qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 7/9] system/dirtylimit: Don't use migration_is_active()


From: Yong Huang
Subject: Re: [PATCH 7/9] system/dirtylimit: Don't use migration_is_active()
Date: Tue, 17 Dec 2024 21:18:17 +0800



On Tue, Dec 17, 2024 at 5:47 PM Avihai Horon <avihaih@nvidia.com> wrote:

On 17/12/2024 5:01, Yong Huang wrote:
> *External email: Use caution opening links or attachments*
>
>
>
>
> On Mon, Dec 16, 2024 at 5:47 PM Avihai Horon <avihaih@nvidia.com> wrote:
>
>     vcpu_dirty_rate_stat_collect() uses migration_is_active() to detect
>     whether migration is running or not, in order to get the correct dirty
>     rate period value.
>
>     However, recently there has been an effort to simplify the migration
>     status API and reduce it to a single migration_is_running() function.
>
>
> Could you post the related links?

Sure, it's here:
https://lore.kernel.org/qemu-devel/20241024213056.1395400-1-peterx@redhat.com/

Peter took only patches 1-5 of this series.

Thanks.

>
>     To accommodate this, and since the same functionality can be achieved
>     with migration_is_running(), use it instead of migration_is_active().
>
>     Signed-off-by: Avihai Horon <avihaih@nvidia.com>
>     ---
>      system/dirtylimit.c | 3 +--
>      1 file changed, 1 insertion(+), 2 deletions(-)
>
>     diff --git a/system/dirtylimit.c b/system/dirtylimit.c
>     index ab20da34bb..d7a855c603 100644
>     --- a/system/dirtylimit.c
>     +++ b/system/dirtylimit.c
>     @@ -80,8 +80,7 @@ static void vcpu_dirty_rate_stat_collect(void)
>          int i = 0;
>          int64_t period = DIRTYLIMIT_CALC_TIME_MS;
>
>     -    if (migrate_dirty_limit() &&
>     -        migration_is_active()) {
>     +    if (migrate_dirty_limit() && migration_is_running()) {
>              period = migrate_vcpu_dirty_limit_period();
>          }
>
>     --
>     2.40.1
>
>
>
> --
> Best regards

Thanks,

Reviewed-by: Hyman Huang <yong.huang@smartx.com>

--
Best regards

reply via email to

[Prev in Thread] Current Thread [Next in Thread]