Re: [PATCH v8 RESEND 3/8] vfs: plumb i_version handling into struct kstat

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

 



On Tue 24-01-23 14:30:20, Jeff Layton wrote:
> The NFS server has a lot of special handling for different types of
> change attribute access, depending on the underlying filesystem. In
> most cases, it's doing a getattr anyway and then fetching that value
> after the fact.
> 
> Rather that do that, add a new STATX_CHANGE_COOKIE flag that is a
> kernel-only symbol (for now). If requested and getattr can implement it,
> it can fill out this field. For IS_I_VERSION inodes, add a generic
> implementation in vfs_getattr_nosec. Take care to mask
> STATX_CHANGE_COOKIE off in requests from userland and in the result
> mask.
> 
> Since not all filesystems can give the same guarantees of monotonicity,
> claim a STATX_ATTR_CHANGE_MONOTONIC flag that filesystems can set to
> indicate that they offer an i_version value that can never go backward.
> 
> Eventually if we decide to make the i_version available to userland, we
> can just designate a field for it in struct statx, and move the
> STATX_CHANGE_COOKIE definition to the uapi header.
> 
> Reviewed-by: NeilBrown <neilb@xxxxxxx>
> Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>

Looks good to me. Feel free to add:

Reviewed-by: Jan Kara <jack@xxxxxxx>

								Honza

> ---
>  fs/stat.c            | 17 +++++++++++++++--
>  include/linux/stat.h |  9 +++++++++
>  2 files changed, 24 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/stat.c b/fs/stat.c
> index d6cc74ca8486..f43afe0081fe 100644
> --- a/fs/stat.c
> +++ b/fs/stat.c
> @@ -18,6 +18,7 @@
>  #include <linux/syscalls.h>
>  #include <linux/pagemap.h>
>  #include <linux/compat.h>
> +#include <linux/iversion.h>
>  
>  #include <linux/uaccess.h>
>  #include <asm/unistd.h>
> @@ -122,6 +123,11 @@ int vfs_getattr_nosec(const struct path *path, struct kstat *stat,
>  	stat->attributes_mask |= (STATX_ATTR_AUTOMOUNT |
>  				  STATX_ATTR_DAX);
>  
> +	if ((request_mask & STATX_CHANGE_COOKIE) && IS_I_VERSION(inode)) {
> +		stat->result_mask |= STATX_CHANGE_COOKIE;
> +		stat->change_cookie = inode_query_iversion(inode);
> +	}
> +
>  	mnt_userns = mnt_user_ns(path->mnt);
>  	if (inode->i_op->getattr)
>  		return inode->i_op->getattr(mnt_userns, path, stat,
> @@ -602,9 +608,11 @@ cp_statx(const struct kstat *stat, struct statx __user *buffer)
>  
>  	memset(&tmp, 0, sizeof(tmp));
>  
> -	tmp.stx_mask = stat->result_mask;
> +	/* STATX_CHANGE_COOKIE is kernel-only for now */
> +	tmp.stx_mask = stat->result_mask & ~STATX_CHANGE_COOKIE;
>  	tmp.stx_blksize = stat->blksize;
> -	tmp.stx_attributes = stat->attributes;
> +	/* STATX_ATTR_CHANGE_MONOTONIC is kernel-only for now */
> +	tmp.stx_attributes = stat->attributes & ~STATX_ATTR_CHANGE_MONOTONIC;
>  	tmp.stx_nlink = stat->nlink;
>  	tmp.stx_uid = from_kuid_munged(current_user_ns(), stat->uid);
>  	tmp.stx_gid = from_kgid_munged(current_user_ns(), stat->gid);
> @@ -643,6 +651,11 @@ int do_statx(int dfd, struct filename *filename, unsigned int flags,
>  	if ((flags & AT_STATX_SYNC_TYPE) == AT_STATX_SYNC_TYPE)
>  		return -EINVAL;
>  
> +	/* STATX_CHANGE_COOKIE is kernel-only for now. Ignore requests
> +	 * from userland.
> +	 */
> +	mask &= ~STATX_CHANGE_COOKIE;
> +
>  	error = vfs_statx(dfd, filename, flags, &stat, mask);
>  	if (error)
>  		return error;
> diff --git a/include/linux/stat.h b/include/linux/stat.h
> index ff277ced50e9..52150570d37a 100644
> --- a/include/linux/stat.h
> +++ b/include/linux/stat.h
> @@ -52,6 +52,15 @@ struct kstat {
>  	u64		mnt_id;
>  	u32		dio_mem_align;
>  	u32		dio_offset_align;
> +	u64		change_cookie;
>  };
>  
> +/* These definitions are internal to the kernel for now. Mainly used by nfsd. */
> +
> +/* mask values */
> +#define STATX_CHANGE_COOKIE		0x40000000U	/* Want/got stx_change_attr */
> +
> +/* file attribute values */
> +#define STATX_ATTR_CHANGE_MONOTONIC	0x8000000000000000ULL /* version monotonically increases */
> +
>  #endif
> -- 
> 2.39.1
> 
-- 
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR



[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux