[llvm-branch-commits] [llvm-branch] r296156 - Merging r296003:
Hans Wennborg via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Fri Feb 24 10:53:54 PST 2017
Author: hans
Date: Fri Feb 24 12:53:54 2017
New Revision: 296156
URL: http://llvm.org/viewvc/llvm-project?rev=296156&view=rev
Log:
Merging r296003:
------------------------------------------------------------------------
r296003 | mcrosier | 2017-02-23 10:49:03 -0800 (Thu, 23 Feb 2017) | 32 lines
[Reassociate] Add negated value of negative constant to the Duplicates list.
In OptimizeAdd, we scan the operand list to see if there are any common factors
between operands that can be factored out to reduce the number of multiplies
(e.g., 'A*A+A*B*C+D' -> 'A*(A+B*C)+D'). For each operand of the operand list, we
only consider unique factors (which is tracked by the Duplicate set). Now if we
find a factor that is a negative constant, we add the negated value as a factor
as well, because we can percolate the negate out. However, we mistakenly don't
add this negated constant to the Duplicates set.
Consider the expression A*2*-2 + B. Obviously, nothing to factor.
For the added value A*2*-2 we over count 2 as a factor without this change,
which causes the assert reported in PR30256. The problem is that this code is
assuming that all the multiply operands of the add are already reassociated.
This change avoids the issue by making OptimizeAdd tolerate multiplies which
haven't been completely optimized; this sort of works, but we're doing wasted
work: we'll end up revisiting the add later anyway.
Another possible approach would be to enforce RPO iteration order more strongly.
If we have RedoInsts, we process them immediately in RPO order, rather than
waiting until we've finished processing the whole function. Intuitively, it
seems like the natural approach: reassociation works on expression trees, so
the optimization only works in one direction. That said, I'm not sure how
practical that is given the current Reassociate; the "optimal" form for an
expression depends on its use list (see all the uses of "user_back()"), so
Reassociate is really an iterative optimization of sorts, so any changes here
would probably get messy.
PR30256
Differential Revision: https://reviews.llvm.org/D30228
------------------------------------------------------------------------
Modified:
llvm/branches/release_40/ (props changed)
llvm/branches/release_40/lib/Transforms/Scalar/Reassociate.cpp
llvm/branches/release_40/test/Transforms/Reassociate/basictest.ll
Propchange: llvm/branches/release_40/
------------------------------------------------------------------------------
--- svn:mergeinfo (original)
+++ svn:mergeinfo Fri Feb 24 12:53:54 2017
@@ -1,3 +1,3 @@
/llvm/branches/Apple/Pertwee:110850,110961
/llvm/branches/type-system-rewrite:133420-134817
-/llvm/trunk:155241,291858-291859,291863,291875,291909,291918,291966,291968,291979,292117,292133,292167,292169-292170,292242,292254-292255,292280,292323,292444,292467,292516,292583,292624-292625,292641,292651,292667,292711-292713,292758,292949,293017,293021,293025,293124,293230,293259,293273,293291,293293,293309,293345,293417,293522,293542,293629,293635,293658,293673,293727,293730,294003,294102,294129,294203,294267,294318,294348-294349,294357,294527,294551,294982,295018,295213,295215,295230,295486,295490,295512,295762,296030
+/llvm/trunk:155241,291858-291859,291863,291875,291909,291918,291966,291968,291979,292117,292133,292167,292169-292170,292242,292254-292255,292280,292323,292444,292467,292516,292583,292624-292625,292641,292651,292667,292711-292713,292758,292949,293017,293021,293025,293124,293230,293259,293273,293291,293293,293309,293345,293417,293522,293542,293629,293635,293658,293673,293727,293730,294003,294102,294129,294203,294267,294318,294348-294349,294357,294527,294551,294982,295018,295213,295215,295230,295486,295490,295512,295762,296003,296030
Modified: llvm/branches/release_40/lib/Transforms/Scalar/Reassociate.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/branches/release_40/lib/Transforms/Scalar/Reassociate.cpp?rev=296156&r1=296155&r2=296156&view=diff
==============================================================================
--- llvm/branches/release_40/lib/Transforms/Scalar/Reassociate.cpp (original)
+++ llvm/branches/release_40/lib/Transforms/Scalar/Reassociate.cpp Fri Feb 24 12:53:54 2017
@@ -1521,8 +1521,8 @@ Value *ReassociatePass::OptimizeAdd(Inst
if (ConstantInt *CI = dyn_cast<ConstantInt>(Factor)) {
if (CI->isNegative() && !CI->isMinValue(true)) {
Factor = ConstantInt::get(CI->getContext(), -CI->getValue());
- assert(!Duplicates.count(Factor) &&
- "Shouldn't have two constant factors, missed a canonicalize");
+ if (!Duplicates.insert(Factor).second)
+ continue;
unsigned Occ = ++FactorOccurrences[Factor];
if (Occ > MaxOcc) {
MaxOcc = Occ;
@@ -1534,8 +1534,8 @@ Value *ReassociatePass::OptimizeAdd(Inst
APFloat F(CF->getValueAPF());
F.changeSign();
Factor = ConstantFP::get(CF->getContext(), F);
- assert(!Duplicates.count(Factor) &&
- "Shouldn't have two constant factors, missed a canonicalize");
+ if (!Duplicates.insert(Factor).second)
+ continue;
unsigned Occ = ++FactorOccurrences[Factor];
if (Occ > MaxOcc) {
MaxOcc = Occ;
Modified: llvm/branches/release_40/test/Transforms/Reassociate/basictest.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/branches/release_40/test/Transforms/Reassociate/basictest.ll?rev=296156&r1=296155&r2=296156&view=diff
==============================================================================
--- llvm/branches/release_40/test/Transforms/Reassociate/basictest.ll (original)
+++ llvm/branches/release_40/test/Transforms/Reassociate/basictest.ll Fri Feb 24 12:53:54 2017
@@ -222,3 +222,23 @@ define i32 @test15(i32 %X1, i32 %X2, i32
; CHECK-LABEL: @test15
; CHECK: and i1 %A, %B
}
+
+; PR30256 - previously this asserted.
+; CHECK-LABEL: @test16
+; CHECK: %[[FACTOR:.*]] = mul i64 %a, -4
+; CHECK-NEXT: %[[RES:.*]] = add i64 %[[FACTOR]], %b
+; CHECK-NEXT: ret i64 %[[RES]]
+define i64 @test16(i1 %cmp, i64 %a, i64 %b) {
+entry:
+ %shl = shl i64 %a, 1
+ %shl.neg = sub i64 0, %shl
+ br i1 %cmp, label %if.then, label %if.end
+
+if.then: ; preds = %entry
+ %add1 = add i64 %shl.neg, %shl.neg
+ %add2 = add i64 %add1, %b
+ ret i64 %add2
+
+if.end: ; preds = %entry
+ ret i64 0
+}
More information about the llvm-branch-commits
mailing list