Re: [Qemu-devel] [PATCH 07/26] block: add the support to drain throttled requests

[Zhi Yong Wu]

On Thu, Apr 12, 2012 at 8:00 PM, Paolo Bonzini <address@hidden> wrote:
> From: Zhi Yong Wu <address@hidden>
>
> Signed-off-by: Zhi Yong Wu <address@hidden>
> [ Iterate until all block devices have processed all requests,
>  add comments. - Paolo ]
> Signed-off-by: Paolo Bonzini <address@hidden>
> ---
>  block.c |   21 ++++++++++++++++++++-
>  1 file changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/block.c b/block.c
> index 4b96654..33efa95 100644
> --- a/block.c
> +++ b/block.c
> @@ -863,12 +863,31 @@ void bdrv_close_all(void)
>  *
>  * This function does not flush data to disk, use bdrv_flush_all() for that
>  * after calling this function.
> + *
> + * Note that completion of an asynchronous I/O operation can trigger any
> + * number of other I/O operations on other devices---for example a coroutine
> + * can be arbitrarily complex and a constant flow of I/O can come until the
> + * coroutine is complete.  Because of this, it is not possible to have a
> + * function to drain a single device's I/O queue.
>  */
>  void bdrv_drain_all(void)
>  {
>     BlockDriverState *bs;
> +    bool busy;
>
> -    qemu_aio_flush();
> +    do {
> +        busy = qemu_aio_wait();
When bdrv_drain_all is invoked, if bs->tracked_requests has some
pending I/O requests, only qemu_aio_wait is called here, not
qemu_aio_flush() to restart them in bs->tracked_requests. It will
cause assert about bs->tracked_requests.
> +
> +        /* FIXME: We do not have timer support here, so this is effectively
> +         * a busy wait.
> +         */
> +        QTAILQ_FOREACH(bs, &bdrv_states, list) {
> +            if (!qemu_co_queue_empty(&bs->throttled_reqs)) {
> +                qemu_co_queue_restart_all(&bs->throttled_reqs);
> +                busy = true;
> +            }
> +        }
> +    } while (busy);
>
>     /* If requests are still pending there is a bug somewhere */
>     QTAILQ_FOREACH(bs, &bdrv_states, list) {
> --
> 1.7.9.3
>
>
>



-- 
Regards,

Zhi Yong Wu