2015-11-25 18:50:50

by Jeff Layton

Subject: [PATCH] nfs: ensure that attrcache is revalidated after a SETATTR

If we get no post-op attributes back from a SETATTR operation, then no
attributes will of course be updated during the call to
nfs_update_inode.

We know however that the attributes are invalid at that point, since we
just changed some of them. At the very least, the ctime will be bogus.
If we get no post-op attributes back on the call, mark the attrcache
invalid to reflect that fact.

Reviewed-by: Steve French <[email protected]>
Signed-off-by: Jeff Layton <[email protected]>
---
fs/nfs/inode.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index ffdf9b9e88ab..31b0a52223a7 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -618,7 +618,10 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
nfs_vmtruncate(inode, attr->ia_size);
}
- nfs_update_inode(inode, fattr);
+ if (fattr->valid)
+ nfs_update_inode(inode, fattr);
+ else
+ NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR;
spin_unlock(&inode->i_lock);
}
EXPORT_SYMBOL_GPL(nfs_setattr_update_inode);
--
2.4.3

2015-11-25 22:13:15

by Steve French

Subject: Re: [PATCH] nfs: ensure that attrcache is revalidated after a SETATTR

I tested it on a previously failing chmod/chattr/stat scenario (where
client was relying on cached mode even though server didn't return
post op attributes, and so reported an incorrect mode in stat) and it
fixes that

Reviewed-and-Tested-by: Steve French <[email protected]>

On Wed, Nov 25, 2015 at 12:50 PM, Jeff Layton <[email protected]> wrote:
> If we get no post-op attributes back from a SETATTR operation, then no
> attributes will of course be updated during the call to
> nfs_update_inode.
>
> We know however that the attributes are invalid at that point, since we
> just changed some of them. At the very least, the ctime will be bogus.
> If we get no post-op attributes back on the call, mark the attrcache
> invalid to reflect that fact.
>
> Reviewed-by: Steve French <[email protected]>
> Signed-off-by: Jeff Layton <[email protected]>
> ---
> fs/nfs/inode.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
> index ffdf9b9e88ab..31b0a52223a7 100644
> --- a/fs/nfs/inode.c
> +++ b/fs/nfs/inode.c
> @@ -618,7 +618,10 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
> nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
> nfs_vmtruncate(inode, attr->ia_size);
> }
> - nfs_update_inode(inode, fattr);
> + if (fattr->valid)
> + nfs_update_inode(inode, fattr);
> + else
> + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR;
> spin_unlock(&inode->i_lock);
> }
> EXPORT_SYMBOL_GPL(nfs_setattr_update_inode);
> --
> 2.4.3
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html



--
Thanks,

Steve