On Tuesday September 26, [EMAIL PROTECTED] wrote:
> Al, HJ,
> please have a look at this patch, for inclusion into 2.4. It adds a
> filesystem revalidate call to the NFS server. This is important if the
> underlying filesystem is itself not a local filesystem.
>
> I also have in-progress patches to ensure that NFS also obeys filesystem
> locking semantics, via file->f_op->lock. Would you be willing to take
> a look at those? They are critical for distributed filesystems to
> work properly using Linux as a scalable/fault-tolerant NFS server.
>
> Cheers, Andreas
>
>
> --- linux-2.3.99-pre3/fs/nfsd/nfsfh.c.orig Thu Aug 3 16:34:44 2000
> +++ linux-2.3.99-pre3/fs/nfsd/nfsfh.c Thu Aug 3 16:36:42 2000
> @@ -631,6 +631,13 @@
>
> inode = dentry->d_inode;
>
> + /* Before using this, we must revalidate with the filesystem. */
> + if (inode->i_op && inode->i_op->revalidate) {
> + error = inode->i_op->revalidate(dentry);
> + if (error)
> + goto out;
> + }
> +
> /* Type check. The correct error return for type mismatches
> * does not seem to be generally agreed upon. SunOS seems to
> * use EISDIR if file isn't S_IFREG; a comment in the NFSv3
> -
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to [EMAIL PROTECTED]
This patch doesn't make much sense to me, but maybe I am missing
something.
NFS does all accesses using a filehandle, which is essentially a
pointer to an inode. By the time you get the the bit of code where
you have inserted a patch, it is the inode that we are really sure of,
and the dentry is just there as a handle to hold the inode, because
Linux/VFS perfers dentrys to inodes. At this point, revalidating the
dentry seems a bit pointless.
What particular non-local filesystem are you trying to access via NFS?
NFS wasn't designed for accessing non-local filesystems (infact, it
looks a lot like it was designed for accessing UFS in particular).
And re-exporting NFS is explicitly a no-no (though I cannot remember
where I read that).
NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [EMAIL PROTECTED]