Trond made a change to the server's tcp logic that allows a fast
client to better take advantage of high bandwidth networks, but
may increase the risk that a single client could starve other
clients; a new sunrpc.svc_rpc_per_connection_limit parameter
should help mitigate this in the (hopefully unlikely) event this
becomes a problem in practice.
Tom Haynes added a minimal flex-layout pnfs server, which is of
no use in production for now--don't build it unless you're doing
client testing or further server development.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=mo54
-----END PGP SIGNATURE-----
Merge tag 'nfsd-4.8' of git://linux-nfs.org/~bfields/linux
Pull nfsd updates from Bruce Fields:
"Highlights:
- Trond made a change to the server's tcp logic that allows a fast
client to better take advantage of high bandwidth networks, but may
increase the risk that a single client could starve other clients;
a new sunrpc.svc_rpc_per_connection_limit parameter should help
mitigate this in the (hopefully unlikely) event this becomes a
problem in practice.
- Tom Haynes added a minimal flex-layout pnfs server, which is of no
use in production for now--don't build it unless you're doing
client testing or further server development"
* tag 'nfsd-4.8' of git://linux-nfs.org/~bfields/linux: (32 commits)
nfsd: remove some dead code in nfsd_create_locked()
nfsd: drop unnecessary MAY_EXEC check from create
nfsd: clean up bad-type check in nfsd_create_locked
nfsd: remove unnecessary positive-dentry check
nfsd: reorganize nfsd_create
nfsd: check d_can_lookup in fh_verify of directories
nfsd: remove redundant zero-length check from create
nfsd: Make creates return EEXIST instead of EACCES
SUNRPC: Detect immediate closure of accepted sockets
SUNRPC: accept() may return sockets that are still in SYN_RECV
nfsd: allow nfsd to advertise multiple layout types
nfsd: Close race between nfsd4_release_lockowner and nfsd4_lock
nfsd/blocklayout: Make sure calculate signature/designator length aligned
xfs: abstract block export operations from nfsd layouts
SUNRPC: Remove unused callback xpo_adjust_wspace()
SUNRPC: Change TCP socket space reservation
SUNRPC: Add a server side per-connection limit
SUNRPC: Micro optimisation for svc_data_ready
SUNRPC: Call the default socket callbacks instead of open coding
SUNRPC: lock the socket while detaching it
...
This modification is useful for debugging issues that happen while
the socket is being initialised.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
We're seeing traces of the following form:
[10952.396347] svc: transport ffff88042ba4a 000 dequeued, inuse=2
[10952.396351] svc: tcp_accept ffff88042ba4 a000 sock ffff88042a6e4c80
[10952.396362] nfsd: connect from 10.2.6.1, port=187
[10952.396364] svc: svc_setup_socket ffff8800b99bcf00
[10952.396368] setting up TCP socket for reading
[10952.396370] svc: svc_setup_socket created ffff8803eb10a000 (inet ffff88042b75b800)
[10952.396373] svc: transport ffff8803eb10a000 put into queue
[10952.396375] svc: transport ffff88042ba4a000 put into queue
[10952.396377] svc: server ffff8800bb0ec000 waiting for data (to = 3600000)
[10952.396380] svc: transport ffff8803eb10a000 dequeued, inuse=2
[10952.396381] svc_recv: found XPT_CLOSE
[10952.396397] svc: svc_delete_xprt(ffff8803eb10a000)
[10952.396398] svc: svc_tcp_sock_detach(ffff8803eb10a000)
[10952.396399] svc: svc_sock_detach(ffff8803eb10a000)
[10952.396412] svc: svc_sock_free(ffff8803eb10a000)
i.e. an immediate close of the socket after initialisation.
The culprit appears to be the test at the end of svc_tcp_init, which
checks if the newly created socket is in the TCP_ESTABLISHED state,
and immediately closes it if not. The evidence appears to suggest that
the socket might still be in the SYN_RECV state at this time.
