[PATCH] D61918: [DAGCombiner] visitShiftByConstant(): drop bogus(?) signbit check
Roman Lebedev via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue May 14 14:44:48 PDT 2019
lebedev.ri created this revision.
lebedev.ri added reviewers: RKSimon, spatel, craig.topper.
lebedev.ri added a project: LLVM.
That check claims that the transform is illegal otherwise.
That isn't true:
1. For `ISD::ADD`, we only process `ISD::SHL` outer shift => sign bit does not matter https://rise4fun.com/Alive/K4A
2. For `ISD::AND`, there is no restriction on constants: https://rise4fun.com/Alive/Wy3
3. For `ISD::OR`, there is no restriction on constants: https://rise4fun.com/Alive/GOH
4. For `ISD::XOR`, there is no restriction on constants: https://rise4fun.com/Alive/ml6
So, why is it there then?
This changes the testcase that was touched by @spatel in rL347478 <https://reviews.llvm.org/rL347478>,
but i'm not sure that test tests anything particular?
Repository:
rL LLVM
https://reviews.llvm.org/D61918
Files:
lib/CodeGen/SelectionDAG/DAGCombiner.cpp
test/CodeGen/X86/xor.ll
Index: test/CodeGen/X86/xor.ll
===================================================================
--- test/CodeGen/X86/xor.ll
+++ test/CodeGen/X86/xor.ll
@@ -480,16 +480,16 @@
;
; X64-LIN-LABEL: test12:
; X64-LIN: # %bb.0:
-; X64-LIN-NEXT: notl %edx
; X64-LIN-NEXT: movslq %edx, %rax
+; X64-LIN-NEXT: notq %rax
; X64-LIN-NEXT: shlq $4, %rax
; X64-LIN-NEXT: addq %rdi, %rax
; X64-LIN-NEXT: retq
;
; X64-WIN-LABEL: test12:
; X64-WIN: # %bb.0:
-; X64-WIN-NEXT: notl %r8d
; X64-WIN-NEXT: movslq %r8d, %rax
+; X64-WIN-NEXT: notq %rax
; X64-WIN-NEXT: shlq $4, %rax
; X64-WIN-NEXT: addq %rcx, %rax
; X64-WIN-NEXT: retq
Index: lib/CodeGen/SelectionDAG/DAGCombiner.cpp
===================================================================
--- lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -6582,11 +6582,16 @@
/// Handle transforms common to the three shifts, when the shift amount is a
/// constant.
+/// We are looking for: (shift being one of shl/sra/srl)
+/// shift (binop X, C0), C1
+/// And want to transform into:
+/// binop (shift X, C1), (shift C0, C1)
SDValue DAGCombiner::visitShiftByConstant(SDNode *N, ConstantSDNode *Amt) {
// Do not turn a 'not' into a regular xor.
if (isBitwiseNot(N->getOperand(0)))
return SDValue();
+ // The inner binop must be one-use, since we want to replace it.
SDNode *LHS = N->getOperand(0).getNode();
if (!LHS->hasOneUse()) return SDValue();
@@ -6594,27 +6599,23 @@
// instead of (shift (and)), likewise for add, or, xor, etc. This sort of
// thing happens with address calculations, so it's important to canonicalize
// it.
- bool HighBitSet = false; // Can we transform this if the high bit is set?
-
switch (LHS->getOpcode()) {
- default: return SDValue();
+ default:
+ return SDValue();
case ISD::OR:
case ISD::XOR:
- HighBitSet = false; // We can only transform sra if the high bit is clear.
- break;
case ISD::AND:
- HighBitSet = true; // We can only transform sra if the high bit is set.
break;
case ISD::ADD:
if (N->getOpcode() != ISD::SHL)
return SDValue(); // only shl(add) not sr[al](add).
- HighBitSet = false; // We can only transform sra if the high bit is clear.
break;
}
// We require the RHS of the binop to be a constant and not opaque as well.
ConstantSDNode *BinOpCst = getAsNonOpaqueConstant(LHS->getOperand(1));
- if (!BinOpCst) return SDValue();
+ if (!BinOpCst)
+ return SDValue();
// FIXME: disable this unless the input to the binop is a shift by a constant
// or is copy/select.Enable this in other cases when figure out it's exactly profitable.
@@ -6634,16 +6635,6 @@
EVT VT = N->getValueType(0);
- // If this is a signed shift right, and the high bit is modified by the
- // logical operation, do not perform the transformation. The highBitSet
- // boolean indicates the value of the high bit of the constant which would
- // cause it to be modified for this operation.
- if (N->getOpcode() == ISD::SRA) {
- bool BinOpRHSSignSet = BinOpCst->getAPIntValue().isNegative();
- if (BinOpRHSSignSet != HighBitSet)
- return SDValue();
- }
-
if (!TLI.isDesirableToCommuteWithShift(N, Level))
return SDValue();
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D61918.199516.patch
Type: text/x-patch
Size: 3325 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190514/cee0bad7/attachment.bin>
More information about the llvm-commits
mailing list