[PATCH v2 0/5] NFS: create blocklayout pipe per network namesapce context

[PATCH v2 0/5] NFS: create blocklayout pipe per network namesapce context

[Stanislav Kinsbursky]

This patch set was created in context of clone of git
branch: git://git.linux-nfs.org/projects/trondmy/nfs-2.6.git.
tag: v3.2

v2:
1) rebased on tag v3.2

This patch set depends on previous patch sets titled:
1) "SUNRPC: initial part of making pipefs work in net ns"
2) "SUNPRC: cleanup PipeFS for network-namespace-aware users"
3) "SUNRPC: make RPC clients use network-namespace-aware PipeFS routines"
4) "NFS: create clients and IDMAP pipes per network namespace"

This patch set is the final part of making SUNRPC PipeFS and it's users work in
network namespace context.


The following series consists of:

---

Stanislav Kinsbursky (5):
      NFS: handle blocklayout pipe PipeFS dentry by network namespace aware routines
      NFS: blocklayout pipe creation per network namespace context introduced
      NFS: blocklayout PipeFS notifier introduced
      NFS: remove RPC PipeFS mount point reference from blocklayout routines
      SUNRPC: kernel PipeFS mount point creation routines removed


 fs/nfs/blocklayout/blocklayout.c    |  160 ++++++++++++++++++++++++++++-------
 fs/nfs/blocklayout/blocklayout.h    |    3 -
 fs/nfs/blocklayout/blocklayoutdev.c |    5 +
 fs/nfs/blocklayout/blocklayoutdm.c  |    7 +-
 fs/nfs/inode.c                      |    1 
 fs/nfs/netns.h                      |    1 
 include/linux/sunrpc/rpc_pipe_fs.h  |    2 
 net/sunrpc/rpc_pipe.c               |   21 -----
 8 files changed, 139 insertions(+), 61 deletions(-)

-- 
Signature

[PATCH v2 1/5] NFS: handle blocklayout pipe PipeFS dentry by network namespace aware routines

[Stanislav Kinsbursky]

This patch makes blocklayout pipe dentry allocated and destroyed in network
namespace context by PipeFS network namespace aware routines.
Network namespace context is obtained from nfs_client structure.

Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>

---
 fs/nfs/blocklayout/blocklayout.c |   61 +++++++++++++++++++++++++++++++-------
 1 files changed, 49 insertions(+), 12 deletions(-)

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 2f4ede1..489f95c 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -965,10 +965,55 @@ static const struct rpc_pipe_ops bl_upcall_ops = {
 	.destroy_msg	= bl_pipe_destroy_msg,
 };
 
+static struct dentry *nfs4blocklayout_register_sb(struct super_block *sb,
+					    struct rpc_pipe *pipe)
+{
+	struct dentry *dir, *dentry;
+
+	dir = rpc_d_lookup_sb(sb, NFS_PIPE_DIRNAME);
+	if (dir == NULL)
+		return ERR_PTR(-ENOENT);
+	dentry = rpc_mkpipe_dentry(dir, "blocklayout", NULL, pipe);
+	dput(dir);
+	return dentry;
+}
+
+static void nfs4blocklayout_unregister_sb(struct super_block *sb,
+					  struct rpc_pipe *pipe)
+{
+	if (pipe->dentry)
+		rpc_unlink(pipe->dentry);
+}
+
+static struct dentry *nfs4blocklayout_register_net(struct net *net,
+						   struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+	struct dentry *dentry;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (!pipefs_sb)
+		return ERR_PTR(-ENOENT);
+	dentry = nfs4blocklayout_register_sb(pipefs_sb, pipe);
+	rpc_put_sb_net(net);
+	return dentry;
+}
+
+static void nfs4blocklayout_unregister_net(struct net *net,
+					   struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (pipefs_sb) {
+		nfs4blocklayout_unregister_sb(pipefs_sb, pipe);
+		rpc_put_sb_net(net);
+	}
+}
+
 static int __init nfs4blocklayout_init(void)
 {
 	struct vfsmount *mnt;
-	struct path path;
 	int ret;
 