The fix is to check for both states, and then to add a check in
svc_tcp_state_change() to ensure we don't close the socket when
it transitions into TCP_ESTABLISHED.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Highlights include:
Stable bugfixes:
- nfs: don't create zero-length requests
- Several LAYOUTGET bugfixes
Features:
- Several performance related features
- More aggressive caching when we can rely on close-to-open cache
consistency
- Remove serialisation of O_DIRECT reads and writes
- Optimise several code paths to not flush to disk unnecessarily. However
allow for the idiosyncracies of pNFS for those layout types that need
to issue a LAYOUTCOMMIT before the metadata can be updated on the server.
- SUNRPC updates to the client data receive path
- pNFS/SCSI support RH/Fedora dm-mpath device nodes
- pNFS files/flexfiles can now use unprivileged ports when the generic NFS
mount options allow it.
Bugfixes:
- Don't use RDMA direct data placement together with data integrity or
privacy security flavours
- Remove the RDMA ALLPHYSICAL memory registration mode as it has potential
security holes.
- Several layout recall fixes to improve NFSv4.1 protocol compliance.
- Fix an Oops in the pNFS files and flexfiles connection setup to the DS
- Allow retry of operations that used a returned delegation stateid
- Don't mark the inode as revalidated if a LAYOUTCOMMIT is outstanding
- Fix writeback races in nfs4_copy_range() and nfs42_proc_deallocate()
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=WNyE
-----END PGP SIGNATURE-----
Merge tag 'nfs-for-4.8-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client updates from Trond Myklebust:
"Highlights include:
Stable bugfixes:
- nfs: don't create zero-length requests
- several LAYOUTGET bugfixes
Features:
- several performance related features
- more aggressive caching when we can rely on close-to-open
cache consistency
- remove serialisation of O_DIRECT reads and writes
- optimise several code paths to not flush to disk unnecessarily.
However allow for the idiosyncracies of pNFS for those layout
types that need to issue a LAYOUTCOMMIT before the metadata can
be updated on the server.
- SUNRPC updates to the client data receive path
- pNFS/SCSI support RH/Fedora dm-mpath device nodes
- pNFS files/flexfiles can now use unprivileged ports when
the generic NFS mount options allow it.
Bugfixes:
- Don't use RDMA direct data placement together with data
integrity or privacy security flavours
- Remove the RDMA ALLPHYSICAL memory registration mode as
it has potential security holes.
- Several layout recall fixes to improve NFSv4.1 protocol
compliance.
- Fix an Oops in the pNFS files and flexfiles connection
setup to the DS
- Allow retry of operations that used a returned delegation
stateid
- Don't mark the inode as revalidated if a LAYOUTCOMMIT is
outstanding
- Fix writeback races in nfs4_copy_range() and
nfs42_proc_deallocate()"
* tag 'nfs-for-4.8-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (104 commits)
pNFS: Actively set attributes as invalid if LAYOUTCOMMIT is outstanding
NFSv4: Clean up lookup of SECINFO_NO_NAME
NFSv4.2: Fix warning "variable ‘stateids’ set but not used"
NFSv4: Fix warning "no previous prototype for ‘nfs4_listxattr’"
SUNRPC: Fix a compiler warning in fs/nfs/clnt.c
pNFS: Remove redundant smp_mb() from pnfs_init_lseg()
pNFS: Cleanup - do layout segment initialisation in one place
pNFS: Remove redundant stateid invalidation
pNFS: Remove redundant pnfs_mark_layout_returned_if_empty()
pNFS: Clear the layout metadata if the server changed the layout stateid
pNFS: Cleanup - don't open code pnfs_mark_layout_stateid_invalid()
NFS: pnfs_mark_matching_lsegs_return() should match the layout sequence id
pNFS: Do not set plh_return_seq for non-callback related layoutreturns
pNFS: Ensure layoutreturn acts as a completion for layout callbacks
pNFS: Fix CB_LAYOUTRECALL stateid verification
pNFS: Always update the layout barrier seqid on LAYOUTGET
pNFS: Always update the layout stateid if NFS_LAYOUT_INVALID_STID is set
pNFS: Clear the layout return tracking on layout reinitialisation
pNFS: LAYOUTRETURN should only update the stateid if the layout is valid
nfs: don't create zero-length requests
...
