[llvm-bugs] [Bug 28888] New: Assertion `Headers.size() >= 2 && "Expected irreducible CFG; -loop-info is likely invalid"' failed.

via llvm-bugs llvm-bugs at lists.llvm.org
Sat Aug 6 22:29:21 PDT 2016


https://llvm.org/bugs/show_bug.cgi?id=28888

            Bug ID: 28888
           Summary: Assertion `Headers.size() >= 2 && "Expected
                    irreducible CFG; -loop-info is likely invalid"'
                    failed.
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Core LLVM classes
          Assignee: unassignedbugs at nondot.org
          Reporter: chisophugis at gmail.com
                CC: dberlin at dberlin.org, llvm-bugs at lists.llvm.org,
                    michael.v.zolotukhin at gmail.com
            Blocks: 28315
    Classification: Unclassified

Created attachment 16905
  --> https://llvm.org/bugs/attachment.cgi?id=16905&action=edit
Failing test case.

Stacktrace here: https://reviews.llvm.org/P7179

opt
'-passes=require<targetir>,require<assumptions>,require<scalar-evolution>,require<opt-remark-emit>,loop(unroll),require<block-freq>'
-disable-output $1

(
Does not reproduce with old PM (loop-vectorize was the thing that used
block-freq):
opt -loop-unroll -loop-vectorize -disable-output $1
)

The test case is still quite large, but it seems like the best that bugpoint
can currently do. DannyB, this might be a useful test case for any new magic in
bugpoint.

Michael Z., if you have a moment, could you look at this?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20160807/baacfe88/attachment.html>


More information about the llvm-bugs mailing list