The hash.h hash_long function, when used on a 64 bit machine, ignores many
of the middle-order bits. (The prime chosen it too bit-sparse).
IP addresses for clients of an NFS server are very likely to differ only in
the low-order bits. As addresses are stored in network-byte-order, these
bits become middle-order bits in a little-endian 64bit 'long', and so do
not contribute to the hash. Thus you can have the situation where all
clients appear on one hash chain.
So, until hash_long is fixed (or maybe forever), us a hash function that
works well on IP addresses - xor the bytes together.
Thanks to "Iozone" <capps@iozone.org> for identifying this problem.
Cc: "Iozone" <capps@iozone.org>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I noticed that some of 'struct proto_ops' used in the kernel may share
a cache line used by locks or other heavily modified data. (default
linker alignement is 32 bytes, and L1_CACHE_LINE is 64 or 128 at
least)
This patch makes sure a 'struct proto_ops' can be declared as const,
so that all cpus can share all parts of it without false sharing.
This is not mandatory : a driver can still use a read/write structure
if it needs to (and eventually a __read_mostly)
I made a global stubstitute to change all existing occurences to make
them const.
This should reduce the possibility of false sharing on SMP, and
speedup some socket system calls.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gss_create_upcall() should not error just because rpc.gssd closed the
pipe on its end. Instead, it should requeue the pending requests and then
retry.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The elements on rpci->in_upcall are tracked by the filp->private_data,
which will ensure that they get released when the file is closed.
The exception is if rpc_close_pipes() gets called first, since that
sets rpci->ops to NULL.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
In __rpc_purge_upcall (net/sunrpc/rpc_pipe.c), the newer code to clean up
the in_upcall list has a typo.
Thanks to Vince Busam <vbusam@google.com> for spotting this!
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Being kernel-threads, nfsd servers don't get pre-empted (depending on
CONFIG). If there is a steady stream of NFS requests that can be served
from cache, an nfsd thread may hold on to a cpu indefinitely, which isn't
very friendly.
So it is good to have a cond_resched in there (just before looking for a
new request to serve), to make sure we play nice.
Signed-off-by: Neil Brown <neilb@suse.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Here is the patch that introduces the generic skb_checksum_complete
which also checks for hardware RX checksum faults. If that happens,
it'll call netdev_rx_csum_fault which currently prints out a stack
trace with the device name. In future it can turn off RX checksum.
I've converted every spot under net/ that does RX checksum checks to
use skb_checksum_complete or __skb_checksum_complete with the
exceptions of:
* Those places where checksums are done bit by bit. These will call
netdev_rx_csum_fault directly.
* The following have not been completely checked/converted:
ipmr
ip_vs
netfilter
dccp
This patch is based on patches and suggestions from Stephen Hemminger
and David S. Miller.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the call to xprt_transmit() fails due to socket buffer space
exhaustion, we do not need to re-encode the RPC message when we
loop back through call_transmit.
Re-encoding can actually end up triggering the WARN_ON() in
call_decode() if we re-encode something like a read() request and
auth->au_rslack has changed.
It can also cause us to increment the RPCSEC_GSS sequence number
beyond the limits of the allowed window.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
->permission and ->lookup have a struct nameidata * argument these days to
pass down lookup intents. Unfortunately some callers of lookup_hash don't
actually pass this one down. For lookup_one_len() we don't have a struct
nameidata to pass down, but as this function is a library function only
used by filesystem code this is an acceptable limitation. All other
callers should pass down the nameidata, so this patch changes the
lookup_hash interface to only take a struct nameidata argument and derives
the other two arguments to __lookup_hash from it. All callers already have
the nameidata argument available so this is not a problem.
At the same time I'd like to deprecate the lookup_hash interface as there
are better exported interfaces for filesystem usage. Before it can
actually be removed I need to fix up rpc_pipefs.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Ram Pai <linuxram@us.ibm.com>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Al Viro <viro@ftp.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
From: Jesper Juhl <jesper.juhl@gmail.com>
This is the net/ part of the big kfree cleanup patch.
Remove pointless checks for NULL prior to calling kfree() in net/.
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Arnaldo Carvalho de Melo <acme@conectiva.com.br>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
If an RPC socket is serving multiple programs, then the pg_authenticate of
the first program in the list is called, instead of pg_authenticate for the
program to be run.
This does not cause a problem with any programs in the current kernel, but
could confuse future code.
Also set pg_authenticate for nfsd_acl_program incase it ever gets used.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix some dprintk's so that NLM, NFS client, and RPC client compile
cleanly if CONFIG_SYSCTL is disabled.
Test plan:
Compile kernel with CONFIG_NFS enabled and CONFIG_SYSCTL disabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The sunrpc module should build properly even when CONFIG_SYSCTL is
disabled.
Reported by Jan-Benedict Glaw.
Test plan:
Compile kernel with CONFIG_NFS as a module and built-in, and CONFIG_SYSCTL
enabled and disabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This patch rewrites various occurences of &sg[0] where sg is an array
of length one to simply sg.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch uses sg_set_buf/sg_init_one in some places where it was
duplicated.
Signed-off-by: David Hardeman <david@2gen.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Greg KH <greg@kroah.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Convert nanoseconds to microseconds correctly.
Spotted by Steve Dickson <SteveD@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Not only are the qop parameters that are passed around throughout the gssapi
unused by any currently implemented mechanism, but there appears to be some
doubt as to whether they will ever be used. Let's just kill them off for now.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add support for privacy to the krb5 rpcsec_gss mechanism.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The code this was originally derived from processed wrap and mic tokens using
the same functions. This required some contortions, and more would be required
with the addition of xdr_buf's, so it's better to separate out the two code
paths.
In preparation for adding privacy support, remove the last vestiges of the
old wrap token code.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Factor out some code that will be shared by privacy crypto routines
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add the code to the client side to handle privacy. This is dead code until
we actually add privacy support to krb5.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Various xdr encode routines use au_rslack to guess where the reply argument
will end up, so we can set up the xdr_buf to recieve data into the right place
for zero copy.
Currently we calculate the au_rslack estimate when we check the verifier.
Normally this only depends on the verifier size. In the integrity case we add
a few bytes to allow for a length and sequence number.
It's a bit simpler to calculate only the verifier size when we check the
verifier, and delay the full calculation till we unwrap.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
For privacy we need to allocate extra pages to hold encrypted page data when
wrapping requests. This allocation may fail, and we handle that case by
waiting and retrying.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
For privacy, we need to allocate pages to store the encrypted data (passed
in pages can't be used without the risk of corrupting data in the page cache).
So we need a way to free that memory after the request has been transmitted.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add support for privacy to generic gss-api code. This is dead code until we
have both a mechanism that supports privacy and code in the client or server
that uses it.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This patch stops the release_pipe() funtion from being called
twice by invalidating the ops pointer in the rpc_inode
when rpc_pipe_release() is called.
Signed-off-by: Steve Dickson <steved@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Currently, call_encode will cause the entire RPC call to abort if it returns
an error. This is unnecessarily rigid, and gets in the way of attempts
to allow the NFSv4 layer to order RPC calls that carry sequence ids.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
After a server crash/reboot, rebinding should always retry, otherwise
requests on "hard" mounts will fail when they shouldn't.
Test plan:
Run a lock-intensive workload against a server while rebooting the server
repeatedly.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
- added typedef unsigned int __nocast gfp_t;
- replaced __nocast uses for gfp flags with gfp_t - it gives exactly
the same warnings as far as sparse is concerned, doesn't change
generated code (from gcc point of view we replaced unsigned int with
typedef) and documents what's going on far better.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix nocast sparse warnings:
net/rxrpc/call.c:2013:25: warning: implicit cast to nocast type
net/rxrpc/connection.c:538:46: warning: implicit cast to nocast type
net/sunrpc/sched.c:730:36: warning: implicit cast to nocast type
net/sunrpc/sched.c:734:56: warning: implicit cast to nocast type
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The in-kernel portmapper does in fact need a reserved port when registering
new services, but not when performing bind queries.
