xfs
[Top] [All Lists]

Re: [PATCH 1/4] xfs: make inode64 as the default allocation mode

To: Carlos Maiolino <cmaiolino@xxxxxxxxxx>
Subject: Re: [PATCH 1/4] xfs: make inode64 as the default allocation mode
From: Eric Sandeen <sandeen@xxxxxxxxxxx>
Date: Fri, 14 Sep 2012 13:06:33 -0500
Cc: xfs@xxxxxxxxxxx
In-reply-to: <1347641381-30050-2-git-send-email-cmaiolino@xxxxxxxxxx>
References: <1347641381-30050-1-git-send-email-cmaiolino@xxxxxxxxxx> <1347641381-30050-2-git-send-email-cmaiolino@xxxxxxxxxx>
User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:15.0) Gecko/20120907 Thunderbird/15.0.1
On 9/14/12 11:49 AM, Carlos Maiolino wrote:
> since 64-bit inodes can be accessed while using inode32, and these can also be
> used on 32-bit kernels, there is no reason to still keep inode32 as the 
> default
> mount option.
> 
> Signed-off-by: Carlos Maiolino <cmaiolino@xxxxxxxxxx>
> ---
>  fs/xfs/xfs_super.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
> index f1f2968..967d024 100644
> --- a/fs/xfs/xfs_super.c
> +++ b/fs/xfs/xfs_super.c
> @@ -88,6 +88,8 @@ mempool_t *xfs_ioend_pool;
>                                        * unwritten extent conversion */
>  #define MNTOPT_NOBARRIER "nobarrier" /* .. disable */
>  #define MNTOPT_64BITINODE   "inode64"        /* inodes can be allocated 
> anywhere */
> +#define MNTOPT_32BITINODE   "inode32"        /* inode allocation limited to
> +                                      * XFS_MAXINUMBER_32 */
>  #define MNTOPT_IKEEP "ikeep"         /* do not free empty inode clusters */
>  #define MNTOPT_NOIKEEP       "noikeep"       /* free empty inode clusters */
>  #define MNTOPT_LARGEIO          "largeio"    /* report large I/O sizes in 
> stat() */
> @@ -198,7 +200,6 @@ xfs_parseargs(
>        */
>       mp->m_flags |= XFS_MOUNT_BARRIER;
>       mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
> -     mp->m_flags |= XFS_MOUNT_SMALL_INUMS;

Self-note-taking.

XFS_BIG_INUMS means the kernel itself is capable of > 32 bit inode nrs in 
struct inode.

XFS_MOUNT_SMALL_INUMS is "users wants 32bit inodes "

XFS_MOUNT_32BITINODES is "do not create inodes above 32 bits"

MNTOPT_32BITINODE is the actual mount option string "inode32"

I think rather than removing the line above, you might need to keep it under
#if !XFS_BIG_INUMS
        mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
#endif

so that "small" systems continue to enforce <= 32 bit inodes.

>       /*
>        * These can be overridden by the mount option parsing.
> @@ -295,6 +296,13 @@ xfs_parseargs(
>                               return EINVAL;
>                       }
>                       dswidth = simple_strtoul(value, &eov, 10);
> +             } else if (!strcmp(this_char, MNTOPT_32BITINODE)) {
> +                     mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
> +#if !XFS_BIG_INUMS
> +                     xfs_warn(mp, "%s option not allowed on this system",
> +                             this_char);
> +                     return EINVAL;
> +#endif

This looks wrong.  If a user has a system which cannot handle 64 bit inodes,
you reject a request to mount with 32-bit inodes?

-Eric

>               } else if (!strcmp(this_char, MNTOPT_64BITINODE)) {
>                       mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
>  #if !XFS_BIG_INUMS
> @@ -493,13 +501,13 @@ xfs_showargs(
>               { XFS_MOUNT_FILESTREAMS,        "," MNTOPT_FILESTREAM },
>               { XFS_MOUNT_GRPID,              "," MNTOPT_GRPID },
>               { XFS_MOUNT_DISCARD,            "," MNTOPT_DISCARD },
> +             { XFS_MOUNT_SMALL_INUMS,        "," MNTOPT_32BITINODE },
>               { 0, NULL }
>       };
>       static struct proc_xfs_info xfs_info_unset[] = {
>               /* the few simple ones we can get from the mount struct */
>               { XFS_MOUNT_COMPAT_IOSIZE,      "," MNTOPT_LARGEIO },
>               { XFS_MOUNT_BARRIER,            "," MNTOPT_NOBARRIER },
> -             { XFS_MOUNT_SMALL_INUMS,        "," MNTOPT_64BITINODE },
>               { 0, NULL }
>       };
>       struct proc_xfs_info    *xfs_infop;
> 

<Prev in Thread] Current Thread [Next in Thread>