Instead of using a fixed number, allow the amount of Send completion
batching to vary based on the client's maximum credit limit.

- A larger default gives a small boost to IOPS throughput

- Reducing it based on max_requests gives a safe result when the
  max credit limit is cranked down (eg. when the device has a small
  max_qp_wr).
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Minor clean-ups I've stumbled on since sendctx was merged last year.
In particular, making Send completion processing more efficient
appears to have a measurable impact on IOPS throughput.

Note: test_and_clear_bit() returns a value, thus an explicit memory
barrier is not necessary.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Record an event when rpcrdma_marshal_req returns a non-zero return
value to help track down why an xprt close might have occurred.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Reflects the change introduced in commit 067c4696 ("NFSv4.1:
Bump the default callback session slot count to 16").
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The Receive handler runs in process context, thus can use on-demand
GFP_KERNEL allocations instead of pre-allocation.

This makes the xprtrdma backchannel independent of the number of
backchannel session slots provisioned by the Upper Layer protocol.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

For code legibility, clean up the function names to be consistent
with the pattern: "rpcrdma" _ object-type _ action

Also rpcrdma_regbuf_alloc and rpcrdma_regbuf_free no longer have any
callers outside of verbs.c, and can thus be made static.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up by providing an API to do this common task.

At this point, the difference between rpcrdma_get_sendbuf and
rpcrdma_get_recvbuf has become tiny. These can be collapsed into a
single helper.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Allocating an rpcrdma_req's regbufs at xprt create time enables
a pair of micro-optimizations:

First, if these regbufs are always there, we can eliminate two
conditional branches from the hot xprt_rdma_allocate path.

Second, by allocating a 1KB buffer, it places a lower bound on the
size of these buffers, without adding yet another conditional
branch. The lower bound reduces the number of hardway re-
allocations. In fact, for some workloads it completely eliminates
hardway allocations.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Allocate the struct rpcrdma_regbuf separately from the I/O buffer
to better guarantee the alignment of the I/O buffer and eliminate
the wasted space between the rpcrdma_regbuf metadata and the buffer
itself.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

For code legibility, clean up the function names to be consistent
with the pattern: "rpcrdma" _ object-type _ action
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Eventually, I'd like to invoke rpcrdma_create_req() during the
call_reserve step. Memory allocation there probably needs to use
GFP_NOIO. Therefore a set of GFP flags needs to be passed in.

As an additional clean up, just return a pointer or NULL, because
the only error return code here is -ENOMEM.

Lastly, clean up the function names to be consistent with the
pattern: "rpcrdma" _ object-type _ action
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

After a DMA map failure in frwr_map, mark the MR so that recycling
won't attempt to DMA unmap it.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Fixes: e2f34e26 ("xprtrdma: Yet another double DMA-unmap")
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Page allocation requests made when the SPARSE_PAGES flag is set are
allowed to fail, and are not critical. No need to spend a rare
resource.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Add the 'softerr' rpc client flag that sets the RPC_TASK_TIMEOUT
flag on all new rpc tasks that are attached to that rpc client.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

In particular, the timeout messages can be very noisy, so we ought to
ratelimit them in order to avoid spamming the syslog.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

When calculating the major timeout for a new task, when we know that the
connection has been broken, use the task->tk_start to ensure that we also
take into account the time spent waiting for a slot or session slot. This
ensures that we fail over soft requests relatively quickly once the
connection has actually been broken, and the first requests have
started to fail.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

If a soft NFSv4 request is sent, then we don't need it to time out unless
the connection breaks. The reason is that as long as the connection is
unbroken, the protocol states that the server is not allowed to drop the
request.  IOW: as long as the connection remains unbroken, the client may
assume that all transmitted RPC requests are being processed by the server,
and that retransmissions and timeouts of those requests are unwarranted.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Add variables to track RPC level errors so that we can distinguish
between issue that arose in the RPC transport layer as opposed to
those arising from the reply message.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Ensure that when in the transport layer, we don't sleep past
a major timeout.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Don't wake idle CPUs only for the purpose of servicing an RPC
queue timeout.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Simplify the setting of queue timeouts by using the timer_reduce()
function.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Add a helper to ensure that debugfs and friends print out the
correct current task timeout value.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up the RPC task sleep interfaces by replacing the task->tk_timeout
'hidden parameter' to rpc_sleep_on() with a new function that takes an
absolute timeout.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

