Clean up: All remaining callers of rpcrdma_deregister_external()
pass NULL as the last argument, so remove that argument.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

If the selected memory registration mode is not supported by the
underlying provider/HCA, the NFS mount command reports that there was
an invalid mount option, and fails. This is misleading.

Reporting a problem allocating memory is a lot closer to the truth.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

An audit of in-kernel RDMA providers that do not support the FRMR
memory registration shows that several of them support MTHCAFMR.
Prefer MTHCAFMR when FRMR is not supported.

If MTHCAFMR is not supported, only then choose ALLPHYSICAL.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

All kernel RDMA providers except amso1100 support either MTHCAFMR
or FRMR, both of which are faster than REGISTER.  amso1100 can
continue to use ALLPHYSICAL.

The only other ULP consumer in the kernel that uses the reg_phys_mr
verb is Lustre.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The MEMWINDOWS and MEMWINDOWS_ASYNC memory registration modes were
intended as stop-gap modes before the introduction of FRMR. They
are now considered obsolete.

MEMWINDOWS_ASYNC is also considered unsafe because it can leave
client memory registered and exposed for an indeterminant time after
each I/O.

At this point, the MEMWINDOWS modes add needless complexity, so
remove them.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up: This memory registration mode is slow and was never
meant for use in production environments. Remove it to reduce
implementation complexity.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

An IB provider can invoke rpcrdma_conn_func() in an IRQ context,
thus rpcrdma_conn_func() cannot be allowed to directly invoke
generic RPC functions like xprt_wake_pending_tasks().
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Two memory region leaks were found during testing:

1. rpcrdma_buffer_create: While allocating RPCRDMA_FRMR's
ib_alloc_fast_reg_mr is called and then ib_alloc_fast_reg_page_list is
called.  If ib_alloc_fast_reg_page_list returns an error it bails out of
the routine dropping the last ib_alloc_fast_reg_mr frmr region creating a
memory leak.  Added code to dereg the last frmr if
ib_alloc_fast_reg_page_list fails.

2. rpcrdma_buffer_destroy: While cleaning up, the routine will only free
the MR's on the rb_mws list if there are rb_send_bufs present.  However, in
rpcrdma_buffer_create while the rb_mws list is being built if one of the MR
allocation requests fail after some MR's have been allocated on the rb_mws
list the routine never gets to create any rb_send_bufs but instead jumps to
the rpcrdma_buffer_destroy routine which will never free the MR's on rb_mws
list because the rb_send_bufs were never created.   This leaks all the MR's
on the rb_mws list that were created prior to one of the MR allocations
failing.

Issue(2) was seen during testing. Our adapter had a finite number of MR's
available and we created enough connections to where we saw an MR
allocation failure on our Nth NFS connection request. After the kernel
cleaned up the resources it had allocated for the Nth connection we noticed
that FMR's had been leaked due to the coding error described above.

Issue(1) was seen during a code review while debugging issue(2).
Signed-off-by: NAllen Andrews <allen.andrews@emulex.com>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Some rdma devices don't support a fast register page list depth of
at least RPCRDMA_MAX_DATA_SEGS.  So xprtrdma needs to chunk its fast
register regions according to the minimum of the device max supported
depth or RPCRDMA_MAX_DATA_SEGS.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The xdr_off value in dma_map_xdr gets passed to ib_dma_map_page as the
offset into the page to be mapped. This calculation does not correctly
take into account the case where the data starts at some offset into
the page. Increment the xdr_off by the page_base to ensure that it is
respected.

Cc: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

There are two entirely separate modules under xprtrdma/ and there's no
reason that enabling one should automatically enable the other. Add
config options for each one so they can be enabled/disabled separately.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The server regression was caused by the addition of rq_next_page
(afc59400). There were a few places that
were missed with the update of the rq_respages array.
Signed-off-by: NTom Tucker <tom@ogc.us>
Tested-by: NSteve Wise <swise@ogc.us>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

It retries in 1s, not 1000 jiffies.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The use of KERN_INFO causes garbage characters to appear when
debugging is enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>

After commit a11a2bf4, "SUNRPC: Optimise away unnecessary data moves
in xdr_align_pages", Thu Aug 2 13:21:43 2012, READs larger than a
few hundred bytes via NFS/RDMA no longer work.  This commit exposed
a long-standing bug in rpcrdma_inline_fixup().

I reproduce this with an rsize=4096 mount using the cthon04 basic
tests.  Test 5 fails with an EIO error.

For my reproducer, kernel log shows:

  NFS: server cheating in read reply: count 4096 > recvd 0

rpcrdma_inline_fixup() is zeroing the xdr_stream::page_len field,
and xdr_align_pages() is now returning that value to the READ XDR
decoder function.

