sunrpc: "Move close processing to a single place"
(d7979ae4) moved the
close processing before the recvfrom method. This may
cause the close processing never to execute. So this
patch moves it to the right place.
Signed-off-by: NWei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

lock_kernel() in knfsd was replaced with a mutex. The later
commit 03cf6c9f ("knfsd:
add file to export stats about nfsd pools") did not follow
that change. This patch fixes the issue.

Also move the get and put of nfsd_serv to the open and close methods
(instead of start and stop methods) to allow atomic check and increment
of reference count in the open method (where we can still return an
error).
Signed-off-by: NRyusei Yamaguchi <mandel59@gmail.com>
Signed-off-by: NIsaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: NYOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Cc: Greg Banks <gnb@fmeh.org>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

* Remove smp_lock.h from files which don't need it (including some headers!)
* Add smp_lock.h to files which do need it
* Make smp_lock.h include conditional in hardirq.h
  It's needed only for one kernel_locked() usage which is under CONFIG_PREEMPT

  This will make hardirq.h inclusion cheaper for every PREEMPT=n config
  (which includes allmodconfig/allyesconfig, BTW)
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The svc_xprt_names() function can overflow its buffer if it's so near
the end of the passed in buffer that the "name too long" string still
doesn't fit.  Of course, it could never tell if it was near the end
of the passed in buffer, since its only caller passes in zero as the
buffer length.

Let's make this API a little safer.

Change svc_xprt_names() so it *always* checks for a buffer overflow,
and change its only caller to pass in the correct buffer length.

If svc_xprt_names() does overflow its buffer, it now fails with an
ENAMETOOLONG errno, instead of trying to write a message at the end
of the buffer.  I don't like this much, but I can't figure out a clean
way that's always safe to return some of the names, *and* an
indication that the buffer was not long enough.

The displayed error when doing a 'cat /proc/fs/nfsd/portlist' is
"File name too long".
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Fix the following sparse warnings in net/sunrpc/svc_xprt.c.

  warning: symbol 'svc_recv' was not declared. Should it be static?
  warning: symbol 'svc_drop' was not declared. Should it be static?
  warning: symbol 'svc_send' was not declared. Should it be static?
  warning: symbol 'svc_close_all' was not declared. Should it be static?
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

On an NFSv4.1 server cache miss that causes an upcall, NFS4ERR_DELAY will be
returned. It is up to the NFSv4.1 client to resend only the operations that
have not been processed.

Initialize rq_usedeferral to 1 in svc_process(). It sill be turned off in
nfsd4_proc_compound() only when NFSv4.1 Sessions are used.

Note: this isn't an adequate solution on its own. It's acceptable as a way
to get some minimal 4.1 up and working, but we're going to have to find a
way to avoid returning DELAY in all common cases before 4.1 can really be
considered ready.
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NBenny Halevy <bhalevy@panasas.com>
[nfsd41: reverse rq_nodeferral negative logic]
Signed-off-by: NBenny Halevy <bhalevy@panasas.com>
[sunrpc: initialize rq_usedeferral]
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NBenny Halevy <bhalevy@panasas.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

The sv_family field is going away.  Pass a protocol family argument to
svc_create_xprt() instead of extracting the family from the passed-in
svc_serv struct.

Again, as this is a listener socket and not an address, we make this
new argument an "int" protocol family, instead of an "sa_family_t."
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Clean up: add documentating comment and use appropriate data types for
svc_find_xprt()'s arguments.

This also eliminates a mixed sign comparison: @port was an int, while
the return value of svc_xprt_local_port() is an unsigned short.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Add /proc/fs/nfsd/pool_stats to export to userspace various
statistics about the operation of rpc server thread pools.

This patch is based on a forward-ported version of
knfsd-add-pool-thread-stats which has been shipping in the SGI
"Enhanced NFS" product since 2006 and which was previously
posted:

http://article.gmane.org/gmane.linux.nfs/10375

It has also been updated thus:

 * moved EXPORT_SYMBOL() to near the function it exports
 * made the new struct struct seq_operations const
 * used SEQ_START_TOKEN instead of ((void *)1)
 * merged fix from SGI PV 990526 "sunrpc: use dprintk instead of
   printk in svc_pool_stats_*()" by Harshula Jayasuriya.
 * merged fix from SGI PV 964001 "Crash reading pool_stats before
   nfsds are started".
Signed-off-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NHarshula Jayasuriya <harshula@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Avoid overloading the CPU scheduler with enormous load averages
when handling high call-rate NFS loads.  When the knfsd bottom half
is made aware of an incoming call by the socket layer, it tries to
choose an nfsd thread and wake it up.  As long as there are idle
threads, one will be woken up.

If there are lot of nfsd threads (a sensible configuration when
the server is disk-bound or is running an HSM), there will be many
more nfsd threads than CPUs to run them.  Under a high call-rate
low service-time workload, the result is that almost every nfsd is
runnable, but only a handful are actually able to run.  This situation
causes two significant problems:

1. The CPU scheduler takes over 10% of each CPU, which is robbing
   the nfsd threads of valuable CPU time.

2. At a high enough load, the nfsd threads starve userspace threads
   of CPU time, to the point where daemons like portmap and rpc.mountd
   do not schedule for tens of seconds at a time.  Clients attempting
   to mount an NFS filesystem timeout at the very first step (opening
   a TCP connection to portmap) because portmap cannot wake up from
   select() and call accept() in time.

