Commit 14b395e3 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge branch 'for-2.6.27' of git://linux-nfs.org/~bfields/linux 

* 'for-2.6.27' of git://linux-nfs.org/~bfields/linux: (51 commits)
  nfsd: nfs4xdr.c do-while is not a compound statement
  nfsd: Use C99 initializers in fs/nfsd/nfs4xdr.c
  lockd: Pass "struct sockaddr *" to new failover-by-IP function
  lockd: get host reference in nlmsvc_create_block() instead of callers
  lockd: minor svclock.c style fixes
  lockd: eliminate duplicate nlmsvc_lookup_host call from nlmsvc_lock
  lockd: eliminate duplicate nlmsvc_lookup_host call from nlmsvc_testlock
  lockd: nlm_release_host() checks for NULL, caller needn't
  file lock: reorder struct file_lock to save space on 64 bit builds
  nfsd: take file and mnt write in nfs4_upgrade_open
  nfsd: document open share bit tracking
  nfsd: tabulate nfs4 xdr encoding functions
  nfsd: dprint operation names
  svcrdma: Change WR context get/put to use the kmem cache
  svcrdma: Create a kmem cache for the WR contexts
  svcrdma: Add flush_scheduled_work to module exit function
  svcrdma: Limit ORD based on client's advertised IRD
  svcrdma: Remove unused wait q from svcrdma_xprt structure
  svcrdma: Remove unneeded spin locks from __svc_rdma_free
  svcrdma: Add dma map count and WARN_ON
  ...
parents 734b397c 5108b276

Documentation/filesystems/nfs-rdma.txt

+59 −44
Original line number Diff line number Diff line
@@ -5,7 +5,7 @@ 
################################################################################



