Credit work contributed by Oracle engineers since 2014.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: C-structure style XDR encoding and decoding logic has
been replaced over the past several merge windows on both the
client and server. These data structures are no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The sendctx circular queue now guarantees that xprtrdma cannot
overflow the Send Queue, so remove the remaining bits of the
original Send WQE counting mechanism.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
When an RPC Call includes a file data payload, that payload can come
from pages in the page cache, or a user buffer (for direct I/O).
If the payload can fit inline, xprtrdma includes it in the Send
using a scatter-gather technique. xprtrdma mustn't allow the RPC
consumer to re-use the memory where that payload resides before the
Send completes. Otherwise, the new contents of that memory would be
exposed by an HCA retransmit of the Send operation.
So, block RPC completion on Send completion, but only in the case
where a separate file data payload is part of the Send. This
prevents the reuse of that memory while it is still part of a Send
operation without an undue cost to other cases.
Waiting is avoided in the common case because typically the Send
will have completed long before the RPC Reply arrives.
These days, an RPC timeout will trigger a disconnect, which tears
down the QP. The disconnect flushes all waiting Sends. This bounds
the amount of time the reply handler has to wait for a Send
completion.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Invoke a common routine for releasing hardware resources (for
example, invalidating MRs). This needs to be done whether an
RPC Reply has arrived or the RPC was terminated early.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
We have one boolean flag in rpcrdma_req today. I'd like to add more
flags, so convert that boolean to a bit flag.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Problem statement:
Recently Sagi Grimberg <sagi@grimberg.me> observed that kernel RDMA-
enabled storage initiators don't handle delayed Send completion
correctly. If Send completion is delayed beyond the end of a ULP
transaction, the ULP may release resources that are still being used
by the HCA to complete a long-running Send operation.
This is a common design trait amongst our initiators. Most Send
operations are faster than the ULP transaction they are part of.
Waiting for a completion for these is typically unnecessary.
Infrequently, a network partition or some other problem crops up
where an ordering problem can occur. In NFS parlance, the RPC Reply
arrives and completes the RPC, but the HCA is still retrying the
Send WR that conveyed the RPC Call. In this case, the HCA can try
to use memory that has been invalidated or DMA unmapped, and the
connection is lost. If that memory has been re-used for something
else (possibly not related to NFS), and the Send retransmission
exposes that data on the wire.
Thus we cannot assume that it is safe to release Send-related
resources just because a ULP reply has arrived.
After some analysis, we have determined that the completion
housekeeping will not be difficult for xprtrdma:
- Inline Send buffers are registered via the local DMA key, and
are already left DMA mapped for the lifetime of a transport
connection, thus no additional handling is necessary for those
- Gathered Sends involving page cache pages _will_ need to
DMA unmap those pages after the Send completes. But like
inline send buffers, they are registered via the local DMA key,
and thus will not need to be invalidated
In addition, RPC completion will need to wait for Send completion
in the latter case. However, nearly always, the Send that conveys
the RPC Call will have completed long before the RPC Reply
arrives, and thus no additional latency will be accrued.
Design notes:
In this patch, the rpcrdma_sendctx object is introduced, and a
lock-free circular queue is added to manage a set of them per
transport.
The RPC client's send path already prevents sending more than one
RPC Call at the same time. This allows us to treat the consumer
side of the queue (rpcrdma_sendctx_get_locked) as if there is a
single consumer thread.
The producer side of the queue (rpcrdma_sendctx_put_locked) is
invoked only from the Send completion handler, which is a single
thread of execution (soft IRQ).
The only care that needs to be taken is with the tail index, which
is shared between the producer and consumer. Only the producer
updates the tail index. The consumer compares the head with the
tail to ensure that the a sendctx that is in use is never handed
out again (or, expressed more conventionally, the queue is empty).
When the sendctx queue empties completely, there are enough Sends
outstanding that posting more Send operations can result in a Send
Queue overflow. In this case, the ULP is told to wait and try again.
This introduces strong Send Queue accounting to xprtrdma.
As a final touch, Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
suggested a mechanism that does not require signaling every Send.
We signal once every N Sends, and perform SGE unmapping of N Send
operations during that one completion.
