, 0(%r3)
+; CHECK: nil{{[lf]}} %r3, 63
+; CHECK: ar %r2, %r3
+ %and = and i32 %sh, 63
+ %shift = shl i32 %a, %and
+ %reuse = add i32 %and, %shift
+ ret i32 %reuse
+}
More information about the llvm-commits
mailing list