Clean up: Make code more readable.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NDevesh Sharma <devesh.sharma@broadcom.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Calling ib_poll_cq() to sort through WCs during a completion is a
common pattern amongst RDMA consumers. Since commit 14d3a3b2
("IB: add a proper completion queue abstraction"), WC sorting can
be handled by the IB core.

By converting to this new API, xprtrdma is made a better neighbor to
other RDMA consumers, as it allows the core to schedule the delivery
of completions more fairly amongst all active consumers.

Because each ib_cqe carries a pointer to a completion method, the
core can now post its own operations on a consumer's QP, and handle
the completions itself, without changes to the consumer.

xprtrdma's reply processing is already handled in a work queue, but
there is some initial order-dependent processing that is done in the
soft IRQ context before a work item is scheduled.

IB_POLL_SOFTIRQ is a direct replacement for the current xprtrdma
receive code path.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NDevesh Sharma <devesh.sharma@broadcom.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Commit fe97b47c ("xprtrdma: Use workqueue to process RPC/RDMA
replies") replaced the reply tasklet with a workqueue that allows
RPC replies to be processed in parallel. Thus the credit values in
RPC-over-RDMA replies can be applied in a different order than in
which the server sent them.

To fix this, revert commit eba8ff66 ("xprtrdma: Move credit
update to RPC reply handler"). Reverting is done by hand to
accommodate code changes that have occurred since then.

Fixes: fe97b47c ("xprtrdma: Use workqueue to process . . .")
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

These are shorter than RPCRDMA_HDRLEN_MIN, and they need to
complete the waiting RPC.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

If ib_post_send() in ro_unmap_sync() fails, the WRs have not been
posted, no completions will fire, and wait_for_completion() will
wait forever. Skip the wait in that case.

To ensure the MRs are invalid, disconnect.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

A single memory allocation is used for the pair of buffers wherein
the RPC client builds an RPC call message and decodes its matching
reply. These buffers are sized based on the maximum possible size
of the RPC call and reply messages for the operation in progress.

This means that as the call buffer increases in size, the start of
the reply buffer is pushed farther into the memory allocation.

RPC requests are growing in size. It used to be that both the call
and reply buffers fit inside a single page.

But these days, thanks to NFSv4 (and especially security labels in
NFSv4.2) the maximum call and reply sizes are large. NFSv4.0 OPEN,
for example, now requires a 6KB allocation for a pair of call and
reply buffers, and NFSv4 LOOKUP is not far behind.

As the maximum size of a call increases, the reply buffer is pushed
far enough into the buffer's memory allocation that a page boundary
can appear in the middle of it.

When the maximum possible reply size is larger than the client's
RDMA receive buffers (currently 1KB), the client has to register a
Reply chunk for the server to RDMA Write the reply into.

The logic in rpcrdma_convert_iovs() assumes that xdr_buf head and
tail buffers would always be contained on a single page. It supplies
just one segment for the head and one for the tail.

FMR, for example, registers up to a page boundary (only a portion of
the reply buffer in the OPEN case above). But without additional
segments, it doesn't register the rest of the buffer.

When the server tries to write the OPEN reply, the RDMA Write fails
with a remote access error since the client registered only part of
the Reply chunk.

rpcrdma_convert_iovs() must split the XDR buffer into multiple
segments, each of which are guaranteed not to contain a page
boundary. That way fmr_op_map is given the proper number of segments
to register the whole reply buffer.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NDevesh Sharma <devesh.sharma@broadcom.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

physical_op_unmap{_sync} don't use mr_nsegs, so don't bother to set
it in physical_op_map.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Some NFSv4.1 OPEN requests were hanging waiting for the NFS server
to finish recalling delegations. Turns out that each NFSv4.1 CB
request on RDMA gets a GARBAGE_ARGS reply from the Linux client.

Commit 756b9b37 added a line in bc_svc_process that
overwrites the incoming rq_rcv_buf's length with the value in
rq_private_buf.len. But rpcrdma_bc_receive_call() does not invoke
xprt_complete_bc_request(), thus rq_private_buf.len is not
initialized. svc_process_common() is invoked with a zero-length
RPC message, and fails.