Pull userns vfs updates from Eric Biederman:
"This tree contains some very long awaited work on generalizing the
user namespace support for mounting filesystems to include filesystems
with a backing store. The real world target is fuse but the goal is
to update the vfs to allow any filesystem to be supported. This
patchset is based on a lot of code review and testing to approach that
goal.
While looking at what is needed to support the fuse filesystem it
became clear that there were things like xattrs for security modules
that needed special treatment. That the resolution of those concerns
would not be fuse specific. That sorting out these general issues
made most sense at the generic level, where the right people could be
drawn into the conversation, and the issues could be solved for
everyone.
At a high level what this patchset does a couple of simple things:
- Add a user namespace owner (s_user_ns) to struct super_block.
- Teach the vfs to handle filesystem uids and gids not mapping into
to kuids and kgids and being reported as INVALID_UID and
INVALID_GID in vfs data structures.
By assigning a user namespace owner filesystems that are mounted with
only user namespace privilege can be detected. This allows security
modules and the like to know which mounts may not be trusted. This
also allows the set of uids and gids that are communicated to the
filesystem to be capped at the set of kuids and kgids that are in the
owning user namespace of the filesystem.
One of the crazier corner casees this handles is the case of inodes
whose i_uid or i_gid are not mapped into the vfs. Most of the code
simply doesn't care but it is easy to confuse the inode writeback path
so no operation that could cause an inode write-back is permitted for
such inodes (aka only reads are allowed).
This set of changes starts out by cleaning up the code paths involved
in user namespace permirted mounts. Then when things are clean enough
adds code that cleanly sets s_user_ns. Then additional restrictions
are added that are possible now that the filesystem superblock
contains owner information.
These changes should not affect anyone in practice, but there are some
parts of these restrictions that are changes in behavior.
- Andy's restriction on suid executables that does not honor the
suid bit when the path is from another mount namespace (think
/proc/[pid]/fd/) or when the filesystem was mounted by a less
privileged user.
- The replacement of the user namespace implicit setting of MNT_NODEV
with implicitly setting SB_I_NODEV on the filesystem superblock
instead.
Using SB_I_NODEV is a stronger form that happens to make this state
user invisible. The user visibility can be managed but it caused
problems when it was introduced from applications reasonably
expecting mount flags to be what they were set to.
There is a little bit of work remaining before it is safe to support
mounting filesystems with backing store in user namespaces, beyond
what is in this set of changes.
- Verifying the mounter has permission to read/write the block device
during mount.
- Teaching the integrity modules IMA and EVM to handle filesystems
mounted with only user namespace root and to reduce trust in their
security xattrs accordingly.
- Capturing the mounters credentials and using that for permission
checks in d_automount and the like. (Given that overlayfs already
does this, and we need the work in d_automount it make sense to
generalize this case).
Furthermore there are a few changes that are on the wishlist:
- Get all filesystems supporting posix acls using the generic posix
acls so that posix_acl_fix_xattr_from_user and
posix_acl_fix_xattr_to_user may be removed. [Maintainability]
- Reducing the permission checks in places such as remount to allow
the superblock owner to perform them.
- Allowing the superblock owner to chown files with unmapped uids and
gids to something that is mapped so the files may be treated
normally.
I am not considering even obvious relaxations of permission checks
until it is clear there are no more corner cases that need to be
locked down and handled generically.
