Prevent svc_rdma_build_writes() from walking off the end of a Write
chunk's segment array. Caught with KASAN.
The test that this fix replaces is invalid, and might have been left
over from an earlier prototype of the PCL work.
Fixes: 7a1cbfa180 ("svcrdma: Use parsed chunk lists to construct RDMA Writes")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
There are currently three separate purposes being served by single
tracepoints. Split them up, as was done with wc_send.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Relieve contention on sc_rw_ctxt_lock by converting rdma->sc_rw_ctxts
to an llist.
The goal is to reduce the average overhead of Send completions,
because a transport's completion handlers are single-threaded on
one CPU core. This change reduces CPU utilization of each 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>
Fix some spelling mistakes in comments:
succes ==> success
Signed-off-by: Zheng Yongjun <zhengyongjun3@huawei.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
These fields are no longer used.
The size of struct svc_rdma_recv_ctxt is now less than 300 bytes on
x86_64, down from 2440 bytes.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Currently the generic RPC server layer calls svc_rdma_recvfrom()
twice to retrieve an RPC message that uses Read chunks. I'm not
exactly sure why this design was chosen originally.
Instead, let's wait for the Read chunk completion inline in the
first call to svc_rdma_recvfrom().
The goal is to eliminate some page allocator churn.
rdma_read_complete() replaces pages in the second svc_rqst by
calling put_page() repeatedly while the upper layer waits for the
request to be constructed, which adds unnecessary NFS WRITE round-
trip latency.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Tom Talpey <tom@talpey.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>
Now that we have an efficient mechanism to update these two stats,
let's start maintaining them again.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
An efficient way to handle multiple Read chunks is to post them all
together and then take a single completion. This is also how the
code is already structured: when the Read completion fires, all
portions of the incoming RPC message are available to be assembled.
The difficult problem is setting up the Read sink buffers so that
the server pulls the client's data into place, making subsequent
pull-up unnecessary. There are several cases:
* No Read chunks. No-op.
* One data item Read chunk. This is the fast case, where the inline
part of the RPC-over-RDMA message becomes the head and tail, and
the data item chunk is placed in buf->pages.
* A Position-zero Read chunk. Treated like TCP: the Read chunk is
pulled into contiguous pages.
+ A Position-zero Read chunk with data item chunks. Treated like
TCP: all of the Read chunks are pulled into contiguous pages.
+ Multiple data item chunks. Treated like TCP: the inline part is
copied and the data item chunks are pulled into contiguous pages.
The "*" cases are already supported. This patch adds support for the
"+" cases.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
As a pre-requisite for handling multiple Read chunks in each Read
list, convert svc_rdma_recv_read_chunk() to use the new parsed Read
chunk list.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
I'm about to change the purpose of ri_chunklen: Instead of tracking
the number of bytes in one Read chunk, it will track the total
number of bytes in the Read list. Rename it for clarity.
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_send_reply_chunk() so that it Sends only the parts
of rq_res that do not contain a result payload.
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>
Refactor for subsequent changes.
Constify the xdr_buf argument to ensure the code here does not
modify it, and to enable callers to pass in a
"const struct xdr_buf *".
At the same time, rename the helper functions, which emit RDMA
Writes, not RDMA Sends, and add documenting comments.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Clean up: Ensure the code in rw.c does not modify the argument, and
enable callers to also use "const struct xdr_buf *".
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
When space in the Reply chunk runs out in the middle of a segment,
we end up passing a zero-length SGL to rdma_rw_ctx_init(), and it
oopses.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Drop duplicate words in net/sunrpc/.
Also fix "Anyone" to be "Any one".
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs@vger.kernel.org
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>
Commit 07d0ff3b0c ("svcrdma: Clean up Read chunk path") moved the
page saver logic so that it gets executed event when an error occurs.
In that case, the I/O is never posted, and those pages are then
leaked. Errors in this path, however, are quite rare.
Fixes: 07d0ff3b0c ("svcrdma: Clean up Read chunk path")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Clean up: Replace a dprintk call site.