Reported-by: Sagi Grimberg <sagi@grimberg.me>
Suggested-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Make rpcrdma_prepare_send_sges() return a negative errno
instead of a bool. Soon callers will want distinct treatments of
different types of failures.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
We need to decode and save the incoming rdma_credits field _after_
we know that the direction of the message is "forward direction
Reply". Otherwise, the credits value in reverse direction Calls is
also used to update the forward direction credits.
It is safe to decode the rdma_credits field in rpcrdma_reply_handler
now that rpcrdma_reply_handler is single-threaded. Receives complete
in the same order as they were sent on the NFS server.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
I noticed that the soft IRQ thread looked pretty busy under heavy
I/O workloads. perf suggested one area that was expensive was the
queue_work() call in rpcrdma_wc_receive. That gave me some ideas.
Instead of scheduling a separate worker to process RPC Replies,
promote the Receive completion handler to IB_POLL_WORKQUEUE, and
invoke rpcrdma_reply_handler directly.
Note that the poll workqueue is single-threaded. In order to keep
memory invalidation from serializing all RPC Replies, handle any
necessary invalidation tasks in a separate multi-threaded workqueue.
This provides a two-tier scheme, similar to OS I/O interrupt
handlers: A fast interrupt handler that schedules the slow handler
and re-enables the interrupt, and a slower handler that is invoked
for any needed heavy lifting.
Benefits include:
- One less context switch for RPCs that don't register memory
- Receive completion handling is moved out of soft IRQ context to
make room for other users of soft IRQ
- The same CPU core now DMA syncs and XDR decodes the Receive buffer
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: I'd like to be able to invoke the tail of
rpcrdma_reply_handler in two different places. Split the tail out
into its own helper function.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Make it easier to pass the decoded XID, vers, credits, and
proc fields around by moving these variables into struct rpcrdma_rep.
Note: the credits field will be handled in a subsequent patch.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: There are no remaining callers of this method.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Adopt the use of xprt_pin_rqst to eliminate contention between
Call-side users of rb_lock and the use of rb_lock in
rpcrdma_reply_handler.
This replaces the mechanism introduced in 431af645cf ("xprtrdma:
Fix client lock-up after application signal fires").
Use recv_lock to quickly find the completing rqst, pin it, then
drop the lock. At that point invalidation and pull-up of the Reply
XDR can be done. Both are often expensive operations.
Finally, take recv_lock again to signal completion to the RPC
layer. It also protects adjustment of "cwnd".
This greatly reduces the amount of time a lock is held by the
reply handler. Comparing lock_stat results shows a marked decrease
in contention on rb_lock and recv_lock.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
[trond.myklebust@primarydata.com: Remove call to rpcrdma_buffer_put() from
the "out_norqst:" path in rpcrdma_reply_handler.]
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
To reduce false cacheline sharing, separate counters that are likely
to be accessed in the Call path from those accessed in the Reply
path.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Re-arrange the pointer arithmetic in the chunk list encoders to
eliminate several more integer multiplication instructions during
Transport Header encoding.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Initialize an xdr_stream at the top of rpcrdma_marshal_req(), and
use it to encode the fixed transport header fields. This xdr_stream
will be used to encode the chunk lists in a subsequent patch.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: The caller already has rpcrdma_xprt, so pass that directly
instead. And provide a documenting comment for this critical
function.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Replace C-structure based XDR decoding for consistency
with other areas.
struct rpcrdma_rep is rearranged slightly so that the relevant fields
are in cache when the Receive completion handler is invoked.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
This field is no longer used outside the Receive completion handler.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Transport header decoding deals with untrusted input data, therefore
decoding this header needs to be hardened.
Adopt the same infrastructure that is used when XDR decoding NFS
replies. This is slightly more CPU-intensive than the replaced code,
but we're not adding new atomics, locking, or context switches. The
cost is manageable.
Start by initializing an xdr_stream in rpcrdma_reply_handler().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
After a signal, the RPC client aborts synchronous RPCs running on
behalf of the signaled application.
The server is still executing those RPCs, and will write the results
back into the client's memory when it's done. By the time the server
writes the results, that memory is likely being used for other
purposes. Therefore xprtrdma has to immediately invalidate all
memory regions used by those aborted RPCs to prevent the server's
writes from clobbering that re-used memory.
