- 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>
o Select the R_key to invalidate while the CPU cache still contains
the received RPC Call transport header, rather than waiting until
we're about to send the RPC Reply.
o Choose Send With Invalidate if there is exactly one distinct R_key
in the received transport header. If there's more than one, the
client will have to perform local invalidation after it has
already waited for remote invalidation.
Signed-off-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-----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=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
...
svc_xprt_release() invokes svc_free_res_pages(), which releases
pages between rq_respages and rq_next_page.
Historically, the RPC/RDMA transport has set these two pointers to
be different by one, which means:
- one page gets released when svc_recv returns 0. This normally
happens whenever one or more RDMA Reads need to be dispatched to
complete construction of an RPC Call.
- one page gets released after every call to svc_send.
In both cases, this released page is immediately refilled by
svc_alloc_arg. There does not seem to be a reason for releasing this
page.
To avoid this unnecessary memory allocator traffic, set rq_next_page
more carefully.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.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>
While sending each RPC Reply, svc_rdma_sendto allocates and DMA-
maps a separate buffer where the RPC/RDMA transport header is
constructed. The buffer is unmapped and released in the Send
completion handler. This is significant per-RPC overhead,
especially for small RPCs.
Instead, allocate and DMA-map a buffer, and cache it in each
svc_rdma_send_ctxt. This buffer and its mapping can be re-used
for each RPC, saving the cost of memory allocation and DMA
mapping.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: No current caller of svc_rdma_send's passes in a chained
WR. The logic that counts the chain length can be replaced with a
constant (1).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: Now that the send_wr is part of the svc_rdma_send_ctxt,
svc_rdma_post_send_wr is nearly empty.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Receive buffers are always the same size, but each Send WR has a
variable number of SGEs, based on the contents of the xdr_buf being
sent.
While assembling a Send WR, keep track of the number of SGEs so that
we don't exceed the device's maximum, or walk off the end of the
Send SGE array.
For now the Send path just fails if it exceeds the maximum.
The current logic in svc_rdma_accept bases the maximum number of
Send SGEs on the largest NFS request that can be sent or received.
In the transport layer, the limit is actually based on the
capabilities of the underlying device, not on properties of the
Upper Layer Protocol.
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.
Introduce a replacement to svc_rdma_op_ctxt's that is built
especially for the svcrdma Send path.
Subsequent patches will take advantage of this new structure 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 send_ctxt are not confused. As an
additional clean up, structure fields are renamed to conform with
kernel coding conventions.
Additional clean ups:
- Handle svc_rdma_send_ctxt_get allocation failure at each call
site, rather than pre-allocating and hoping we guessed correctly
- All send_ctxt_put call-sites request page freeing, so remove
the @free_pages argument
- All send_ctxt_put call-sites unmap SGEs, so fold that into
svc_rdma_send_ctxt_put
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: Since there's already a svc_rdma_op_ctxt being passed
around with the running count of mapped SGEs, drop unneeded
parameters to svc_rdma_post_send_wr().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: svc_rdma_dma_map_buf does mostly the same thing as
svc_rdma_dma_map_page, so let's fold these together.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.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>
Rather than releasing the incoming svc_rdma_recv_ctxt at the end of
svc_rdma_recvfrom, hold onto it until svc_rdma_sendto.
This permits the contents of the Receive buffer to be preserved
through svc_process and then referenced directly in sendto as it
constructs Write and Reply chunks to return to the client.
The real changes will come in subsequent patches.
Note: I cannot use ->xpo_release_rqst for this purpose because that
is called _before_ ->xpo_sendto. svc_rdma_sendto uses information in
the received Call transport header to construct the Reply transport
header, which is preserved in the RPC's Receive buffer.
The historical comment in svc_send() isn't helpful: it is already
obvious that ->xpo_release_rqst is being called before ->xpo_sendto,
but there is no explanation for this ordering going back to the
beginning of the git era.
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>
This change improves Receive efficiency by posting Receives only
on the same CPU that handles Receive completion. Improved latency
and throughput has been noted with this change.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The current check will always be true and will always jump to
err1, this looks dubious to me. I believe && should be used
instead of ||.
Detected by CoverityScan, CID#1450120 ("Logically Dead Code")
Fixes: 107c1d0a99 ("svcrdma: Avoid Send Queue overflow")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Sanity case: Catch the case where more Work Requests are being
posted to the Send Queue than there are Send Queue Entries.
