maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[InstCombine] improve fold for icmp_eq_and to icmp_ult 

In D95959, the improve analysis for "C >> X" broken the fold
((%x & C) == 0) --> %x u< (-C) iff (-C) is power of two.

It simplifies C, but fails to satisfy the fold condition.
This patch try to restore C before the fold.

Reviewed By: spatel

Differential Revision: https://reviews.llvm.org/D128790
This commit is contained in:
Chenbing Zheng 2022-07-05 17:14:22 +08:00
parent 220366d5b5
commit b43dd2f6c4
2 changed files with 34 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
View File

@ -1777,11 +1777,16 @@ Instruction *InstCombinerImpl::foldICmpAndConstConst(ICmpInst &Cmp,
return new ICmpInst(NewPred, X, Zero);
}
APInt NewC2 = *C2;
KnownBits Know = computeKnownBits(And->getOperand(0), 0, And);
// Set high zeros of C2 to allow matching negated power-of-2.
NewC2 = *C2 + APInt::getHighBitsSet(C2->getBitWidth(),
Know.countMinLeadingZeros());
// Restrict this fold only for single-use 'and' (PR10267).
// ((%x & C) == 0) --> %x u< (-C) iff (-C) is power of two.
if (C2->isNegatedPowerOf2()) {
Constant *NegBOC =
ConstantExpr::getNeg(cast<Constant>(And->getOperand(1)));
if (NewC2.isNegatedPowerOf2()) {
Constant *NegBOC = ConstantInt::get(And->getType(), -NewC2);
auto NewPred = isICMP_NE ? ICmpInst::ICMP_UGE : ICmpInst::ICMP_ULT;
return new ICmpInst(NewPred, X, NegBOC);
}

37
llvm/test/Transforms/InstCombine/lshr-and-negC-icmpeq-zero.ll
View File

@ -177,15 +177,10 @@ define i1 @scalar_lshr_and_negC_eq_extra_use_lshr_and(i32 %x, i32 %y, i32 %z, i3
ret i1 %r
}
; Negative tests
; TODO: This could be reduced to lshr+icmp ult.
define i1 @scalar_i32_lshr_and_negC_eq_X_is_constant1(i32 %y) {
; CHECK-LABEL: @scalar_i32_lshr_and_negC_eq_X_is_constant1(
; CHECK-NEXT: [[LSHR:%.*]] = lshr i32 12345, [[Y:%.*]]
; CHECK-NEXT: [[AND:%.*]] = and i32 [[LSHR]], 16376
; CHECK-NEXT: [[R:%.*]] = icmp eq i32 [[AND]], 0
; CHECK-NEXT: [[R:%.*]] = icmp ult i32 [[LSHR]], 8
; CHECK-NEXT: ret i1 [[R]]
;
%lshr = lshr i32 12345, %y
@ -194,13 +189,9 @@ define i1 @scalar_i32_lshr_and_negC_eq_X_is_constant1(i32 %y) {
ret i1 %r
}
; TODO: This could be reduced to lshr+icmp ult.
define i1 @scalar_i32_lshr_and_negC_eq_X_is_constant2(i32 %y) {
; CHECK-LABEL: @scalar_i32_lshr_and_negC_eq_X_is_constant2(
; CHECK-NEXT: [[LSHR:%.*]] = lshr i32 268435456, [[Y:%.*]]
; CHECK-NEXT: [[AND:%.*]] = and i32 [[LSHR]], 536870904
; CHECK-NEXT: [[R:%.*]] = icmp eq i32 [[AND]], 0
; CHECK-NEXT: [[R:%.*]] = icmp ugt i32 [[Y:%.*]], 25
; CHECK-NEXT: ret i1 [[R]]
;
%lshr = lshr i32 268435456, %y
@ -208,7 +199,31 @@ define i1 @scalar_i32_lshr_and_negC_eq_X_is_constant2(i32 %y) {
%r = icmp eq i32 %and, 0
ret i1 %r
}
define i1 @scalar_i32_udiv_and_negC_eq_X_is_constant3(i32 %y) {
; CHECK-LABEL: @scalar_i32_udiv_and_negC_eq_X_is_constant3(
; CHECK-NEXT: [[R:%.*]] = icmp ult i32 [[Y:%.*]], 1544
; CHECK-NEXT: ret i1 [[R]]
;
%lshr = udiv i32 12345, %y
%and = and i32 %lshr, 16376 ; 0x3ff8
%r = icmp ne i32 %and, 0
ret i1 %r
}
; Negative test
define i1 @scalar_i32_lshr_and_negC_eq_X_is_constant_negtive(i32 %y) {
; CHECK-LABEL: @scalar_i32_lshr_and_negC_eq_X_is_constant_negtive(
; CHECK-NEXT: [[LSHR:%.*]] = lshr i32 16384, [[Y:%.*]]
; CHECK-NEXT: [[AND:%.*]] = and i32 [[LSHR]], 16376
; CHECK-NEXT: [[R:%.*]] = icmp eq i32 [[AND]], 0
; CHECK-NEXT: ret i1 [[R]]
;
%lshr = lshr i32 16384, %y ; 0x4000
%and = and i32 %lshr, 16376 ; 0x3ff8
%r = icmp eq i32 %and, 0
ret i1 %r
}
; Check 'slt' predicate
define i1 @scalar_i32_lshr_and_negC_slt(i32 %x, i32 %y) {