Fixes: 756b9b37 ('SUNRPC: Fix callback channel')
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

We now alwasy have a per-PD local_dma_lkey available.  Make use of that
fact in svc_rdma and stop registering our own MR.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Reviewed-by: NJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSteve Wise <swise@opengridcomputing.com>
Acked-by: NJ. Bruce Fields <bfields@redhat.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

To support the server-side of an NFSv4.1 backchannel on RDMA
connections, add a transport class that enables backward
direction messages on an existing forward channel connection.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Extra resources for handling backchannel requests have to be
pre-allocated when a transport instance is created. Set up
additional fields in svcxprt_rdma to track these resources.

The max_requests fields are elements of the RPC-over-RDMA
protocol, so they should be u32. To ensure that unsigned
arithmetic is used everywhere, some other fields in the
svcxprt_rdma struct are updated.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Pre-requisite to use map_xdr in the backchannel code.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Clean up.

These functions can otherwise fail, so check for page allocation
failures too.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

svc_rdma_post_recv() allocates pages for receive buffers on-demand.
It uses GFP_KERNEL so the allocator tries hard, and may sleep. But
I'm about to add a call to svc_rdma_post_recv() from a function
that may not sleep.

Since all svc_rdma_post_recv() call sites can tolerate its failure,
allow it to fail if the page allocator returns nothing. Longer term,
receive buffers, being a finite resource per-connection, should be
pre-allocated and re-used.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Clean up.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

To ensure this allocation cannot fail and will not sleep,
pre-allocate the req_map structures per-connection.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

When the maximum payload size of NFS READ and WRITE was increased
by commit cc9a903d ("svcrdma: Change maximum server payload back
to RPCSVC_MAXPAYLOAD"), the size of struct svc_rdma_op_ctxt
increased to over 6KB (on x86_64). That makes allocating one of
these from a kmem_cache more likely to fail in situations when
system memory is exhausted.

Since I'm about to add a caller where this allocation must always
work _and_ it cannot sleep, pre-allocate ctxts for each connection.

Another motivation for this change is that NFSv4.x servers are
required by specification not to drop NFS requests. Pre-allocating
memory resources reduces the likelihood of a drop.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Be sure the completed ctxt is put in every path.

The xprt enqueue can take a while, so put the completed ctxt back
in circulation _before_ enqueuing the xprt.

Remove/disable debugging.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

kzalloc is used here, so setting the atomic fields to zero is
unnecessary. sc_ord is set again in handle_connect_req. The other
fields are re-initialized in svc_rdma_accept().
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: NBruce Fields <bfields@fieldses.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

This reverts commit 6f18dc89.

Just as one example, it appears this code could do the wrong thing in
the case of a two-byte NFS READ that crosses a page boundary.

Chuck says: "In that case, nfsd would pass down an xdr_buf that has one
byte in a page, one byte in another page, and a two-byte XDR pad. The
logic introduced by this optimization would be fooled, and neither the
second byte nor the XDR pad would be written to the client."

Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Instead, use the cached copy of the attributes present on the device.
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

The root of the problem was that sends (especially unsignalled
FASTREG and LOCAL_INV Work Requests) were not properly flow-
controlled, which allowed a send queue overrun.

Now that the RPC/RDMA reply handler waits for invalidation to
complete, the send queue is properly flow-controlled. Thus this
limit is no longer necessary.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

There is a window between the time the RPC reply handler wakes the
waiting RPC task and when xprt_release() invokes ops->buf_free.
During this time, memory regions containing the data payload may
still be accessed by a broken or malicious server, but the RPC
application has already been allowed access to the memory containing
the RPC request's data payloads.

The server should be fenced from client memory containing RPC data
payloads _before_ the RPC application is allowed to continue.

This change also more strongly enforces send queue accounting. There
is a maximum number of RPC calls allowed to be outstanding. When an
RPC/RDMA transport is set up, just enough send queue resources are
allocated to handle registration, Send, and invalidation WRs for
each those RPCs at the same time.

Before, additional RPC calls could be dispatched while invalidation
WRs were still consuming send WQEs. When invalidation WRs backed
up, dispatching additional RPCs resulted in a send queue overrun.

Now, the reply handler prevents RPC dispatch until invalidation is
complete. This prevents RPC call dispatch until there are enough
send queue resources to proceed.

Still to do: If an RPC exits early (say, ^C), the reply handler has
no opportunity to perform invalidation. Currently, xprt_rdma_free()
still frees remaining RDMA resources, which could deadlock.
Additional changes are needed to handle invalidation properly in this
case.
Reported-by: NJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

physical's ro_unmap is synchronous already. The new ro_unmap_sync
method just has to DMA unmap all MRs associated with the RPC
request.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

FMR's ro_unmap method is already synchronous because ib_unmap_fmr()
is a synchronous verb. However, some improvements can be made here.

1. Gather all the MRs for the RPC request onto a list, and invoke
   ib_unmap_fmr() once with that list. This reduces the number of
   doorbells when there is more than one MR to invalidate

2. Perform the DMA unmap _after_ the MRs are unmapped, not before.
   This is critical after invalidating a Write chunk.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

FRWR's ro_unmap is asynchronous. The new ro_unmap_sync posts
LOCAL_INV Work Requests and waits for them to complete before
returning.

Note also, DMA unmapping is now done _after_ invalidation.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

In the current xprtrdma implementation, some memreg strategies
implement ro_unmap synchronously (the MR is knocked down before the
method returns) and some asynchonously (the MR will be knocked down
and returned to the pool in the background).

To guarantee the MR is truly invalid before the RPC consumer is
allowed to resume execution, we need an unmap method that is
always synchronous, invoked from the RPC/RDMA reply handler.

The new method unmaps all MRs for an RPC. The existing ro_unmap
method unmaps only one MR at a time.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

For FRWR FASTREG and LOCAL_INV, move the ib_*_wr structure off
the stack. This allows frwr_op_map and frwr_op_unmap to chain
WRs together without limit to register or invalidate a set of MRs
with a single ib_post_send().

(This will be for chaining LOCAL_INV requests).
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up.

Fixes: 63cae470 ('xprtrdma: Handle incoming backward direction')
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Preserve any rpcrdma_req that is attached to rpc_rqst's allocated
for the backchannel. Otherwise, after all the pre-allocated
backchannel req's are consumed, incoming backward calls start
writing on freed memory.

Somehow this hunk got lost.

Fixes: f531a5db ('xprtrdma: Pre-allocate backward rpc_rqst')
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up.

rb_lock critical sections added in rpcrdma_ep_post_extra_recv()
should have first been converted to use normal spin_lock now that
the reply handler is a work queue.

The backchannel set up code should use the appropriate helper
instead of open-coding a rb_recv_bufs list add.

Problem introduced by glib patch re-ordering on my part.

Fixes: f531a5db ('xprtrdma: Pre-allocate backward rpc_rqst')
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Tested-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The rpcrdma_create_req() function returns error pointers or success.  It
never returns NULL.

Fixes: f531a5db ('xprtrdma: Pre-allocate backward rpc_rqst and send/receive buffers')
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

It doesn't matter either way, but the curly braces were clearly intended
here.  It causes a Smatch warning.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Minor optimization: when dealing with write chunk XDR roundup, do
not post a Write WR for the zero bytes in the pad. Simply update
the write segment in the RPC-over-RDMA header to reflect the extra
pad bytes.

The Reply chunk is also a write chunk, but the server does not use
send_write_chunks() to send the Reply chunk. That's OK in this case:
the server Upper Layer typically marshals the Reply chunk contents
in a single contiguous buffer, without a separate tail for the XDR
pad.

The comments and the variable naming refer to "chunks" but what is
really meant is "segments." The existing code sends only one
xdr_write_chunk per RPC reply.

The fix assumes this as well. When the XDR pad in the first write
chunk is reached, the assumption is the Write list is complete and
send_write_chunks() returns.

That will remain a valid assumption until the server Upper Layer can
support multiple bulk payload results per RPC.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Forechannel transports get their own "bc_up" method to create an
endpoint for the backchannel service.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
[Anna Schumaker: Add forward declaration of struct net to xprt.h]
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

On NFSv4.1 mount points, the Linux NFS client uses this transport
endpoint to receive backward direction calls and route replies back
to the NFSv4.1 server.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Acked-by: N"J. Bruce Fields" <bfields@fieldses.org>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Tested-By: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Introduce a code path in the rpcrdma_reply_handler() to catch
incoming backward direction RPC calls and route them to the ULP's
backchannel server.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Tested-By: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Backward direction RPC replies are sent via the client transport's
send_request method, the same way forward direction RPC calls are
sent.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Tested-By: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Pre-allocate extra send and receive Work Requests needed to handle
backchannel receives and sends.

The transport doesn't know how many extra WRs to pre-allocate until
the xprt_setup_backchannel() call, but that's long after the WRs are
allocated during forechannel setup.

So, use a fixed value for now.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Tested-By: NDevesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>