The maximum size of a backchannel message on RPC-over-RDMA depends
on the connection's inline threshold. Today that threshold is
typically 1024 bytes, making the maximum message size 996 bytes.
The Linux server's CREATE_SESSION operation checks that the size
of callback Calls can be as large as 1044 bytes, to accommodate
RPCSEC_GSS. Thus CREATE_SESSION fails if a client advertises the
true message size maximum of 996 bytes.
But the server's backchannel currently does not support RPCSEC_GSS.
The actual maximum size it needs is much smaller. It is safe to
reduce the limit to enable NFSv4.1 on RDMA backchannel operation.
Signed-off-by: Chuck Lever <chuck.lever@oracle.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>
The qword_get() function NUL-terminates its output buffer. If the input
string is in hex format \xXXXX... and the same length as the output
buffer, there is an off-by-one:
int qword_get(char **bpp, char *dest, int bufsize)
{
...
while (len < bufsize) {
...
*dest++ = (h << 4) | l;
len++;
}
...
*dest = '\0';
return len;
}
This patch ensures the NUL terminator doesn't fall outside the output
buffer.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
- Remove usage of ib_query_device and instead store attributes in
ib_device struct
- Move iopoll out of block and into lib, rename to irqpoll, and use
in several places in the rdma stack as our new completion queue
polling library mechanism. Update the other block drivers that
already used iopoll to use the new mechanism too.
- Replace the per-entry GID table locks with a single GID table lock
- IPoIB multicast cleanup
- Cleanups to the IB MR facility
- Add support for 64bit extended IB counters
- Fix for netlink oops while parsing RDMA nl messages
- RoCEv2 support for the core IB code
- mlx4 RoCEv2 support
- mlx5 RoCEv2 support
- Cross Channel support for mlx5
- Timestamp support for mlx5
- Atomic support for mlx5
- Raw QP support for mlx5
- MAINTAINERS update for mlx4/mlx5
- Misc ocrdma, qib, nes, usNIC, cxgb3, cxgb4, mlx4, mlx5 updates
- Add support for remote invalidate to the iSER driver (pushed through the
RDMA tree due to dependencies, acknowledged by nab)
- Update to NFSoRDMA (pushed through the RDMA tree due to dependencies,
acknowledged by Bruce)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=5oO7
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Pull rdma updates from Doug Ledford:
"Initial roundup of 4.5 merge window patches
- Remove usage of ib_query_device and instead store attributes in
ib_device struct
- Move iopoll out of block and into lib, rename to irqpoll, and use
in several places in the rdma stack as our new completion queue
polling library mechanism. Update the other block drivers that
already used iopoll to use the new mechanism too.
- Replace the per-entry GID table locks with a single GID table lock
- IPoIB multicast cleanup
- Cleanups to the IB MR facility
- Add support for 64bit extended IB counters
- Fix for netlink oops while parsing RDMA nl messages
- RoCEv2 support for the core IB code
- mlx4 RoCEv2 support
- mlx5 RoCEv2 support
- Cross Channel support for mlx5
- Timestamp support for mlx5
- Atomic support for mlx5
- Raw QP support for mlx5
- MAINTAINERS update for mlx4/mlx5
- Misc ocrdma, qib, nes, usNIC, cxgb3, cxgb4, mlx4, mlx5 updates
- Add support for remote invalidate to the iSER driver (pushed
through the RDMA tree due to dependencies, acknowledged by nab)
- Update to NFSoRDMA (pushed through the RDMA tree due to
dependencies, acknowledged by Bruce)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (169 commits)
IB/mlx5: Unify CQ create flags check
IB/mlx5: Expose Raw Packet QP to user space consumers
{IB, net}/mlx5: Move the modify QP operation table to mlx5_ib
IB/mlx5: Support setting Ethernet priority for Raw Packet QPs
IB/mlx5: Add Raw Packet QP query functionality
IB/mlx5: Add create and destroy functionality for Raw Packet QP
IB/mlx5: Refactor mlx5_ib_qp to accommodate other QP types
IB/mlx5: Allocate a Transport Domain for each ucontext
net/mlx5_core: Warn on unsupported events of QP/RQ/SQ
net/mlx5_core: Add RQ and SQ event handling
net/mlx5_core: Export transport objects
IB/mlx5: Expose CQE version to user-space
IB/mlx5: Add CQE version 1 support to user QPs and SRQs
IB/mlx5: Fix data validation in mlx5_ib_alloc_ucontext
IB/sa: Fix netlink local service GFP crash
IB/srpt: Remove redundant wc array
IB/qib: Improve ipoib UD performance
IB/mlx4: Advertise RoCE v2 support
IB/mlx4: Create and use another QP1 for RoCEv2
IB/mlx4: Enable send of RoCE QP1 packets with IP/UDP headers
...
parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested},
inode_foo(inode) being mutex_foo(&inode->i_mutex).
Please, use those for access to ->i_mutex; over the coming cycle
->i_mutex will become rwsem, with ->lookup() done with it held
only shared.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
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>
To support the server-side of an NFSv4.1 backchannel on RDMA
connections, add a transport class that enables backward
direction messages on an existing forward channel connection.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Extra resources for handling backchannel requests have to be
pre-allocated when a transport instance is created. Set up
additional fields in svcxprt_rdma to track these resources.
The max_requests fields are elements of the RPC-over-RDMA
protocol, so they should be u32. To ensure that unsigned
arithmetic is used everywhere, some other fields in the
svcxprt_rdma struct are updated.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
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>
When the maximum payload size of NFS READ and WRITE was increased
by commit cc9a903d91 ("svcrdma: Change maximum server payload back
to RPCSVC_MAXPAYLOAD"), the size of struct svc_rdma_op_ctxt
increased to over 6KB (on x86_64). That makes allocating one of
these from a kmem_cache more likely to fail in situations when
system memory is exhausted.
Since I'm about to add a caller where this allocation must always
work _and_ it cannot sleep, pre-allocate ctxts for each connection.
Another motivation for this change is that NFSv4.x servers are
required by specification not to drop NFS requests. Pre-allocating
memory resources reduces the likelihood of a drop.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Be sure the completed ctxt is put in every path.
The xprt enqueue can take a while, so put the completed ctxt back
in circulation _before_ enqueuing the xprt.
Remove/disable debugging.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
kzalloc is used here, so setting the atomic fields to zero is
unnecessary. sc_ord is set again in handle_connect_req. The other
fields are re-initialized in svc_rdma_accept().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Bruce Fields <bfields@fieldses.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
kerberized NFSv4.1 mounts, and Scott Mayhew's work addressing ACK storms
that can affect some high-availability NFS setups.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=G7V3
-----END PGP SIGNATURE-----
Merge tag 'nfsd-4.5' of git://linux-nfs.org/~bfields/linux
Pull nfsd updates from Bruce Fields:
"Smaller bugfixes and cleanup, including a fix for a failures of
kerberized NFSv4.1 mounts, and Scott Mayhew's work addressing ACK
storms that can affect some high-availability NFS setups"
* tag 'nfsd-4.5' of git://linux-nfs.org/~bfields/linux:
nfsd: add new io class tracepoint
nfsd: give up on CB_LAYOUTRECALLs after two lease periods
nfsd: Fix nfsd leaks sunrpc module references
lockd: constify nlmsvc_binding structure
lockd: use to_delayed_work
nfsd: use to_delayed_work
Revert "svcrdma: Do not send XDR roundup bytes for a write chunk"
lockd: Register callbacks on the inetaddr_chain and inet6addr_chain
nfsd: Register callbacks on the inetaddr_chain and inet6addr_chain
sunrpc: Add a function to close temporary transports immediately
nfsd: don't base cl_cb_status on stale information
nfsd4: fix gss-proxy 4.1 mounts for some AD principals
nfsd: fix unlikely NULL deref in mach_creds_match
nfsd: minor consolidation of mach_cred handling code
nfsd: helper for dup of possibly NULL string
svcrpc: move some initialization to common code
nfsd: fix a warning message
nfsd: constify nfsd4_callback_ops structure
nfsd: recover: constify nfsd4_client_tracking_ops structures
svcrdma: Do not send XDR roundup bytes for a write chunk
Merge first patch-bomb from Andrew Morton:
- A few hotfixes which missed 4.4 becasue I was asleep. cc'ed to
-stable
- A few misc fixes
- OCFS2 updates
- Part of MM. Including pretty large changes to page-flags handling
and to thp management which have been buffered up for 2-3 cycles now.