Disclaimer: these effects were observed on a SLES9 kernel, modern
kernels' schedulers may behave more gracefully.

The solution is simple: keep in each svc_pool a counter of the number
of threads which have been woken but have not yet run, and do not wake
any more if that count reaches an arbitrary small threshold.

Testing was on a 4 CPU 4 NIC Altix using 4 IRIX clients, each with 16
synthetic client threads simulating an rsync (i.e. recursive directory
listing) workload reading from an i386 RH9 install image (161480
regular files in 10841 directories) on the server.  That tree is small
enough to fill in the server's RAM so no disk traffic was involved.
This setup gives a sustained call rate in excess of 60000 calls/sec
before being CPU-bound on the server.  The server was running 128 nfsds.

Profiling showed schedule() taking 6.7% of every CPU, and __wake_up()
taking 5.2%.  This patch drops those contributions to 3.0% and 2.2%.
Load average was over 120 before the patch, and 20.9 after.

This patch is a forward-ported version of knfsd-avoid-nfsd-overload
which has been shipping in the SGI "Enhanced NFS" product since 2006.
It has been posted before:

http://article.gmane.org/gmane.linux.nfs/10374Signed-off-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

A race between svc_revisit and svc_delete_xprt can result in
deferred requests holding references on a transport that can never be
recovered because dead transports are not enqueued for subsequent
processing.

Check for XPT_DEAD in revisit to clean up completing deferrals on a dead
transport and sweep a transport's deferred queue to do the same for queued
but unprocessed deferrals.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

The rqstp structure has a pointer to a svc_deferred_req record
that is allocated when requests are deferred. This record is common
to all transports and can be freed in common code.

Move the kfree of the rq_deferred to the common svc_xprt_release
function.

This also fixes a memory leak in the RDMA transport which does not
kfree the dr structure in it's version of the xpo_release_rqst callback.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

svc_check_conn_limits() attempts to prevent denial of service attacks
by having the service close old connections once it reaches a
threshold. This threshold is based on the number of threads in the
service:

	(serv->sv_nrthreads + 3) * 20

Once we reach this, we drop the oldest connections and a printk pops
to warn the admin that they should increase the number of threads.

Increasing the number of threads isn't an option however for services
like lockd. We don't want to eliminate this check entirely for such
services but we need some way to increase this limit.

This patch adds a sv_maxconn field to the svc_serv struct. When it's
set to 0, we use the current method to calculate the max number of
connections. RPC services can then set this on an as-needed basis.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Acked-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Teach svc_create_xprt() to use the correct ANY address for AF_INET6 based
RPC services.

No caller uses AF_INET6 yet.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

This cosmetic patch removes unused header files that svc_xprt.c
inherited from svcsock.c
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>

Remove a redundant check for the XPT_DEAD bit in the svc_xprt_enqueue
function. This same bit is checked below while holding the pool lock
and prints a debug message if found to be dead.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>

svc_recv() calls alloc_page(), and if it fails it does a 500ms
uninterruptible sleep and then reattempts. There doesn't seem to be any
real reason for this to be uninterruptible, so change it to an
interruptible sleep. Also check for kthread_stop() and signalled() after
setting the task state to avoid races that might lead to sleeping after
kthread_stop() wakes up the task.

I've done some very basic smoke testing with this, but obviously it's
hard to test the actual changes since this all depends on an
alloc_page() call failing.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

When using kthreads that call into svc_recv, we want to make sure that
they do not block there for a long time when we're trying to take down
the kthread.

This patch changes svc_recv() to check kthread_should_stop() at the same
places that it checks to see if it's signalled(). Also check just before
svc_recv() tries to schedule(). By making sure that we check it just
after setting the task state we can avoid having to use any locking or
signalling to ensure it doesn't block for a long time.

There's still a chance of a 500ms sleep if alloc_page() fails, but
that should be a rare occurrence and isn't a terribly long time in
the context of a kthread being taken down.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

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

Do it for the server code...
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Create a transport independent version of the svc_sock_names function.

The toclose capability of the svc_sock_names service can be implemented
using the svc_xprt_find and svc_xprt_close services.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Update the write handler for the portlist file to allow creating new
listening endpoints on a transport. The general form of the string is:

<transport_name><space><port number>

For example:

echo "tcp 2049" > /proc/fs/nfsd/portlist

This is intended to support the creation of a listening endpoint for
RDMA transports without adding #ifdef code to the nfssvc.c file.

Transports can also be removed as follows:

'-'<transport_name><space><port number>

For example:

echo "-tcp 2049" > /proc/fs/nfsd/portlist

Attempting to add a listener with an invalid transport string results
in EPROTONOSUPPORT and a perror string of "Protocol not supported".

