Actually implement multiple pools.  On NUMA machines, allocate a svc_pool per
NUMA node; on SMP a svc_pool per CPU; otherwise a single global pool.  Enqueue
sockets on the svc_pool corresponding to the CPU on which the socket bh is run
(i.e.  the NIC interrupt CPU).  Threads have their cpu mask set to limit them
to the CPUs in the svc_pool that owns them.

This is the patch that allows an Altix to scale NFS traffic linearly
beyond 4 CPUs and 4 NICs.

Incorporates changes and feedback from Neil Brown, Trond Myklebust, and
Christoph Hellwig.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Currently knfsd keeps its own list of all nfsd threads in nfssvc.c; add a new
way of managing the list of all threads in a svc_serv.  Add
svc_create_pooled() to allow creation of a svc_serv whose threads are managed
by the sunrpc code.  Add svc_set_num_threads() to manage the number of threads
in a service, either per-pool or globally across the service.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Split out the list of idle threads and pending sockets from svc_serv into a
new svc_pool structure, and allocate a fixed number (in this patch, 1) of
pools per svc_serv.  The new structure contains a lock which takes over
several of the duties of svc_serv->sv_lock, which is now relegated to
protecting only sv_tempsocks, sv_permsocks, and sv_tmpcnt in svc_serv.

The point is to move the hottest fields out of svc_serv and into svc_pool,
allowing a following patch to arrange for a svc_pool per NUMA node or per CPU.
 This is a major step towards making the NFS server NUMA-friendly.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The SK_BUSY bit in svc_sock->sk_flags ensures that we do not attempt to
enqueue a socket twice.  Currently, setting and clearing the bit is protected
by svc_serv->sv_lock.  As I intend to reduce the data that the lock protects
so it's not held when svc_sock_enqueue() tests and sets SK_BUSY, that test and
set needs to be atomic.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Convert the svc_sock->sk_reserved variable from an int protected by
svc_serv->sv_lock, to an atomic.  This reduces (by 1) the number of places we
need to take the (effectively global) svc_serv->sv_lock.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Protect the svc_sock->sk_deferred list with a new lock svc_sock->sk_defer_lock
instead of svc_serv->sv_lock.  Using the more fine-grained lock reduces the
number of places we need to take the svc_serv lock.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Convert the svc_sock->sk_inuse counter from an int protected by
svc_serv->sv_lock, to an atomic.  This reduces the number of places we need to
take the (effectively global) svc_serv->sv_lock.
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Following are 11 patches from Greg Banks which combine to make knfsd more
Numa-aware.  They reduce hitting on 'global' data structures, and create some
data-structures that can be node-local.

knfsd threads are bound to a particular node, and the thread to handle a new
request is chosen from the threads that are attach to the node that received
the interrupt.

The distribution of threads across nodes can be controlled by a new file in
the 'nfsd' filesystem, though the default approach of an even spread is
probably fine for most sites.

Some (old) numbers that show the efficacy of these patches: N == number of
NICs == number of CPUs == nmber of clients.  Number of NUMA nodes == N/2

N	Throughput, MiB/s	CPU usage, % (max=N*100)
	Before	After		Before	After
	---	------	----		-----	-----
	4	312	435		350	228
	6	500	656		501	418
	8	562	804		690	589

This patch:

Move the aging of RPC/TCP connection sockets from the main svc_recv() loop to
a timer which uses a mark-and-sweep algorithm every 6 minutes.  This reduces
the amount of work that needs to be done in the main RPC loop and the length
of time we need to hold the (effectively global) svc_serv->sv_lock.

[akpm@osdl.org: cleanup]
Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

It isn't needed as it is available in rqstp->rq_server, and dropping it allows
some local vars to be dropped.

[akpm@osdl.org: build fix]
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Userspace should create and bind a socket (but not connectted) and write the
'fd' to portlist.  This will cause the nfs server to listen on that socket.

To close a socket, the name of the socket - as read from 'portlist' can be
written to 'portlist' with a preceding '-'.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This file will list all ports that nfsd has open.
Default when TCP enabled will be
   ipv4 udp 0.0.0.0 2049
   ipv4 tcp 0.0.0.0 2049

Later, the list of ports will be settable.

'portlist' chosen rather than 'ports', to avoid unnecessary confusion with
non-mainline patches which created 'ports' with different semantics.

[akpm@osdl.org: cleanups, build fix]
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

nfsd has some cleanup that it wants to do when the last thread exits, and
there will shortly be some more.  So collect this all into one place and
define a callback for an rpc service to call when the service is about to be
destroyed.

[akpm@osdl.org: cleanups, build fix]
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NGreg Banks <gnb@melbourne.sgi.com>
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This is mostly included for parity with dec_nlink(), where we will have some
more hooks.  This one should stay pretty darn straightforward for now.
Signed-off-by: NDave Hansen <haveblue@us.ibm.com>
Acked-by: NChristoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

coverity spotted this one as possible dereference in the dprintk(),
but since there is only one caller of svc_create_socket(), which always
passes a valid sin, we dont need this check.
Signed-off-by: NEric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

pure s/u32/__be32/

[AV: large part based on Alexey's patches]
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

* add svc_getnl():
	Take network-endian value from buffer, convert to host-endian
	and return it.
* add svc_putnl():
	Take host-endian value, convert to network-endian and put it
	into a buffer.
* annotate svc_getu32()/svc_putu32() as dealing with network-endian.
* convert to svc_getnl(), svc_putnl().