I have a lot of MM material this time.
[ It turns out the THP part wasn't quite ready, so that got dropped from
this series - Linus ]
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (117 commits)
zsmalloc: reorganize struct size_class to pack 4 bytes hole
mm/zbud.c: use list_last_entry() instead of list_tail_entry()
zram/zcomp: do not zero out zcomp private pages
zram: pass gfp from zcomp frontend to backend
zram: try vmalloc() after kmalloc()
zram/zcomp: use GFP_NOIO to allocate streams
mm: add tracepoint for scanning pages
drivers/base/memory.c: fix kernel warning during memory hotplug on ppc64
mm/page_isolation: use macro to judge the alignment
mm: fix noisy sparse warning in LIBCFS_ALLOC_PRE()
mm: rework virtual memory accounting
include/linux/memblock.h: fix ordering of 'flags' argument in comments
mm: move lru_to_page to mm_inline.h
Documentation/filesystems: describe the shared memory usage/accounting
memory-hotplug: don't BUG() in register_memory_resource()
hugetlb: make mm and fs code explicitly non-modular
mm/swapfile.c: use list_for_each_entry_safe in free_swap_count_continuations
mm: /proc/pid/clear_refs: no need to clear VM_SOFTDIRTY in clear_soft_dirty_pmd()
mm: make sure isolate_lru_page() is never called for tail page
vmstat: make vmstat_updater deferrable again and shut down on idle
...
Mark those kmem allocations that are known to be easily triggered from
userspace as __GFP_ACCOUNT/SLAB_ACCOUNT, which makes them accounted to
memcg. For the list, see below:
- threadinfo
- task_struct
- task_delay_info
- pid
- cred
- mm_struct
- vm_area_struct and vm_region (nommu)
- anon_vma and anon_vma_chain
- signal_struct
- sighand_struct
- fs_struct
- files_struct
- fdtable and fdtable->full_fds_bits
- dentry and external_name
- inode for all filesystems. This is the most tedious part, because
most filesystems overwrite the alloc_inode method.
The list is far from complete, so feel free to add more objects.
Nevertheless, it should be close to "account everything" approach and
keep most workloads within bounds. Malevolent users will be able to
breach the limit, but this was possible even with the former "account
everything" approach (simply because it did not account everything in
fact).
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Greg Thelen <gthelen@google.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* bugfixes:
SUNRPC: Fixup socket wait for memory
SUNRPC: Fix a missing break in rpc_anyaddr()
pNFS/flexfiles: Fix an Oopsable typo in ff_mirror_match_fh()
NFS: Fix attribute cache revalidation
NFS: Ensure we revalidate attributes before using execute_ok()
NFS: Flush reclaim writes using FLUSH_COND_STABLE
NFS: Background flush should not be low priority
NFSv4.1/pnfs: Fixup an lo->plh_block_lgets imbalance in layoutreturn
NFSv4: Don't perform cached access checks before we've OPENed the file
NFS: Allow the combination pNFS and labeled NFS
NFS42: handle layoutstats stateid error
nfs: Fix race in __update_open_stateid()
nfs: fix missing assignment in nfs4_sequence_done tracepoint
This reverts commit 6f18dc8939.
Just as one example, it appears this code could do the wrong thing in
the case of a two-byte NFS READ that crosses a page boundary.
Chuck says: "In that case, nfsd would pass down an xdr_buf that has one
byte in a page, one byte in another page, and a two-byte XDR pad. The
logic introduced by this optimization would be fooled, and neither the
second byte nor the XDR pad would be written to the client."
Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
We're seeing hangs in the NFS client code, with loops of the form:
RPC: 30317 xmit incomplete (267368 left of 524448)
RPC: 30317 call_status (status -11)
RPC: 30317 call_transmit (status 0)
RPC: 30317 xprt_prepare_transmit
RPC: 30317 xprt_transmit(524448)
RPC: xs_tcp_send_request(267368) = -11
RPC: 30317 xmit incomplete (267368 left of 524448)
RPC: 30317 call_status (status -11)
RPC: 30317 call_transmit (status 0)
RPC: 30317 xprt_prepare_transmit
RPC: 30317 xprt_transmit(524448)
Turns out commit ceb5d58b21 ("net: fix sock_wake_async() rcu protection")
moved SOCKWQ_ASYNC_NOSPACE out of sock->flags and into sk->sk_wq->flags,
however it never tried to fix up the code in net/sunrpc.
The new idiom is to use the flags in the RCU protected struct socket_wq.
While we're at it, clear out the now redundant places where we set/clear
SOCKWQ_ASYNC_NOSPACE and SOCK_NOSPACE. In principle, sk_stream_wait_memory()
is supposed to set these for us, so we only need to clear them in the
particular case of our ->write_space() callback.
Fixes: ceb5d58b21 ("net: fix sock_wake_async() rcu protection")
Cc: Eric Dumazet <edumazet@google.com>
Cc: stable@vger.kernel.org # 4.4
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The missing break means that we always return EAFNOSUPPORT when
faced with a request for an IPv6 loopback address.
Reported-by: coverity (CID 401987)
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
These patches mostly fix send queue ordering issues inside the NFSoRDMA
client, but there are also two patches from Dan Carpenter fixing up smatch
warnings.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=kwhB
-----END PGP SIGNATURE-----
Merge tag 'nfs-rdma-4.5' of git://git.linux-nfs.org/projects/anna/nfs-rdma
NFS: NFSoRDMA Client Side Changes
These patches mostly fix send queue ordering issues inside the NFSoRDMA
client, but there are also two patches from Dan Carpenter fixing up smatch
warnings.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
* tag 'nfs-rdma-4.5' of git://git.linux-nfs.org/projects/anna/nfs-rdma:
xprtrdma: Revert commit e7104a2a96 ('xprtrdma: Cap req_cqinit').
xprtrdma: Invalidate in the RPC reply handler
xprtrdma: Add ro_unmap_sync method for all-physical registration
xprtrdma: Add ro_unmap_sync method for FMR
xprtrdma: Add ro_unmap_sync method for FRWR
xprtrdma: Introduce ro_unmap_sync method
xprtrdma: Move struct ib_send_wr off the stack
xprtrdma: Disable RPC/RDMA backchannel debugging messages
xprtrdma: xprt_rdma_free() must not release backchannel reqs
xprtrdma: Fix additional uses of spin_lock_irqsave(rb_lock)
xprtrdma: checking for NULL instead of IS_ERR()
xprtrdma: clean up some curly braces
xs_reclassify_socket4() and friends used to be called directly.
xs_reclassify_socket() is called instead nowadays.
The xs_reclassify_socketX() helper functions are empty when
CONFIG_DEBUG_LOCK_ALLOC is not defined. Drop them since they have no
callers.
Note that AF_LOCAL still calls xs_reclassify_socketu() directly but is
easily converted to generic xs_reclassify_socket().
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Add a function svc_age_temp_xprts_now() to close temporary transports
whose xpt_local matches the address passed in server_addr immediately
instead of waiting for them to be closed by the timer function.
The function is intended to be used by notifier_blocks that will be
added to nfsd and lockd that will run when an ip address is deleted.
This will eliminate the ACK storms and client hangs that occur in
HA-NFS configurations where nfsd & lockd is left running on the cluster
nodes all the time and the NFS 'service' is migrated back and forth
within a short timeframe.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Instead, use the cached copy of the attributes present on the device.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The root of the problem was that sends (especially unsignalled
FASTREG and LOCAL_INV Work Requests) were not properly flow-
controlled, which allowed a send queue overrun.
