CEPH filesystem development
 help / color / mirror / Atom feed
From: Alex Elder <elder@ieee.org>
To: Ilya Dryomov <ilya.dryomov@inktank.com>, ceph-devel@vger.kernel.org
Subject: Re: [PATCH 4/8] rbd: split rbd_dev_spec_update() into two functions
Date: Thu, 24 Jul 2014 07:55:43 -0500	[thread overview]
Message-ID: <53D1024F.6000500@ieee.org> (raw)
In-Reply-To: <1406191369-6746-5-git-send-email-ilya.dryomov@inktank.com>

On 07/24/2014 03:42 AM, Ilya Dryomov wrote:
> rbd_dev_spec_update() has two modes of operation, with nothing in
> common between them.  Split it into two functions, one for each mode
> and make our expectations more clear.
> 
> Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>

Looks good.

Reviewed-by: Alex Elder <elder@linaro.org>

> ---
>  drivers/block/rbd.c |   79 +++++++++++++++++++++++++++++++--------------------
>  1 file changed, 48 insertions(+), 31 deletions(-)
> 
> diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
> index 4541f6027e4a..23df1773ef77 100644
> --- a/drivers/block/rbd.c
> +++ b/drivers/block/rbd.c
> @@ -3798,6 +3798,9 @@ static struct rbd_spec *rbd_spec_alloc(void)
>  	spec = kzalloc(sizeof (*spec), GFP_KERNEL);
>  	if (!spec)
>  		return NULL;
> +
> +	spec->pool_id = CEPH_NOPOOL;
> +	spec->snap_id = CEPH_NOSNAP;
>  	kref_init(&spec->kref);
>  
>  	return spec;
> @@ -4257,18 +4260,38 @@ static u64 rbd_snap_id_by_name(struct rbd_device *rbd_dev, const char *name)
>  }
>  
>  /*
> - * When an rbd image has a parent image, it is identified by the
> - * pool, image, and snapshot ids (not names).  This function fills
> - * in the names for those ids.  (It's OK if we can't figure out the
> - * name for an image id, but the pool and snapshot ids should always
> - * exist and have names.)  All names in an rbd spec are dynamically
> - * allocated.
> + * An image being mapped will have everything but the snap id.
> + */
> +static int rbd_spec_fill_snap_id(struct rbd_device *rbd_dev)
> +{
> +	struct rbd_spec *spec = rbd_dev->spec;
> +
> +	rbd_assert(spec->pool_id != CEPH_NOPOOL && spec->pool_name);
> +	rbd_assert(spec->image_id && spec->image_name);
> +	rbd_assert(spec->snap_name);
> +
> +	if (strcmp(spec->snap_name, RBD_SNAP_HEAD_NAME)) {
> +		u64 snap_id;
> +
> +		snap_id = rbd_snap_id_by_name(rbd_dev, spec->snap_name);
> +		if (snap_id == CEPH_NOSNAP)
> +			return -ENOENT;
> +
> +		spec->snap_id = snap_id;
> +	} else {
> +		spec->snap_id = CEPH_NOSNAP;
> +	}
> +
> +	return 0;
> +}
> +
> +/*
> + * A parent image will have all ids but none of the names.
>   *
> - * When an image being mapped (not a parent) is probed, we have the
> - * pool name and pool id, image name and image id, and the snapshot
> - * name.  The only thing we're missing is the snapshot id.
> + * All names in an rbd spec are dynamically allocated.  It's OK if we
> + * can't figure out the name for an image id.
>   */
> -static int rbd_dev_spec_update(struct rbd_device *rbd_dev)
> +static int rbd_spec_fill_names(struct rbd_device *rbd_dev)
>  {
>  	struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
>  	struct rbd_spec *spec = rbd_dev->spec;
> @@ -4277,24 +4300,9 @@ static int rbd_dev_spec_update(struct rbd_device *rbd_dev)
>  	const char *snap_name;
>  	int ret;
>  
> -	/*
> -	 * An image being mapped will have the pool name (etc.), but
> -	 * we need to look up the snapshot id.
> -	 */
> -	if (spec->pool_name) {
> -		if (strcmp(spec->snap_name, RBD_SNAP_HEAD_NAME)) {
> -			u64 snap_id;
> -
> -			snap_id = rbd_snap_id_by_name(rbd_dev, spec->snap_name);
> -			if (snap_id == CEPH_NOSNAP)
> -				return -ENOENT;
> -			spec->snap_id = snap_id;
> -		} else {
> -			spec->snap_id = CEPH_NOSNAP;
> -		}
> -
> -		return 0;
> -	}
> +	rbd_assert(spec->pool_id != CEPH_NOPOOL);
> +	rbd_assert(spec->image_id);
> +	rbd_assert(spec->snap_id != CEPH_NOSNAP);
>  
>  	/* Get the pool name; we have to make our own copy of this */
>  
> @@ -4313,7 +4321,7 @@ static int rbd_dev_spec_update(struct rbd_device *rbd_dev)
>  	if (!image_name)
>  		rbd_warn(rbd_dev, "unable to get image name");
>  
> -	/* Look up the snapshot name, and make a copy */
> +	/* Fetch the snapshot name */
>  
>  	snap_name = rbd_snap_name(rbd_dev, spec->snap_id);
>  	if (IS_ERR(snap_name)) {
> @@ -4326,10 +4334,10 @@ static int rbd_dev_spec_update(struct rbd_device *rbd_dev)
>  	spec->snap_name = snap_name;
>  
>  	return 0;
> +
>  out_err:
>  	kfree(image_name);
>  	kfree(pool_name);
> -
>  	return ret;
>  }
>  
> @@ -5158,7 +5166,16 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping)
>  	if (ret)
>  		goto err_out_watch;
>  
> -	ret = rbd_dev_spec_update(rbd_dev);
> +	/*
> +	 * If this image is the one being mapped, we have pool name and
> +	 * id, image name and id, and snap name - need to fill snap id.
> +	 * Otherwise this is a parent image, identified by pool, image
> +	 * and snap ids - need to fill in names for those ids.
> +	 */
> +	if (mapping)
> +		ret = rbd_spec_fill_snap_id(rbd_dev);
> +	else
> +		ret = rbd_spec_fill_names(rbd_dev);
>  	if (ret)
>  		goto err_out_probe;
>  
> 