 Author: NetApp and Open Grid Computing

 Date: April 15, 2008

 Date: May 29, 2008



Table of Contents

~~~~~~~~~~~~~~~~~
@@ -60,16 +60,18 @@ Installation 
    The procedures described in this document have been tested with

    distributions from Red Hat's Fedora Project (http://fedora.redhat.com/).



  - Install nfs-utils-1.1.1 or greater on the client

  - Install nfs-utils-1.1.2 or greater on the client



    An NFS/RDMA mount point can only be obtained by using the mount.nfs

    command in nfs-utils-1.1.1 or greater. To see which version of mount.nfs

    you are using, type:

    An NFS/RDMA mount point can be obtained by using the mount.nfs command in

    nfs-utils-1.1.2 or greater (nfs-utils-1.1.1 was the first nfs-utils

    version with support for NFS/RDMA mounts, but for various reasons we

    recommend using nfs-utils-1.1.2 or greater). To see which version of

    mount.nfs you are using, type:



    > /sbin/mount.nfs -V

    $ /sbin/mount.nfs -V



    If the version is less than 1.1.1 or the command does not exist,

    then you will need to install the latest version of nfs-utils.

    If the version is less than 1.1.2 or the command does not exist,

    you should install the latest version of nfs-utils.



    Download the latest package from:
@@ -77,22 +79,33 @@ Installation 


    Uncompress the package and follow the installation instructions.



    If you will not be using GSS and NFSv4, the installation process

    can be simplified by disabling these features when running configure:

    If you will not need the idmapper and gssd executables (you do not need

    these to create an NFS/RDMA enabled mount command), the installation

    process can be simplified by disabling these features when running

    configure:



    > ./configure --disable-gss --disable-nfsv4

    $ ./configure --disable-gss --disable-nfsv4



    For more information on this see the package's README and INSTALL files.

    To build nfs-utils you will need the tcp_wrappers package installed. For

    more information on this see the package's README and INSTALL files.



    After building the nfs-utils package, there will be a mount.nfs binary in

    the utils/mount directory. This binary can be used to initiate NFS v2, v3,

    or v4 mounts. To initiate a v4 mount, the binary must be called mount.nfs4.

    The standard technique is to create a symlink called mount.nfs4 to mount.nfs.

    or v4 mounts. To initiate a v4 mount, the binary must be called

    mount.nfs4.  The standard technique is to create a symlink called

    mount.nfs4 to mount.nfs.



    NOTE: mount.nfs and therefore nfs-utils-1.1.1 or greater is only needed

    This mount.nfs binary should be installed at /sbin/mount.nfs as follows:



    $ sudo cp utils/mount/mount.nfs /sbin/mount.nfs



    In this location, mount.nfs will be invoked automatically for NFS mounts

    by the system mount commmand.



    NOTE: mount.nfs and therefore nfs-utils-1.1.2 or greater is only needed

    on the NFS client machine. You do not need this specific version of

    nfs-utils on the server. Furthermore, only the mount.nfs command from

    nfs-utils-1.1.1 is needed on the client.

    nfs-utils-1.1.2 is needed on the client.



  - Install a Linux kernel with NFS/RDMA
@@ -156,8 +169,8 @@ Check RDMA and NFS Setup 
    this time. For example, if you are using a Mellanox Tavor/Sinai/Arbel

    card:



    > modprobe ib_mthca

    > modprobe ib_ipoib

    $ modprobe ib_mthca

    $ modprobe ib_ipoib



    If you are using InfiniBand, make sure there is a Subnet Manager (SM)

    running on the network. If your IB switch has an embedded SM, you can
@@ -166,7 +179,7 @@ Check RDMA and NFS Setup 


    If an SM is running on your network, you should see the following:



    > cat /sys/class/infiniband/driverX/ports/1/state

    $ cat /sys/class/infiniband/driverX/ports/1/state

    4: ACTIVE



    where driverX is mthca0, ipath5, ehca3, etc.
@@ -174,10 +187,10 @@ Check RDMA and NFS Setup 
    To further test the InfiniBand software stack, use IPoIB (this

    assumes you have two IB hosts named host1 and host2):



    host1> ifconfig ib0 a.b.c.x

    host2> ifconfig ib0 a.b.c.y

    host1> ping a.b.c.y

    host2> ping a.b.c.x

    host1$ ifconfig ib0 a.b.c.x

    host2$ ifconfig ib0 a.b.c.y

    host1$ ping a.b.c.y

    host2$ ping a.b.c.x



    For other device types, follow the appropriate procedures.
@@ -202,11 +215,11 @@ NFS/RDMA Setup 
    /vol0   192.168.0.47(fsid=0,rw,async,insecure,no_root_squash)

    /vol0   192.168.0.0/255.255.255.0(fsid=0,rw,async,insecure,no_root_squash)



    The IP address(es) is(are) the client's IPoIB address for an InfiniBand HCA or the

    cleint's iWARP address(es) for an RNIC.

