summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
authorChuck Lever <chuck.lever@oracle.com>2010-07-06 16:53:34 -0400
committerJ. Bruce Fields <bfields@citi.umich.edu>2010-07-07 17:12:32 -0400
commit43a9aa64a2f4330a9cb59aaf5c5636566bce067c (patch)
tree98139046627c7fa354a106f41af55d524b7e6174 /fs
parent6a85d6c76962db769bb2f2cb11b17b16f32c4158 (diff)
downloadlinux-43a9aa64a2f4330a9cb59aaf5c5636566bce067c.tar.gz
linux-43a9aa64a2f4330a9cb59aaf5c5636566bce067c.tar.xz
NFSD: Fill in WCC data for REMOVE, RMDIR, MKNOD, and MKDIR
Some well-known NFSv3 clients drop their directory entry caches when they receive replies with no WCC data. Without this data, they employ extra READ, LOOKUP, and GETATTR requests to ensure their directory entry caches are up to date, causing performance to suffer needlessly. In order to return WCC data, our server has to have both the pre-op and the post-op attribute data on hand when a reply is XDR encoded. The pre-op data is filled in when the incoming fh is locked, and the post-op data is filled in when the fh is unlocked. Unfortunately, for REMOVE, RMDIR, MKNOD, and MKDIR, the directory fh is not unlocked until well after the reply has been XDR encoded. This means that encode_wcc_data() does not have wcc_data for the parent directory, so none is returned to the client after these operations complete. By unlocking the parent directory fh immediately after the internal operations for each NFS procedure is complete, the post-op data is filled in before XDR encoding starts, so it can be returned to the client properly. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Diffstat (limited to 'fs')
-rw-r--r--fs/nfsd/nfs3proc.c6
1 files changed, 4 insertions, 2 deletions
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 3d68f45a37b9..9ae93317abe4 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -271,7 +271,7 @@ nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp,
fh_init(&resp->fh, NFS3_FHSIZE);
nfserr = nfsd_create(rqstp, &resp->dirfh, argp->name, argp->len,
&argp->attrs, S_IFDIR, 0, &resp->fh);
-
+ fh_unlock(&resp->dirfh);
RETURN_STATUS(nfserr);
}
@@ -327,7 +327,7 @@ nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp,
type = nfs3_ftypes[argp->ftype];
nfserr = nfsd_create(rqstp, &resp->dirfh, argp->name, argp->len,
&argp->attrs, type, rdev, &resp->fh);
-
+ fh_unlock(&resp->dirfh);
RETURN_STATUS(nfserr);
}
@@ -348,6 +348,7 @@ nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp,
/* Unlink. -S_IFDIR means file must not be a directory */
fh_copy(&resp->fh, &argp->fh);
nfserr = nfsd_unlink(rqstp, &resp->fh, -S_IFDIR, argp->name, argp->len);
+ fh_unlock(&resp->fh);
RETURN_STATUS(nfserr);
}
@@ -367,6 +368,7 @@ nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp,
fh_copy(&resp->fh, &argp->fh);
nfserr = nfsd_unlink(rqstp, &resp->fh, S_IFDIR, argp->name, argp->len);
+ fh_unlock(&resp->fh);
RETURN_STATUS(nfserr);
}