Many thanks to Seth Forshee who kept this code alive, and putting up
with me rewriting substantial portions of what he did to handle more
corner cases, and for his diligent testing and reviewing of my
changes"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (30 commits)
fs: Call d_automount with the filesystems creds
fs: Update i_[ug]id_(read|write) to translate relative to s_user_ns
evm: Translate user/group ids relative to s_user_ns when computing HMAC
dquot: For now explicitly don't support filesystems outside of init_user_ns
quota: Handle quota data stored in s_user_ns in quota_setxquota
quota: Ensure qids map to the filesystem
vfs: Don't create inodes with a uid or gid unknown to the vfs
vfs: Don't modify inodes with a uid or gid unknown to the vfs
cred: Reject inodes with invalid ids in set_create_file_as()
fs: Check for invalid i_uid in may_follow_link()
vfs: Verify acls are valid within superblock's s_user_ns.
userns: Handle -1 in k[ug]id_has_mapping when !CONFIG_USER_NS
fs: Refuse uid/gid changes which don't map into s_user_ns
selinux: Add support for unprivileged mounts from user namespaces
Smack: Handle labels consistently in untrusted mounts
Smack: Add support for unprivileged mounts from user namespaces
fs: Treat foreign mounts as nosuid
fs: Limit file caps to the user namespace of the super block
userns: Remove the now unnecessary FS_USERNS_DEV_MOUNT flag
userns: Remove implicit MNT_NODEV fragility.
...
Fix the report:
net/sunrpc/clnt.c:2580:1: warning: ‘static’ is not at beginning of declaration [-Wold-style-declaration]
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The current min/max resvport settings are independently limited
by the entire range of allowed ports, so max_resvport can be
set to a port lower than min_resvport.
Prevent inversion of min/max values when set through sysfs and
module parameter by setting the limits dependent on each other.
Signed-off-by: Frank Sorenson <sorenson@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The current min/max resvport settings are independently limited
by the entire range of allowed ports, so max_resvport can be
set to a port lower than min_resvport.
Prevent inversion of min/max values when set through sysctl by
setting the limits dependent on each other.
Signed-off-by: Frank Sorenson <sorenson@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The range calculation for choosing the random reserved port will panic
with divide-by-zero when min_resvport == max_resvport, a range of one
port, not zero.
Fix the reserved port range calculation by adding one to the difference.
Signed-off-by: Frank Sorenson <sorenson@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Author: Frank Sorenson <sorenson@redhat.com>
Date: 2016-06-27 13:55:48 -0500
sunrpc: Fix bit count when setting hashtable size to power-of-two
The hashtable size is incorrectly calculated as the next higher
power-of-two when being set to a power-of-two. fls() returns the
bit number of the most significant set bit, with the least
significant bit being numbered '1'. For a power-of-two, fls()
will return a bit number which is one higher than the number of bits
required, leading to a hashtable which is twice the requested size.
In addition, the value of (1 << nbits) will always be at least num,
so the test will never be true.
Fix the hash table size calculation to correctly set hashtable
size, and eliminate the unnecessary check.
Signed-off-by: Frank Sorenson <sorenson@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
A generic_cred can be used to look up a unx_cred or a gss_cred, so it's
not really safe to use the the generic_cred->acred->ac_flags to store
the NO_CRKEY_TIMEOUT flag. A lookup for a unx_cred triggered while the
KEY_EXPIRE_SOON flag is already set will cause both NO_CRKEY_TIMEOUT and
KEY_EXPIRE_SOON to be set in the ac_flags, leaving the user associated
with the auth_cred to be in a state where they're perpetually doing 4K
NFS_FILE_SYNC writes.
This can be reproduced as follows:
1. Mount two NFS filesystems, one with sec=krb5 and one with sec=sys.
They do not need to be the same export, nor do they even need to be from
the same NFS server. Also, v3 is fine.