    The IP address(es) is(are) the client's IPoIB address for an InfiniBand

    HCA or the cleint's iWARP address(es) for an RNIC.



    NOTE: The "insecure" option must be used because the NFS/RDMA client does not

    use a reserved port.

    NOTE: The "insecure" option must be used because the NFS/RDMA client does

    not use a reserved port.



 Each time a machine boots:
@@ -214,43 +227,45 @@ NFS/RDMA Setup 


    For InfiniBand using a Mellanox adapter:



    > modprobe ib_mthca

    > modprobe ib_ipoib

    > ifconfig ib0 a.b.c.d

    $ modprobe ib_mthca

    $ modprobe ib_ipoib

    $ ifconfig ib0 a.b.c.d



    NOTE: use unique addresses for the client and server



  - Start the NFS server



    If the NFS/RDMA server was built as a module (CONFIG_SUNRPC_XPRT_RDMA=m in kernel config),

    load the RDMA transport module:

    If the NFS/RDMA server was built as a module (CONFIG_SUNRPC_XPRT_RDMA=m in

    kernel config), load the RDMA transport module:



    > modprobe svcrdma

    $ modprobe svcrdma



    Regardless of how the server was built (module or built-in), start the server:

    Regardless of how the server was built (module or built-in), start the

    server:



    > /etc/init.d/nfs start

    $ /etc/init.d/nfs start



    or



    > service nfs start

    $ service nfs start



    Instruct the server to listen on the RDMA transport:



    > echo rdma 2050 > /proc/fs/nfsd/portlist

    $ echo rdma 2050 > /proc/fs/nfsd/portlist



  - On the client system



    If the NFS/RDMA client was built as a module (CONFIG_SUNRPC_XPRT_RDMA=m in kernel config),

    load the RDMA client module:

    If the NFS/RDMA client was built as a module (CONFIG_SUNRPC_XPRT_RDMA=m in

    kernel config), load the RDMA client module:



    > modprobe xprtrdma.ko

    $ modprobe xprtrdma.ko



    Regardless of how the client was built (module or built-in), issue the mount.nfs command:

    Regardless of how the client was built (module or built-in), use this

    command to mount the NFS/RDMA server:



    > /path/to/your/mount.nfs <IPoIB-server-name-or-address>:/<export> /mnt -i -o rdma,port=2050

    $ mount -o rdma,port=2050 <IPoIB-server-name-or-address>:/<export> /mnt



    To verify that the mount is using RDMA, run "cat /proc/mounts" and check the

    "proto" field for the given mount.

    To verify that the mount is using RDMA, run "cat /proc/mounts" and check

    the "proto" field for the given mount.



  Congratulations! You're using NFS/RDMA!

fs/lockd/svc.c

+13 −20
Original line number Diff line number Diff line
@@ -50,7 +50,7 @@ EXPORT_SYMBOL(nlmsvc_ops); 
static DEFINE_MUTEX(nlmsvc_mutex);

static unsigned int		nlmsvc_users;

static struct task_struct	*nlmsvc_task;

static struct svc_serv		*nlmsvc_serv;

static struct svc_rqst		*nlmsvc_rqst;

int				nlmsvc_grace_period;

unsigned long			nlmsvc_timeout;
@@ -194,20 +194,11 @@ lockd(void *vrqstp) 


		svc_process(rqstp);

	}



	flush_signals(current);

	if (nlmsvc_ops)

		nlmsvc_invalidate_all();

	nlm_shutdown_hosts();



	unlock_kernel();



	nlmsvc_task = NULL;

	nlmsvc_serv = NULL;



	/* Exit the RPC thread */

	svc_exit_thread(rqstp);



	return 0;

}
@@ -254,16 +245,15 @@ int 
lockd_up(int proto) /* Maybe add a 'family' option when IPv6 is supported ?? */

{

	struct svc_serv *serv;

	struct svc_rqst *rqstp;

	int		error = 0;



	mutex_lock(&nlmsvc_mutex);

	/*

	 * Check whether we're already up and running.

	 */

	if (nlmsvc_serv) {

	if (nlmsvc_rqst) {

		if (proto)

			error = make_socks(nlmsvc_serv, proto);

			error = make_socks(nlmsvc_rqst->rq_server, proto);

		goto out;

	}
@@ -288,9 +278,10 @@ lockd_up(int proto) /* Maybe add a 'family' option when IPv6 is supported ?? */ 
	/*

	 * Create the kernel thread and wait for it to start.

	 */

	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0]);

	if (IS_ERR(rqstp)) {

		error = PTR_ERR(rqstp);

	nlmsvc_rqst = svc_prepare_thread(serv, &serv->sv_pools[0]);

	if (IS_ERR(nlmsvc_rqst)) {

		error = PTR_ERR(nlmsvc_rqst);

		nlmsvc_rqst = NULL;

		printk(KERN_WARNING

			"lockd_up: svc_rqst allocation failed, error=%d\n",

			error);
@@ -298,16 +289,15 @@ lockd_up(int proto) /* Maybe add a 'family' option when IPv6 is supported ?? */ 
	}



	svc_sock_update_bufs(serv);

	nlmsvc_serv = rqstp->rq_server;



	nlmsvc_task = kthread_run(lockd, rqstp, serv->sv_name);

	nlmsvc_task = kthread_run(lockd, nlmsvc_rqst, serv->sv_name);

	if (IS_ERR(nlmsvc_task)) {

		error = PTR_ERR(nlmsvc_task);

		svc_exit_thread(nlmsvc_rqst);

		nlmsvc_task = NULL;

		nlmsvc_serv = NULL;

		nlmsvc_rqst = NULL;

		printk(KERN_WARNING

			"lockd_up: kthread_run failed, error=%d\n", error);

		svc_exit_thread(rqstp);

		goto destroy_and_out;

	}
@@ -346,6 +336,9 @@ lockd_down(void) 
		BUG();

	}

	kthread_stop(nlmsvc_task);