This might happen if a client sends a chunk with more segments than
there are SQEs for the transport. The server can't send that reply,
so the transport will deadlock unless the client drops the RPC.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The server displays "svcrdma: failed to post Send WR (-107)" in the
kernel log when the client disconnects. This could flood the server's
log, so remove the message.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
req_maps are no longer used by the send path and can thus be removed.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Observed at Connectathon 2017.
If a client has underestimated the size of a Write or Reply chunk,
the Linux server writes as much payload data as it can, then it
recognizes there was a problem and closes the connection without
sending the transport header.
This creates a couple of problems:
<> The client never receives indication of the server-side failure,
so it continues to retransmit the bad RPC. Forward progress on
the transport is blocked.
<> The reply payload pages are not moved out of the svc_rqst, thus
they can be released by the RPC server before the RDMA Writes
have completed.
The new rdma_rw-ized helpers return a distinct error code when a
Write/Reply chunk overrun occurs, so it's now easy for the caller
(svc_rdma_sendto) to recognize this case.
Instead of dropping the connection, post an RDMA_ERROR message. The
client now sees an RDMA_ERROR and can properly terminate the RPC
transaction.
As part of the new logic, set up the same delayed release for these
payload pages as would have occurred in the normal case.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Now that svc_rdma_sendto has been renovated, svc_rdma_send_error can
be refactored to reduce code duplication and remove C structure-
based XDR encoding. It is also relocated to the source file that
contains its only caller.
This is a refactoring change only.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The current svcrdma sendto code path posts one RDMA Write WR at a
time. Each of these Writes typically carries a small number of pages
(for instance, up to 30 pages for mlx4 devices). That means a 1MB
NFS READ reply requires 9 ib_post_send() calls for the Write WRs,
and one for the Send WR carrying the actual RPC Reply message.
Instead, use the new rdma_rw API. The details of Write WR chain
construction and memory registration are taken care of in the RDMA
core. svcrdma can focus on the details of the RPC-over-RDMA
protocol. This gives three main benefits:
1. All Write WRs for one RDMA segment are posted in a single chain.
As few as one ib_post_send() for each Write chunk.
2. The Write path can now use FRWR to register the Write buffers.
If the device's maximum page list depth is large, this means a
single Write WR is needed for each RPC's Write chunk data.
3. The new code introduces support for RPCs that carry both a Write
list and a Reply chunk. This combination can be used for an NFSv4
READ where the data payload is large, and thus is removed from the
Payload Stream, but the Payload Stream is still larger than the
inline threshold.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Replace C structure-based XDR decoding with more portable code that
instead uses pointer arithmetic.
This is a refactoring change only.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: extract the logic to save pages under I/O into a helper to
add a big documenting comment without adding clutter in the send
path.
This is a refactoring change only.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Introduce a helper to DMA-map a reply's transport header before
sending it. This will in part replace the map vector cache.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: Move the ib_send_wr off the stack, and move common code
to post a Send Work Request into a helper.
This is a refactoring change only.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Replace C structure-based XDR decoding with pointer arithmetic.
Pointer arithmetic is considered more portable, and is used
throughout the kernel's existing XDR encoders. The gcc optimizer
generates similar assembler code either way.
Byte-swapping before a memory store on x86 typically results in an
instruction pipeline stall. Avoid byte-swapping when encoding a new
header.
svcrdma currently doesn't alter a connection's credit grant value
after the connection has been accepted, so it is effectively a
constant. Cache the byte-swapped value in a separate field.
Christoph suggested pulling the header encoding logic into the only
function that uses it.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Commit 5fdca65314 ("svcrdma: Renovate sendto chunk list parsing")
missed a spot. svc_rdma_xdr_get_reply_hdr_len() also assumes the
Write list has only one Write chunk. There's no harm in making this
code more general.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
svcrdma's current SQ accounting algorithm takes sc_lock and disables
bottom-halves while posting all RDMA Read, Write, and Send WRs.
This is relatively heavyweight serialization. And note that Write and
Send are already fully serialized by the xpt_mutex.