$ sudo mount -o v3,sec=krb5 server1:/export /mnt/krb5
$ sudo mount -o v3,sec=sys server2:/export /mnt/sys
2. As the normal user, before accessing the kerberized mount, kinit with
a short lifetime (but not so short that renewing the ticket would leave
you within the 4-minute window again by the time the original ticket
expires), e.g.
$ kinit -l 10m -r 60m
3. Do some I/O to the kerberized mount and verify that the writes are
wsize, UNSTABLE:
$ dd if=/dev/zero of=/mnt/krb5/file bs=1M count=1
4. Wait until you're within 4 minutes of key expiry, then do some more
I/O to the kerberized mount to ensure that RPC_CRED_KEY_EXPIRE_SOON gets
set. Verify that the writes are 4K, FILE_SYNC:
$ dd if=/dev/zero of=/mnt/krb5/file bs=1M count=1
5. Now do some I/O to the sec=sys mount. This will cause
RPC_CRED_NO_CRKEY_TIMEOUT to be set:
$ dd if=/dev/zero of=/mnt/sys/file bs=1M count=1
6. Writes for that user will now be permanently 4K, FILE_SYNC for that
user, regardless of which mount is being written to, until you reboot
the client. Renewing the kerberos ticket (assuming it hasn't already
expired) will have no effect. Grabbing a new kerberos ticket at this
point will have no effect either.
Move the flag to the auth->au_flags field (which is currently unused)
and rename it slightly to reflect that it's no longer associated with
the auth_cred->ac_flags. Add the rpc_auth to the arg list of
rpcauth_cred_key_to_expire and check the au_flags there too. Finally,
add the inode to the arg list of nfs_ctx_key_to_expire so we can
determine the rpc_auth to pass to rpcauth_cred_key_to_expire.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
If there were less than 2 entries in the multipath list, then
xprt_iter_next_entry_multiple() would never advance beyond the
first entry, which is correct for round robin behaviour, but not
for the list iteration.
The end result would be infinite looping in rpc_clnt_iterate_for_each_xprt()
as we would never see the xprt == NULL condition fulfilled.
Reported-by: Oleg Drokin <green@linuxhacker.ru>
Fixes: 80b14d5e61 ("SUNRPC: Add a structure to track multiple transports")
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The current server rpc tcp code attempts to predict how much writeable
socket space will be available to a given RPC call before accepting it
for processing. On a 40GigE network, we've found this throttles
individual clients long before the network or disk is saturated. The
server may handle more clients easily, but the bandwidth of individual
clients is still artificially limited.
Instead of trying (and failing) to predict how much writeable socket space
will be available to the RPC call, just fall back to the simple model of
deferring processing until the socket is uncongested.
This may increase the risk of fast clients starving slower clients; in
such cases, the previous patch allows setting a hard per-connection
limit.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Allow the user to limit the number of requests serviced through a single
connection, to help prevent faster clients from starving slower clients.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Don't call svc_xprt_enqueue() if the XPT_DATA flag is already set.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Rather than code up our own versions of the socket callbacks, just
call the defaults.
This also allows us to merge svc_udp_data_ready() and svc_tcp_data_ready().
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Prevent callbacks from triggering while we're detaching the socket.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Dropping and/or deferring requests has an impact on performance. Let's
make sure we can trace those events.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Add a tracepoint to track when the processing of incoming RPC data gets
deferred due to out-of-space issues on the outgoing transport.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
GSS-Proxy doesn't produce very much debug logging at all. Printing out
the gss minor status will aid in troubleshooting if the
GSS_Accept_sec_context upcall fails.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Before commit 778be232a2 ("NFS do not find client in NFSv4
pg_authenticate"), the Linux callback server replied with
RPC_AUTH_ERROR / RPC_AUTH_BADCRED, instead of dropping the CB
request. Let's restore that behavior so the server has a chance to
do something useful about it, and provide a warning that helps
admins correct the problem.