	svc_exit_thread(nlmsvc_rqst);

	nlmsvc_task = NULL;

	nlmsvc_rqst = NULL;

out:

	mutex_unlock(&nlmsvc_mutex);

}

fs/lockd/svc4proc.c

+3 −4
Original line number Diff line number Diff line
@@ -58,7 +58,6 @@ nlm4svc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, 
	return 0;



no_locks:

	if (host)

	nlm_release_host(host);

 	if (error)

		return error;
@@ -100,7 +99,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, 
		return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success;



	/* Now check for conflicting locks */

	resp->status = nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie);

	resp->status = nlmsvc_testlock(rqstp, file, host, &argp->lock, &resp->lock, &resp->cookie);

	if (resp->status == nlm_drop_reply)

		rc = rpc_drop_reply;

	else
@@ -146,7 +145,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, 
#endif



	/* Now try to lock the file */

	resp->status = nlmsvc_lock(rqstp, file, &argp->lock,

	resp->status = nlmsvc_lock(rqstp, file, host, &argp->lock,

					argp->block, &argp->cookie);

	if (resp->status == nlm_drop_reply)

		rc = rpc_drop_reply;

fs/lockd/svclock.c

+10 −23
Original line number Diff line number Diff line
@@ -180,6 +180,7 @@ nlmsvc_create_block(struct svc_rqst *rqstp, struct nlm_host *host, 
	struct nlm_block	*block;

	struct nlm_rqst		*call = NULL;



	nlm_get_host(host);

	call = nlm_alloc_call(host);

	if (call == NULL)

		return NULL;
@@ -358,10 +359,10 @@ nlmsvc_defer_lock_rqst(struct svc_rqst *rqstp, struct nlm_block *block) 
 */

__be32

nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,

			struct nlm_lock *lock, int wait, struct nlm_cookie *cookie)

	    struct nlm_host *host, struct nlm_lock *lock, int wait,

	    struct nlm_cookie *cookie)

{

	struct nlm_block	*block = NULL;

	struct nlm_host		*host;

	int			error;

	__be32			ret;
@@ -373,11 +374,6 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, 
				(long long)lock->fl.fl_end,

				wait);



	/* Create host handle for callback */

	host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len);

	if (host == NULL)

		return nlm_lck_denied_nolocks;



	/* Lock file against concurrent access */

	mutex_lock(&file->f_mutex);

	/* Get existing block (in case client is busy-waiting)
@@ -385,8 +381,7 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, 
	 */

	block = nlmsvc_lookup_block(file, lock);

	if (block == NULL) {

		block = nlmsvc_create_block(rqstp, nlm_get_host(host), file,

				lock, cookie);

		block = nlmsvc_create_block(rqstp, host, file, lock, cookie);

		ret = nlm_lck_denied_nolocks;

		if (block == NULL)

			goto out;
@@ -450,7 +445,6 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, 
out:

	mutex_unlock(&file->f_mutex);

	nlmsvc_release_block(block);

	nlm_release_host(host);

	dprintk("lockd: nlmsvc_lock returned %u\n", ret);

	return ret;

}
@@ -460,8 +454,8 @@ out: 
 */

__be32

nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,

		struct nlm_lock *lock, struct nlm_lock *conflock,

		struct nlm_cookie *cookie)

		struct nlm_host *host, struct nlm_lock *lock,

		struct nlm_lock *conflock, struct nlm_cookie *cookie)

{

	struct nlm_block 	*block = NULL;

	int			error;
@@ -479,16 +473,9 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file, 


	if (block == NULL) {

		struct file_lock *conf = kzalloc(sizeof(*conf), GFP_KERNEL);

		struct nlm_host	*host;



		if (conf == NULL)

			return nlm_granted;

		/* Create host handle for callback */

		host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len);

		if (host == NULL) {

			kfree(conf);

			return nlm_lck_denied_nolocks;

		}

		block = nlmsvc_create_block(rqstp, host, file, lock, cookie);

		if (block == NULL) {

			kfree(conf);

fs/lockd/svcproc.c

+3 −4
Original line number Diff line number Diff line
@@ -87,7 +87,6 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, 
	return 0;



no_locks:

	if (host)

	nlm_release_host(host);

	if (error)

		return error;
@@ -129,7 +128,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, 
		return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success;



	/* Now check for conflicting locks */

	resp->status = cast_status(nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie));

	resp->status = cast_status(nlmsvc_testlock(rqstp, file, host, &argp->lock, &resp->lock, &resp->cookie));

	if (resp->status == nlm_drop_reply)

		rc = rpc_drop_reply;

	else
@@ -176,7 +175,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, 
#endif



	/* Now try to lock the file */

	resp->status = cast_status(nlmsvc_lock(rqstp, file, &argp->lock,

	resp->status = cast_status(nlmsvc_lock(rqstp, file, host, &argp->lock,

					       argp->block, &argp->cookie));

	if (resp->status == nlm_drop_reply)

		rc = rpc_drop_reply;

CRA Git | Maintained and supported by SUSTech CRA and CCSE