Ensure that we distinguish between the two cases.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Currently rpc_mkdir/rpc_rmdir and rpc_mkpipe/mk_unlink have an API that's
a little unfortunate. They take a path relative to the rpc_pipefs root and
thus need to perform a full lookup. If you look at debugfs or usbfs they
always store the dentry for directories they created and thus can pass in
a dentry + single pathname component pair into their equivalents of the
above functions.
And in fact rpc_pipefs actually stores a dentry for all but one component so
this change not only simplifies the core rpc_pipe code but also the callers.
Unfortuntately this code path is only used by the NFS4 idmapper and
AUTH_GSSAPI for which I don't have a test enviroment. Could someone give
it a spin? It's the last bit needed before we can rework the
lookup_hash API
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
In fact, ->set_buffer_size should be completely functionless for non-UDP.
Test-plan:
Check socket buffer size on UDP sockets over time.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Each transport implementation can now set unique bind, connect,
reestablishment, and idle timeout values. These are variables,
allowing the values to be modified dynamically. This permits
exponential backoff of any of these values, for instance.
As an example, we implement exponential backoff for the connection
reestablishment timeout.
Test-plan:
Destructive testing (unplugging the network temporarily). Connectathon
with UDP and TCP.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Implement a best practice: if the remote end drops our connection, try to
reconnect using the same port number. This is important because the NFS
server's Duplicate Reply Cache often hashes on the source port number.
If the client reuses the port number when it reconnects, the server's DRC
will be more effective.
Based on suggestions by Mike Eisler, Olaf Kirch, and Alexey Kuznetsky.
Test-plan:
Destructive testing.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Select an RPC client source port between 650 and 1023 instead of between
1 and 800. The old range conflicts with a number of network services.
Provide sysctls to allow admins to select a different port range.
Note that this doesn't affect user-level RPC library behavior, which
still uses 1 to 800.
Based on a suggestion by Olaf Kirch <okir@suse.de>.
Test-plan:
Repeated mount and unmount. Destructive testing. Idle timeouts.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Clean-up: Move some macros that are specific to the Van Jacobson
implementation into xprt.c. Get rid of the cong_wait field in
rpc_xprt, which is no longer used. Get rid of xprt_clear_backlog.
Test-plan:
Compile with CONFIG_NFS enabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Get rid of the "xprt->nocong" variable.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss with UDP mounts.
Look for significant regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The final place where congestion control state is adjusted is in
xprt_release, where each request is finally released. Add a callout
there to allow transports to perform additional processing when a
request is about to be released.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss. Look for significant
regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
A new interface that allows transports to adjust their congestion window
using the Van Jacobson implementation in xprt.c is provided.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss. Look for
significant regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Allow transports to hook the retransmit timer interrupt. Some transports
calculate their congestion window here so that a retransmit timeout has
immediate effect on the congestion window.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss. Look for significant
regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The next method we abstract is the one that releases a transport,
allowing another task to have access to the transport.
Again, one generic version of this is provided for transports that
don't need the RPC client to perform congestion control, and one
version is for transports that can use the original Van Jacobson
implementation in xprt.c.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss. Look for
significant regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The next several patches introduce an API that allows transports to
choose whether the RPC client provides congestion control or whether
the transport itself provides it.
The first method we abstract is the one that serializes access to the
RPC transport to prevent the bytes from different requests from mingling
together. This method provides proper request serialization and the
opportunity to prevent new requests from being started because the
transport is congested.
The normal situation is for the transport to handle congestion control
itself. Although NFS over UDP was first, it has been recognized after
years of experience that having the transport provide congestion control
is much better than doing it in the RPC client. Thus TCP, and probably
every future transport implementation, will use the default method,
xprt_lock_write, provided in xprt.c, which does not provide any kind
of congestion control. UDP can continue using the xprt.c-provided
Van Jacobson congestion avoidance implementation.
Test-plan:
Use WAN simulation to cause sporadic bursty packet loss. Look for significant
regression in performance or client stability.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
NeilBrown
Eric Dumazet
Trond Myklebust
Herbert Xu
Trond Myklebust
Christoph Hellwig
Jesper Juhl
Chuck Lever
David Hardeman
Andrew Morton
J. Bruce Fields
Steve Dickson
Al Viro
Randy Dunlap
Chuck Lever
Christoph Hellwig