Fixes: 778be232a2 ("NFS do not find client in NFSv4 ...")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
If an RPC program does not set vs_dispatch and pc_func() returns
rpc_drop_reply, the server sends a reply anyway containing a single
word containing the value RPC_DROP_REPLY (in network byte-order, of
course). This is a nonsense RPC message.
Fixes: 9e701c6109 ("svcrpc: simpler request dropping")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Direct data placement is not allowed when using flavors that
guarantee integrity or privacy. When such security flavors are in
effect, don't allow the use of Read and Write chunks for moving
individual data items. All messages larger than the inline threshold
are sent via Long Call or Long Reply.
On my systems (CX-3 Pro on FDR), for small I/O operations, the use
of Long messages adds only around 5 usecs of latency in each
direction.
Note that when integrity or encryption is used, the host CPU touches
every byte in these messages. Even if it could be used, data
movement offload doesn't buy much in this case.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
fixup_copy_count should count only the number of bytes copied to the
page list. The head and tail are now always handled without a data
copy.
And the debugging at the end of rpcrdma_inline_fixup() is also no
longer necessary, since copy_len will be non-zero when there is reply
data in the tail (a normal and valid case).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Now that rpcrdma_inline_fixup() updates only two fields in
rq_rcv_buf, a full memcpy of that structure to rq_private_buf is
unwarranted. Updating rq_private_buf fields only where needed also
better documents what is going on.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
While trying NFSv4.0/RDMA with sec=krb5p, I noticed small NFS READ
operations failed. After the client unwrapped the NFS READ reply
message, the NFS READ XDR decoder was not able to decode the reply.
The message was "Server cheating in reply", with the reported
number of received payload bytes being zero. Applications reported
a read(2) that returned -1/EIO.
The problem is rpcrdma_inline_fixup() sets the tail.iov_len to zero
when the incoming reply fits entirely in the head iovec. The zero
tail.iov_len confused xdr_buf_trim(), which then mangled the actual
reply data instead of simply removing the trailing GSS checksum.
As near as I can tell, RPC transports are not supposed to update the
head.iov_len, page_len, or tail.iov_len fields in the receive XDR
buffer when handling an incoming RPC reply message. These fields
contain the length of each component of the XDR buffer, and hence
the maximum number of bytes of reply data that can be stored in each
XDR buffer component. I've concluded this because:
- This is how xdr_partial_copy_from_skb() appears to behave
- rpcrdma_inline_fixup() already does not alter page_len
- call_decode() compares rq_private_buf and rq_rcv_buf and WARNs
if they are not exactly the same
Unfortunately, as soon as I tried the simple fix to just remove the
line that sets tail.iov_len to zero, I saw that the logic that
appends the implicit Write chunk pad inline depends on inline_fixup
setting tail.iov_len to zero.
To address this, re-organize the tail iovec handling logic to use
the same approach as with the head iovec: simply point tail.iov_base
to the correct bytes in the receive buffer.
While I remember all this, write down the conclusion in documenting
comments.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
When the remaining length of an incoming reply is longer than the
XDR buf's page_len, switch over to the tail iovec instead of
copying more than page_len bytes into the page list.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Currently, all three chunk list encoders each use a portion of the
one rl_segments array in rpcrdma_req. This is because the MWs for
each chunk list were preserved in rl_segments so that ro_unmap could
find and invalidate them after the RPC was complete.
However, now that MWs are placed on a per-req linked list as they
are registered, there is no longer any information in rpcrdma_mr_seg
that is shared between ro_map and ro_unmap_{sync,safe}, and thus
nothing in rl_segments needs to be preserved after
rpcrdma_marshal_req is complete.
Thus the rl_segments array can be used now just for the needs of
each rpcrdma_convert_iovs call. Once each chunk list is encoded, the
next chunk list encoder is free to re-use all of rl_segments.
This means all three chunk lists in one RPC request can now each
encode a full size data payload with no increase in the size of
rl_segments.