That field is set up by xdr_inline_pages() by the READ XDR encoder
function.  As far as I can tell, it is supposed to be left alone
after that, as it describes the dimensions of the reply xdr_stream,
not the contents of that stream.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=68391Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>

My static checker marks everything from ntohl() as untrusted and it
complains we could have an underflow problem doing:

	return (u32 *)&ary->wc_array[nchunks];

Also on 32 bit systems the upper bound check could overflow.

Cc: stable@vger.kernel.org
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Reduce the uses of this unnecessary typedef.

Done via perl script:

$ git grep --name-only -w ctl_table net | \
  xargs perl -p -i -e '\
	sub trim { my ($local) = @_; $local =~ s/(^\s+|\s+$)//g; return $local; } \
        s/\b(?<!struct\s)ctl_table\b(\s*\*\s*|\s+\w+)/"struct ctl_table " . trim($1)/ge'

Reflow the modified lines that now exceed 80 columns.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch enhances the IB core support for Memory Windows (MWs).

MWs allow an application to have better/flexible control over remote
access to memory.

Two types of MWs are supported, with the second type having two flavors:

    Type 1  - associated with PD only
    Type 2A - associated with QPN only
    Type 2B - associated with PD and QPN

Applications can allocate a MW once, and then repeatedly bind the MW
to different ranges in MRs that are associated to the same PD. Type 1
windows are bound through a verb, while type 2 windows are bound by
posting a work request.

The 32-bit memory key is composed of a 24-bit index and an 8-bit
key. The key is changed with each bind, thus allowing more control
over the peer's use of the memory key.

The changes introduced are the following:

* add memory window type enum and a corresponding parameter to ib_alloc_mw.
* type 2 memory window bind work request support.
* create a struct that contains the common part of the bind verb struct
  ibv_mw_bind and the bind work request into a single struct.
* add the ib_inc_rkey helper function to advance the tag part of an rkey.

Consumer interface details:

* new device capability flags IB_DEVICE_MEM_WINDOW_TYPE_2A and
  IB_DEVICE_MEM_WINDOW_TYPE_2B are added to indicate device support
  for these features.

  Devices can set either IB_DEVICE_MEM_WINDOW_TYPE_2A or
  IB_DEVICE_MEM_WINDOW_TYPE_2B if it supports type 2A or type 2B
  memory windows. It can set neither to indicate it doesn't support
  type 2 windows at all.

* modify existing provides and consumers code to the new param of
  ib_alloc_mw and the ib_mw_bind_info structure
Signed-off-by: NHaggai Eran <haggaie@mellanox.com>
Signed-off-by: NShani Michaeli <shanim@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>

These routines are used by server and client code, so having them in a
separate header would be best.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Acked-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Avoid another RCU dereference by passing the pointer to struct rpc_xprt
from the caller.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

tk_xprt is just a shortcut for tk_client->cl_xprt, however cl_xprt is
defined as an __rcu variable. Replace dereferences of tk_xprt with
non-rcu dereferences where it is safe to do so.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

It may be a matter of personal taste, but I find this makes the code
clearer.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

It is only set after everyone has dereferenced the transport,
and serves no useful purpose: setting it is racy, so all the
socket code, etc still needs to be able to cope with the cases
where they miss reading it.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Commit 43cedbf0 (SUNRPC: Ensure that
we grab the XPRT_LOCK before calling xprt_alloc_slot) is causing
hangs in the case of NFS over UDP mounts.

Since neither the UDP or the RDMA transport mechanism use dynamic slot
allocation, we can skip grabbing the socket lock for those transports.
Add a new rpc_xprt_op to allow switching between the TCP and UDP/RDMA
case.

Note that the NFSv4.1 back channel assigns the slot directly
through rpc_run_bc_task, so we can ignore that case.
Reported-by: NDick Streefland <dick.streefland@altium.nl>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@vger.kernel.org [>= 3.1]

Note this isn't used outside svc_xprt.c.

May as well move it so we don't need a declaration while we're here.

Also remove an outdated comment.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

We've had some reports of a deadlock where rpciod ends up with a stack
trace like this:

