Red Hat Bugzilla – Bug 191948
gdb segfaults on ia64
Last modified: 2007-11-30 17:11:32 EST
Description of problem:
gdb on rawhide ia64 hits a segmentation fault when doing any debugging. Simply
running a simple hello world program through it will hit this.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. compile a hello world with -g
2. gdb a.out
A little more info...
I also tried compiling with -static with the same results. Also tried compiling
both normal and static on a RHEL4 system and running on rawhide with the same
Works (sort of) (on a slightly older?) RHEL4 installation.
$ gdb --version
GNU gdb Red Hat Linux (126.96.36.199-0.31rh)
$ rpm -q -f /usr/bin/gdb
I do see a "Failed to read a valid object file image from memory." on startup
of gdb, and when I type "run", but no SIGSEGV, and things otherwise seem to be
working (can set breakpoints, gdb stops at them and prints reasonable location
information). This was with a kernel.org kernel. Running with RHEL4 kernel
2.6.9-34.EL, I see similar results ... gdb works, but there are some warning
messages (e.g. "warning: cannot close "shared object read from target memory":
File in wrong format").
Tony's note above gave me the idea to try the RHEL4 gdb on rawhide. I removed
gdb and installed gdb-188.8.131.52-1.96.ia64.rpm from the RHEL4 tree. This version
of gdb works with no problem. I don't see the error that Tony mentions here.
Created attachment 129253 [details]
stack trace of gdb segvault
This was created by setting "ulimit -c" so that gdb would create a core file
then using the RHEL4 gdb to read that core and get a stack trace.
FYI, gdb-184.108.40.206-1.122 shipped with the ia64 FC5 ISO DVD image posted by Prarit
on oss.sgi.com is running fine too.
Thanks for the report, this should be fixed in tomorrow's rawhide,
I confirm that today's Rawhide gdb-220.127.116.11-1.129.FC6 is working again. Many
thanks for this quick fix.
gdb-18.104.22.168-1.134.fc5 has been pushed for fc5, which should resolve this issue. If these problems are still present in this version, then please make note of it in this bug report.