Re: [Qemu-devel] [PATCH] migration: use XBZRLE only after bulk stage

[Peter Lieven <pl@dlhnet.de>]

On 01.03.2013 14:52, Eric Blake wrote:
> On 03/01/2013 06:31 AM, Peter Lieven wrote:
>> at the beginning of migration all pages are marked dirty and
>> in the first round a bulk migration of all pages is performed.
>>
>> currently all these pages are copied to the page cache regardless
>> if there are frequently updated or not. this doesn't make sense
>> since most of these pages are never transferred again.
>>
>> this patch changes the XBZRLE transfer to only be used after
>> the bulk stage has been completed. that means a page is added
>> to the page cache the second time it is transferred and XBZRLE
>> can benefit from the third time of transfer.
>>
>> since the page cache is likely smaller than the number of pages
>> its also likely that in the second round the page is missing in the
>> cache due to collisions in the bulk phase.
>>
>> on the other hand a lot of unneccssary mallocs, memdups and frees
>
> s/unneccssary/unnecessary/
>
>> are saved.
>>
>> Signed-off-by: Peter Lieven <pl@kamp.de>
>
> Do you have any benchmark numbers?  At any rate, the explanation seems
> sound, so a benchmark should show this.

Do you have a particular test pattern in mind? If there is nothing going on
in the VM XBZRLE will not be better than normal copy at all.

Otherwise you will have N xbzrle misses and 0 xbzrle pages without the patch
and 0 xbzrle misses and 0 xbzrle pages with the patch.

Peter

>
>> ---
>>   arch_init.c |    5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
>
> Reviewed-by: Eric Blake <eblake@redhat.com>
>