This is a key requirement for Kerberos support, since both the Call
and Reply for a single RPC transaction are conveyed via Long
messages (RDMA Read/Write). Both can be large.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Instead of placing registered MWs sparsely into the rl_segments
array, place these MWs on a per-req list.
ro_unmap_{sync,safe} can then simply pull those MWs off the list
instead of walking through the array.
This change significantly reduces the size of struct rpcrdma_req
by removing nsegs and rl_mw from every array element.
As an additional clean-up, chunk co-ordinates are returned in the
"*mw" output argument so they are no longer needed in every
array element.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Instead of leaving orphaned MRs to be released when the transport
is destroyed, release them immediately. The MR free list can now be
replenished if it becomes exhausted.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Frequent MR list exhaustion can impact I/O throughput, so enough MRs
are always created during transport set-up to prevent running out.
This means more MRs are created than most workloads need.
Commit 94f58c58c0 ("xprtrdma: Allow Read list and Reply chunk
simultaneously") introduced support for sending two chunk lists per
RPC, which consumes more MRs per RPC.
Instead of trying to provision more MRs, introduce a mechanism for
allocating MRs on demand. A few MRs are allocated during transport
set-up to kick things off.
This significantly reduces the average number of MRs per transport
while allowing the MR count to grow for workloads or devices that
need more MRs.
FRWR with mlx4 allocated almost 400 MRs per transport before this
patch. Now it starts with 32.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up, based on code audit: Remove the possibility that the
chunk list XDR encoders can return zero, which would be interpreted
as a NULL.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Commit c93c62231c ("xprtrdma: Disconnect on registration failure")
added a disconnect for some RPC marshaling failures. This is needed
only in a handful of cases, but it was triggering for simple stuff
like temporary resource shortages. Try to straighten this out.
Fix up the lower layers so they don't return -ENOMEM or other error
codes that the RPC client's FSM doesn't explicitly recognize.
Also fix up the places in the send_request path that do want a
disconnect. For example, when ib_post_send or ib_post_recv fail,
this is a sign that there is a send or receive queue resource
miscalculation. That should be rare, and is a sign of a software
bug. But xprtrdma can recover: disconnect to reset the transport and
start over.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Not having an rpcrdma_rep at call_allocate time can be a problem.
It means that send_request can't post a receive buffer to catch
the RPC's reply. Possible consequences are RPC timeouts or even
transport deadlock.
Instead of allowing an RPC to proceed if an rpcrdma_rep is
not available, return NULL to force call_allocate to wait and
try again.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: ALLPHYSICAL is gone and FMR has been converted to use
scatterlists. There are no more users of these functions.
This patch shrinks the size of struct rpcrdma_req by about 3500
bytes on x86_64. There is one of these structs for each RPC credit
(128 credits per transport connection).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
No HCA or RNIC in the kernel tree requires the use of ALLPHYSICAL.
ALLPHYSICAL advertises in the clear on the network fabric an R_key
that is good for all of the client's memory. No known exploit
exists, but theoretically any user on the server can use that R_key
on the client's QP to read or update any part of the client's memory.
ALLPHYSICAL exposes the client to server bugs, including:
o base/bounds errors causing data outside the i/o buffer to be
accessed
o RDMA access after reply causing data corruption and/or integrity
fail
ALLPHYSICAL can't protect application memory regions from server
update after a local signal or soft timeout has terminated an RPC.
ALLPHYSICAL chunks are no larger than a page. Special cases to
handle small chunks and long chunk lists have been a source of
implementation complexity and bugs.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Based on code audit.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
I found that commit ead3f26e35 ("xprtrdma: Add ro_unmap_safe
memreg method"), which introduces ro_unmap_safe, never wired up the
FMR recovery worker.
The FMR and FRWR recovery work queues both do the same thing.
Instead of setting up separate individual work queues for this,
schedule a delayed worker to deal with them, since recovering MRs is
not performance-critical.
