Commit 9597c13b forbade opens with O_APPEND|O_DIRECT for NFSv4:
nfs: verify open flags before allowing an atomic open
Currently, you can open a NFSv4 file with O_APPEND|O_DIRECT, but cannot
fcntl(F_SETFL,...) with those flags. This flag combination is explicitly
forbidden on NFSv3 opens, and it seems like it should also be on NFSv4.
However, you can still open a file with O_DIRECT|O_APPEND if there exists a
cached dentry for the file because nfs4_file_open() is used instead of
nfs_atomic_open() and the check is bypassed. Add the check in
nfs4_file_open() as well.
Signed-off-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
A ds can be associated with more than one mirror, but we currently skip
setting a mirror's credentials if we find that it's already set up with
a connected client.
The upshot is that we can end up sending DS writes with MDS credentials
instead of properly setting them up. Fix nfs4_ff_layout_prepare_ds to
always verify that the mirror's credentials are set up, even when we
have a DS that's already connected.
Reported-by: Tom Haynes <thomas.haynes@primarydata.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Cc: stable@vger.kernel.org # 4.0+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
If we have two tasks racing to update a mirror's credentials, then they
can end up leaking one (or more) sets of credentials. The first task
will set mirror->cred and then the second task will just overwrite it.
Use a cmpxchg to ensure that the creds are only set once. If we get to
the point where we would set mirror->cred and find that they're already
set, then we just release the creds that were just found.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Cc: stable@vger.kernel.org # 4.0+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
* bugfixes:
NFS: Ensure we set NFS_CONTEXT_RESEND_WRITES when requeuing writes
pNFS: Fix a memory leak when attempted pnfs fails
NFS: Ensure that we update the sequence id under the slot table lock
nfs: Initialize cb_sequenceres information before validate_seqid()
nfs: Only update callback sequnce id when CB_SEQUENCE success
NFSv4: nfs4_handle_delegation_recall_error should ignore EAGAIN
If a write attempt fails, and the write is queued up for resending to
the server, as opposed to being dropped, then we need to set the
appropriate flag so that nfs_file_fsync() does the right thing.
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
pnfs_do_write() expects the call to pnfs_write_through_mds() to free the
pgio header and to release the layout segment before exiting. The problem
is that nfs_pgio_data_destroy() doesn't actually do this; it only frees
the memory allocated by nfs_generic_pgio().
Ditto for pnfs_do_read()...
Fix in both cases is to add a call to hdr->release(hdr).
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Client can receives stateid-type error (eg., BAD_STATEID) on SETATTR when
delegation stateid was used. When no open state exists, in case of application
calling truncate() on the file, client has no state to recover and fails with
EIO.
Instead, upon such error, return the bad delegation and then resend the
SETATTR with a zero stateid.
Signed-off: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
A truncated fsid showing from /proc/fs/nfsfs/volumes as,
NV SERVER PORT DEV FSID FSC
v4 c0a80881 801 0:43 34931f044c2a439b no
It should be as,
NV SERVER PORT DEV FSID FSC
v4 c0a80881 801 0:43 34931f044c2a439b:954c5d830fa4be8c no
The max buffer length for storing "%llx:%llx" format should be
16 + 1 + 16 + 1 = 34 (16 for %llx, 1 for ':', 1 for '\0').
Also, for storing "%u:%u" of MAJOR() and MINOR() should be
8 + 1 + 3 + 1 = 13 (8 for 2^24, 1 for ':', 3 for 2^8, 1 for '\0').
v2, add comments for dev/fsid buffer and use sizeof in snprintf.
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Change the uniform client string generator to dynamically allocate the
NFSv4 client name string buffer. With this patch, we can eliminate the
buffers that are embedded within the "args" structs and simply use the
name string that is hanging off the client.
This uniform string case is a little simpler than the nonuniform since
we don't need to deal with RCU, but we do have two different cases,
depending on whether there is a uniquifier or not.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The way the *_client_string functions work is a little goofy. They build
the string in an on-stack buffer and then use kstrdup to copy it. This
is not only stack-heavy but artificially limits the size of the client
name string. Change it so that we determine the length of the string,
allocate it and then scnprintf into it.
Since the contents of the nonuniform string depend on rcu-managed data
structures, it's possible that they'll change between when we allocate
the string and when we go to fill it. If that happens, free the string,
recalculate the length and try again. If it the mismatch isn't resolved
on the second try then just give up and return -EINVAL.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The spec allows for up to NFS4_OPAQUE_LIMIT (1k). While we'll almost
certainly never use that much, these ops are generally the only ones
in the compound so we might as well allow for them to be that large.
