Am 17.10.2011 17:47, schrieb Stefan Hajnoczi: > When copy-on-read is enabled it is necessary to wait for overlapping > requests before issuing new requests. This prevents races between the > copy-on-read and a write request. > > Signed-off-by: Stefan Hajnoczi <stefa...@linux.vnet.ibm.com>
This doesn't only order guest request against COR requests, but also makes guest requests wait on each other. It's probably not a big problem, but if we had to optimise performance with COR later, this is something to remember. Doing an optimisation that only requests of different type are ordered wouldn't be too hard, though maybe avoiding starvation of the other type could get a bit harder. Let's leave it as it is for now. Kevin