    PID: 2507   TASK: ffff88103691ab40  CPU: 14  COMMAND: "rpciod/14"
     #0 [ffff8810343bf2f0] schedule at ffffffff814dabd9
     #1 [ffff8810343bf3b8] nfs_wait_bit_killable at ffffffffa038fc04 [nfs]
     #2 [ffff8810343bf3c8] __wait_on_bit at ffffffff814dbc2f
     #3 [ffff8810343bf418] out_of_line_wait_on_bit at ffffffff814dbcd8
     #4 [ffff8810343bf488] nfs_commit_inode at ffffffffa039e0c1 [nfs]
     #5 [ffff8810343bf4f8] nfs_release_page at ffffffffa038bef6 [nfs]
     #6 [ffff8810343bf528] try_to_release_page at ffffffff8110c670
     #7 [ffff8810343bf538] shrink_page_list.clone.0 at ffffffff81126271
     #8 [ffff8810343bf668] shrink_inactive_list at ffffffff81126638
     #9 [ffff8810343bf818] shrink_zone at ffffffff8112788f
    #10 [ffff8810343bf8c8] do_try_to_free_pages at ffffffff81127b1e
    #11 [ffff8810343bf958] try_to_free_pages at ffffffff8112812f
    #12 [ffff8810343bfa08] __alloc_pages_nodemask at ffffffff8111fdad
    #13 [ffff8810343bfb28] kmem_getpages at ffffffff81159942
    #14 [ffff8810343bfb58] fallback_alloc at ffffffff8115a55a
    #15 [ffff8810343bfbd8] ____cache_alloc_node at ffffffff8115a2d9
    #16 [ffff8810343bfc38] kmem_cache_alloc at ffffffff8115b09b
    #17 [ffff8810343bfc78] sk_prot_alloc at ffffffff81411808
    #18 [ffff8810343bfcb8] sk_alloc at ffffffff8141197c
    #19 [ffff8810343bfce8] inet_create at ffffffff81483ba6
    #20 [ffff8810343bfd38] __sock_create at ffffffff8140b4a7
    #21 [ffff8810343bfd98] xs_create_sock at ffffffffa01f649b [sunrpc]
    #22 [ffff8810343bfdd8] xs_tcp_setup_socket at ffffffffa01f6965 [sunrpc]
    #23 [ffff8810343bfe38] worker_thread at ffffffff810887d0
    #24 [ffff8810343bfee8] kthread at ffffffff8108dd96
    #25 [ffff8810343bff48] kernel_thread at ffffffff8100c1ca

rpciod is trying to allocate memory for a new socket to talk to the
server. The VM ends up calling ->releasepage to get more memory, and it
tries to do a blocking commit. That commit can't succeed however without
a connected socket, so we deadlock.

Fix this by setting PF_FSTRANS on the workqueue task prior to doing the
socket allocation, and having nfs_release_page check for that flag when
deciding whether to do a commit call. Also, set PF_FSTRANS
unconditionally in rpc_async_schedule since that function can also do
allocations sometimes.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@vger.kernel.org

The xprtrdma FRMR mapping logic assumes that a segment is <= PAGE_SIZE.
This is not true for NFS4.
Signed-off-by: NTom Tucker <tom@ogc.us>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The client side RDMA transport will bug check if it receives a duplicate
reply, instead we should simply drop the duplicate reply.
Signed-off-by: NTom Tucker <tom@ogc.us>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NCong Wang <amwang@redhat.com>

The svcrdma transport was un-marshalling requests in-place. This resulted
in sparse warnings due to __beXX data containing both NBO and HBO data.

The code has been restructured to do byte-swapping as the header is
parsed instead of when the header is validated immediately after receipt.

Also moved extern declarations for the workqueue and memory pools to the
private header file.
Signed-off-by: NTom Tucker <tom@ogc.us>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This patch makes svc_xprt inherit network namespace link from its socket.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

These files are non modular, but need to export symbols using
the macros now living in export.h -- call out the include so
that things won't break when we remove the implicit presence
of module.h from everywhere.
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>

This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: NArun Sharma <asharma@fb.com>
Reviewed-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Our performance team has noticed that increasing
RPCRDMA_MAX_DATA_SEGS from 8 to 64 significantly
increases throughput when using the RDMA transport.
Signed-off-by: NSteve Dickson <steved@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Allow the number of available slots to grow with the TCP window size.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

This throttles the allocation of new slots when the socket is busy
reconnecting and/or is out of buffer space.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

* remove interrupt.g inclusion from netdevice.h -- not needed
* fixup fallout, add interrupt.h and hardirq.h back where needed.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The RDMA CM currently infers the QP type from the port space selected
by the user.  In the future (eg with RDMA_PS_IB or XRC), there may not
be a 1-1 correspondence between port space and QP type.  For netlink
export of RDMA CM state, we want to export the QP type to userspace,
so it is cleaner to explicitly associate a QP type to an ID.

Modify rdma_create_id() to allow the user to specify the QP type, and
use it to make our selections of datagram versus connected mode.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>

- kenrel -> kernel
- whetehr -> whether
- ttt -> tt
- sss -> ss
Signed-off-by: NJustin P. Mattock <justinmattock@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

We leak the memory allocated to 'ctxt' when we return after
'ib_dma_mapping_error()' returns !=0.
Signed-off-by: NJesper Juhl <jj@chaosbits.net>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>