 	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
@@ -984,21 +1029,13 @@ static int __init nfs4blocklayout_init(void)
 		ret = PTR_ERR(mnt);
 		goto out_remove;
 	}
-
-	ret = vfs_path_lookup(mnt->mnt_root,
-			      mnt,
-			      NFS_PIPE_DIRNAME, 0, &path);
-	if (ret)
-		goto out_putrpc;
-
 	bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0);
-	path_put(&path);
 	if (IS_ERR(bl_device_pipe)) {
 		ret = PTR_ERR(bl_device_pipe);
 		goto out_putrpc;
 	}
-	bl_device_pipe->dentry = rpc_mkpipe_dentry(path.dentry, "blocklayout",
-						   NULL, bl_device_pipe);
+	bl_device_pipe->dentry = nfs4blocklayout_register_net(&init_net,
+							      bl_device_pipe);
 	if (IS_ERR(bl_device_pipe->dentry)) {
 		ret = PTR_ERR(bl_device_pipe->dentry);
 		goto out_destroy_pipe;
@@ -1021,7 +1058,7 @@ static void __exit nfs4blocklayout_exit(void)
 	       __func__);
 
 	pnfs_unregister_layoutdriver(&blocklayout_type);
-	rpc_unlink(bl_device_pipe->dentry);
+	nfs4blocklayout_unregister_net(&init_net, bl_device_pipe);
 	rpc_destroy_pipe_data(bl_device_pipe);
 	rpc_put_mount();
 }

[PATCH v2 2/5] NFS: blocklayout pipe creation per network namespace context introduced

[Stanislav Kinsbursky]

This patch implements blocklayout pipe creation and registration per each
existent network namespace.
This was achived by registering NFS per-net operations, responsible for
blocklayout pipe allocation/register and unregister/destruction instead of
initialization and destruction of static "bl_device_pipe" pipe (this one was
removed).
Note, than pointer to network blocklayout pipe is stored in per-net "nfs_net"
structure, because allocating of one more per-net structure for blocklayout
module looks redundant.
This patch also changes dev_remove() function prototype (and all it's callers,
where it' requied) by adding network namespace pointer parameter, which is used
to discover proper blocklayout pipe for rpc_queue_upcall() call.

Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>

---
 fs/nfs/blocklayout/blocklayout.c    |   54 +++++++++++++++++++++++------------
 fs/nfs/blocklayout/blocklayout.h    |    3 +-
 fs/nfs/blocklayout/blocklayoutdev.c |    5 +++
 fs/nfs/blocklayout/blocklayoutdm.c  |    7 +++--
 fs/nfs/inode.c                      |    1 +
 fs/nfs/netns.h                      |    1 +
 6 files changed, 47 insertions(+), 24 deletions(-)

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 489f95c..ce76d05 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -46,7 +46,6 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Andy Adamson <andros@citi.umich.edu>");
 MODULE_DESCRIPTION("The NFSv4.1 pNFS Block layout driver");
 
-struct rpc_pipe *bl_device_pipe;
 wait_queue_head_t bl_wq;
 
 static void print_page(struct page *page)
@@ -1011,6 +1010,37 @@ static void nfs4blocklayout_unregister_net(struct net *net,
 	}
 }
 
