Re: [Qemu-block] [PATCH v2 2/5] qcow2: Implement .bdrv_co_preadv()

[Eric Blake]

On 06/06/2016 08:59 AM, Kevin Wolf wrote:
> Reading from qcow2 images is now byte granularity.
> 
> Most of the affected code in qcow2 actually gets simpler with this
> change. The only exception is encryption, which is fixed on 512 bytes
> blocks; in order to keep this working, bs->request_alignment is set for
> encrypted images.
> 
> Signed-off-by: Kevin Wolf <address@hidden>
> ---
>  block/qcow2-cluster.c |  27 ++++++-------
>  block/qcow2.c         | 108 
> +++++++++++++++++++++++++++-----------------------
>  block/qcow2.h         |   2 +-
>  3 files changed, 72 insertions(+), 65 deletions(-)
> 

> - * on exit, *num is the number of contiguous sectors we can read.
> + * On exit, *bytes is the number of bytes staring at offset that have the 
> same

s/staring/starting/

(although "staring" does make for a rather interesting thought :)

> @@ -1389,26 +1402,24 @@ static coroutine_fn int 
> qcow2_co_readv(BlockDriverState *bs, int64_t sector_num,
>  
>      qemu_co_mutex_lock(&s->lock);
>  
> -    while (remaining_sectors != 0) {
> +    while (bytes != 0) {
>  
>          /* prepare next request */
> -        cur_nr_sectors = remaining_sectors;
> +        cur_bytes = MIN(bytes, INT_MAX);
>          if (s->cipher) {
> -            cur_nr_sectors = MIN(cur_nr_sectors,
> -                QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors);
> +            cur_bytes = MIN(cur_bytes,
> +                            QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size);
>          }

Can this ever result in cur_bytes that is not aligned, even though
encryption requires aligned reads?  Do you need to round anything to a
sector boundary in this case?

>              qemu_co_mutex_unlock(&s->lock);
> -            ret = bdrv_co_readv(bs->file->bs,
> -                                (cluster_offset >> 9) + index_in_cluster,
> -                                cur_nr_sectors, &hd_qiov);
> +            ret = bdrv_co_preadv(bs->file->bs,
> +                                 cluster_offset + offset_in_cluster,
> +                                 cur_bytes, &hd_qiov, 0);
>              qemu_co_mutex_lock(&s->lock);
>              if (ret < 0) {
>                  goto fail;
>              }
>              if (bs->encrypted) {
>                  assert(s->cipher);
> +                assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
> +                assert((cur_bytes & (BDRV_SECTOR_SIZE - 1)) == 0);

Hmm, you do check that later on.

So, other than the typo (easy to fix on commit),
Reviewed-by: Eric Blake <address@hidden>

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

signature.asc
Description: OpenPGP digital signature