On Wed, Feb 26, 2020 at 12:23:05PM -0800, Christoph Hellwig wrote:
> No need to allocate the max size if we can just allocate the easily
> known actual ACL size.
>
> Suggested-by: Dave Chinner <dchinner@xxxxxxxxxx>
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
Looks reasonable,
Reviewed-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
--D
> ---
> fs/xfs/xfs_acl.c | 7 +------
> 1 file changed, 1 insertion(+), 6 deletions(-)
>
> diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
> index 552258399648..5807f11aed3e 100644
> --- a/fs/xfs/xfs_acl.c
> +++ b/fs/xfs/xfs_acl.c
> @@ -187,16 +187,11 @@ __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
> args.namelen = strlen(args.name);
>
> if (acl) {
> - args.valuelen = XFS_ACL_MAX_SIZE(ip->i_mount);
> + args.valuelen = XFS_ACL_SIZE(acl->a_count);
> args.value = kmem_zalloc_large(args.valuelen, 0);
> if (!args.value)
> return -ENOMEM;
> -
> xfs_acl_to_disk(args.value, acl);
> -
> - /* subtract away the unused acl entries */
> - args.valuelen -= sizeof(struct xfs_acl_entry) *
> - (XFS_ACL_MAX_ENTRIES(ip->i_mount) - acl->a_count);
> }
>
> error = xfs_attr_set(&args);
> --
> 2.24.1
>