Attempting to remove an non-existent listener (.e.g. bad proto or port)
results in ENOTCONN and a perror string of
"Transport endpoint is not connected"
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Add a new svc function that allows a service to query whether a
transport instance has already been created. This is used in lockd
to determine whether or not a transport needs to be created when
a lockd instance is brought up.

Specifying 0 for the address family or port is effectively a wild-card,
and will result in matching the first transport in the service's list
that has a matching class name.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Add a file that when read lists the set of registered svc
transports.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Some transports have a header in front of the RPC header. The current
defer/revisit processing considers only the iov_len and arg_len to
determine how much to back up when saving the original request
to revisit. Add a field to the rqstp structure to save the size
of the transport header so svc_defer can correctly compute
the start of a request.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

This functionally trivial patch moves all of the transport independent
functions from the svcsock.c file to the transport independent svc_xprt.c
file.

In addition the following formatting changes were made:
- White space cleanup
- Function signatures on single line
- The inline directive was removed
- Lines over 80 columns were reformatted
- The term 'socket' was changed to 'transport' in comments
- The SMP comment was moved and updated.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Move the svc transport list logic into common transport creation code.
Refactor this code path to make the flow of control easier to read.

Move the setting and clearing of the BUSY_BIT during transport creation
to common code.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

This patch moves the transport sockaddr to the svc_xprt
structure.  Convenience functions are added to set and
get the local and remote addresses of a transport from
the transport provider as well as determine the length
of a sockaddr.

A transport is responsible for setting the xpt_local
and xpt_remote addresses in the svc_xprt structure as
part of transport creation and xpo_accept processing. This
cannot be done in a generic way and in fact varies
between TCP, UDP and RDMA. A set of xpo_ functions
(e.g. getlocalname, getremotename) could have been
added but this would have resulted in additional
caching and copying of the addresses around.  Note that
the xpt_local address should also be set on listening
endpoints; for TCP/RDMA this is done as part of
endpoint creation.

For connected transports like TCP and RDMA, the addresses
never change and can be set once and copied into the
rqstp structure for each request. For UDP, however, the
local and remote addresses may change for each request. In
this case, the address information is obtained from the
UDP recvmsg info and copied into the rqstp structure from
there.

A svc_xprt_local_port function was also added that returns
the local port given a transport. This is used by
svc_create_xprt when returning the port associated with
a newly created transport, and later when creating a
generic find transport service to check if a service is
already listening on a given port.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

This patch moves the transport independent sk_deferred list to the svc_xprt
structure and updates the svc_deferred_req structure to keep pointers to
svc_xprt's directly. The deferral processing code is also moved out of the
transport dependent recvfrom functions and into the generic svc_recv path.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Move the authinfo cache to svc_xprt. This allows both the TCP and RDMA
transports to share this logic. A flag bit is used to determine if
auth information is to be cached or not. Previously, this code looked
at the transport protocol.

I've also changed the spin_lock/unlock logic so that a lock is not taken for
transports that are not caching auth info.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Move the sk_mutex field to the transport independent svc_xprt structure.
Now all the fields that svc_send touches are transport neutral. Change the
svc_send function to use the transport independent svc_xprt directly instead
of the transport dependent svc_sock structure.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Move sk_list and sk_ready to svc_xprt. This involves close because these
lists are walked by svcs when closing all their transports. So I combined
the moving of these lists to svc_xprt with making close transport independent.

The svc_force_sock_close has been changed to svc_close_all and takes a list
as an argument. This removes some svc internals knowledge from the svcs.

This code races with module removal and transport addition.

Thanks to Simon Holm Thøgersen for a compile fix.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>
Cc: Simon Holm Thøgersen <odie@cs.aau.dk>

This is another incremental change that moves transport independent
fields from svc_sock to the svc_xprt structure. The changes
should be functionally null.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

Change the atomic_t reference count to a kref and move it to the
transport indepenent svc_xprt structure. Change the reference count
wrapper names to be generic.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

The svc_create_xprt function is a transport independent version
of the svc_makesock function.

Since transport instance creation contains transport dependent and
independent components, add an xpo_create transport function. The
transport implementation of this function allocates the memory for the
endpoint, implements the transport dependent initialization logic, and
calls svc_xprt_init to initialize the transport independent field (svc_xprt)
in it's data structure.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>

The transport class (svc_xprt_class) represents a type of transport, e.g.
udp, tcp, rdma.  A transport class has a unique name and a set of transport
operations kept in the svc_xprt_ops structure.

A transport class can be dynamically registered and unregisterd. The
svc_xprt_class represents the module that implements the transport
type and keeps reference counts on the module to avoid unloading while
there are active users.

The endpoint (svc_xprt) is a generic, transport independent endpoint that can
be used to send and receive data for an RPC service. It inherits it's
operations from the transport class.

A transport driver module registers and unregisters itself with svc sunrpc
by calling svc_reg_xprt_class, and svc_unreg_xprt_class respectively.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Acked-by: NNeil Brown <neilb@suse.de>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NGreg Banks <gnb@sgi.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>