With FMR memory registration, invalidation takes a relatively long
time. In fact, the invalidation is often still running when the
server tries to write the results into the memory regions that are
being invalidated.
This sets up a race between two processes:
1. After the signal, xprt_rdma_free calls ro_unmap_safe.
2. While ro_unmap_safe is still running, the server replies and
rpcrdma_reply_handler runs, calling ro_unmap_sync.
Both processes invoke ib_unmap_fmr on the same FMR.
The mlx4 driver allows two ib_unmap_fmr calls on the same FMR at
the same time, but HCAs generally don't tolerate this. Sometimes
this can result in a system crash.
If the HCA happens to survive, rpcrdma_reply_handler continues. It
removes the rpc_rqst from rq_list and releases the transport_lock.
This enables xprt_rdma_free to run in another process, and the
rpc_rqst is released while rpcrdma_reply_handler is still waiting
for the ib_unmap_fmr call to finish.
But further down in rpcrdma_reply_handler, the transport_lock is
taken again, and "rqst" is dereferenced. If "rqst" has already been
released, this triggers a general protection fault. Since bottom-
halves are disabled, the system locks up.
Address both issues by reversing the order of the xprt_lookup_rqst
call and the ro_unmap_sync call. Introduce a separate lookup
mechanism for rpcrdma_req's to enable calling ro_unmap_sync before
xprt_lookup_rqst. Now the handler takes the transport_lock once
and holds it for the XID lookup and RPC completion.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=305
Fixes: 68791649a7 ('xprtrdma: Invalidate in the RPC reply ... ')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: I'm about to use the rl_free field for purposes other than
a free list. So use a more generic name.
This is a refactoring change only.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=305
Fixes: 68791649a7 ('xprtrdma: Invalidate in the RPC reply ... ')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
There are rare cases where an rpcrdma_req can be re-used (via
rpcrdma_buffer_put) while the RPC reply handler is still running.
This is due to a signal firing at just the wrong instant.
Since commit 9d6b040978 ("xprtrdma: Place registered MWs on a
per-req list"), rpcrdma_mws are self-contained; ie., they fully
describe an MR and scatterlist, and no part of that information is
stored in struct rpcrdma_req.
As part of closing the above race window, pass only the req's list
of registered MRs to ro_unmap_sync, rather than the rpcrdma_req
itself.
Some extra transport header sanity checking is removed. Since the
client depends on its own recollection of what memory had been
registered, there doesn't seem to be a way to abuse this change.
And, the check was not terribly effective. If the client had sent
Read chunks, the "list_empty" test is negative in both of the
removed cases, which are actually looking for Write or Reply
chunks.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=305
Fixes: 68791649a7 ('xprtrdma: Invalidate in the RPC reply ... ')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
There are rare cases where an rpcrdma_req and its matched
rpcrdma_rep can be re-used, via rpcrdma_buffer_put, while the RPC
reply handler is still using that req. This is typically due to a
signal firing at just the wrong instant.
As part of closing this race window, avoid using the wrong
rpcrdma_rep to detect remotely invalidated MRs. Mark MRs as
invalidated while we are sure the rep is still OK to use.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=305
Fixes: 68791649a7 ('xprtrdma: Invalidate in the RPC reply ... ')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Since commit 1e465fd4ff ("xprtrdma: Replace send and receive
arrays"), this field is no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The device driver for the underlying physical device associated
with an RPC-over-RDMA transport can be removed while RPC-over-RDMA
transports are still in use (ie, while NFS filesystems are still
mounted and active). The IB core performs a connection event upcall
to request that consumers free all RDMA resources associated with
a transport.
There may be pending RPCs when this occurs. Care must be taken to
release associated resources without leaving references that can
trigger a subsequent crash if a signal or soft timeout occurs. We
rely on the caller of the transport's ->close method to ensure that
the previous RPC task has invoked xprt_release but the transport
remains write-locked.
A DEVICE_REMOVE upcall forces a disconnect then sleeps. When ->close
is invoked, it destroys the transport's H/W resources, then wakes
the upcall, which completes and allows the core driver unload to
continue.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=266
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
When the underlying device driver is reloaded, ia->ri_device will be
replaced. All cached copies of that device pointer have to be
updated as well.
Commit 54cbd6b0c6 ("xprtrdma: Delay DMA mapping Send and Receive
buffers") added the rg_device field to each regbuf. As part of
handling a device removal, rpcrdma_dma_unmap_regbuf is invoked on
all regbufs for a transport.
Simply calling rpcrdma_dma_map_regbuf for each Receive buffer after
the driver has been reloaded should reinitialize rg_device correctly
for every case except rpcrdma_wc_receive, which still uses
rpcrdma_rep::rr_device.
Ensure the same device that was used to map a Receive buffer is also
used to sync it in rpcrdma_wc_receive by using rg_device there
instead of rr_device.
This is the only use of rr_device, so it can be removed.
The use of regbufs in the send path is also updated, for
completeness.
Fixes: 54cbd6b0c6 ("xprtrdma: Delay DMA mapping Send and ... ")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
In order to unload a device driver and reload it, xprtrdma will need
to close a transport's interface adapter, and then call
rpcrdma_ia_open again, possibly finding a different interface
adapter.
Make rpcrdma_ia_open safe to call on the same transport multiple
times.
This is a refactoring change only.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
We no longer need to accommodate an xdr_buf whose pages start at an
offset and cross extra page boundaries. If there are more partial or
whole pages to send than there are available SGEs, the marshaling
logic is now smart enough to use a Read chunk instead of failing.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The MAX_SEND_SGES check introduced in commit 655fec6987
("xprtrdma: Use gathered Send for large inline messages") fails
for devices that have a small max_sge.
Instead of checking for a large fixed maximum number of SGEs,
check for a minimum small number. RPC-over-RDMA will switch to
using a Read chunk if an xdr_buf has more pages than can fit in
the device's max_sge limit. This is considerably better than
failing all together to mount the server.
This fix supports devices that have as few as three send SGEs
available.
Reported-by: Selvin Xavier <selvin.xavier@broadcom.com>
Reported-by: Devesh Sharma <devesh.sharma@broadcom.com>
Reported-by: Honggang Li <honli@redhat.com>
Reported-by: Ram Amrani <Ram.Amrani@cavium.com>
Fixes: 655fec6987 ("xprtrdma: Use gathered Send for large ...")
Cc: stable@vger.kernel.org # v4.9+
Tested-by: Honggang Li <honli@redhat.com>
Tested-by: Ram Amrani <Ram.Amrani@cavium.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Pad optimization is changed by echoing into
/proc/sys/sunrpc/rdma_pad_optimize. This is a global setting,
affecting all RPC-over-RDMA connections to all servers.
The marshaling code picks up that value and uses it for decisions
about how to construct each RPC-over-RDMA frame. Having it change
suddenly in mid-operation can result in unexpected failures. And
some servers a client mounts might need chunk round-up, while
others don't.
So instead, copy the pad_optimize setting into each connection's
rpcrdma_ia when the transport is created, and use the copy, which
can't change during the life of the connection, instead.
This also removes a hack: rpcrdma_convert_iovs was using
the remote-invalidation-expected flag to predict when it could leave
out Write chunk padding. This is because the Linux server handles
implicit XDR padding on Write chunks correctly, and only Linux
servers can set the connection's remote-invalidation-expected flag.
It's more sensible to use the pad optimization setting instead.
Fixes: 677eb17e94 ("xprtrdma: Fix XDR tail buffer marshalling")
Cc: stable@vger.kernel.org # v4.9+
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Some devices (such as the Mellanox CX-4) can register, under a
single R_key, a set of memory regions that are not contiguous. When
this is done, all the segments in a Reply list, say, can then be
invalidated in a single LocalInv Work Request (or via Remote
Invalidation, which can invalidate exactly one R_key when completing
a Receive).
This means a single FastReg WR is used to register, and one or zero
LocalInv WRs can invalidate, the memory involved with RDMA transfers
on behalf of an RPC.
In addition, xprtrdma constructs some Reply chunks from three or
more segments. By registering them with SG_GAP, only one segment
is needed for the Reply chunk, allowing the whole chunk to be
invalidated remotely.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Verbs providers may perform house-keeping on the Send Queue during
each signaled send completion. It is necessary therefore for a verbs
consumer (like xprtrdma) to occasionally force a signaled send
completion if it runs unsignaled most of the time.
xprtrdma does not require signaled completions for Send or FastReg
Work Requests, but does signal some LocalInv Work Requests. To
ensure that Send Queue house-keeping can run before the Send Queue
is more than half-consumed, xprtrdma forces a signaled completion
on occasion by counting the number of Send Queue Entries it
consumes. It currently does this by counting each ib_post_send as
one Entry.
Commit c9918ff56d ("xprtrdma: Add ro_unmap_sync method for FRWR")
introduced the ability for frwr_op_unmap_sync to post more than one
Work Request with a single post_send. Thus the underlying assumption
of one Send Queue Entry per ib_post_send is no longer true.
Also, FastReg Work Requests are currently never signaled. They
should be signaled once in a while, just as Send is, to keep the
accounting of consumed SQEs accurate.
While we're here, convert the CQCOUNT macros to the currently
preferred kernel coding style, which is inline functions.
Fixes: c9918ff56d ("xprtrdma: Add ro_unmap_sync method for FRWR")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
When a LOCALINV WR is flushed, the frmr is marked STALE, then
frwr_op_unmap_sync DMA-unmaps the frmr's SGL. These STALE frmrs
are then recovered when frwr_op_map hunts for an INVALID frmr to
use.
All other cases that need frmr recovery leave that SGL DMA-mapped.
The FRMR recovery path unconditionally DMA-unmaps the frmr's SGL.
To avoid DMA unmapping the SGL twice for flushed LOCAL_INV WRs,
alter the recovery logic (rather than the hot frwr_op_unmap_sync
path) to distinguish among these cases. This solution also takes
care of the case where multiple LOCAL_INV WRs are issued for the
same rpcrdma_req, some complete successfully, but some are flushed.
Reported-by: Vasco Steinmetz <linux@kyberraum.net>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Vasco Steinmetz <linux@kyberraum.net>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: the extra layer of indirection doesn't add value.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
An RPC Call message that is sent inline but that has a data payload
(ie, one or more items in rq_snd_buf's page list) must be "pulled
up:"
- call_allocate has to reserve enough RPC Call buffer space to
accommodate the data payload
- call_transmit has to memcopy the rq_snd_buf's page list and tail
into its head iovec before it is sent
As the inline threshold is increased beyond its current 1KB default,
however, this means data payloads of more than a few KB are copied
by the host CPU. For example, if the inline threshold is increased
just to 4KB, then NFS WRITE requests up to 4KB would involve a
memcpy of the NFS WRITE's payload data into the RPC Call buffer.
This is an undesirable amount of participation by the host CPU.
The inline threshold may be much larger than 4KB in the future,
after negotiation with a peer server.
Instead of copying the components of rq_snd_buf into its head iovec,
construct a gather list of these components, and send them all in
place. The same approach is already used in the Linux server's
RPC-over-RDMA reply path.
This mechanism also eliminates the need for rpcrdma_tail_pullup,
which is used to manage the XDR pad and trailing inline content when
a Read list is present.
This requires that the pages in rq_snd_buf's page list be DMA-mapped
during marshaling, and unmapped when a data-bearing RPC is
completed. This is slightly less efficient for very small I/O
payloads, but significantly more efficient as data payload size and
inline threshold increase past a kilobyte.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Have frwr's ro_unmap_sync recognize an invalidated rkey that appears
as part of a Receive completion. Local invalidation can be skipped
for that rkey.
Use an out-of-band signaling mechanism to indicate to the server
that the client is prepared to receive RDMA Send With Invalidate.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Send an RDMA-CM private message on connect, and look for one during
a connection-established event.
Both sides can communicate their various implementation limits.
Implementations that don't support this sideband protocol ignore it.
Once the client knows the server's inline threshold maxima, it can
adjust the use of Reply chunks, and eliminate most use of Position
Zero Read chunks. Moderately-sized I/O can be done using a pure
inline RDMA Send instead of RDMA operations that require memory
registration.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: The fields in the recv_wr do not vary. There is no need to
initialize them before each ib_post_recv(). This removes a large-ish
data structure from the stack.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Most of the fields in each send_wr do not vary. There is
no need to initialize them before each ib_post_send(). This removes
a large-ish data structure from the stack.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up.
Since commit fc66448549 ("xprtrdma: Split the completion queue"),
rpcrdma_ep_post_recv() no longer uses the "ep" argument.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up. The "ia" argument is no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Currently, each regbuf is allocated and DMA mapped at the same time.
This is done during transport creation.
When a device driver is unloaded, every DMA-mapped buffer in use by
a transport has to be unmapped, and then remapped to the new
device if the driver is loaded again. Remapping will have to be done
_after_ the connect worker has set up the new device.
But there's an ordering problem:
call_allocate, which invokes xprt_rdma_allocate which calls
rpcrdma_alloc_regbuf to allocate Send buffers, happens _before_
the connect worker can run to set up the new device.
Instead, at transport creation, allocate each buffer, but leave it
unmapped. Once the RPC carries these buffers into ->send_request, by
which time a transport connection should have been established,
check to see that the RPC's buffers have been DMA mapped. If not,
map them there.
When device driver unplug support is added, it will simply unmap all
the transport's regbufs, but it doesn't have to deallocate the
underlying memory.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The use of DMA_BIDIRECTIONAL is discouraged by DMA-API.txt.
Fortunately, xprtrdma now knows which direction I/O is going as
soon as it allocates each regbuf.
The RPC Call and Reply buffers are no longer the same regbuf. They
can each be labeled correctly now. The RPC Reply buffer is never
part of either a Send or Receive WR, but it can be part of Reply
chunk, which is mapped and registered via ->ro_map . So it is not
DMA mapped when it is allocated (DMA_NONE), to avoid a double-
mapping.
Since Receive buffers are no longer DMA_BIDIRECTIONAL and their
contents are never modified by the host CPU, DMA-API-HOWTO.txt
suggests that a DMA sync before posting each buffer should be
unnecessary. (See my_card_interrupt_handler).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Commit 949317464b ("xprtrdma: Limit number of RDMA segments in
RPC-over-RDMA headers") capped the number of chunks that may appear
in RPC-over-RDMA headers. The maximum header size can be estimated
and fixed to avoid allocating buffer space that is never used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
RPC-over-RDMA needs to separate its RPC call and reply buffers.
o When an RPC Call is sent, rq_snd_buf is DMA mapped for an RDMA
Send operation using DMA_TO_DEVICE
o If the client expects a large RPC reply, it DMA maps rq_rcv_buf
as part of a Reply chunk using DMA_FROM_DEVICE
The two mappings are for data movement in opposite directions.
DMA-API.txt suggests that if these mappings share a DMA cacheline,
bad things can happen. This could occur in the final bytes of
rq_snd_buf and the first bytes of rq_rcv_buf if the two buffers
happen to share a DMA cacheline.
On x86_64 the cacheline size is typically 8 bytes, and RPC call
messages are usually much smaller than the send buffer, so this
hasn't been a noticeable problem. But the DMA cacheline size can be
larger on other platforms.
Also, often rq_rcv_buf starts most of the way into a page, thus
an additional RDMA segment is needed to map and register the end of
that buffer. Try to avoid that scenario to reduce the cost of
registering and invalidating Reply chunks.
Instead of carrying a single regbuf that covers both rq_snd_buf and
rq_rcv_buf, each struct rpcrdma_req now carries one regbuf for
rq_snd_buf and one regbuf for rq_rcv_buf.
Some incidental changes worth noting:
- To clear out some spaghetti, refactor xprt_rdma_allocate.
- The value stored in rg_size is the same as the value stored in
the iov.length field, so eliminate rg_size
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Currently there's a hidden and indirect mechanism for finding the
rpcrdma_req that goes with an rpc_rqst. It depends on getting from
the rq_buffer pointer in struct rpc_rqst to the struct
rpcrdma_regbuf that controls that buffer, and then to the struct
rpcrdma_req it goes with.
This was done back in the day to avoid the need to add a per-rqst
pointer or to alter the buf_free API when support for RPC-over-RDMA
was introduced.
I'm about to change the way regbuf's work to support larger inline
thresholds. Now is a good time to replace this indirect mechanism
with something that is more straightforward. I guess this should be
considered a clean up.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