+static int nfs4blocklayout_net_init(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct dentry *dentry;
+
+	nn->bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0);
+	if (IS_ERR(nn->bl_device_pipe))
+		return PTR_ERR(nn->bl_device_pipe);
+	dentry = nfs4blocklayout_register_net(net, nn->bl_device_pipe);
+	if (IS_ERR(dentry)) {
+		rpc_destroy_pipe_data(nn->bl_device_pipe);
+		return PTR_ERR(dentry);
+	}
+	nn->bl_device_pipe->dentry = dentry;
+	return 0;
+}
+
+static void nfs4blocklayout_net_exit(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	nfs4blocklayout_unregister_net(net, nn->bl_device_pipe);
+	rpc_destroy_pipe_data(nn->bl_device_pipe);
+	nn->bl_device_pipe = NULL;
+}
+
+static struct pernet_operations nfs4blocklayout_net_ops = {
+	.init = nfs4blocklayout_net_init,
+	.exit = nfs4blocklayout_net_exit,
+};
+
 static int __init nfs4blocklayout_init(void)
 {
 	struct vfsmount *mnt;
@@ -1029,24 +1059,12 @@ static int __init nfs4blocklayout_init(void)
 		ret = PTR_ERR(mnt);
 		goto out_remove;
 	}
-	bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0);
-	if (IS_ERR(bl_device_pipe)) {
-		ret = PTR_ERR(bl_device_pipe);
-		goto out_putrpc;
-	}
-	bl_device_pipe->dentry = nfs4blocklayout_register_net(&init_net,
-							      bl_device_pipe);
-	if (IS_ERR(bl_device_pipe->dentry)) {
-		ret = PTR_ERR(bl_device_pipe->dentry);
-		goto out_destroy_pipe;
-	}
+	ret = register_pernet_subsys(&nfs4blocklayout_net_ops);
+	if (ret)
+		goto out_remove;
 out:
 	return ret;
 
-out_destroy_pipe:
-	rpc_destroy_pipe_data(bl_device_pipe);
-out_putrpc:
-	rpc_put_mount();
 out_remove:
 	pnfs_unregister_layoutdriver(&blocklayout_type);
 	return ret;
@@ -1057,10 +1075,8 @@ static void __exit nfs4blocklayout_exit(void)
 	dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n",
 	       __func__);
 
+	unregister_pernet_subsys(&nfs4blocklayout_net_ops);
 	pnfs_unregister_layoutdriver(&blocklayout_type);
-	nfs4blocklayout_unregister_net(&init_net, bl_device_pipe);
-	rpc_destroy_pipe_data(bl_device_pipe);
-	rpc_put_mount();
 }
 
 MODULE_ALIAS("nfs-layouttype4-3");
diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h
index 046b513..10e0a62 100644
--- a/fs/nfs/blocklayout/blocklayout.h
+++ b/fs/nfs/blocklayout/blocklayout.h
@@ -37,6 +37,7 @@
 #include <linux/sunrpc/rpc_pipe_fs.h>
 
 #include "../pnfs.h"
+#include "../netns.h"
 
 #define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT)
 #define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)
@@ -50,6 +51,7 @@ struct pnfs_block_dev {
 	struct list_head		bm_node;
 	struct nfs4_deviceid		bm_mdevid;    /* associated devid */
 	struct block_device		*bm_mdev;     /* meta device itself */
+	struct net			*net;
 };
 
 enum exstate4 {
@@ -159,7 +161,6 @@ struct bl_msg_hdr {
 	u16 totallen; /* length of entire message, including hdr itself */
 };
 
-extern struct rpc_pipe *bl_device_pipe;
 extern wait_queue_head_t bl_wq;
 
 #define BL_DEVICE_UMOUNT               0x0 /* Umount--delete devices */
diff --git a/fs/nfs/blocklayout/blocklayoutdev.c b/fs/nfs/blocklayout/blocklayoutdev.c
index 949b624..94ed978 100644
--- a/fs/nfs/blocklayout/blocklayoutdev.c
+++ b/fs/nfs/blocklayout/blocklayoutdev.c
@@ -120,6 +120,8 @@ nfs4_blk_decode_device(struct nfs_server *server,
 	DECLARE_WAITQUEUE(wq, current);
 	struct bl_dev_msg *reply = &bl_mount_reply;
 	int offset, len, i, rc;
+	struct net *net = server->nfs_client->net;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
 
 	dprintk("%s CREATING PIPEFS MESSAGE\n", __func__);
 	dprintk("%s: deviceid: %s, mincount: %d\n", __func__, dev->dev_id.data,
@@ -146,7 +148,7 @@ nfs4_blk_decode_device(struct nfs_server *server,
 
 	dprintk("%s CALLING USERSPACE DAEMON\n", __func__);
 	add_wait_queue(&bl_wq, &wq);
-	rc = rpc_queue_upcall(bl_device_pipe, &msg);
+	rc = rpc_queue_upcall(nn->bl_device_pipe, &msg);
 	if (rc < 0) {
 		remove_wait_queue(&bl_wq, &wq);
 		rv = ERR_PTR(rc);
@@ -181,6 +183,7 @@ nfs4_blk_decode_device(struct nfs_server *server,
 
 	rv->bm_mdev = bd;
 	memcpy(&rv->bm_mdevid, &dev->dev_id, sizeof(struct nfs4_deviceid));
+	rv->net = net;
 	dprintk("%s Created device %s with bd_block_size %u\n",
 		__func__,
 		bd->bd_disk->disk_name,
diff --git a/fs/nfs/blocklayout/blocklayoutdm.c b/fs/nfs/blocklayout/blocklayoutdm.c
index 631f254..970490f 100644
--- a/fs/nfs/blocklayout/blocklayoutdm.c
+++ b/fs/nfs/blocklayout/blocklayoutdm.c
@@ -38,7 +38,7 @@
 
 #define NFSDBG_FACILITY         NFSDBG_PNFS_LD
 
-static void dev_remove(dev_t dev)
+static void dev_remove(struct net *net, dev_t dev)
 {
 	struct rpc_pipe_msg msg;
 	struct bl_dev_msg bl_umount_request;
@@ -48,6 +48,7 @@ static void dev_remove(dev_t dev)
 	};
 	uint8_t *dataptr;
 	DECLARE_WAITQUEUE(wq, current);
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
 
 	dprintk("Entering %s\n", __func__);
 
@@ -66,7 +67,7 @@ static void dev_remove(dev_t dev)
 	msg.len = sizeof(bl_msg) + bl_msg.totallen;
 
 	add_wait_queue(&bl_wq, &wq);
-	if (rpc_queue_upcall(bl_device_pipe, &msg) < 0) {
+	if (rpc_queue_upcall(nn->bl_device_pipe, &msg) < 0) {
 		remove_wait_queue(&bl_wq, &wq);
 		goto out;
 	}
@@ -93,7 +94,7 @@ static void nfs4_blk_metadev_release(struct pnfs_block_dev *bdev)
 		printk(KERN_ERR "%s nfs4_blkdev_put returns %d\n",
 				__func__, rv);
 
-	dev_remove(bdev->bm_mdev->bd_dev);
+	dev_remove(bdev->net, bdev->bm_mdev->bd_dev);
 }
 
 void bl_free_block_dev(struct pnfs_block_dev *bdev)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 9590da3..60cf71a 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1552,6 +1552,7 @@ static void nfsiod_stop(void)
 }
 
 int nfs_net_id;
+EXPORT_SYMBOL_GPL(nfs_net_id);
 
 static int nfs_net_init(struct net *net)
 {
diff --git a/fs/nfs/netns.h b/fs/nfs/netns.h
index 8c1f130..39ae4ca 100644
--- a/fs/nfs/netns.h
+++ b/fs/nfs/netns.h
@@ -6,6 +6,7 @@
 
 struct nfs_net {
 	struct cache_detail *nfs_dns_resolve;
+	struct rpc_pipe *bl_device_pipe;
 };
 
 extern int nfs_net_id;

[PATCH v2 3/5] NFS: blocklayout PipeFS notifier introduced

[Stanislav Kinsbursky]

This patch subscribes blocklayout pipes to RPC pipefs notifications. Notifier
is registering on blocklayout module load. This notifier callback is
responsible for creation/destruction of PipeFS blocklayout pipe dentry.
Note that no locking required in notifier callback because PipeFS superblock
pointer is passed as an argument from it's creation or destruction routine and
thus we can be sure about it's validity.

Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>

---
 fs/nfs/blocklayout/blocklayout.c |   50 +++++++++++++++++++++++++++++++++++++-
 1 files changed, 49 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index ce76d05..322d920 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -984,6 +984,48 @@ static void nfs4blocklayout_unregister_sb(struct super_block *sb,
 		rpc_unlink(pipe->dentry);
 }
 
+static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,
+			   void *ptr)
+{
+	struct super_block *sb = ptr;
+	struct net *net = sb->s_fs_info;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct dentry *dentry;
+	int ret = 0;
+
+	if (!try_module_get(THIS_MODULE))
+		return 0;
+
+	if (nn->bl_device_pipe == NULL) {
+		module_put(THIS_MODULE);
+		return 0;
+	}
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		dentry = nfs4blocklayout_register_sb(sb, nn->bl_device_pipe);
+		if (IS_ERR(dentry)) {
+			ret = PTR_ERR(dentry);
+			break;
+		}
+		nn->bl_device_pipe->dentry = dentry;
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		if (nn->bl_device_pipe->dentry)
+			nfs4blocklayout_unregister_sb(sb, nn->bl_device_pipe);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+static struct notifier_block nfs4blocklayout_block = {
+	.notifier_call = rpc_pipefs_event,
+};
+
 static struct dentry *nfs4blocklayout_register_net(struct net *net,
 						   struct rpc_pipe *pipe)
 {
@@ -1059,12 +1101,17 @@ static int __init nfs4blocklayout_init(void)
 		ret = PTR_ERR(mnt);
 		goto out_remove;
 	}
-	ret = register_pernet_subsys(&nfs4blocklayout_net_ops);
+	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
 	if (ret)
 		goto out_remove;
+	ret = register_pernet_subsys(&nfs4blocklayout_net_ops);
+	if (ret)
+		goto out_notifier;
 out:
 	return ret;
 
+out_notifier:
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
 out_remove:
 	pnfs_unregister_layoutdriver(&blocklayout_type);
 	return ret;
@@ -1075,6 +1122,7 @@ static void __exit nfs4blocklayout_exit(void)
 	dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n",
 	       __func__);
 
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
 	unregister_pernet_subsys(&nfs4blocklayout_net_ops);
 	pnfs_unregister_layoutdriver(&blocklayout_type);
 }

[PATCH v2 4/5] NFS: remove RPC PipeFS mount point reference from blocklayout routines

[Stanislav Kinsbursky]

This is a cleanup patch. We don't need this reference anymore, because
blocklayout pipes dentries now creates and destroys in per-net operations and
on PipeFS mount/umount notification.
Note that nfs4blocklayout_register_net() now returns 0 instead of -ENOENT in
case of PipeFS superblock absence. This is ok, because blocklayout pipe dentry
will be created on PipeFS mount event.

Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>

---
 fs/nfs/blocklayout/blocklayout.c |    9 +--------
 1 files changed, 1 insertions(+), 8 deletions(-)

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 322d920..2c70202 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -1034,7 +1034,7 @@ static struct dentry *nfs4blocklayout_register_net(struct net *net,
 
 	pipefs_sb = rpc_get_sb_net(net);
 	if (!pipefs_sb)
-		return ERR_PTR(-ENOENT);
+		return 0;
 	dentry = nfs4blocklayout_register_sb(pipefs_sb, pipe);
 	rpc_put_sb_net(net);
 	return dentry;
@@ -1085,7 +1085,6 @@ static struct pernet_operations nfs4blocklayout_net_ops = {
 
 static int __init nfs4blocklayout_init(void)
 {
-	struct vfsmount *mnt;
 	int ret;
 