Fixes: ead3f26e35 ("xprtrdma: Add ro_unmap_safe memreg method")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The use of a scatterlist for handling DMA mapping and unmapping
was recently introduced in frwr_ops.c in commit 4143f34e01
("xprtrdma: Port to new memory registration API"). That commit did
not make a similar update to xprtrdma's FMR support because the
core ib_map_phys_fmr() and ib_unmap_fmr() APIs have not been changed
to take a scatterlist argument.
However, FMR still needs to do DMA mapping and unmapping. It appears
that RDS, for example, uses a scatterlist for this, then builds the
DMA addr array for the ib_map_phys_fmr call separately. I see that
SRP also utilizes a scatterlist for DMA mapping. xprtrdma can do
something similar.
This modernization is used immediately to properly defer DMA
unmapping during fmr_unmap_safe (a FIXME). It separates the DMA
unmapping coordinates from the rl_segments array. This array, being
part of an rpcrdma_req, is always re-used immediately when an RPC
exits. A scatterlist is allocated in memory independent of the
rl_segments array, so it can be preserved indefinitely (ie, until
the MR invalidation and DMA unmapping can actually be done by a
worker thread).
The FRWR and FMR DMA mapping code are slightly different from each
other now, and will diverge further when the "Check for holes" logic
can be removed from FRWR (support for SG_GAP MRs). So I chose not to
create helpers for the common-looking code.
Fixes: ead3f26e35 ("xprtrdma: Add ro_unmap_safe memreg method")
Suggested-by: Sagi Grimberg <sagi@lightbits.io>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Use the same naming convention used in other
RPC/RDMA-related data structures.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: Moving these helpers in a separate patch makes later
patches more readable.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: FMR is about to replace the rpcrdma_map_one code with
scatterlists. Move the scatterlist fields out of the FRWR-specific
union and into the generic part of rpcrdma_mw.
One minor change: -EIO is now returned if FRWR registration fails.
The RPC is terminated immediately, since the problem is likely due
to a software bug, thus retrying likely won't help.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
ib_unmap_fmr() takes a list of FMRs to unmap. However, it does not
remove the FMRs from this list as it processes them. Other
ib_unmap_fmr() call sites are careful to remove FMRs from the list
after ib_unmap_fmr() returns.
Since commit 7c7a5390dc ("xprtrdma: Add ro_unmap_sync method for FMR")
fmr_op_unmap_sync passes more than one FMR to ib_unmap_fmr(), but
it didn't bother to remove the FMRs from that list once the call was
complete.
I've noticed some instability that could be related to list
tangling by the new fmr_op_unmap_sync() logic. In an abundance
of caution, add some defensive logic to clean up properly after
ib_unmap_fmr().
Fixes: 7c7a5390dc ("xprtrdma: Add ro_unmap_sync method for FMR")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Today what is normally called data (the mount options) is not passed
to fill_super through mount_ns.
Pass the mount options and the namespace separately to mount_ns so
that filesystems such as proc that have mount options, can use
mount_ns.
Pass the user namespace to mount_ns so that the standard permission
check that verifies the mounter has permissions over the namespace can
be performed in mount_ns instead of in each filesystems .mount method.
Thus removing the duplication between mqueuefs and proc in terms of
permission checks. The extra permission check does not currently
affect the rpc_pipefs filesystem and the nfsd filesystem as those
filesystems do not currently allow unprivileged mounts. Without
unpvileged mounts it is guaranteed that the caller has already passed
capable(CAP_SYS_ADMIN) which guarantees extra permission check will
pass.
Update rpc_pipefs and the nfsd filesystem to ensure that the network
namespace reference is always taken in fill_super and always put in kill_sb
so that the logic is simpler and so that errors originating inside of
fill_super do not cause a network namespace leak.
Acked-by: Seth Forshee <seth.forshee@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Linus Torvalds