[LLVMbugs] [Bug 2134] New: llc -relocation-model=pic: hangs while generating code for glibc-2.7/posix/wordexp.c

[bugzilla-daemon at cs.uiuc.edu]

http://llvm.org/bugs/show_bug.cgi?id=2134

           Summary: llc -relocation-model=pic: hangs while generating code
                    for glibc-2.7/posix/wordexp.c
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=1513)
 --> (http://llvm.org/bugs/attachment.cgi?id=1513)
bugpoint reduced testcase

Using SVN r48109 llvm, and llvm-gcc4.2 on x86 compilation hangs while building
glibc-2.7/posix/wordexp.c. 
I narrowed this down to be a problem in llc.

It was tricky to get bugpoint to reduce the testcase, since I was dealing with
a hang and not a crash, so I put an alarm(1) in llc.cpp, and I got bugpoint
reducing to a small testcase.

To reproduce:
$ llc bugpoint-reduced-simplified.bc -f -relocation-model=pic

gdb shows this (sorry this is a release build, without symbols):

(gdb) r bugpoint-reduced-simplified.bc -f -relocation-model=pic
Starting program: /home/edwin/llvm/llvm-svn/trunk/Release/bin/llc
bugpoint-reduced-simplified.bc -f -relocation-model=pic
[Thread debugging using libthread_db enabled]
^C[New Thread 0xb7d586c0 (LWP 11110)]

Program received signal SIGINT, Interrupt.
[Switching to Thread 0xb7d586c0 (LWP 11110)]
0x0864fe1f in (anonymous namespace)::RALinScan::assignRegOrStackSlotAtInterval
()
(gdb) bt
#0  0x0864fe1f in (anonymous
namespace)::RALinScan::assignRegOrStackSlotAtInterval ()
#1  0x08f54ba0 in ?? ()
#2  0x00000001 in ?? ()
#3  0xb801eff4 in ?? () from /lib/ld-linux.so.2
#4  0x7c9c61e1 in ?? ()
#5  0xb801f820 in ?? ()
#6  0xbfc1ecd0 in ?? ()
#7  0xb800d5bf in ?? () from /lib/ld-linux.so.2
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
(gdb) c

I will follow-up when the debug build is complete with a better stacktrace.

I am attaching the bugpoint reduced testcase, and the preprocessed C source
too.


-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.