maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Analysis/MemorySSA/pr43438.ll

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

102 lines
3.3 KiB
LLVM
Raw Normal View History

[MSSA][NewPM] Handle tests with -print-memoryssa -print-memoryssa in legacy PM is print<memoryssa> in NPM. Pin tests with -print-memoryssa to legacy PM. Add corresponding tests for NPM where missing. This fixes "unknown pass name 'print-memoryssa'". Some tests still fail in Analysis/MemorySSA due to other passes that haven't been ported. pr43427.ll and pr43438.ll required adding -aa-pipeline=basic-aa, -loop-simplify (since it doesn't run on legacy PM by default), and decrementing some of the MemoryPhi numbers. Reviewed By: asbirlea Differential Revision: https://reviews.llvm.org/D85333
2020-08-06 02:12:29 +08:00
; RUN: opt -disable-output -loop-simplify -licm -enable-new-pm=0 -print-memoryssa -enable-mssa-loop-dependency=true < %s 2>&1 | FileCheck %s
; RUN: opt -disable-output -aa-pipeline=basic-aa -passes='loop-mssa(licm),print<memoryssa>' < %s 2>&1 | FileCheck %s
[MemorySSA] Avoid adding Phis in the presence of unreachable blocks. Summary: If a block has all incoming values with the same MemoryAccess (ignoring incoming values from unreachable blocks), then use that incoming MemoryAccess and do not create a Phi in the first place. Revert IDF work-around added in rL372673; it should not be required unless the Def inserted is the first in its block. The patch also cleans up a series of tests, added during the many iterations on insertDef. The patch also fixes PR43438. The same issue that occurs in insertDef with "adding phis, hence the IDF of Phis is needed", can also occur in fixupDefs: the `getPreviousRecursive` call only adds Phis walking on the predecessor edges, which means there may be the case of a Phi added walking the CFG "backwards" which triggers the needs for an additional Phi in successor blocks. Such Phis are added during fixupDefs only in the presence of unreachable blocks. Hence this highlights the need to avoid adding Phis in blocks with unreachable predecessors in the first place. Reviewers: george.burgess.iv Subscribers: Prazek, sanjoy.google, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D67995 llvm-svn: 372932
2019-09-26 07:24:39 +08:00
target triple = "x86_64-unknown-linux-gnu"
; CHECK-LABEL: @main()
; CHECK: 5 = MemoryPhi(
; CHECK-NOT: 7 = MemoryPhi(
@v_67 = external dso_local global i32, align 1
@v_76 = external dso_local global i16, align 1
@v_86 = external dso_local global i16 *, align 1
define dso_local void @main() {
entry:
%v_59 = alloca i16, align 2
br label %for.cond
for.cond: ; preds = %for.body, %entry
br i1 undef, label %for.body, label %for.end
for.body: ; preds = %for.cond
store i16 undef, i16* %v_59, align 2
br label %for.cond
for.end: ; preds = %for.cond
br i1 undef, label %if.else568, label %cond.end82
cond.false69: ; No predecessors!
br label %cond.end82
cond.end82: ; preds = %cond.false69, %cond.true55
br i1 undef, label %if.else568, label %land.lhs.true87
land.lhs.true87: ; preds = %cond.end82
br i1 undef, label %if.then88, label %if.else568
if.then88: ; preds = %land.lhs.true87
store i16 * @v_76, i16 ** @v_86, align 1
br label %if.end569
if.else568: ; preds = %land.lhs.true87, %cond.end82, %for.end
store volatile i32 undef, i32 * @v_67, align 1
br label %if.end569
if.end569: ; preds = %if.else568, %if.then88
ret void
}
[MemorySSA] Update last_access_in_block check. The check for "was there an access in this block" should be: is the last access in this block and is it not a newly inserted phi. Resolves new test in PR43438. Also fix a typo when simplifying trivial Phis to match the comment. llvm-svn: 373380
2019-10-02 02:34:39 +08:00
; CHECK-LABEL: @f()
; CHECK: 7 = MemoryPhi(
[MSSA][NewPM] Handle tests with -print-memoryssa -print-memoryssa in legacy PM is print<memoryssa> in NPM. Pin tests with -print-memoryssa to legacy PM. Add corresponding tests for NPM where missing. This fixes "unknown pass name 'print-memoryssa'". Some tests still fail in Analysis/MemorySSA due to other passes that haven't been ported. pr43427.ll and pr43438.ll required adding -aa-pipeline=basic-aa, -loop-simplify (since it doesn't run on legacy PM by default), and decrementing some of the MemoryPhi numbers. Reviewed By: asbirlea Differential Revision: https://reviews.llvm.org/D85333
2020-08-06 02:12:29 +08:00
; CHECK: 6 = MemoryPhi(
[MemorySSA] Update last_access_in_block check. The check for "was there an access in this block" should be: is the last access in this block and is it not a newly inserted phi. Resolves new test in PR43438. Also fix a typo when simplifying trivial Phis to match the comment. llvm-svn: 373380
2019-10-02 02:34:39 +08:00
; CHECK: 10 = MemoryPhi(
; CHECK: 9 = MemoryPhi(
[MSSA][NewPM] Handle tests with -print-memoryssa -print-memoryssa in legacy PM is print<memoryssa> in NPM. Pin tests with -print-memoryssa to legacy PM. Add corresponding tests for NPM where missing. This fixes "unknown pass name 'print-memoryssa'". Some tests still fail in Analysis/MemorySSA due to other passes that haven't been ported. pr43427.ll and pr43438.ll required adding -aa-pipeline=basic-aa, -loop-simplify (since it doesn't run on legacy PM by default), and decrementing some of the MemoryPhi numbers. Reviewed By: asbirlea Differential Revision: https://reviews.llvm.org/D85333
2020-08-06 02:12:29 +08:00
; CHECK: 8 = MemoryPhi(
[MemorySSA] Update last_access_in_block check. The check for "was there an access in this block" should be: is the last access in this block and is it not a newly inserted phi. Resolves new test in PR43438. Also fix a typo when simplifying trivial Phis to match the comment. llvm-svn: 373380
2019-10-02 02:34:39 +08:00
define void @f() {
entry:
%e = alloca i16, align 1
br label %lbl1
lbl1: ; preds = %if.else, %for.end5, %entry
store i16 undef, i16* %e, align 1
%0 = load i16, i16* %e, align 1
%call = call i16 @g(i16 %0)
br i1 undef, label %for.end, label %if.else
for.end: ; preds = %if.then
br i1 true, label %for.cond2, label %lbl2
lbl2: ; preds = %for.body4, %if.end
br label %for.cond2
for.cond2: ; preds = %lbl3
br i1 undef, label %for.body4, label %for.end5
for.body4: ; preds = %for.cond2
br label %lbl2
for.end5: ; preds = %for.cond2
switch i32 undef, label %unreachable [
i32 0, label %if.end12
i32 2, label %lbl1
]
if.else: ; preds = %lbl1
switch i32 undef, label %unreachable [
i32 0, label %if.end12
i32 2, label %lbl1
]
if.end12: ; preds = %cleanup.cont11s, %cleanup.cont
call void @llvm.lifetime.end.p0i8(i64 1, i8* undef)
ret void
unreachable: ; preds = %if.else, %for.end5
unreachable
}
declare i16 @g(i16)
; Function Attrs: argmemonly nounwind willreturn
declare void @llvm.lifetime.end.p0i8(i64 immarg, i8* nocapture)