24.08.2010 15:02, Stefan Hajnoczi wrote: [] > Sometimes an improvement has a side effect and it makes sense to hold > back the improvement until the side effect can be resolved. The > period of time in which users could rely on qcow2 data integrity is > small to none, I feel reverting the commit makes sense.
And I agree with this. > QEMU 0.12.5 has qcow2 sync metadata writes in commit > 37060c28e522843fbf6f7e59af745dfcb05b132c. Was the performance > regression spotted on 0.12.5 or 0.13? It started as a debian bugreport ( http://bugs.debian.org/594069 ) which were reported once 0.12.5 entered debian testing. So it were spotted in 0.12.5, not 0.13. I verfied the bug locally and indeed, it makes _huge_ difference (0.12.4 vs 0.12.5), -- e.g 600mb mem windows7 hybernation on my machine now takes about 40 minutes to complete instead of ~20 sec in 0.12.4. I'm reverting whole series (there are 5 patches in total, covering qcow, qcow2, vmdk and vpc plus infrastructure) for debian 0.12 package. It is sad but there's no other option for now. /mjt
