See 32-bit arithmetic on 'position' in the seek_to_sector() function in block/parallels.c No wonder it gives me crap. Making 'position' a uint64_t and ensuring that the arithmetic on it is also suitably cast to uint64_t gives me a more reasonable converted image.
Fix is committed upstream now http://repo.or.cz/w/qemu.git/commitdiff/c34d2451ed dwmw2: do you need cherry-picked into F-12, or are you happy to mark this as CLOSED UPSTREAM ?
Turns out I can't get our Parallels images (of Windows XP) to boot in qemu/KVM anyway -- I can't seem to solve the problems described at http://www.virtualbox.org/wiki/Migrate_Windows So for now I'm not advising people to convert our existing Parallels images -- if I do get it working, I'd probably have to give them a 'fixed' image, which might as well be in qcow2 format already. So I'm not personally desperate to get it, but it _would_ be nice to drop it into the next update if it's not too much trouble -- a conversion tool which can only handle images smaller than 4GiB isn't really something we should be shipping these days :)
Building this now: * Fri Nov 20 2009 Mark McLoughlin <markmc> - 2:0.11.0-12 - Fix a use-after-free crasher in the slirp code (#539583) - Fix overflow in the parallels image format support (#533573)
qemu-0.11.0-12.fc12 has been submitted as an update for Fedora 12. http://admin.fedoraproject.org/updates/qemu-0.11.0-12.fc12
qemu-0.11.0-12.fc12 has been pushed to the Fedora 12 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update qemu'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F12/FEDORA-2009-11901
qemu-0.11.0-12.fc12 has been pushed to the Fedora 12 stable repository. If problems still persist, please make note of it in this bug report.