On Sat, Aug 04, 2007 at 04:07:14PM +0100, Colin Watson wrote: > >I noticed this on the Ubuntu master CD building system after it was >upgraded to Ubuntu 6.10, whose cdrtools package is based on cdrtools >2.01+01a03-5. The problem is that the code that constructs linked lists >for JTE includes and excludes doesn't always initialise the next pointer >in those linked lists. The patch is simple, and is attached to this >mail. > >> I don't know if this is reproducible while building standard Debian images, >> but >> it's quite possible. Steve mentionned that CD images are built on a sarge >> system. If that's the case, then this bug might need to be upgraded to >> serious >> as it's really release-critical to be able to build etch CD images on a etch >> system. > >The buggy code doesn't look like it's been changed since its cdrtools >incarnation, so I expect this bug is present in sarge too; but of course >it's dependent on stack layout and so on. valgrind should show it up >pretty easily: > > mkdir -p /tmp/mkisofs-test > cd /tmp/mkisofs-test > mkdir -p t/pool > dd if=/dev/zero of=t/file bs=1024 count=100 > touch md5 > valgrind genisoimage -o t.iso -jigdo-jigdo t.jigdo -jigdo-template > t.template -md5-list md5 -jigdo-force-md5 /pool/ t > >... producing output including an error similar to the following: > > ==8000== Conditional jump or move depends on uninitialised value(s) > ==8000== at 0x42D2F8: list_file_in_jigdo (in /usr/bin/genisoimage) > ==8000== by 0x42A2CF: (within /usr/bin/genisoimage) > ==8000== by 0x42A77C: (within /usr/bin/genisoimage) > ==8000== by 0x40D5A5: main (in /usr/bin/genisoimage) > >FWIW, I plan to issue stable release updates for Ubuntu to fix this.
Cool, thanks! -- Steve McIntyre, Cambridge, UK. [EMAIL PROTECTED] You raise the blade, you make the change... You re-arrange me 'til I'm sane... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