Using a single atomic_t should be all that is necessary to guarantee
that ib_post_send() is called only when there is enough space on the
send queue. This is what the other RDMA-enabled storage targets do.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The current sendto code appears to support clients that provide only
one of a Read list, a Write list, or a Reply chunk. My reading of
that code is that it doesn't support the following cases:
- Read list + Write list
- Read list + Reply chunk
- Write list + Reply chunk
- Read list + Write list + Reply chunk
The protocol allows more than one Read or Write chunk in those
lists. Some clients do send a Read list and Reply chunk
simultaneously. NFSv4 WRITE uses a Read list for the data payload,
and a Reply chunk because the GETATTR result in the reply can
contain a large object like an ACL.
Generalize one of the sendto code paths needed to support all of
the above cases, and attempt to ensure that only one pass is done
through the RPC Call's transport header to gather chunk list
information for building the reply.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Support Remote Invalidation. A private message is exchanged with
the client upon RDMA transport connect that indicates whether
Send With Invalidation may be used by the server to send RPC
replies. The invalidate_rkey is arbitrarily chosen from among
rkeys present in the RPC-over-RDMA header's chunk lists.
Send With Invalidate improves performance only when clients can
recognize, while processing an RPC reply, that an rkey has already
been invalidated. That has been submitted as a separate change.
In the future, the RPC-over-RDMA protocol might support Remote
Invalidation properly. The protocol needs to enable signaling
between peers to indicate when Remote Invalidation can be used
for each individual RPC.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Message from syslogd@klimt at Aug 18 17:00:37 ...
kernel:page:ffffea0020639b00 count:0 mapcount:0 mapping: (null) index:0x0
Aug 18 17:00:37 klimt kernel: flags: 0x2fffff80000000()
Aug 18 17:00:37 klimt kernel: page dumped because: VM_BUG_ON_PAGE(page_ref_count(page) == 0)
Aug 18 17:00:37 klimt kernel: kernel BUG at /home/cel/src/linux/linux-2.6/include/linux/mm.h:445!
Aug 18 17:00:37 klimt kernel: RIP: 0010:[<ffffffffa05c21c1>] svc_rdma_sendto+0x641/0x820 [rpcrdma]
send_reply() assigns its page argument as the first page of ctxt. On
error, send_reply() already invokes svc_rdma_put_context(ctxt, 1);
which does a put_page() on that very page. No need to do that again
as svc_rdma_sendto exits.
Fixes: 3e1eeb9808 ("svcrdma: Close connection when a send error occurs")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The ctxt's count field is overloaded to mean the number of pages in
the ctxt->page array and the number of SGEs in the ctxt->sge array.
Typically these two numbers are the same.
However, when an inline RPC reply is constructed from an xdr_buf
with a tail iovec, the head and tail often occupy the same page,
but each are DMA mapped independently. In that case, ->count equals
the number of pages, but it does not equal the number of SGEs.
There's one more SGE, for the tail iovec. Hence there is one more
DMA mapping than there are pages in the ctxt->page array.
This isn't a real problem until the server's iommu is enabled. Then
each RPC reply that has content in that iovec orphans a DMA mapping
that consists of real resources.
krb5i and krb5p always populate that tail iovec. After a couple
million sent krb5i/p RPC replies, the NFS server starts behaving
erratically. Reboot is needed to clear the problem.
Fixes: 9d11b51ce7 ("svcrdma: Fix send_reply() scatter/gather set-up")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Get a fresh op_ctxt in send_reply() instead of in svc_rdma_sendto().
This ensures that svc_rdma_put_context() is invoked only once if
send_reply() fails.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Calling ib_poll_cq() to sort through WCs during a completion is a
common pattern amongst RDMA consumers. Since commit 14d3a3b249
("IB: add a proper completion queue abstraction"), WC sorting can
be handled by the IB core.
By converting to this new API, svcrdma 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.
This new API also aims each completion at a function that is
specific to the WR's opcode. Thus the ctxt->wr_op field and the
switch in process_context is replaced by a set of methods that
handle each completion type.
Because each ib_cqe carries a pointer to a completion method, the
core can now post operations on a consumer's QP, and handle the
completions itself.
The server's rdma_stat_sq_poll and rdma_stat_sq_prod metrics are no
longer updated.