Now that the RPC/RDMA reply handler waits for invalidation to
complete, the send queue is properly flow-controlled. Thus this
limit is no longer necessary.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
There is a window between the time the RPC reply handler wakes the
waiting RPC task and when xprt_release() invokes ops->buf_free.
During this time, memory regions containing the data payload may
still be accessed by a broken or malicious server, but the RPC
application has already been allowed access to the memory containing
the RPC request's data payloads.
The server should be fenced from client memory containing RPC data
payloads _before_ the RPC application is allowed to continue.
This change also more strongly enforces send queue accounting. There
is a maximum number of RPC calls allowed to be outstanding. When an
RPC/RDMA transport is set up, just enough send queue resources are
allocated to handle registration, Send, and invalidation WRs for
each those RPCs at the same time.
Before, additional RPC calls could be dispatched while invalidation
WRs were still consuming send WQEs. When invalidation WRs backed
up, dispatching additional RPCs resulted in a send queue overrun.
Now, the reply handler prevents RPC dispatch until invalidation is
complete. This prevents RPC call dispatch until there are enough
send queue resources to proceed.
Still to do: If an RPC exits early (say, ^C), the reply handler has
no opportunity to perform invalidation. Currently, xprt_rdma_free()
still frees remaining RDMA resources, which could deadlock.
Additional changes are needed to handle invalidation properly in this
case.
Reported-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
physical's ro_unmap is synchronous already. The new ro_unmap_sync
method just has to DMA unmap all MRs associated with the RPC
request.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
FMR's ro_unmap method is already synchronous because ib_unmap_fmr()
is a synchronous verb. However, some improvements can be made here.
1. Gather all the MRs for the RPC request onto a list, and invoke
ib_unmap_fmr() once with that list. This reduces the number of
doorbells when there is more than one MR to invalidate
2. Perform the DMA unmap _after_ the MRs are unmapped, not before.
This is critical after invalidating a Write chunk.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
FRWR's ro_unmap is asynchronous. The new ro_unmap_sync posts
LOCAL_INV Work Requests and waits for them to complete before
returning.
Note also, DMA unmapping is now done _after_ invalidation.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
In the current xprtrdma implementation, some memreg strategies
implement ro_unmap synchronously (the MR is knocked down before the
method returns) and some asynchonously (the MR will be knocked down
and returned to the pool in the background).
To guarantee the MR is truly invalid before the RPC consumer is
allowed to resume execution, we need an unmap method that is
always synchronous, invoked from the RPC/RDMA reply handler.
The new method unmaps all MRs for an RPC. The existing ro_unmap
method unmaps only one MR at a time.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
For FRWR FASTREG and LOCAL_INV, move the ib_*_wr structure off
the stack. This allows frwr_op_map and frwr_op_unmap to chain
WRs together without limit to register or invalidate a set of MRs
with a single ib_post_send().
(This will be for chaining LOCAL_INV requests).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Preserve any rpcrdma_req that is attached to rpc_rqst's allocated
for the backchannel. Otherwise, after all the pre-allocated
backchannel req's are consumed, incoming backward calls start
writing on freed memory.
Somehow this hunk got lost.
Fixes: f531a5dbc4 ('xprtrdma: Pre-allocate backward rpc_rqst')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up.
rb_lock critical sections added in rpcrdma_ep_post_extra_recv()
should have first been converted to use normal spin_lock now that
the reply handler is a work queue.
The backchannel set up code should use the appropriate helper
instead of open-coding a rb_recv_bufs list add.
Problem introduced by glib patch re-ordering on my part.
Fixes: f531a5dbc4 ('xprtrdma: Pre-allocate backward rpc_rqst')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The rpcrdma_create_req() function returns error pointers or success. It
never returns NULL.