 	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
@@ -1095,12 +1094,6 @@ static int __init nfs4blocklayout_init(void)
 		goto out;
 
 	init_waitqueue_head(&bl_wq);
-
-	mnt = rpc_get_mount();
-	if (IS_ERR(mnt)) {
-		ret = PTR_ERR(mnt);
-		goto out_remove;
-	}
 	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
 	if (ret)
 		goto out_remove;

[PATCH v2 5/5] SUNRPC: kernel PipeFS mount point creation routines removed

[Stanislav Kinsbursky]

This patch removes static rpc_mnt variable and its creation and destruction
routines, because they are not used anymore.

Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>

---
 include/linux/sunrpc/rpc_pipe_fs.h |    2 --
 net/sunrpc/rpc_pipe.c              |   21 ---------------------
 2 files changed, 0 insertions(+), 23 deletions(-)

diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h
index 7eb0160..a28d2de 100644
--- a/include/linux/sunrpc/rpc_pipe_fs.h
+++ b/include/linux/sunrpc/rpc_pipe_fs.h
@@ -90,8 +90,6 @@ void rpc_destroy_pipe_data(struct rpc_pipe *pipe);
 extern struct dentry *rpc_mkpipe_dentry(struct dentry *, const char *, void *,
 					struct rpc_pipe *);
 extern int rpc_unlink(struct dentry *);
-extern struct vfsmount *rpc_get_mount(void);
-extern void rpc_put_mount(void);
 extern int register_rpc_pipefs(void);
 extern void unregister_rpc_pipefs(void);
 
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index cd565d7..b53bb3b 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -18,7 +18,6 @@
 #include <linux/kernel.h>
 
 #include <asm/ioctls.h>
-#include <linux/fs.h>
 #include <linux/poll.h>
 #include <linux/wait.h>
 #include <linux/seq_file.h>
@@ -37,9 +36,6 @@
 
 #define NET_NAME(net)	((net == &init_net) ? " (init_net)" : "")
 
-static struct vfsmount *rpc_mnt __read_mostly;
-static int rpc_mount_count;
-
 static struct file_system_type rpc_pipe_fs_type;
 
 
@@ -450,23 +446,6 @@ struct rpc_filelist {
 	umode_t mode;
 };
 
-struct vfsmount *rpc_get_mount(void)
-{
-	int err;
-
-	err = simple_pin_fs(&rpc_pipe_fs_type, &rpc_mnt, &rpc_mount_count);
-	if (err != 0)
-		return ERR_PTR(err);
-	return rpc_mnt;
-}
-EXPORT_SYMBOL_GPL(rpc_get_mount);
-
-void rpc_put_mount(void)
-{
-	simple_release_fs(&rpc_mnt, &rpc_mount_count);
-}
-EXPORT_SYMBOL_GPL(rpc_put_mount);
-
 static int rpc_delete_dentry(const struct dentry *dentry)
 {
 	return 1;

Re: [PATCH v2 4/5] NFS: remove RPC PipeFS mount point reference from blocklayout routines

[Trond Myklebust]

On Tue, 2012-01-10 at 17:04 +0400, Stanislav Kinsbursky wrote: 
> This is a cleanup patch. We don't need this reference anymore, because
> blocklayout pipes dentries now creates and destroys in per-net operations and
> on PipeFS mount/umount notification.
> Note that nfs4blocklayout_register_net() now returns 0 instead of -ENOENT in
> case of PipeFS superblock absence. This is ok, because blocklayout pipe dentry
> will be created on PipeFS mount event.
> 
> Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
> 
> ---
>  fs/nfs/blocklayout/blocklayout.c |    9 +--------
>  1 files changed, 1 insertions(+), 8 deletions(-)
> 
> diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
> index 322d920..2c70202 100644
> --- a/fs/nfs/blocklayout/blocklayout.c
> +++ b/fs/nfs/blocklayout/blocklayout.c
> @@ -1034,7 +1034,7 @@ static struct dentry *nfs4blocklayout_register_net(struct net *net,
>  
>  	pipefs_sb = rpc_get_sb_net(net);
>  	if (!pipefs_sb)
> -		return ERR_PTR(-ENOENT);
> +		return 0;

This looks like a typo of some sort. The caller in
nfs4blocklayout_net_init() checks for IS_ERR(), not NULL. I'll just
remove this hunk...