This is the last remaining dprintk call site in svc_rdma_rw.c, so
remove dprintk infrastructure as well.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
- De-duplicate code
- Rename the tracepoint with "_err" to allow enabling via glob
- Report the sg_cnt for the failing rw_ctx
- Fix a dumb signage issue
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
These trace points are misnamed:
trace_svcrdma_encode_wseg
trace_svcrdma_encode_write
trace_svcrdma_encode_reply
trace_svcrdma_encode_rseg
trace_svcrdma_encode_read
trace_svcrdma_encode_pzr
Because they actually trace posting on the Send Queue. Let's rename
them so that I can add trace points in the chunk list encoders that
actually do trace chunk list encoding events.
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>
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>
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 7649773293 ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
sg_alloc_table_chained() currently allows the caller to provide one
preallocated SGL and returns if the requested number isn't bigger than
size of that SGL. This is used to inline an SGL for an IO request.
However, scattergather code only allows that size of the 1st preallocated
SGL to be SG_CHUNK_SIZE(128). This means a substantial amount of memory
(4KB) is claimed for the SGL for each IO request. If the I/O is small, it
would be prudent to allocate a smaller SGL.
Introduce an extra parameter to sg_alloc_table_chained() and
sg_free_table_chained() for specifying size of the preallocated SGL.
Both __sg_free_table() and __sg_alloc_table() assume that each SGL has the
same size except for the last one. Change the code to allow both functions
to accept a variable size for the 1st preallocated SGL.
[mkp: attempted to clarify commit desc]
Cc: Christoph Hellwig <hch@lst.de>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Ewan D. Milne <emilne@redhat.com>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: netdev@vger.kernel.org
Cc: linux-nvme@lists.infradead.org
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.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>
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
struct boo entry[];
};
instance = kmalloc(sizeof(struct foo) + count * sizeof(struct boo), GFP_KERNEL);
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
instance = kmalloc(struct_size(instance, entry, count), GFP_KERNEL);
This code was detected with the help of Coccinelle.
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
In the rpc server, When something happens that might be reason to wake
up a thread to do something, what we do is
- modify xpt_flags, sk_sock->flags, xpt_reserved, or
xpt_nr_rqsts to indicate the new situation
- call svc_xprt_enqueue() to decide whether to wake up a thread.
svc_xprt_enqueue may require multiple conditions to be true before
queueing up a thread to handle the xprt. In the SMP case, one of the
other CPU's may have set another required condition, and in that case,
although both CPUs run svc_xprt_enqueue(), it's possible that neither
call sees the writes done by the other CPU in time, and neither one
recognizes that all the required conditions have been set. A socket
could therefore be ignored indefinitely.
Add memory barries to ensure that any svc_xprt_enqueue() call will
always see the conditions changed by other CPUs before deciding to
ignore a socket.
I've never seen this race reported. In the unlikely event it happens,
another event will usually come along and the problem will fix itself.
So I don't think this is worth backporting to stable.
Chuck tried this patch and said "I don't see any performance
regressions, but my server has only a single last-level CPU cache."
Tested-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
missing Chuck's fixes for the problem with callbacks over GSS from
multi-homed servers, and a smaller fix from Laura Abbott.
-----BEGIN PGP SIGNATURE-----
iQIcBAABAgAGBQJbftA8AAoJECebzXlCjuG+QPMQALieEKkX0YoqRhPz5G+RrWFy
KgOBFAoiRcjFQD6wMt9FzD6qYEZqSJ+I2b+K5N3BkdyDDQu845iD0wK0zBGhMgLm
7ith85nphIMbe18+5jPorqAsI9RlfBQjiSGw1MEx5dicLQQzTObHL5q+l5jcWna4
jWS3yUKv1URpOsR1hIryw74ktSnhuH8n//zmntw8aWrCkq3hnXOZK/agtYxZ7Viv
V3kiQsiNpL2FPRcHN7ejhLUTnRkkuD2iYKrzP/SpTT/JfdNEUXlMhKkAySogNpus
nvR9X7hwta8Lgrt7PSB9ibFTXtCupmuICg5mbDWy6nXea2NvpB01QhnTzrlX17Eh
Yfk/18z95b6Qs1v4m3SI8ESmyc6l5dMZozLudtHzifyCqooWZriEhCR1PlQfQ/FJ
4cYQ8U/qiMiZIJXL7N2wpSoSaWR5bqU1rXen29Np1WEDkiv4Nf5u2fsCXzv0ZH2C
ReWpNkbnNxsNiKpp4geBZtlcSEU1pk+1PqE0MagTdBV3iptiUHRSP4jR7qLnc0zT
J1lCvU7Fodnt9vNSxMpt2Jd6XxQ6xtx7n6aMQAiYFnXDs+hP2hPnJVCScnYW3L6R
2r1sHRKKeoOzCJ2thw+zu4lOwMm7WPkJPWAYfv90reWkiKoy2vG0S9P7wsNGoJuW
fuEjB2b9pow1Ffynat6q
=JnLK
-----END PGP SIGNATURE-----
Merge tag 'nfsd-4.19-1' of git://linux-nfs.org/~bfields/linux
Pull nfsd updates from Bruce Fields:
"Chuck Lever fixed a problem with NFSv4.0 callbacks over GSS from
multi-homed servers.