None of the callers set the 'action' argument, so let's just remove it.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

rpc_sleep_on() does not need to set the task->tk_callback under the
queue lock, so move that out.
Also refactor the check for whether the task is active.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Convert the transport callback to actually put the request to sleep
instead of just setting a timeout. This is in preparation for
rpc_sleep_on_timeout().
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The RPC_TASK_KILLED flag should really not be set from another context
because it can clobber data in the struct task when task->tk_flags is
changed non-atomically.
Let's therefore swap out RPC_TASK_KILLED with an atomic flag, and add
a function to set that flag and safely wake up the task.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

If a request transmission fails due to write space or slot unavailability
errors, but the queued task then gets transmitted before it has time to
process the error in call_transmit_status() or call_bc_transmit_status(),
we need to suppress the transmission error code to prevent it from leaking
out of the RPC layer.
Reported-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
Tested-by: NChuck Lever <chuck.lever@oracle.com>

This reverts commit 009a82f6.

The ability to optimise here relies on compiler being able to optimise
away tail calls to avoid stack overflows. Unfortunately, we are seeing
reports of problems, so let's just revert.
Reported-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

We want to drain only the RQ first. Otherwise the transport can
deadlock on ->close if there are outstanding Send completions.

Fixes: 6d2d0ee2 ("xprtrdma: Replace rpcrdma_receive_wq ... ")
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Cc: stable@vger.kernel.org # v5.0+
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

Avoid following compiler warning on uninitialized variable

net/sunrpc/xprtsock.c: In function ‘xs_read_stream_request.constprop’:
net/sunrpc/xprtsock.c:525:10: warning: ‘read’ may be used uninitialized in this function [-Wmaybe-uninitialized]
   return read;
          ^~~~
net/sunrpc/xprtsock.c:529:23: warning: ‘ret’ may be used uninitialized in this function [-Wmaybe-uninitialized]
  return ret < 0 ? ret : read;
         ~~~~~~~~~~~~~~^~~~~~
Signed-off-by: NAlakesh Haloi <alakesh.haloi@gmail.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

If the transport is still connected, then we do want to allow
RPC_SOFTCONN tasks to retry. They should time out if and only if
the connection is broken.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

Now that we're using the xdr_stream functions to decode the header,
the test for the minimum reply length is redundant.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

Handle the SYSTEM_ERR rpc error by retrying the RPC call as if it
were a garbage argument.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

Ensure that when the "garbage args" case falls through, we do set
an error of EIO.

Fixes: a0584ee9 ("SUNRPC: Use struct xdr_stream when decoding...")
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

When the socket is closed, we currently send an EAGAIN error to all
pending requests in order to ask them to retransmit. Use ENOTCONN
instead, to ensure that they try to reconnect before attempting to
transmit.
This also helps SOFTCONN tasks to behave correctly in this
situation.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

We must at minimum allocate enough memory to be able to see any auth
errors in the reply from the server.

Fixes: 2c94b8ec ("SUNRPC: Use au_rslack when computing reply...")
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

If the server sends a reply that is larger than the pre-allocated
buffer, then the current code may fail to register how much of
the stream that it has finished reading. This again can lead to
hangs.

Fixes: e92053a5 ("SUNRPC: Handle zero length fragments correctly")
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>

Before trying to bind a port, ensure we grab the send lock to
ensure that we don't change the port while another task is busy
transmitting requests.
The connect code already takes the send lock in xprt_connect(),
but it is harmless to take it before that.
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>