On 01/12/2015 16:25, Eduardo Habkost wrote: > > I think it's easier to use small guests (i.e. kvm-unit-tests) to test > > this code. > > I agree it's easier, but how likely it is to catch bugs in the > save/load code? If the code corrupts a register, we need to > trigger a save/load cycle at the exact moment the guest code is > using that register. Do we have something that helps us > repeatedly save/load CPU state while kvm-unit-tests is running? A vmware magic port read should do that. Put VMPORT_MAGIC in EAX and VMPORT_CMD_GETVERSION in ECX, then do a 32-bit in from port 0x5658. Paolo -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
