Re: [Qemu-devel] [PATCH v2 04/13] blockjob: allow block_job_throttle to take delay_ns

[Max Reitz]

On 2018-01-19 21:58, John Snow wrote:
> Instead of only sleeping for 0ms when we've hit a timeout, optionally
> take a longer more explicit delay_ns that always forces the sleep.
> 
> Signed-off-by: John Snow <address@hidden>
> ---
>  block/mirror.c               |  4 ++--
>  blockjob.c                   |  9 ++++-----
>  include/block/blockjob_int.h | 10 +++++++---
>  3 files changed, 13 insertions(+), 10 deletions(-)

[...]

> diff --git a/blockjob.c b/blockjob.c
> index 6f2e709b51..51c0eb5d9e 100644
> --- a/blockjob.c
> +++ b/blockjob.c
> @@ -906,12 +906,11 @@ void block_job_yield(BlockJob *job)
>      block_job_pause_point(job);
>  }
>  
> -void block_job_relax(BlockJob *job)
> +void block_job_relax(BlockJob *job, int64_t delay_ns)
>  {
> -    int64_t now = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
> -
> -    if (now - job->last_enter_ns > SLICE_TIME) {
> -        block_job_sleep_ns(job, 0);
> +    if (delay_ns || (qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - \
> +                     job->last_enter_ns > SLICE_TIME)) {
> +        block_job_sleep_ns(job, delay_ns);

I can't say I like the readability of that any better...

(And I'd argue that if delay_ns > 0, the one superfluous call to
qemu_clock_get_ns() isn't going to harm performance.)

>      } else {
>          block_job_pause_point(job);
>      }
> diff --git a/include/block/blockjob_int.h b/include/block/blockjob_int.h
> index 553784d86f..5f1520fab7 100644
> --- a/include/block/blockjob_int.h
> +++ b/include/block/blockjob_int.h
> @@ -160,11 +160,15 @@ void block_job_yield(BlockJob *job);
>  /**
>   * block_job_relax:
>   * @job: The job that calls the function.
> + * @delay_ns: The amount of time to sleep for
>   *
> - * Yield if it has been SLICE_TIME nanoseconds since the last yield.
> - * Otherwise, check if we need to pause, and yield if so.
> + * Sleep for delay_ns nanoseconds.
> + *
> + * If delay_ns is 0, yield if it has been SLICE_TIME
> + * nanoseconds since the last yield. Otherwise, check
> + * if we need to yield for a pause event.

That "Otherwise" now sounds as if it refers to the "If delay_ns is 0".

Code change is OK, but this comment needs some fixing.

Max

>   */
> -void block_job_relax(BlockJob *job);
> +void block_job_relax(BlockJob *job, int64_t delay_ns);
>  
>  /**
>   * block_job_pause_all:
>

signature.asc
Description: OpenPGP digital signature