As a clean up, the cq_event_handler, the dto_tasklet, and all
associated locking is removed, as they are no longer referenced or
used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Fix several issues with svc_rdma_send_error():
- Post a receive buffer to replace the one that was consumed by
the incoming request
- Posting a send should use DMA_TO_DEVICE, not DMA_FROM_DEVICE
- No need to put_page _and_ free pages in svc_rdma_put_context
- Make sure the sge is set up completely in case the error
path goes through svc_rdma_unmap_dma()
- Replace the use of ENOSYS, which has a reserved meaning
Related fixes in svc_rdma_recvfrom():
- Don't leak the ctxt associated with the incoming request
- Don't close the connection after sending an error reply
- Let svc_rdma_send_error() figure out the right header error code
As a last clean up, move svc_rdma_send_error() to svc_rdma_sendto.c
with other similar functions. There is some common logic in these
functions that could someday be combined to reduce code duplication.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Devesh Sharma <devesh.sharma@broadcom.com>
Tested-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Clean up: Most svc_rdma_post_recv() call sites close the transport
connection when a receive cannot be posted. Wrap that in a common
helper.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Devesh Sharma <devesh.sharma@broadcom.com>
Tested-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The NFS server's XDR encoders adds an XDR pad for content in the
xdr_buf page list at the beginning of the xdr_buf's tail buffer.
On RDMA transports, Write chunks are sent separately and without an
XDR pad.
If a Write chunk is being sent, strip off the pad in the tail buffer
so that inline content following the Write chunk remains XDR-aligned
when it is sent to the client.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=294
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
When the Linux NFS server writes an odd-length data item into a
Write chunk, it finishes with XDR pad bytes. If the data item is
smaller than the Write chunk, the pad bytes are written at the end
of the data item, but still inside the chunk (ie, in the
application's buffer). Since this is direct data placement, that
exposes the pad bytes.
XDR pad bytes are inserted in order to preserve the XDR alignment
of the next XDR data item in an XDR stream. But Write chunks do not
appear in the payload XDR stream, and only one data item is allowed
in each chunk. Thus XDR padding is not needed in a Write chunk.
With NFSv4, the Linux NFS server places the results of any
operations that follow an NFSv4 READ or READLINK in the xdr_buf's
tail. Those results also should never be sent as a part of a Write
chunk. The current logic in send_write_chunks() appears to assume
that the xdr_buf's tail contains only pad bytes (ie, NFSv3).
The server should write only the contents of the xdr_buf's page list
in a Write chunk. If there's more than an XDR pad in the tail, that
needs to go inline or in the Reply chunk.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=294
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The client provides the location of Write chunks into which the
server writes bulk payload. The client provides these when the
Upper Layer Protocol wants direct data placement and the Binding
allows it. (For NFS, this is READ and READLINK operations).
The client also provides the location of a Reply chunk into which
the server writes the non-bulk part of an RPC reply. The client
provides this chunk whenever it believes the reply can be larger
than its receive buffers.
The server then uses the presence of these chunks to determine how
it will form its reply message.
svc_rdma_sendto() was looking for Write and Reply chunks multiple
times for every reply message. It would be more efficient to do it
just once.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.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: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Acked-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Pre-requisite to use map_xdr in the backchannel code.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Clean up.
These functions can otherwise fail, so check for page allocation
failures too.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug 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: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug 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: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This patch split up struct ib_send_wr so that all non-trivial verbs
use their own structure which embedds struct ib_send_wr. This dramaticly
shrinks the size of a WR for most common operations:
sizeof(struct ib_send_wr) (old): 96
sizeof(struct ib_send_wr): 48
sizeof(struct ib_rdma_wr): 64
sizeof(struct ib_atomic_wr): 96
sizeof(struct ib_ud_wr): 88
sizeof(struct ib_fast_reg_wr): 88
sizeof(struct ib_bind_mw_wr): 96
sizeof(struct ib_sig_handover_wr): 80
And with Sagi's pending MR rework the fast registration WR will also be
down to a reasonable size:
sizeof(struct ib_fastreg_wr): 64
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> [srp, srpt]
Reviewed-by: Chuck Lever <chuck.lever@oracle.com> [sunrpc]
Tested-by: Haggai Eran <haggaie@mellanox.com>
Tested-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Kernel coding conventions frown upon having large nontrivial
functions in header files, and the preference these days is to
allow the compiler to make inlining decisions if possible.
As these functions are re-homed into a .c file, be sure that
comparisons with fields in struct rpcrdma_msg are with be32
constants.
This is a refactoring change; no behavior change is intended.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The Linux NFS server returns garbage in the data payload of inline
NFS/RDMA READ replies. These are READs of under 1000 bytes or so
where the client has not provided either a reply chunk or a write
list.