The only new feature is a minor bit of protocol (change_attr_type)
which the client doesn't even use yet.
Other than that, various bugfixes and cleanup"
* tag 'nfsd-4.19-1' of git://linux-nfs.org/~bfields/linux: (27 commits)
sunrpc: Add comment defining gssd upcall API keywords
nfsd: Remove callback_cred
nfsd: Use correct credential for NFSv4.0 callback with GSS
sunrpc: Extract target name into svc_cred
sunrpc: Enable the kernel to specify the hostname part of service principals
sunrpc: Don't use stack buffer with scatterlist
rpc: remove unneeded variable 'ret' in rdma_listen_handler
nfsd: use true and false for boolean values
nfsd: constify write_op[]
fs/nfsd: Delete invalid assignment statements in nfsd4_decode_exchange_id
NFSD: Handle full-length symlinks
NFSD: Refactor the generic write vector fill helper
svcrdma: Clean up Read chunk path
svcrdma: Avoid releasing a page in svc_xprt_release()
nfsd: Mark expected switch fall-through
sunrpc: remove redundant variables 'checksumlen','blocksize' and 'data'
nfsd: fix leaked file lock with nfs exported overlayfs
nfsd: don't advertise a SCSI layout for an unsupported request_queue
nfsd: fix corrupted reply to badly ordered compound
nfsd: clarify check_op_ordering
...
Simplify the error handling at the tail of recv_read_chunk() by
re-arranging rq_pages[] housekeeping and documenting it properly.
NB: In this path, svc_rdma_recvfrom returns zero. Therefore no
subsequent reply processing is done on the svc_rqstp, and thus the
rq_respages field does not need to be updated.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Since neither ib_post_send() nor ib_post_recv() modify the data structure
their second argument points at, declare that argument const. This change
makes it necessary to declare the 'bad_wr' argument const too and also to
modify all ULPs that call ib_post_send(), ib_post_recv() or
ib_post_srq_recv(). This patch does not change any functionality but makes
it possible for the compiler to verify whether the
ib_post_(send|recv|srq_recv) really do not modify the posted work request.
To make this possible, only one cast had to be introduce that casts away
constness, namely in rpcrdma_post_recvs(). The only way I can think of to
avoid that cast is to introduce an additional loop in that function or to
change the data type of bad_wr from struct ib_recv_wr ** into int
(an index that refers to an element in the work request list). However,
both approaches would require even more extensive changes than this
patch.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The call in svc_rdma_post_chunk_ctxt() does actually use bad_wr.
Fixes: ed288d74a9 ("net/xprtrdma: Simplify ib_post_(send|recv|srq_recv)() calls")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Instead of declaring and passing a dummy 'bad_wr' pointer, pass NULL
as third argument to ib_post_(send|recv|srq_recv)().
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Anna Schumaker <Anna.Schumaker@netapp.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The current Receive path uses an array of pages which are allocated
and DMA mapped when each Receive WR is posted, and then handed off
to the upper layer in rqstp::rq_arg. The page flip releases unused
pages in the rq_pages pagelist. This mechanism introduces a
significant amount of overhead.
So instead, kmalloc the Receive buffer, and leave it DMA-mapped
while the transport remains connected. This confers a number of
benefits:
* Each Receive WR requires only one receive SGE, no matter how large
the inline threshold is. This helps the server-side NFS/RDMA
transport operate on less capable RDMA devices.
* The Receive buffer is left allocated and mapped all the time. This
relieves svc_rdma_post_recv from the overhead of allocating and
DMA-mapping a fresh buffer.
* svc_rdma_wc_receive no longer has to DMA unmap the Receive buffer.
