On 11/10/2011 04:43 PM, Jim Fehlig wrote: >>> In the 'virsh save --bypass-cache' case, we pass a pipe instead of a >>> regular fd, but using a pipe adds I/O overhead, so always passing a >>> pipe just so qemu can see EAGAIN doesn't seem nice. >>> >>> The ultimate fix needs to come from qemu - background migration must >>> >> >> ACK, it's a bit of a shame, but until we have a fix, it's more >> important to be able to abort a save to disk than to be able to save >> at a speed of more than 32MB/s (if the storage system support this) >> > > Agreed. > > BTW, I don't have access to the bug to see details, but I think I > understand given Eric's description. I've gone ahead and pushed my quick fix, but hopefully this isn't the end of the story (see Dan's thread: https://www.redhat.com/archives/libvir-list/2011-November/msg00572.html). -- Eric Blake eblake@xxxxxxxxxx +1-801-349-2682 Libvirt virtualization library http://libvirt.org
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list