The NFS server delivers the data payload for an NFS READ reply to
the transport in an xdr_buf page list. If the NFS client did not
provide a reply chunk or a write list, send_reply() is supposed to
set up a separate sge for the page containing the READ data, and
another sge for XDR padding if needed, then post all of the sges via
a single SEND Work Request.
The problem is send_reply() does not advance through the xdr_buf
when setting up scatter/gather entries for SEND WR. It always calls
dma_map_xdr with xdr_off set to zero. When there's more than one
sge, dma_map_xdr() sets up the SEND sge's so they all point to the
xdr_buf's head.
The current Linux NFS/RDMA client always provides a reply chunk or
a write list when performing an NFS READ over RDMA. Therefore, it
does not exercise this particular case. The Linux server has never
had to use more than one extra sge for building RPC/RDMA replies
with a Linux client.
However, an NFS/RDMA client _is_ allowed to send small NFS READs
without setting up a write list or reply chunk. The NFS READ reply
fits entirely within the inline reply buffer in this case. This is
perhaps a more efficient way of performing NFS READs that the Linux
NFS/RDMA client may some day adopt.
Fixes: b432e6b3d9 ('svcrdma: Change DMA mapping logic to . . .')
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=285
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
At the 2015 LSF/MM, it was requested that memory allocation
call sites that request GFP_KERNEL allocations in a loop should be
annotated with __GFP_NOFAIL.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
In send_write_chunks(), we have:
for (xdr_off = rqstp->rq_res.head[0].iov_len, chunk_no = 0;
xfer_len && chunk_no < arg_ary->wc_nchunks;
chunk_no++) {
. . .
}
Note that arg_ary->wc_nchunk is in network byte-order. For the
comparison to work correctly, both have to be in native byte-order.
In send_reply_chunks, we have:
write_len = min(xfer_len, htonl(ch->rs_length));
xfer_len is in native byte-order, and ch->rs_length is in
network byte-order. be32_to_cpu() is the correct byte swap
for ch->rs_length.
As an additional clean up, replace ntohl() with be32_to_cpu() in
a few other places.
This appears to address a problem with large rsize hangs while
using PHYSICAL memory registration. I suspect that is the only
registration mode that uses more than one chunk element.
BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=248
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
xdr_start() can return the wrong rmsgp address if an assumption
about how the xdr_buf was constructed changes. When it gets it
wrong, the client receives a reply that has gibberish in the
RPC/RDMA header, preventing it from matching a waiting RPC request.
Instead, make (and document) just one assumption: that the RDMA
header for the client's RPC call is at the start of the first page
in rq_pages.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Current convention is to avoid using BUG_ON() in places where an
oops could cause complete system failure.
Replace BUG_ON() call sites in svcrdma with an assertion error
message and allow execution to continue safely.
Some BUG_ON() calls are removed because they have never fired in
production (that we are aware of).
Some WARN_ON() calls are also replaced where a back trace is not
helpful; e.g., in a workqueue task.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Function send_write() must stop creating sges when it reaches the device
max and return the amount sent in the RDMA Write to the caller.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This patch refactors the NFSRDMA server marshalling logic to
remove the intermediary map structures. It also fixes an existing bug
where the NFSRDMA server was not minding the device fast register page
list length limitations.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
The xdr_off value in dma_map_xdr gets passed to ib_dma_map_page as the
offset into the page to be mapped. This calculation does not correctly
take into account the case where the data starts at some offset into
the page. Increment the xdr_off by the page_base to ensure that it is
respected.
Cc: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The server regression was caused by the addition of rq_next_page
(afc59400d6). There were a few places that
were missed with the update of the rq_respages array.
Signed-off-by: Tom Tucker <tom@ogc.us>
Tested-by: Steve Wise <swise@ogc.us>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The svcrdma transport was un-marshalling requests in-place. This resulted
in sparse warnings due to __beXX data containing both NBO and HBO data.
The code has been restructured to do byte-swapping as the header is
parsed instead of when the header is validated immediately after receipt.
Also moved extern declarations for the workqueue and memory pools to the
private header file.
Signed-off-by: Tom Tucker <tom@ogc.us>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
There are several error paths in the code that do not unmap DMA. This
patch adds calls to svc_rdma_unmap_dma to free these DMA contexts.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
There was logic in the send path that assumed that a page containing data
to send to the client has a KVA. This is not always the case and can result
in data corruption when page_address returns zero and we end up DMA mapping
zero.