Fixes: f531a5dbc4 ('xprtrdma: Pre-allocate backward rpc_rqst and send/receive buffers')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
It doesn't matter either way, but the curly braces were clearly intended
here. It causes a Smatch warning.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Jan Stancek reported that I wrecked things for him by fixing things for
Vladimir :/
His report was due to an UNINTERRUPTIBLE wait getting -EINTR, which
should not be possible, however my previous patch made this possible by
unconditionally checking signal_pending().
We cannot use current->state as was done previously, because the
instruction after the store to that variable it can be changed. We must
instead pass the initial state along and use that.
Fixes: 68985633bc ("sched/wait: Fix signal handling in bit wait helpers")
Reported-by: Jan Stancek <jstancek@redhat.com>
Reported-by: Chris Mason <clm@fb.com>
Tested-by: Jan Stancek <jstancek@redhat.com>
Tested-by: Vladimir Murzin <vladimir.murzin@arm.com>
Tested-by: Chris Mason <clm@fb.com>
Reviewed-by: Paul Turner <pjt@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: tglx@linutronix.de
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: hpa@zytor.com
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The NFSv4.1 callback channel is currently broken because the receive
message will keep shrinking because the backchannel receive buffer size
never gets reset.
The easiest solution to this problem is instead of changing the receive
buffer, to rather adjust the copied request.
Fixes: 38b7631fbe ("nfs4: limit callback decoding to received bytes")
Cc: Benjamin Coddington <bcodding@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Pull networking fixes from David Miller:
"A lot of Thanksgiving turkey leftovers accumulated, here goes:
1) Fix bluetooth l2cap_chan object leak, from Johan Hedberg.
2) IDs for some new iwlwifi chips, from Oren Givon.
3) Fix rtlwifi lockups on boot, from Larry Finger.
4) Fix memory leak in fm10k, from Stephen Hemminger.
5) We have a route leak in the ipv6 tunnel infrastructure, fix from
Paolo Abeni.
6) Fix buffer pointer handling in arm64 bpf JIT,f rom Zi Shen Lim.
7) Wrong lockdep annotations in tcp md5 support, fix from Eric
Dumazet.
8) Work around some middle boxes which prevent proper handling of TCP
Fast Open, from Yuchung Cheng.
9) TCP repair can do huge kmalloc() requests, build paged SKBs
instead. From Eric Dumazet.
10) Fix msg_controllen overflow in scm_detach_fds, from Daniel
Borkmann.
11) Fix device leaks on ipmr table destruction in ipv4 and ipv6, from
Nikolay Aleksandrov.
12) Fix use after free in epoll with AF_UNIX sockets, from Rainer
Weikusat.
13) Fix double free in VRF code, from Nikolay Aleksandrov.
14) Fix skb leaks on socket receive queue in tipc, from Ying Xue.
15) Fix ifup/ifdown crach in xgene driver, from Iyappan Subramanian.
16) Fix clearing of persistent array maps in bpf, from Daniel
Borkmann.
17) In TCP, for the cross-SYN case, we don't initialize tp->copied_seq
early enough. From Eric Dumazet.
18) Fix out of bounds accesses in bpf array implementation when
updating elements, from Daniel Borkmann.
19) Fill gaps in RCU protection of np->opt in ipv6 stack, from Eric
Dumazet.
20) When dumping proxy neigh entries, we have to accomodate NULL
device pointers properly, from Konstantin Khlebnikov.
21) SCTP doesn't release all ipv6 socket resources properly, fix from
Eric Dumazet.
22) Prevent underflows of sch->q.qlen for multiqueue packet
schedulers, also from Eric Dumazet.
23) Fix MAC and unicast list handling in bnxt_en driver, from Jeffrey
Huang and Michael Chan.
