Skip to contentSkip to conversation

misc server patches

PATCH New J. Bruce Fields 1 participant 4 messages Mar 9, 2012

Subsystems (2)
NFS client NFS server
lore
Share & tools
lore

J. Bruce Fields

author
#0 lore raw

These are a few more server patches I'm queueing up for 3.4, absent objections.

--b.

author
#1 lore raw

PATCH 1/3 nfsd4: delay setting current filehandle till success

From: "J. Bruce Fields" <[email protected]>

Compound processing stops on error, so the current filehandle won't be used on error. Thus the order here doesn't really matter. It'll be more convenient to do it later, though.

Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/nfsd/nfs4proc.c |    8 +++-----
 1 files changed, 3 insertions(+), 5 deletions(-)
fs/nfsd/nfs4proc.c +3 -5
index cdb7ca3..bdb71a5 100644
@@ -247,16 +247,14 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 	if (is_create_with_attrs(open) && open->op_acl != NULL)
 		do_set_nfs4_acl(rqstp, &resfh, open->op_acl, open->op_bmval);
 
-	set_change_info(&open->op_cinfo, current_fh);
-	fh_dup2(current_fh, &resfh);
-
 	/* set reply cache */
 	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
 			&resfh.fh_handle);
 	if (!open->op_created)
-		status = do_open_permission(rqstp, current_fh, open,
+		status = do_open_permission(rqstp, &resfh, open,
 					    NFSD_MAY_NOP);
-
+	set_change_info(&open->op_cinfo, current_fh);
+	fh_dup2(current_fh, &resfh);
 out:
 	fh_put(&resfh);
 	return status;
Signature · 1 line
1.7.5.4
author
#2 lore raw

PATCH 2/3 nfsd4: reduce do_open_lookup() stack usage

In reply to J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

I get 320 bytes for struct svc_fh on x86_64, really a little large to be putting on the stack; kmalloc() instead.

Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/nfsd/nfs4proc.c |   24 ++++++++++++++----------
 1 files changed, 14 insertions(+), 10 deletions(-)
fs/nfsd/nfs4proc.c +14 -10
index bdb71a5..abe808a 100644
@@ -193,10 +193,13 @@ static __be32 nfsd_check_obj_isreg(struct svc_fh *fh)
 static __be32
 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open)
 {
-	struct svc_fh resfh;
+	struct svc_fh *resfh;
 	__be32 status;
 
-	fh_init(&resfh, NFS4_FHSIZE);
+	resfh = kmalloc(sizeof(struct svc_fh), GFP_KERNEL);
+	if (!resfh)
+		return nfserr_jukebox;
+	fh_init(resfh, NFS4_FHSIZE);
 	open->op_truncate = 0;
 
 	if (open->op_create) {

@@ -221,7 +224,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 		 */
 		status = do_nfsd_create(rqstp, current_fh, open->op_fname.data,
 					open->op_fname.len, &open->op_iattr,
-					&resfh, open->op_createmode,
+					resfh, open->op_createmode,
 					(u32 *)open->op_verf.data,
 					&open->op_truncate, &open->op_created);
 

@@ -235,28 +238,29 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 						FATTR4_WORD1_TIME_MODIFY);
 	} else {
 		status = nfsd_lookup(rqstp, current_fh,
-				     open->op_fname.data, open->op_fname.len, &resfh);
+				     open->op_fname.data, open->op_fname.len, resfh);
 		fh_unlock(current_fh);
 		if (status)
 			goto out;
-		status = nfsd_check_obj_isreg(&resfh);
+		status = nfsd_check_obj_isreg(resfh);
 	}
 	if (status)
 		goto out;
 
 	if (is_create_with_attrs(open) && open->op_acl != NULL)
-		do_set_nfs4_acl(rqstp, &resfh, open->op_acl, open->op_bmval);
+		do_set_nfs4_acl(rqstp, resfh, open->op_acl, open->op_bmval);
 
 	/* set reply cache */
 	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
-			&resfh.fh_handle);
+			&resfh->fh_handle);
 	if (!open->op_created)
-		status = do_open_permission(rqstp, &resfh, open,
+		status = do_open_permission(rqstp, resfh, open,
 					    NFSD_MAY_NOP);
 	set_change_info(&open->op_cinfo, current_fh);
-	fh_dup2(current_fh, &resfh);
+	fh_dup2(current_fh, resfh);
 out:
-	fh_put(&resfh);
+	fh_put(resfh);
+	kfree(resfh);
 	return status;
 }
Signature · 1 line
1.7.5.4
author
#3 lore raw

PATCH 3/3 nfsd4: make sure set CB_PATH_DOWN sequence flag set

In reply to J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

Make sure this is set whenever there is no callback channel.

If a client does not set up a callback channel at all, then it will get this flag set from the very start. That's OK, it can just ignore the flag if it doesn't care. If a client does care, I think it's better to inform it of the problem as early as possible.

Reported-by Rick Macklem <[email protected]>
Signed-off-by J. Bruce Fields <[email protected]>
---
 fs/nfsd/nfs4callback.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
fs/nfsd/nfs4callback.c +1 -1
index 6f3ebb4..24b6bcf 100644
@@ -986,7 +986,7 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 
 	err = setup_callback_client(clp, &conn, ses);
 	if (err) {
-		warn_no_callback_path(clp, err);
+		nfsd4_mark_cb_down(clp, err);
 		return;
 	}
 	/* Yay, the callback channel's back! Restart any callbacks: */
Signature · 1 line
1.7.5.4