It has to DMA sync only the number of bytes that were received.
* svc_rdma_build_arg_xdr no longer has to free a page in rq_pages
for each page in the Receive buffer, making it a constant-time
function.
* The Receive buffer is now plugged directly into the rq_arg's
head[0].iov_vec, and can be larger than a page without spilling
over into rq_arg's page list. This enables simplification of
the RDMA Read path in subsequent patches.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Currently svc_rdma_recv_ctxt_put's callers have to know whether they
want to free the ctxt's pages or not. This means the human
developers have to know when and why to set that free_pages
argument.
Instead, the ctxt should carry that information with it so that
svc_rdma_recv_ctxt_put does the right thing no matter who is
calling.
We want to keep track of the number of pages in the Receive buffer
separately from the number of pages pulled over by RDMA Read. This
is so that the correct number of pages can be freed properly and
that number is well-documented.
So now, rc_hdr_count is the number of pages consumed by head[0]
(ie., the page index where the Read chunk should start); and
rc_page_count is always the number of pages that need to be released
when the ctxt is put.
The @free_pages argument is no longer needed.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
svc_rdma_op_ctxt's are pre-allocated and maintained on a per-xprt
free list. This eliminates the overhead of calling kmalloc / kfree,
both of which grab a globally shared lock that disables interrupts.
To reduce contention further, separate the use of these objects in
the Receive and Send paths in svcrdma.
Subsequent patches will take advantage of this separation by
allocating real resources which are then cached in these objects.
The allocations are freed when the transport is torn down.
I've renamed the structure so that static type checking can be used
to ensure that uses of op_ctxt and recv_ctxt are not confused. As an
additional clean up, structure fields are renamed to conform with
kernel coding conventions.
As a final clean up, helpers related to recv_ctxt are moved closer
to the functions that use them.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This includes:
* Posting on the Send and Receive queues
* Send, Receive, Read, and Write completion
* Connect upcalls
* QP errors
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This includes:
* Transport accept and tear-down
* Decisions about using Write and Reply chunks
* Each RDMA segment that is handled
* Whenever an RDMA_ERR is sent
As a clean-up, I've standardized the order of the includes, and
removed some now redundant dprintk call sites.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
A single NFSv4 WRITE compound can often have three operations:
PUTFH, WRITE, then GETATTR.
When the WRITE payload is sent in a Read chunk, the client places
the GETATTR in the inline part of the RPC/RDMA message, just after
the WRITE operation (sans payload). The position value in the Read
chunk enables the receiver to insert the Read chunk at the correct
place in the received XDR stream; that is between the WRITE and
GETATTR.
According to RFC 8166, an NFS/RDMA client does not have to add XDR
round-up to the Read chunk that carries the WRITE payload. The
receiver adds XDR round-up padding if it is absent and the
receiver's XDR decoder requires it to be present.
Commit 193bcb7b37 ("svcrdma: Populate tail iovec when receiving")
attempted to add support for receiving such a compound so that just
the WRITE payload appears in rq_arg's page list, and the trailing
GETATTR is placed in rq_arg's tail iovec. (TCP just strings the
whole compound into the head iovec and page list, without regard
to the alignment of the WRITE payload).
The server transport logic also had to accommodate the optional XDR
round-up of the Read chunk, which it did simply by lengthening the
tail iovec when round-up was needed. This approach is adequate for
the NFSv2 and NFSv3 WRITE decoders.
Unfortunately it is not sufficient for nfsd4_decode_write. When the
Read chunk length is a couple of bytes less than PAGE_SIZE, the
computation at the end of nfsd4_decode_write allows argp->pagelen to
go negative, which breaks the logic in read_buf that looks for the
tail iovec.
The result is that a WRITE operation whose payload length is just
less than a multiple of a page succeeds, but the subsequent GETATTR
in the same compound fails with NFS4ERR_OP_ILLEGAL because the XDR
decoder can't find it. Clients ignore the error, but they must
update their attribute cache via a separate round trip.
As nfsd4_decode_write appears to expect the payload itself to always
have appropriate XDR round-up, have svc_rdma_build_normal_read_chunk
add the Read chunk XDR round-up to the page_len rather than
lengthening the tail iovec.
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Fixes: 193bcb7b37 ("svcrdma: Populate tail iovec when receiving")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
