On 06/01/2015 03:26 PM, Michael S. Tsirkin wrote:
> On Mon, Jun 01, 2015 at 01:53:35PM +0800, Jason Wang wrote:
>> >
>> >
>> > On 06/01/2015 02:12 AM, Michael S. Tsirkin wrote:
>>> > > On Tue, May 26, 2015 at 01:54:09AM -0400, Jason Wang wrote:
> >> We allocate one vhost log per vhost devi
On Mon, Jun 01, 2015 at 01:53:35PM +0800, Jason Wang wrote:
>
>
> On 06/01/2015 02:12 AM, Michael S. Tsirkin wrote:
> > On Tue, May 26, 2015 at 01:54:09AM -0400, Jason Wang wrote:
> >> We allocate one vhost log per vhost device. This is sub optimal when:
> >>
> >> - Guest has several device with
On 06/01/2015 02:12 AM, Michael S. Tsirkin wrote:
> On Tue, May 26, 2015 at 01:54:09AM -0400, Jason Wang wrote:
>> We allocate one vhost log per vhost device. This is sub optimal when:
>>
>> - Guest has several device with vhost as backend
>> - Guest has multiqueue devices
>>
>> Since each vhost
On Tue, May 26, 2015 at 01:54:09AM -0400, Jason Wang wrote:
> We allocate one vhost log per vhost device. This is sub optimal when:
>
> - Guest has several device with vhost as backend
> - Guest has multiqueue devices
>
> Since each vhost device will allocate and use their private log, this
> cou
We allocate one vhost log per vhost device. This is sub optimal when:
- Guest has several device with vhost as backend
- Guest has multiqueue devices
Since each vhost device will allocate and use their private log, this
could cause very huge unnecessary memory allocation. We can in fact
avoid ext