24) Don't actively scan radar channels, from Antonio Quartulli"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (110 commits)
net: phy: reset only targeted phy
bnxt_en: Setup uc_list mac filters after resetting the chip.
bnxt_en: enforce proper storing of MAC address
bnxt_en: Fixed incorrect implementation of ndo_set_mac_address
net: lpc_eth: remove irq > NR_IRQS check from probe()
net_sched: fix qdisc_tree_decrease_qlen() races
openvswitch: fix hangup on vxlan/gre/geneve device deletion
ipv4: igmp: Allow removing groups from a removed interface
ipv6: sctp: implement sctp_v6_destroy_sock()
arm64: bpf: add 'store immediate' instruction
ipv6: kill sk_dst_lock
ipv6: sctp: add rcu protection around np->opt
net/neighbour: fix crash at dumping device-agnostic proxy entries
sctp: use GFP_USER for user-controlled kmalloc
sctp: convert sack_needed and sack_generation to bits
ipv6: add complete rcu protection around np->opt
bpf: fix allocation warnings in bpf maps and integer overflow
mvebu: dts: enable IP checksum with jumbo frames for Armada 38x on Port0
net: mvneta: enable setting custom TX IP checksum limit
net: mvneta: fix error path for building skb
...
This patch is a cleanup to make following patch easier to
review.
Goal is to move SOCK_ASYNC_NOSPACE and SOCK_ASYNC_WAITDATA
from (struct socket)->flags to a (struct socket_wq)->flags
to benefit from RCU protection in sock_wake_async()
To ease backports, we rename both constants.
Two new helpers, sk_set_bit(int nr, struct sock *sk)
and sk_clear_bit(int net, struct sock *sk) are added so that
following patch can change their implementation.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The principal name on a gss cred is used to setup the NFSv4.0 callback,
which has to have a client principal name to authenticate to.
That code wants the name to be in the form servicetype@hostname.
rpc.svcgssd passes down such names (and passes down no principal name at
all in the case the principal isn't a service principal).
gss-proxy always passes down the principal name, and passes it down in
the form servicetype/hostname@REALM. So we've been munging the name
gss-proxy passes down into the format the NFSv4.0 callback code expects,
or throwing away the name if we can't.
Since the introduction of the MACH_CRED enforcement in NFSv4.1, we've
also been using the principal name to verify that certain operations are
done as the same principal as was used on the original EXCHANGE_ID call.
For that application, the original name passed down by gss-proxy is also
useful.
Lack of that name in some cases was causing some kerberized NFSv4.1
mount failures in an Active Directory environment.
This fix only works in the gss-proxy case. The fix for legacy
rpc.svcgssd would be more involved, and rpc.svcgssd already has other
problems in the AD case.
Reported-and-tested-by: James Ralston <ralston@pobox.com>
Acked-by: Simo Sorce <simo@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
A truncated cb_compound request will cause the client to decode null or
data from a previous callback for nfs4.1 backchannel case, or uninitialized
data for the nfs4.0 case. This is because the path through
svc_process_common() advances the request's iov_base and decrements iov_len
without adjusting the overall xdr_buf's len field. That causes
xdr_init_decode() to set up the xdr_stream with an incorrect length in
nfs4_callback_compound().
Fixing this for the nfs4.1 backchannel case first requires setting the
correct iov_len and page_len based on the length of received data in the
same manner as the nfs4.0 case.
Then the request's xdr_buf length can be adjusted for both cases based upon
the remaining iov_len and page_len.
Signed-off-by: Benjamin Coddington <bcodding@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Minor optimization: when dealing with write chunk XDR roundup, do
not post a Write WR for the zero bytes in the pad. Simply update
the write segment in the RPC-over-RDMA header to reflect the extra
pad bytes.
The Reply chunk is also a write chunk, but the server does not use
send_write_chunks() to send the Reply chunk. That's OK in this case:
the server Upper Layer typically marshals the Reply chunk contents
in a single contiguous buffer, without a separate tail for the XDR
pad.
The comments and the variable naming refer to "chunks" but what is
really meant is "segments." The existing code sends only one
xdr_write_chunk per RPC reply.
The fix assumes this as well. When the XDR pad in the first write
chunk is reached, the assumption is the Write list is complete and
send_write_chunks() returns.
That will remain a valid assumption until the server Upper Layer can
support multiple bulk payload results per RPC.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