  reply	other threads:[~2014-07-24 12:55 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-24  8:42 [PATCH 0/8] wip-overlap Ilya Dryomov
2014-07-24  8:42 ` [PATCH 1/8] rbd: show the entire chain of parent images Ilya Dryomov
2014-07-24 12:31   ` Alex Elder
2014-07-24 12:45     ` Ilya Dryomov
2014-07-24  8:42 ` [PATCH 2/8] rbd: introduce rbd_dev_header_info() Ilya Dryomov
2014-07-24 12:34   ` Alex Elder
2014-07-24  8:42 ` [PATCH 3/8] rbd: remove unnecessary asserts in rbd_dev_image_probe() Ilya Dryomov
2014-07-24 12:40   ` Alex Elder
2014-07-24  8:42 ` [PATCH 4/8] rbd: split rbd_dev_spec_update() into two functions Ilya Dryomov
2014-07-24 12:55   ` Alex Elder [this message]
2014-07-24  8:42 ` [PATCH 5/8] rbd: harden rbd_dev_refresh() caller Ilya Dryomov
2014-07-24 13:09   ` Alex Elder
2014-07-24  8:42 ` [PATCH 6/8] rbd: update mapping size only on refresh Ilya Dryomov
2014-07-24 13:25   ` Alex Elder
2014-07-24 13:46     ` Ilya Dryomov
2014-07-24 15:10       ` Ilya Dryomov
2014-07-25 13:31         ` Alex Elder
2014-07-24 17:59       ` Alex Elder
2014-07-24  8:42 ` [PATCH 7/8] rbd: do not read in parent info before snap context Ilya Dryomov
2014-07-25  8:14   ` Alex Elder
2014-07-25  8:36     ` Ilya Dryomov
2014-07-25 12:46       ` Alex Elder
2014-07-24  8:42 ` [PATCH 8/8] rbd: take snap_id into account when reading in parent info Ilya Dryomov
2014-07-24 18:43   ` Alex Elder

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=53D1024F.6000500@ieee.org \
    --to=elder@ieee.org \
    --cc=ceph-devel@vger.kernel.org \
    --cc=ilya.dryomov@inktank.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox