There are currently three separate purposes being served by a single
tracepoint here. They need to be split up.
svcrdma_wc_send:
- status is always zero, so there's no value in recording it.
- vendor_err is meaningless unless status is not zero, so
there's no value in recording it.
- This tracepoint is needed only when developing modifications,
so it should be left disabled most of the time.
svcrdma_wc_send_flush:
- As above, needed only rarely, and not an error.
svcrdma_wc_send_err:
- This tracepoint can be left persistently enabled because
completion errors are run-time problems (except for FLUSHED_ERR).
- Tracepoint name now ends in _err to reflect its purpose.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
/proc/lock_stat indicates the the sc_send_lock is heavily
contended when the server is under load from a single client.
To address this, convert the send_ctxt free list to an llist.
Returning an item to the send_ctxt cache is now waitless, which
reduces the instruction path length in the single-threaded Send
handler (svc_rdma_wc_send).
The goal is to enable the ib_comp_wq worker to handle a higher
RPC/RDMA Send completion rate given the same CPU resources. This
change reduces CPU utilization of Send completion by 2-3% on my
server.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-By: Tom Talpey <tom@talpey.com>
Because wake_up() takes an IRQ-safe lock, it can be expensive,
especially to call inside of a single-threaded completion handler.
What's more, the Send wait queue almost never has waiters, so
most of the time, this is an expensive no-op.
As always, the goal is to reduce the average overhead of each
completion, because a transport's completion handlers are single-
threaded on one CPU core. This change reduces CPU utilization of
the Send completion thread by 2-3% on my server.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-By: Tom Talpey <tom@talpey.com>
Capture error codes in @ret, which is passed to the send_err
tracepoint, so that they can be logged when something goes awry.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
svc_rdma_sendto() now waits for the NIC hardware to finish with
the pages backing rq_res. We still have to release the page array
in some cases, but now it's always safe to immediately re-use the
page backing rq_res's head buffer.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Currently svc_rdma_sendto() migrates xdr_buf pages into a separate
page list and NULLs out a bunch of entries in rq_pages while the
pages are under I/O. The Send completion handler then frees those
pages later.
Instead, let's wait for the Send completion, then handle page
releasing in the nfsd thread. I'd like to avoid the cost of 250+
put_page() calls in the Send completion handler, which is single-
threaded.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Refactor a bit of commonly used logic so that every site that wants
a close deferred to an nfsd thread does all the right things
(set_bit(XPT_CLOSE) then enqueue).
Also, once XPT_CLOSE is set on a transport, it is never cleared. If
XPT_CLOSE is already set, then the close is already being handled
and the enqueue can be skipped.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
We already have trace_svcrdma_decode_rseg(), which records each
ingress Read segment. Instead of reporting those again when they
are about to be posted as RDMA Reads, let's fire one tracepoint
before posting each type of chunk.
So we'll get:
nfsd-1998 [002] 321.666615: svcrdma_decode_rseg: cq.id=4 cid=42 segno=0 position=0 192@0x013ca9ebfae14000:0xb0010b05
nfsd-1998 [002] 321.666615: svcrdma_decode_rseg: cq.id=4 cid=42 segno=1 position=0 7688@0x013ca9ebf914e000:0xb0010a05
nfsd-1998 [002] 321.666615: svcrdma_decode_rseg: cq.id=4 cid=42 segno=2 position=0 28@0x013ca9ebfae15000:0xb0010905
nfsd-1998 [002] 321.666622: svcrdma_decode_rqst: cq.id=4 cid=42 xid=0x013ca9eb vers=1 credits=128 proc=RDMA_NOMSG hdrlen=100
nfsd-1998 [002] 321.666642: svcrdma_post_read_chunk: cq.id=3 cid=112 sqecount=3
kworker/2:1H-221 [002] 321.673949: svcrdma_wc_read: cq.id=3 cid=112 status=SUCCESS (0/0x0)
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Refactor: svc_rdma_map_reply_msg() is restructured to DMA map only
the parts of rq_res that do not contain a result payload.
This change has been tested to confirm that it does not cause a
regression in the no Write chunk and single Write chunk cases.
Multiple Write chunks have not been tested.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
When counting the number of SGEs needed to construct a Send request,
do not count result payloads. And, when copying the Reply message
into the pull-up buffer, result payloads are not to be copied to the
Send buffer.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Refactor: Instead of re-parsing the ingress RPC Call transport
header when constructing the egress RPC Reply transport header, use
the new parsed Write list and Reply chunk, which are version-
agnostic and already XDR decoded.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Refactor: Instead of re-parsing the ingress RPC Call transport
header when constructing RDMA Writes, use the new parsed chunk lists
for the Write list and Reply chunk, which are version-agnostic and
already XDR-decoded.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The only RPC/RDMA ordering requirement between RDMA Writes and RDMA
Sends is that the responder must post the Writes on the Send queue
before posting the Send that conveys the RPC Reply for that Write
payload.
The Linux NFS server implementation now has a transport method that
can post result Payload Writes earlier than svc_rdma_sendto:
->xpo_result_payload()
This gets RDMA Writes going earlier so they are more likely to be
complete at the remote end before the Send completes.
Some care must be taken with pulled-up Replies. We don't want to
push the Write chunk and then send the same payload data via Send.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Have the NFSD encoders annotate the boundaries of every
direct-data-placement eligible result data payload. Then change
svcrdma to use that annotation instead of the xdr->page_len
when handling Write chunks.
For NFSv4 on RDMA, that enables the ability to recognize multiple
result payloads per compound. This is a pre-requisite for supporting
multiple Write chunks per RPC transaction.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Clean up: "result payload" is a less confusing name for these
payloads. "READ payload" reflects only the NFS usage.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
This was discovered using O_DIRECT at the client side, with small
unaligned file offsets or IOs that span multiple file pages.
Fixes: e248aa7be8 ("svcrdma: Remove max_sge check at connect time")
Signed-off-by: Dan Aloni <dan@kernelim.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Jason tells me that a ULP cannot rely on getting an ESTABLISHED
and DISCONNECTED event pair for each connection, so transport
reference counting in the CM event handler will never be reliable.
Now that we have ib_drain_qp(), svcrdma should no longer need to
hold transport references while Sends and Receives are posted. So
remove the get/put call sites in the CM event handlers.
This eliminates a significant source of locked memory bus traffic.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
First, refactor: Dereference the svc_rdma_send_ctxt inside
svc_rdma_send() instead of at every call site.
Then, it can be passed into trace_svcrdma_post_send() to get the
proper completion ID.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Set up a completion ID in each svc_rdma_send_ctxt. The ID is used
to match an incoming Send completion to a transport and to a
previous ib_post_send().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
- Use the _err naming convention instead
- Remove display of kernel memory address of the controlling xprt
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Like svc_rdma_send_error(), have svc_rdma_send_error_msg() handle
any error conditions internally, rather than duplicating that
recovery logic at every call site.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The common "send RDMA_ERR" function should be in svc_rdma_sendto.c,
since that is where the other Send-related functions are located.
So from here, I will beef up svc_rdma_send_error_msg() and deprecate
svc_rdma_send_error().
A generic svc_rdma_send_error_msg() will need to handle both
ERR_CHUNK and ERR_VERS. Copy that logic from svc_rdma_send_error()
to svc_rdma_send_error_msg().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Another step towards making svc_rdma_send_error_msg() and
svc_rdma_send_error() similar enough to eliminate one of them.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Commit 4757d90b15 ("svcrdma: Report Write/Reply chunk overruns")
made an effort to preserve I/O pages until RDMA Write completion.
In a subsequent patch, I intend to de-duplicate the two functions
that send ERR_CHUNK responses. Pull the save_io_pages() call out of
svc_rdma_send_error_msg() to make it more like
svc_rdma_send_error().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
It appears that the RPC/RDMA transport does not need serialization
of calls to its xpo_sendto method. Move the mutex into the socket
methods that still need that serialization.
Tail latencies are unambiguously better with this patch applied.
fio randrw 8KB 70/30 on NFSv3, smaller numbers are better:
clat percentiles (usec):
With xpt_mutex:
r | 99.99th=[ 8848]
w | 99.99th=[ 9634]
Without xpt_mutex:
r | 99.99th=[ 8586]
w | 99.99th=[ 8979]
Serializing the construction of RPC/RDMA transport headers is not
really necessary at this point, because the Linux NFS server
implementation never changes its credit grant on a connection. If
that should change, then svc_rdma_sendto will need to serialize
access to the transport's credit grant fields.
Reported-by: kbuild test robot <lkp@intel.com>
[ cel: fix uninitialized variable warning ]
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Utilize the xpo_release_rqst transport method to ensure that each
rqstp's svc_rdma_recv_ctxt object is released even when the server
cannot return a Reply for that rqstp.
Without this fix, each RPC whose Reply cannot be sent leaks one
svc_rdma_recv_ctxt. This is a 2.5KB structure, a 4KB DMA-mapped
Receive buffer, and any pages that might be part of the Reply
message.
The leak is infrequent unless the network fabric is unreliable or
Kerberos is in use, as GSS sequence window overruns, which result
in connection loss, are more common on fast transports.
Fixes: 3a88092ee3 ("svcrdma: Preserve Receive buffer until svc_rdma_sendto")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
On some platforms, DMA mapping part of a page is more costly than
copying bytes. Indeed, not involving the I/O MMU can help the
RPC/RDMA transport scale better for tiny I/Os across more RDMA
devices. This is because interaction with the I/O MMU is eliminated
for each of these small I/Os. Without the explicit unmapping, the
NIC no longer needs to do a costly internal TLB shoot down for
buffers that are just a handful of bytes.
Since pull-up is now a more a frequent operation, I've introduced a
trace point in the pull-up path. It can be used for debugging or
user-space tools that count pull-up frequency.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Performance optimization: Avoid syncing the transport buffer twice
when Reply buffer pull-up is necessary.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Same idea as the receive-side changes I did a while back: use
xdr_stream helpers rather than open-coding the XDR chunk list
encoders. This builds the Reply transport header from beginning to
end without backtracking.
As additional clean-ups, fill in documenting comments for the XDR
encoders and sprinkle some trace points in the new encoding
functions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Preparing for subsequent patches, no behavior change expected.
Pass the RPC Call's svc_rdma_recv_ctxt deeper into the sendto()
path. This enables passing more information about Requester-
provided Write and Reply chunks into those lower-level functions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Preparing for subsequent patches, no behavior change expected.
Pass the RPC Call's svc_rdma_recv_ctxt deeper into the sendto()
path. This enables passing more information about Requester-
provided Write and Reply chunks into those lower-level functions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Preparing for subsequent patches, no behavior change expected.
Pass the RPC Call's svc_rdma_recv_ctxt deeper into the sendto()
path. This enables passing more information about Requester-
provided Write and Reply chunks into the lower-level send
functions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Cache the locations of the Requester-provided Write list and Reply
chunk so that the Send path doesn't need to parse the Call header
again.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Introduce a helper function to compute the XDR pad size of a
variable-length XDR object.
Clean up: Replace open-coded calculation of XDR pad sizes.
I'm sure I haven't found every instance of this calculation.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
This error path is almost never executed. Found by code inspection.
Fixes: 99722fe4d5 ("svcrdma: Persistently allocate and DMA-map Send buffers")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
svcrdma expects that the payload falls precisely into the xdr_buf
page vector. This does not seem to be the case for
nfsd4_encode_readv().
This code is called only when fops->splice_read is missing or when
RQ_SPLICE_OK is clear, so it's not a noticeable problem in many
common cases.
Add new transport method: ->xpo_read_payload so that when a READ
payload does not fit exactly in rq_res's page vector, the XDR
encoder can inform the RPC transport exactly where that payload is,
without the payload's XDR pad.
That way, when a Write chunk is present, the transport knows what
byte range in the Reply message is supposed to be matched with the
chunk.
Note that the Linux NFS server implementation of NFS/RDMA can
currently handle only one Write chunk per RPC-over-RDMA message.
This simplifies the implementation of this fix.
Fixes: b042098063 ("nfsd4: allow exotic read compounds")
Buglink: https://bugzilla.kernel.org/show_bug.cgi?id=198053
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Capture the total size of Sends, the size of DMA map and the
matching DMA unmap to ensure operation is correct.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
These can result in a lot of log noise, and are able to be triggered
by client misbehavior. Since there are trace points in these
handlers now, there's no need to spam the log.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Two and a half years ago, the client was changed to use gathered
Send for larger inline messages, in commit 655fec6987 ("xprtrdma:
Use gathered Send for large inline messages"). Several fixes were
required because there are a few in-kernel device drivers whose
max_sge is 3, and these were broken by the change.
Apparently my memory is going, because some time later, I submitted
commit 25fd86eca1 ("svcrdma: Don't overrun the SGE array in
svc_rdma_send_ctxt"), and after that, commit f3c1fd0ee2 ("svcrdma:
Reduce max_send_sges"). These too incorrectly assumed in-kernel
device drivers would have more than a few Send SGEs available.
The fix for the server side is not the same. This is because the
fundamental problem on the server is that, whether or not the client
has provisioned a chunk for the RPC reply, the server must squeeze
even the most complex RPC replies into a single RDMA Send. Failing
in the send path because of Send SGE exhaustion should never be an
option.
Therefore, instead of failing when the send path runs out of SGEs,
switch to using a bounce buffer mechanism to handle RPC replies that
are too complex for the device to send directly. That allows us to
remove the max_sge check to enable drivers with small max_sge to
work again.
Reported-by: Don Dutile <ddutile@redhat.com>
Fixes: 25fd86eca1 ("svcrdma: Don't overrun the SGE array in ...")
Cc: stable@vger.kernel.org
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
xpo_prep_reply_hdr are not used now.
It was defined for tcp transport only, however it cannot be
called indirectly, so let's move it to its caller and
remove unused callback.
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
