>>> On 01.06.15 at 17:26, wrote:
> --- a/xen/common/hvm/save.c
> +++ b/xen/common/hvm/save.c
This being an x86-specific problem I would think this would better be
addressed in x86-specific code. If it was to stay here, a proper
arch hook should be introduced (even if ARM isn't using this code
rig
An MTRR record is processed for each vCPU during hvm_load. Each MTRR
record sets several mtrrs, each of which flushes the cache on all pCPUs.
This can take some time and trip the watchdog for HVM guests with many
CPUs.
To fix this, introduce a flag which prevents flushing the cache when
doing MTRR