> 	dentry = nfs4blocklayout_register_sb(pipefs_sb, pipe);
>  	rpc_put_sb_net(net);
>  	return dentry;
> @@ -1085,7 +1085,6 @@ static struct pernet_operations nfs4blocklayout_net_ops = {
>  
>  static int __init nfs4blocklayout_init(void)
>  {
> -	struct vfsmount *mnt;
>  	int ret;
>  
>  	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
> @@ -1095,12 +1094,6 @@ static int __init nfs4blocklayout_init(void)
>  		goto out;
>  
>  	init_waitqueue_head(&bl_wq);
> -
> -	mnt = rpc_get_mount();
> -	if (IS_ERR(mnt)) {
> -		ret = PTR_ERR(mnt);
> -		goto out_remove;
> -	}
>  	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
>  	if (ret)
>  		goto out_remove;
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Trond Myklebust
Linux NFS client maintainer

NetApp
Trond.Myklebust@netapp.com
www.netapp.com

Re: [PATCH v2 4/5] NFS: remove RPC PipeFS mount point reference from blocklayout routines

[Trond Myklebust]

On Tue, 2012-01-10 at 17:04 +0400, Stanislav Kinsbursky wrote: 
> This is a cleanup patch. We don't need this reference anymore, because
> blocklayout pipes dentries now creates and destroys in per-net operations and
> on PipeFS mount/umount notification.
> Note that nfs4blocklayout_register_net() now returns 0 instead of -ENOENT in
> case of PipeFS superblock absence. This is ok, because blocklayout pipe dentry
> will be created on PipeFS mount event.
> 
> Signed-off-by: Stanislav Kinsbursky <skinsbursky-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
> 
> ---
>  fs/nfs/blocklayout/blocklayout.c |    9 +--------
>  1 files changed, 1 insertions(+), 8 deletions(-)
> 
> diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
> index 322d920..2c70202 100644
> --- a/fs/nfs/blocklayout/blocklayout.c
> +++ b/fs/nfs/blocklayout/blocklayout.c
> @@ -1034,7 +1034,7 @@ static struct dentry *nfs4blocklayout_register_net(struct net *net,
>  
>  	pipefs_sb = rpc_get_sb_net(net);
>  	if (!pipefs_sb)
> -		return ERR_PTR(-ENOENT);
> +		return 0;

This looks like a typo of some sort. The caller in
nfs4blocklayout_net_init() checks for IS_ERR(), not NULL. I'll just
remove this hunk...

> 	dentry = nfs4blocklayout_register_sb(pipefs_sb, pipe);
>  	rpc_put_sb_net(net);
>  	return dentry;
> @@ -1085,7 +1085,6 @@ static struct pernet_operations nfs4blocklayout_net_ops = {
>  
>  static int __init nfs4blocklayout_init(void)
>  {
> -	struct vfsmount *mnt;
>  	int ret;
>  
>  	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
> @@ -1095,12 +1094,6 @@ static int __init nfs4blocklayout_init(void)
>  		goto out;
>  
>  	init_waitqueue_head(&bl_wq);
> -
> -	mnt = rpc_get_mount();
> -	if (IS_ERR(mnt)) {
> -		ret = PTR_ERR(mnt);
> -		goto out_remove;
> -	}
>  	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
>  	if (ret)
>  		goto out_remove;
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Trond Myklebust
Linux NFS client maintainer

