http://llvm.org/bugs/show_bug.cgi?id=18955
Bug ID: 18955
Summary: Infinite loop in AlwaysInliner pass
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: Interprocedural Optimizations
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
Classification: Unclassified
Attached is a tarball of two LLVM code modules which are nearly identical. When
running
opt -O2 -o bad.bc bad.ll
opt rapidly consumes GiB's of memory and seems to run indefinitely (>10min).
When running
opt -O2 -o good.bc good.ll
opt exits quickly. The difference between the two modules is 599 removals of
the alwaysinline function attribute; therefore I believe the problem is in the
AlwaysInliner pass. I also wrote code to only run the AlwaysInliner pass, and
that produced the same result.
I have also experimentally observed this bug in the LLVM 3.4 and LLVM 3.0
releases.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
LLVMbugs mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
