Re: [PATCH v3] romfs: use different way to generate fsid for BLOCK or MTD

[Coly Li <i@coly.li>]

Hi Andrew,

This fix is verified by Nong Li, it works well in all three conditions,
- CONFIG_ROMFS_ON_BLOCK defined	
- CONFIG_ROMFS_ON_BLOCK undefined and CONFIG_ROMFS_ON_MTD defined
- both CONFIG_ROMFS_ON_BLOCK and CONFIG_ROMFS_ON_MTD undefined

Last time the original patch was picked by you, could you please take
care it again this time.

Thanks in advance.

Coly Li

On 2016/12/28 下午8:36, Coly Li wrote:
> 'Commit 8a59f5d25265 ("fs/romfs: return f_fsid for statfs(2)")' generates
> a 64bit id from sb->s_bdev->bd_dev. This is only correct when romfs is
> defined with CONFIG_ROMFS_ON_BLOCK. If romfs is only defined with
> CONFIG_ROMFS_ON_MTD, sb->s_bdev is NULL, referencing sb->s_bdev->bd_dev
> will triger an oops.
> 
> Richard Weinberger points out that when CONFIG_ROMFS_BACKED_BY_BOTH=y,
> both CONFIG_ROMFS_ON_BLOCK and CONFIG_ROMFS_ON_MTD are defined. Therefore
> when calling huge_encode_dev() to generate a 64bit id, I use the follow
> order to choose parameter,
> - CONFIG_ROMFS_ON_BLOCK defined
>   use sb->s_bdev->bd_dev
> - CONFIG_ROMFS_ON_BLOCK undefined and CONFIG_ROMFS_ON_MTD defined
>   use sb->s_dev when,
> - both CONFIG_ROMFS_ON_BLOCK and CONFIG_ROMFS_ON_MTD undefined
>   leave id as 0
> 
> When CONFIG_ROMFS_ON_MTD is defined and sb->s_mtd is not NULL, sb->s_dev
> is set to a device ID generated by MTD_BLOCK_MAJOR and mtd index, otherwise
> sb->s_dev is 0.
> 
> This is a try-best effort to generate a uniq file system ID, if all the
> above conditions are not meet, f_fsid of this romfs instance will be 0.
> Generally only one romfs can be built on single MTD block device, this
> method is enough to identify multiple romfs instances in a computer.
> 
> Signed-off-by: Coly Li <colyli@suse.de>
> Reported-by: Nong Li <nongli1031@gmail.com>
> Tested-by: Nong Li <nongli1031@gmail.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Richard Weinberger <richard.weinberger@gmail.com>
> Cc: linux-fsdevel@vger.kernel.org
> Cc: linux-mtd@lists.infradead.org
> ---
>  fs/romfs/super.c | 23 ++++++++++++++++++++++-
>  1 file changed, 22 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/romfs/super.c b/fs/romfs/super.c
> index d0f8a38..0186fe6 100644
> --- a/fs/romfs/super.c
> +++ b/fs/romfs/super.c
> @@ -74,6 +74,7 @@
>  #include <linux/highmem.h>
>  #include <linux/pagemap.h>
>  #include <linux/uaccess.h>
> +#include <linux/major.h>
>  #include "internal.h"
>  
>  static struct kmem_cache *romfs_inode_cachep;
> @@ -416,7 +417,22 @@ static void romfs_destroy_inode(struct inode *inode)
>  static int romfs_statfs(struct dentry *dentry, struct kstatfs *buf)
>  {
>  	struct super_block *sb = dentry->d_sb;
> -	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
> +	u64 id = 0;
> +
> +	/* When calling huge_encode_dev(),
> +	 * use sb->s_bdev->bd_dev when,
> +	 *   - CONFIG_ROMFS_ON_BLOCK defined
> +	 * use sb->s_dev when,
> +	 *   - CONFIG_ROMFS_ON_BLOCK undefined and
> +	 *   - CONFIG_ROMFS_ON_MTD defined
> +	 * leave id as 0 when,
> +	 *   - CONFIG_ROMFS_ON_BLOCK undefined and
> +	 *   - CONFIG_ROMFS_ON_MTD undefined
> +	 */
> +	if (sb->s_bdev)
> +		id = huge_encode_dev(sb->s_bdev->bd_dev);
> +	else if (sb->s_dev)
> +		id = huge_encode_dev(sb->s_dev);
>  
>  	buf->f_type = ROMFS_MAGIC;
>  	buf->f_namelen = ROMFS_MAXFN;
> @@ -489,6 +505,11 @@ static int romfs_fill_super(struct super_block *sb, void *data, int silent)
>  	sb->s_flags |= MS_RDONLY | MS_NOATIME;
>  	sb->s_op = &romfs_super_ops;
>  
> +#ifdef CONFIG_ROMFS_ON_MTD
> +	/* Use same dev ID from the underlying mtdblock device */
> +	if (sb->s_mtd)
> +		sb->s_dev = MKDEV(MTD_BLOCK_MAJOR, sb->s_mtd->index);
> +#endif
>  	/* read the image superblock and check it */
>  	rsb = kmalloc(512, GFP_KERNEL);
>  	if (!rsb)
> 


-- 
Coly Li