linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nikolay Borisov <nborisov@suse.com>
To: Anand Jain <anand.jain@oracle.com>, linux-btrfs@vger.kernel.org
Subject: Re: [PATCH 1/2] btrfs: match btrfs_device->mode same as it used for open
Date: Fri, 20 Oct 2017 15:26:38 +0300	[thread overview]
Message-ID: <dd850164-a7fa-b7d4-9c5c-4bf497289a66@suse.com> (raw)
In-Reply-To: <20171020103325.9595-1-anand.jain@oracle.com>



On 20.10.2017 13:33, Anand Jain wrote:
> We aren't saving the fmode_t flags at %struct btrfs_device->mode same
> as it used to open the device at device replace, so fix that.

I think this could use a bit of rewording to better explain what the
problem and what it can potentially cause.

> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  fs/btrfs/volumes.c | 16 ++++++++--------
>  1 file changed, 8 insertions(+), 8 deletions(-)
> 
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index b39737568c22..765c2bd2d8d9 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -2323,12 +2323,12 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
>  	u64 tmp;
>  	int seeding_dev = 0;
>  	int ret = 0;
> +	fmode_t flag = FMODE_WRITE | FMODE_EXCL;
>  
>  	if (sb_rdonly(sb) && !fs_info->fs_devices->seeding)
>  		return -EROFS;
>  
> -	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
> -				  fs_info->bdev_holder);
> +	bdev = blkdev_get_by_path(device_path, flag, fs_info->bdev_holder);
>  	if (IS_ERR(bdev))
>  		return PTR_ERR(bdev);
>  
> @@ -2392,7 +2392,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
>  	device->bdev = bdev;
>  	device->in_fs_metadata = 1;
>  	device->is_tgtdev_for_dev_replace = 0;
> -	device->mode = FMODE_EXCL;
> +	device->mode = flag;

So the actual bug being fixed by this patch is that currently we aren't
setting the FMODE_WRITE when initializing btrfs_device structure and
when calling blkdev_put, correct?

>  	device->dev_stats_valid = 1;
>  	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
>  
> @@ -2506,7 +2506,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
>  	btrfs_sysfs_rm_device_link(fs_info->fs_devices, device);
>  	kfree(device);
>  error:
> -	blkdev_put(bdev, FMODE_EXCL);
> +	blkdev_put(bdev, flag);
>  	if (seeding_dev) {
>  		mutex_unlock(&uuid_mutex);
>  		up_write(&sb->s_umount);
> @@ -2526,6 +2526,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
>  	struct rcu_string *name;
>  	u64 devid = BTRFS_DEV_REPLACE_DEVID;
>  	int ret = 0;
> +	fmode_t flags = FMODE_WRITE | FMODE_EXCL;
>  
>  	*device_out = NULL;
>  	if (fs_info->fs_devices->seeding) {
> @@ -2533,8 +2534,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
>  		return -EINVAL;
>  	}
>  
> -	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
> -				  fs_info->bdev_holder);
> +	bdev = blkdev_get_by_path(device_path, flags, fs_info->bdev_holder);
>  	if (IS_ERR(bdev)) {
>  		btrfs_err(fs_info, "target device %s is invalid!", device_path);
>  		return PTR_ERR(bdev);
> @@ -2595,7 +2595,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
>  	device->bdev = bdev;
>  	device->in_fs_metadata = 1;
>  	device->is_tgtdev_for_dev_replace = 1;
> -	device->mode = FMODE_EXCL;
> +	device->mode = flags;
>  	device->dev_stats_valid = 1;
>  	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
>  	device->fs_devices = fs_info->fs_devices;
> @@ -2608,7 +2608,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
>  	return ret;
>  
>  error:
> -	blkdev_put(bdev, FMODE_EXCL);
> +	blkdev_put(bdev, flags);
>  	return ret;
>  }
>  
> 

  parent reply	other threads:[~2017-10-20 12:26 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-20 10:33 [PATCH 1/2] btrfs: match btrfs_device->mode same as it used for open Anand Jain
2017-10-20 10:33 ` [PATCH 2/2] btrfs: maintain consistent fmode flags for device open Anand Jain
2017-10-20 12:26 ` Nikolay Borisov [this message]
2017-10-20 14:00   ` [PATCH 1/2] btrfs: match btrfs_device->mode same as it used for open Anand Jain
2017-10-20 14:07 ` [PATCH v2 " Anand Jain
2017-10-30 14:39   ` David Sterba
2017-10-31  2:11     ` Anand Jain
2017-10-31 14:21       ` Nikolay Borisov
2017-11-02  5:55         ` Anand Jain
2017-11-02  6:11 ` [PATCH v3 " Anand Jain
2017-11-02  8:47   ` Nikolay Borisov

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=dd850164-a7fa-b7d4-9c5c-4bf497289a66@suse.com \
    --to=nborisov@suse.com \
    --cc=anand.jain@oracle.com \
    --cc=linux-btrfs@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).