maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/polly/test/ScopDetect/expand-region-correctly-2.ll

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

54 lines
2.1 KiB
LLVM
Raw Normal View History

tests: Add test cases for LLVM commit r251267 This fixes llvm.org/PR25242 llvm-svn: 251268
2015-10-26 06:56:42 +08:00
; RUN: opt %loadPolly -polly-detect \
; RUN: -analyze < %s | FileCheck %s
[FIX] Normalize loops outside the SCoP during schedule generation We build the schedule based on a traversal of the region and accumulate information for each loop in it. The total schedule is associated with the loop surrounding the SCoP, though it can happen that there are blocks in the SCoP which are part of loops that are only partially in the SCoP. Instead of associating information with them (they are not part of the SCoP and consequently are not modeled) we have to associate the schedule information with the surrounding loop if any. This fixes bug 25240. llvm-svn: 250668
2015-10-19 05:17:11 +08:00
;
tests: Add test cases for LLVM commit r251267 This fixes llvm.org/PR25242 llvm-svn: 251268
2015-10-26 06:56:42 +08:00
; CHECK: Valid Region for Scop: if.end.1631 => for.cond.1647.outer
[FIX] Normalize loops outside the SCoP during schedule generation We build the schedule based on a traversal of the region and accumulate information for each loop in it. The total schedule is associated with the loop surrounding the SCoP, though it can happen that there are blocks in the SCoP which are part of loops that are only partially in the SCoP. Instead of associating information with them (they are not part of the SCoP and consequently are not modeled) we have to associate the schedule information with the surrounding loop if any. This fixes bug 25240. llvm-svn: 250668
2015-10-19 05:17:11 +08:00
;
target triple = "x86_64-unknown-linux-gnu"
; Function Attrs: nounwind uwtable
define void @qtm_decompress() #0 {
entry:
br label %if.end.1631
if.end.1631: ; preds = %entry
br i1 false, label %for.cond.1647.preheader, label %if.then.1635
if.then.1635: ; preds = %if.end.1631
br label %for.cond.1647.preheader
for.cond.1647.preheader: ; preds = %if.then.1635, %if.end.1631
br label %for.cond.1647.outer
for.cond.1647.outer: ; preds = %do.end.1685, %for.cond.1647.preheader
%bits_needed.5.ph = phi i8 [ 8, %for.cond.1647.preheader ], [ undef, %do.end.1685 ]
br label %for.cond.1647
for.cond.1647: ; preds = %do.cond.1718, %for.cond.1647.outer
%bits_needed.5 = phi i8 [ 8, %do.cond.1718 ], [ %bits_needed.5.ph, %for.cond.1647.outer ]
br i1 undef, label %do.cond.1718, label %if.then.1659
if.then.1659: ; preds = %for.cond.1647
br i1 false, label %do.end.1685, label %cleanup.1785
do.end.1685: ; preds = %if.then.1659
br label %for.cond.1647.outer
do.cond.1718: ; preds = %for.cond.1647
br i1 false, label %land.lhs.true.1736, label %for.cond.1647
land.lhs.true.1736: ; preds = %do.cond.1718
br label %if.then.1742
if.then.1742: ; preds = %land.lhs.true.1736
unreachable
cleanup.1785: ; preds = %if.then.1659
ret void
}
Revert "[NFC] remove explicit default value for strboolattr attribute in tests" This reverts commit bda6e5bee04c75b1f1332b4fd1ac4e8ef6c3c247. See https://lab.llvm.org/buildbot/#/builders/109/builds/15424 for instance
2021-05-25 01:43:40 +08:00
attributes #0 = { nounwind uwtable "disable-tail-calls"="false" "less-precise-fpmad"="false" "frame-pointer"="none" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+mmx,+sse,+sse2" "unsafe-fp-math"="false" "use-soft-float"="false" }
[FIX] Normalize loops outside the SCoP during schedule generation We build the schedule based on a traversal of the region and accumulate information for each loop in it. The total schedule is associated with the loop surrounding the SCoP, though it can happen that there are blocks in the SCoP which are part of loops that are only partially in the SCoP. Instead of associating information with them (they are not part of the SCoP and consequently are not modeled) we have to associate the schedule information with the surrounding loop if any. This fixes bug 25240. llvm-svn: 250668
2015-10-19 05:17:11 +08:00
!llvm.ident = !{!0}
!0 = !{!"clang version 3.8.0 (trunk 250010) (llvm/trunk 250018)"}