Also, the existing code didn't add in a word for the opaque length
field for either name string. Fix that while we're in there.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
...instead of buffers that are part of their arg structs. We already
hold a reference to the client, so we might as well use the allocated
buffer. In the event that we can't allocate the clp->cl_owner_id, then
just return -ENOMEM.
Note too that we switch from a GFP_KERNEL allocation here to GFP_NOFS.
It's possible we could end up trying to do a SETCLIENTID or EXCHANGE_ID
in order to reclaim some memory, and the GFP_KERNEL allocations in the
existing code could cause recursion back into NFS reclaim.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Use kernel.h macro definition.
Thanks to Julia Lawall for Coccinelle scripting support.
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Fix a callback slot table regression.
Fixes: e937ee714b ("nfs: Only update callback sequnce id when CB_SEQUENCE success")
Cc: Kinglong Mee <kinglongmee@gmail.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
For a cb_layoutrecall replay, nfsd got CB_SEQUENCE status of zero,
but all informations of cb_sequenceres are zero too !!!
validate_seqid() return NFS4ERR_RETRY_UNCACHED_REP for a replay,
and skip the initlize cb_sequenceres.
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
A drop should really only be done when the frame is malformed or we have
reason to think that there is some sort of DoS going on. When we get an
RPC with bad auth, we should send back an error instead.
Cc: Andy Adamson <William.Adamson@netapp.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
When testing pnfs layout, nfsd got error NFS4ERR_SEQ_MISORDERED.
It is caused by nfs return NFS4ERR_DELAY before validate_seqid(),
don't update the sequnce id, but nfsd updates the sequnce id !!!
According to RFC5661 20.9.3,
" If CB_SEQUENCE returns an error, then the state of the slot
(sequence ID, cached reply) MUST NOT change. "
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
In little endian cases, the macro htonl unfolds to __swab32 which
provides special case for constants. In big endian cases,
__constant_htonl and htonl expand directly to the same expression.
So, replace __constant_htonl with htonl with the goal of getting
rid of the definition of __constant_htonl completely.
The semantic patch that performs this transformation is as follows:
@@expression x;@@
- __constant_htonl(x)
+ htonl(x)
Signed-off-by: Vaishali Thakkar <vthakkar1994@gmail.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This was only ever set to nfs_writeback_release_common(), a function
which is completely empty. Let's just drop this function pointer and
simplify the code a bit.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
nfs4_proc_lookup_common is supposed to return a posix error, we have to
handle any error returned that isn't errno
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Frank S. Filz <ffilzlnx@mindspring.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Jerome reported seeing a warning pop when working with a swapfile on
NFS. The nfs_swap_activate can end up calling sk_set_memalloc while
holding the rcu_read_lock and that function can sleep.
To fix that, we need to take a reference to the xprt while holding the
rcu_read_lock, set the socket up for swapping and then drop that
reference. But, xprt_put is not exported and having NFS deal with the
underlying xprt is a bit of layering violation anyway.
Fix this by adding a set of activate/deactivate functions that take a
rpc_clnt pointer instead of an rpc_xprt, and have nfs_swap_activate and
nfs_swap_deactivate call those.
Also, add a per-rpc_clnt atomic counter to keep track of the number of
active swapfiles associated with it. When the counter does a 0->1
transition, we enable swapping on the xprt, when we do a 1->0 transition
we disable swapping on it.
This also allows us to be a bit more selective with the RPC_TASK_SWAPPER
flag. If non-swapper and swapper clnts are sharing a xprt, then we only
need to flag the tasks from the swapper clnt with that flag.
Acked-by: Mel Gorman <mgorman@suse.de>
Reported-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
EAGAIN is a valid return code from nfs4_open_recover(), and should
be handled by nfs4_handle_delegation_recall_error by simply passing
it through.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Delete jump to a label on the next line, when that label is not
used elsewhere.
A simplified version of the semantic patch that makes this change is as
follows: (http://coccinelle.lip6.fr/)
// <smpl>
@r@
identifier l;
@@
-if (...) goto l;
-l:
// </smpl>
Also drop the unnecessary ret variable.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
When encoding the NFSACL SETACL operation, reserve just the estimated
size of the ACL rather than a fixed maximum. This eliminates needless
zero padding on the wire that the server ignores.
