8b5de48d62
Summary: Previously, in the NewPM pipeline, TailCallElim recalculates the DomTree when it modifies any instruction in the Function. For example, ``` CallInst *CI = dyn_cast<CallInst>(&I); ... CI->setTailCall(); Modified = true; ... if (!Modified || ...) return PreservedAnalyses::all(); ``` After applying this patch, the DomTree only recalculates if needed (plus an extra insertEdge() + an extra deleteEdge() call). When optimizing SQLite with `-passes="default<O3>"` pipeline of the newPM, the number of DomTree recalculation decreases by 6.2%, the number of nodes visited by DFS decreases by 2.9%. The time used by DomTree will decrease approximately 1%~2.5% after applying the patch. Statistics: ``` Before the patch: 23010 dom-tree-stats - Number of DomTree recalculations 489264 dom-tree-stats - Number of nodes visited by DFS -- DomTree After the patch: 21581 dom-tree-stats - Number of DomTree recalculations 475088 dom-tree-stats - Number of nodes visited by DFS -- DomTree ``` Reviewers: kuhar, dmgreen, brzycki, grosser, davide Reviewed By: kuhar, brzycki Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D49982 llvm-svn: 338954
55 lines
1.2 KiB
LLVM
55 lines
1.2 KiB
LLVM
; RUN: opt < %s -tailcallelim -verify-dom-info -S | FileCheck %s
|
|
|
|
; Don't turn this into an infinite loop, this is probably the implementation
|
|
; of fabs and we expect the codegen to lower fabs.
|
|
; CHECK: @fabs(double %f)
|
|
; CHECK: call
|
|
; CHECK: ret
|
|
|
|
define double @fabs(double %f) {
|
|
entry:
|
|
%tmp2 = call double @fabs( double %f ) ; <double> [#uses=1]
|
|
ret double %tmp2
|
|
}
|
|
|
|
; Do turn other calls into infinite loops though.
|
|
|
|
; CHECK-LABEL: define double @foo(
|
|
; CHECK-NOT: call
|
|
; CHECK: }
|
|
define double @foo(double %f) {
|
|
%t= call double @foo(double %f)
|
|
ret double %t
|
|
}
|
|
|
|
; CHECK-LABEL: define float @fabsf(
|
|
; CHECK-NOT: call
|
|
; CHECK: }
|
|
define float @fabsf(float %f) {
|
|
%t= call float @fabsf(float 2.0)
|
|
ret float %t
|
|
}
|
|
|
|
declare x86_fp80 @fabsl(x86_fp80 %f)
|
|
|
|
; Don't crash while transforming a function with infinite recursion.
|
|
define i32 @PR22704(i1 %bool) {
|
|
entry:
|
|
br i1 %bool, label %t, label %f
|
|
|
|
t:
|
|
%call1 = call i32 @PR22704(i1 1)
|
|
br label %return
|
|
|
|
f:
|
|
%call = call i32 @PR22704(i1 1)
|
|
br label %return
|
|
|
|
return:
|
|
ret i32 0
|
|
|
|
; CHECK-LABEL: @PR22704(
|
|
; CHECK: %bool.tr = phi i1 [ %bool, %entry ], [ true, %t ], [ true, %f ]
|
|
; CHECK: br i1 %bool.tr, label %t, label %f
|
|
}
|