This patch changes the bus mapping logic to avoid page_address() where
necessary and converts all calls from ib_dma_map_single to ib_dma_map_page
in order to keep the map/unmap calls symmetric.
Signed-off-by: Tom Tucker <tom@ogc.us>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The svcrdma module was incorrectly unmapping the RPCRDMA header page.
On IBM pserver systems this causes a resource leak that results in
running out of bus address space (10 cthon iterations will reproduce it).
The code was mapping the full page but only unmapping the actual header
length. The fix is to only map the header length.
I also cleaned up the use of ib_dma_map_page() calls since the unmap
logic always uses ib_dma_unmap_single(). I made these symmetrical.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
These fixes resolved crashes due to resource leak BUG_ON checks. The
resource leaks were detected by introducing asynchronous transport errors.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
printk formats in prior commit were reversed/incorrect.
Compiled without warning on x86 and x86_64, but detected on ppc.
Signed-off-by: Tom Talpey <tmtalpey@gmail.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Certain client-provided RPCRDMA chunk alignments result in an
additional scatter/gather entry, which triggered nfs/rdma server
assertions incorrectly. OpenSolaris nfs/rdma client connectathon
testing was blocked by these in the special/locking section.
Signed-off-by: Tom Talpey <tmtalpey@gmail.com>
Cc: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Use FRMR to map local RPC reply data. This allows RDMA_WRITE to send reply
data using a single WR. The FRMR is invalidated by linking the LOCAL_INV WR
to the RDMA_SEND message used to complete the reply.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Add per-device dma_mapping_ops support for CONFIG_X86_64 as POWER
architecture does:
This enables us to cleanly fix the Calgary IOMMU issue that some devices
are not behind the IOMMU (http://lkml.org/lkml/2008/5/8/423).
I think that per-device dma_mapping_ops support would be also helpful for
KVM people to support PCI passthrough but Andi thinks that this makes it
difficult to support the PCI passthrough (see the above thread). So I
CC'ed this to KVM camp. Comments are appreciated.
A pointer to dma_mapping_ops to struct dev_archdata is added. If the
pointer is non NULL, DMA operations in asm/dma-mapping.h use it. If it's
NULL, the system-wide dma_ops pointer is used as before.
If it's useful for KVM people, I plan to implement a mechanism to register
a hook called when a new pci (or dma capable) device is created (it works
with hot plugging). It enables IOMMUs to set up an appropriate
dma_mapping_ops per device.
The major obstacle is that dma_mapping_error doesn't take a pointer to the
device unlike other DMA operations. So x86 can't have dma_mapping_ops per
device. Note all the POWER IOMMUs use the same dma_mapping_error function
so this is not a problem for POWER but x86 IOMMUs use different
dma_mapping_error functions.
The first patch adds the device argument to dma_mapping_error. The patch
is trivial but large since it touches lots of drivers and dma-mapping.h in
all the architecture.
This patch:
dma_mapping_error() doesn't take a pointer to the device unlike other DMA
operations. So we can't have dma_mapping_ops per device.
Note that POWER already has dma_mapping_ops per device but all the POWER
IOMMUs use the same dma_mapping_error function. x86 IOMMUs use device
argument.
[akpm@linux-foundation.org: fix sge]
[akpm@linux-foundation.org: fix svc_rdma]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix bnx2x]
[akpm@linux-foundation.org: fix s2io]
[akpm@linux-foundation.org: fix pasemi_mac]
[akpm@linux-foundation.org: fix sdhci]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix sparc]
[akpm@linux-foundation.org: fix ibmvscsi]
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Avi Kivity <avi@qumranet.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add a dma map count in order to verify that all DMA mapping resources
have been freed when the transport is closed.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Use the new svc_rdma_req_map data type for mapping the client side memory
to the server side memory. Move the DMA mapping to the context pointed to
by each WR individually so that it is unmapped after the WR completes.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
If an error is encountered trying to post a recv buffer in send_reply,
free the passed in context. Return an error to the caller so it is
aware that the request was not posted.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
The svcrdma transport provider currently allocates receive buffers
to the RQ through the xpo_release_rqst method. This approach is overly
complicated since it means that the rqstp rq_xprt_ctxt has to be
selectively set based on whether the RPC is going to be processed
immediately or deferred. Instead, just post the receive buffer when
we are certain that we are replying in the send_reply function.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Chuck Lever