On Fri, Apr 13, 2012 at 5:49 PM, Paolo Bonzini <pbonz...@redhat.com> wrote: > Il 13/04/2012 09:51, Zhi Yong Wu ha scritto: >> Sorry, i hadn't check the code. thanks for pointing out this. In the >> past, i tested this patch, and found it will assert about >> bs->tracked_requests. Do you think that only qemu_aio_wait can make >> sure that bs->tracked_requests is empty here? > > FWIW, live snapshot during guest boot (block_set_io_throttle ide0-hd0 0 > 500000 500000 0 500000 500000) died immediately without the patches. > With the patch I could do 20 snapshots before I stopped. Why do you not use qemu_aio_flush before loop start? it can fix the issue said by me. > > Paolo
-- Regards, Zhi Yong Wu