NetApp
Trond.Myklebust-HgOvQuBEEgTQT0dZR+AlfA@public.gmane.org
www.netapp.com

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v2 4/5] NFS: remove RPC PipeFS mount point reference from blocklayout routines

[Trond Myklebust]

On Wed, 2012-01-11 at 13:33 -0500, Trond Myklebust wrote: 
> On Tue, 2012-01-10 at 17:04 +0400, Stanislav Kinsbursky wrote: 
> > This is a cleanup patch. We don't need this reference anymore, because
> > blocklayout pipes dentries now creates and destroys in per-net operations and
> > on PipeFS mount/umount notification.
> > Note that nfs4blocklayout_register_net() now returns 0 instead of -ENOENT in
> > case of PipeFS superblock absence. This is ok, because blocklayout pipe dentry
> > will be created on PipeFS mount event.
> > 
> > Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
> > 
> > ---
> >  fs/nfs/blocklayout/blocklayout.c |    9 +--------
> >  1 files changed, 1 insertions(+), 8 deletions(-)
> > 
> > diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
> > index 322d920..2c70202 100644
> > --- a/fs/nfs/blocklayout/blocklayout.c
> > +++ b/fs/nfs/blocklayout/blocklayout.c
> > @@ -1034,7 +1034,7 @@ static struct dentry *nfs4blocklayout_register_net(struct net *net,
> >  
> >  	pipefs_sb = rpc_get_sb_net(net);
> >  	if (!pipefs_sb)
> > -		return ERR_PTR(-ENOENT);
> > +		return 0;
> 
> This looks like a typo of some sort. The caller in
> nfs4blocklayout_net_init() checks for IS_ERR(), not NULL. I'll just
> remove this hunk...

Never mind. I see what's going on now... It should be NULL, and not 0
for coding style reasons, though...

-- 
Trond Myklebust
Linux NFS client maintainer

NetApp
Trond.Myklebust@netapp.com
www.netapp.com

Re: [PATCH v2 4/5] NFS: remove RPC PipeFS mount point reference from blocklayout routines

[Stanislav Kinsbursky]

11.01.2012 22:36, Trond Myklebust пишет:
> On Wed, 2012-01-11 at 13:33 -0500, Trond Myklebust wrote:
>> On Tue, 2012-01-10 at 17:04 +0400, Stanislav Kinsbursky wrote:
>>> This is a cleanup patch. We don't need this reference anymore, because
>>> blocklayout pipes dentries now creates and destroys in per-net operations and
>>> on PipeFS mount/umount notification.
>>> Note that nfs4blocklayout_register_net() now returns 0 instead of -ENOENT in
>>> case of PipeFS superblock absence. This is ok, because blocklayout pipe dentry
>>> will be created on PipeFS mount event.
>>>
>>> Signed-off-by: Stanislav Kinsbursky<skinsbursky@parallels.com>
>>>
>>> ---
>>>   fs/nfs/blocklayout/blocklayout.c |    9 +--------
>>>   1 files changed, 1 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
>>> index 322d920..2c70202 100644
>>> --- a/fs/nfs/blocklayout/blocklayout.c
>>> +++ b/fs/nfs/blocklayout/blocklayout.c
>>> @@ -1034,7 +1034,7 @@ static struct dentry *nfs4blocklayout_register_net(struct net *net,
>>>
>>>   	pipefs_sb = rpc_get_sb_net(net);
>>>   	if (!pipefs_sb)
>>> -		return ERR_PTR(-ENOENT);
>>> +		return 0;
>>
>> This looks like a typo of some sort. The caller in
>> nfs4blocklayout_net_init() checks for IS_ERR(), not NULL. I'll just
>> remove this hunk...
>
> Never mind. I see what's going on now... It should be NULL, and not 0
> for coding style reasons, though...
>

Yep, NULL of course, sorry.

-- 
Best regards,
Stanislav Kinsbursky