qemu-block
[Top][All Lists]
Advanced

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

Re: [Qemu-block] [PATCH 3/9] block: Track discard requests


From: Kevin Wolf
Subject: Re: [Qemu-block] [PATCH 3/9] block: Track discard requests
Date: Wed, 28 Oct 2015 10:54:21 +0100
User-agent: Mutt/1.5.21 (2010-09-15)

Am 26.10.2015 um 07:24 hat Fam Zheng geschrieben:
> Both bdrv_discard and bdrv_aio_discard will call into bdrv_co_discard,
> so add tracked_request_begin/end calls around the loop.
> 
> Signed-off-by: Fam Zheng <address@hidden>
> ---
>  block/io.c | 14 ++++++++++----
>  1 file changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/block/io.c b/block/io.c
> index 223c4e9..abb3aaa 100644
> --- a/block/io.c
> +++ b/block/io.c
> @@ -2415,7 +2415,8 @@ static void coroutine_fn bdrv_discard_co_entry(void 
> *opaque)
>  int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num,
>                                   int nb_sectors)
>  {
> -    int max_discard, ret;
> +    BdrvTrackedRequest req;
> +    int max_discard, ret = 0;
>  
>      if (!bs->drv) {
>          return -ENOMEDIUM;
> @@ -2437,6 +2438,8 @@ int coroutine_fn bdrv_co_discard(BlockDriverState *bs, 
> int64_t sector_num,
>          return 0;
>      }
>  
> +    tracked_request_begin(&req, bs, sector_num, nb_sectors,
> +                          BDRV_TRACKED_DISCARD);
>      bdrv_set_dirty(bs, sector_num, nb_sectors);
>  
>      max_discard = MIN_NON_ZERO(bs->bl.max_discard, BDRV_REQUEST_MAX_SECTORS);
> @@ -2470,20 +2473,23 @@ int coroutine_fn bdrv_co_discard(BlockDriverState 
> *bs, int64_t sector_num,
>              acb = bs->drv->bdrv_aio_discard(bs, sector_num, nb_sectors,
>                                              bdrv_co_io_em_complete, &co);
>              if (acb == NULL) {
> -                return -EIO;
> +                ret = -EIO;
> +                goto out;
>              } else {
>                  qemu_coroutine_yield();
>                  ret = co.ret;
>              }
>          }
>          if (ret && ret != -ENOTSUP) {
> -            return ret;
> +            goto out;
>          }
>  
>          sector_num += num;
>          nb_sectors -= num;
>      }
> -    return 0;
> +out:
> +    tracked_request_end(&req);
> +    return ret;
>  }

I would prefer an explicit ret = 0 before the out label because
otherwise you're relying on the previous value that has been set
somewhere in the loop. As far as I can tell, it's still correct, but it
makes it needlessly hard to tell whether success is 0 or >= 0.

Kevin



reply via email to

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