[AV: in large part it's a carved-up Alexey's patch]
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This eliminates the i_blksize field from struct inode.  Filesystems that want
to provide a per-inode st_blksize can do so by providing their own getattr
routine instead of using the generic_fillattr() function.

Note that some filesystems were providing pretty much random (and incorrect)
values for i_blksize.

[bunk@stusta.de: cleanup]
[akpm@osdl.org: generic_fillattr() fix]
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

* Rougly half of callers already do it by not checking return value
* Code in drivers/acpi/osl.c does the following to be sure:

	(void)kmem_cache_destroy(cache);

* Those who check it printk something, however, slab_error already printed
  the name of failed cache.
* XFS BUGs on failed kmem_cache_destroy which is not the decision
  low-level filesystem driver should make. Converted to ignore.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch stop rpc_mkpipe from create S_IFSOCK nodes what don't
have associated sk buffers attached (which causes SELinux to oops
during NFSv4 mounts). Instead the S_IFIFO mode bit is set which
probably make more sense and seems to work just fine during
my connectathon and fsx testing...
Signed-off-by: NSteve Dickson <steved@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

There is no guarantee that the parent task still exists when we exit from
the portmapper. Save the xprt instead.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

In a subsequent patch, this will allow the portmapper to take a reference
to the rpc_xprt for which it is updating the port number, fixing an Oops.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

- Ensure that the task aborts the RPC call only when it has actually timed out.
 - Ensure that req->rq_majortimeo is initialised correctly.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

In case of any of the above errors occuring, delay for 3 seconds, then
handle as if it were a timeout error.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Doing so prevents stuff like call_encode() from working correctly.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

This patch is optional.

It has been suggested that the RPC client internal functions used by upper
layer protocols (such as NFS) be exported via EXPORT_SYMBOL_GPL.  This
patch does that.

Test plan:
Compile kernel with CONFIG_NFS enabled as a module.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The two function call API for creating a new RPC client is now obsolete.
Remove it.

Also, remove an unnecessary check to see whether the caller is capable of
using privileged network services.  The kernel RPC client always uses a
privileged ephemeral port by default; callers are responsible for checking
the authority of users to make use of any RPC service, or for specifying
that a nonprivileged port is acceptable.

Test plan:
Repeated runs of Connectathon locking suite.  Check network trace to ensure
correctness of NLM requests and replies.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Replace xprt_create_proto/rpc_create_client calls in pmap_clnt.c with new
rpc_create() API.

Test plan:
Repeated runs of Connectathon locking suite.  Check network trace for
proper PMAP calls and replies.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Prepare for more generic transport endpoint handling needed by transports
that might use different forms of addressing, such as IPv6.

Introduce a single function call to replace the two-call
xprt_create_proto/rpc_create_client API.  Define a new rpc_create_args
structure that allows callers to pass in remote endpoint addresses of
varying length.

Test-plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

IPv6 addresses are big (128 bytes).  Now that no RPC client consumers treat
the addr field in rpc_xprt structs as an opaque, and access it only via the
API calls, we can safely widen the field in the rpc_xprt struct to
accomodate larger addresses.

Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Hide the details of how the RPC client stores remote peer addresses from
the RPC pipefs implementation.

Test plan:
Connectathon with Kerberos 5 authentication.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Provide an API for formatting the remote peer address for printing without
exposing its internal structure.  The address could be dynamic, so we
support a function call to get the address rather than reading it straight
out of a structure.

Test-plan:
Destructive testing (unplugging the network temporarily).  Probably need
to rig a server where certain services aren't running, or that returns an
error for some typical operation.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Add a new method to the transport switch API to provide a way to convert
the opaque contents of xprt->addr to a human-readable string.

Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

include/linux/sunrpc/clnt.h already includes include/linux/sunrpc/xprt.h.
We can remove xprt.h from source files that already include clnt.h.
Likewise include/linux/sunrpc/timer.h.

Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Hide the details of how the RPC client stores remote peer addresses from
the RPC portmapper.

Test plan:
Destructive testing (unplugging the network temporarily).  Connectathon
with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Provide an API for retrieving the remote peer address without allowing
direct access to the rpc_xprt struct.

Test-plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Introduce a clean transport switch API for plugging in different types of
rpcbind mechanisms.  For instance, rpcbind can cleanly replace the
existing portmapper client, or a transport can choose to implement RPC
binding any way it likes.

Test plan:
Destructive testing (unplugging the network temporarily).  Connectathon
with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
Probably need to rig a server where certain services aren't running, or
that returns an error for some typical operation.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The previous patches removed the last user of RPC child tasks, so we can
remove support for child tasks from net/sunrpc/sched.c now.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>