Re: [Qemu-devel] [PATCH 1/2] rdma: remove check on time_spent when calculating mbs

Re: [Qemu-devel] [PATCH 1/2] rdma: remove check on time_spent when calculating mbs

[Michael Tokarev]

24.01.2016 17:09, Wei Yang wrote:
> Within the if statement, time_spent is assured to be non-zero.
> 
> This patch just removes the check on time_spent when calculating mbs.

The if statement is this one:

        if (current_time >= initial_time + BUFFER_DELAY) {

Note that we use time_spent as a divisor to calculate
bandwidth too.

This is indeed a trivial patch, I'm applying it to -trivial,
but please the next time post it to qemu-devel (the main
mailinglist), and Cc qemu-trivial if it is trivial.

Thanks!

/mjt

> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
> ---
>  migration/migration.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/migration/migration.c b/migration/migration.c
> index c842499..40b87f2 100644
> --- a/migration/migration.c
> +++ b/migration/migration.c
> @@ -1677,8 +1677,8 @@ static void *migration_thread(void *opaque)
>              double bandwidth = (double)transferred_bytes / time_spent;
>              max_size = bandwidth * migrate_max_downtime() / 1000000;
>  
> -            s->mbps = time_spent ? (((double) transferred_bytes * 8.0) /
> -                    ((double) time_spent / 1000.0)) / 1000.0 / 1000.0 : -1;
> +            s->mbps = (((double) transferred_bytes * 8.0) /
> +                    ((double) time_spent / 1000.0)) / 1000.0 / 1000.0;
>  
>              trace_migrate_transferred(transferred_bytes, time_spent,
>                                        bandwidth, max_size);
> 

Re: [Qemu-devel] [PATCH 2/2] qmp-spec: fix index in doc

[Michael Tokarev]

24.01.2016 17:09, Wei Yang wrote:
> The index is duplicated. Just change it.

It is indeed, with previous being 2.5 as well
and the next being 3. Applying to -trivial.

Please the next time send to qemu-devel@ and Cc to -trivial.

Thanks!

/mjt

> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
> ---
>  docs/qmp-spec.txt | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/docs/qmp-spec.txt b/docs/qmp-spec.txt
> index 4fb10a5..8e4bc3d 100644
> --- a/docs/qmp-spec.txt
> +++ b/docs/qmp-spec.txt
> @@ -180,7 +180,7 @@ Some events are rate-limited to at most one per second.  If additional
>  dropped, and the last one is delayed.  "Similar" normally means same
>  event type.  See qmp-events.txt for details.
>  
> -2.5 QGA Synchronization
> +2.6 QGA Synchronization
>  -----------------------
>  
>  When using QGA, an additional synchronization feature is built into
> 

Re: [Qemu-devel] [PATCH 2/2] qmp-spec: fix index in doc

[Wei Yang]

On Sat, Jan 30, 2016 at 12:46:58PM +0300, Michael Tokarev wrote:
>24.01.2016 17:09, Wei Yang wrote:
>> The index is duplicated. Just change it.
>
>It is indeed, with previous being 2.5 as well
>and the next being 3. Applying to -trivial.
>
>Please the next time send to qemu-devel@ and Cc to -trivial.
>

Thanks, got the correct procedure.

>Thanks!
>
>/mjt
>
>> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
>> ---
>>  docs/qmp-spec.txt | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>> 
>> diff --git a/docs/qmp-spec.txt b/docs/qmp-spec.txt
>> index 4fb10a5..8e4bc3d 100644
>> --- a/docs/qmp-spec.txt
>> +++ b/docs/qmp-spec.txt
>> @@ -180,7 +180,7 @@ Some events are rate-limited to at most one per second.  If additional
>>  dropped, and the last one is delayed.  "Similar" normally means same
>>  event type.  See qmp-events.txt for details.
>>  
>> -2.5 QGA Synchronization
>> +2.6 QGA Synchronization
>>  -----------------------
>>  
>>  When using QGA, an additional synchronization feature is built into
>> 

-- 
Wei Yang
Help you, Help me

Re: [Qemu-devel] [PATCH 1/2] rdma: remove check on time_spent when calculating mbs

[Wei Yang]

On Sat, Jan 30, 2016 at 12:45:27PM +0300, Michael Tokarev wrote:
>24.01.2016 17:09, Wei Yang wrote:
>> Within the if statement, time_spent is assured to be non-zero.
>> 
>> This patch just removes the check on time_spent when calculating mbs.
>
>The if statement is this one:
>
>        if (current_time >= initial_time + BUFFER_DELAY) {
>
>Note that we use time_spent as a divisor to calculate
>bandwidth too.
>
>This is indeed a trivial patch, I'm applying it to -trivial,
>but please the next time post it to qemu-devel (the main
>mailinglist), and Cc qemu-trivial if it is trivial.
>

To Amit and Michael,

Yep, thanks. I thought these two are too trivial to post to qemu-devel :)

Will do this next time.

>Thanks!
>
>/mjt
>
>> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
>> ---
>>  migration/migration.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>> 
>> diff --git a/migration/migration.c b/migration/migration.c
>> index c842499..40b87f2 100644
>> --- a/migration/migration.c
>> +++ b/migration/migration.c
>> @@ -1677,8 +1677,8 @@ static void *migration_thread(void *opaque)
>>              double bandwidth = (double)transferred_bytes / time_spent;
>>              max_size = bandwidth * migrate_max_downtime() / 1000000;
>>  
>> -            s->mbps = time_spent ? (((double) transferred_bytes * 8.0) /
>> -                    ((double) time_spent / 1000.0)) / 1000.0 / 1000.0 : -1;
>> +            s->mbps = (((double) transferred_bytes * 8.0) /
>> +                    ((double) time_spent / 1000.0)) / 1000.0 / 1000.0;
>>  
>>              trace_migrate_transferred(transferred_bytes, time_spent,
>>                                        bandwidth, max_size);
>> 

-- 
Wei Yang
Help you, Help me