Fixes: ee5dc7732b ('NFS: Fix "kernel BUG at fs/nfs/nfs3xdr.c:1338!"')
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
In glibc 2.21 (and several previous), a call to opendir() will
result in a 32K (BUFSIZ*4) buffer being allocated and passed to
getdents.
However a call to fdopendir() results in an 'fstat' request to
determine block size and a matching buffer allocated for subsequent
use with getdents. This will typically be 1M.
The first getdents call on an NFS directory will always use
READDIR_PLUS (or NFSv4 equivalent) if available. Subsequent getdents
calls only use this more expensive version if some 'stat' requests are
made between the getdents calls.
For this reason it is good to keep at least that first getdents call
relatively short. When fdopendir() and readdir() is used on a large
directory, it takes approximately 32 times as long to complete as
using "opendir". Current versions of 'find' use fdopendir() and
demonstrate this slowness.
'stat' on a directory currently returns the 'wsize'. This number has
no meaning on directories.
Actual READDIR requests are limited to ->dtsize, which itself is
capped at 4 pages, coincidently the same as BUFSIZ*4.
So this is a meaningful number to use as the blocksize on directories,
and has the effect of making 'find' on large directories go a lot
faster.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
While the NFSv4.1 code has always drained the slot tables in order to stop
non-recovery related RPC calls when doing lease recovery, the NFSv4 code
did not.
The reason for the difference in behaviour is that NFSv4 does not have
session state, and so RPC calls can in theory proceed while recovery is
happening. In practice, however, anything I/O or state related needs to
wait until recovery is over.
This patch changes the behaviour of NFSv4 to match that of NFSv4.1 so that
we can simplify the state recovery code by assuming that we do not have to
deal with races between recovery and ordinary I/O.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Problem: When an operation like WRITE receives a BAD_STATEID, even though
recovery code clears the RECLAIM_NOGRACE recovery flag before recovering
the open state, because of clearing delegation state for the associated
inode, nfs_inode_find_state_and_recover() gets called and it makes the
same state with RECLAIM_NOGRACE flag again. As a results, when we restart
looking over the open states, we end up in the infinite loop instead of
breaking out in the next test of state flags.
Solution: unset the RECLAIM_NOGRACE set because of
calling of nfs_inode_find_state_and_recover() after returning from calling
recover_open() function.
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
We had a report of a crash while stress testing the NFS client:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000150
IP: [<ffffffff8127b698>] locks_get_lock_context+0x8/0x90
PGD 0
Oops: 0000 [#1] SMP
Modules linked in: rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 xt_conntrack ebtable_nat ebtable_filter ebtable_broute bridge stp llc ebtables ip6table_security ip6table_mangle ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_raw ip6table_filter ip6_tables iptable_security iptable_mangle iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_raw coretemp crct10dif_pclmul ppdev crc32_pclmul crc32c_intel ghash_clmulni_intel vmw_balloon serio_raw vmw_vmci i2c_piix4 shpchp parport_pc acpi_cpufreq parport nfsd auth_rpcgss nfs_acl lockd grace sunrpc vmwgfx drm_kms_helper ttm drm mptspi scsi_transport_spi mptscsih mptbase e1000 ata_generic pata_acpi
CPU: 1 PID: 399 Comm: kworker/1:1H Not tainted 4.1.0-0.rc1.git0.1.fc23.x86_64 #1
Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 07/30/2013
Workqueue: rpciod rpc_async_schedule [sunrpc]
task: ffff880036aea7c0 ti: ffff8800791f4000 task.ti: ffff8800791f4000
RIP: 0010:[<ffffffff8127b698>] [<ffffffff8127b698>] locks_get_lock_context+0x8/0x90
RSP: 0018:ffff8800791f7c00 EFLAGS: 00010293
RAX: ffff8800791f7c40 RBX: ffff88001f2ad8c0 RCX: ffffe8ffffc80305
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffff8800791f7c88 R08: ffff88007fc971d8 R09: 279656d600000000
R10: 0000034a01000000 R11: 279656d600000000 R12: ffff88001f2ad918
R13: ffff88001f2ad8c0 R14: 0000000000000000 R15: 0000000100e73040
FS: 0000000000000000(0000) GS:ffff88007fc80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000150 CR3: 0000000001c0b000 CR4: 00000000000407e0
Stack:
ffffffff8127c5b0 ffff8800791f7c18 ffffffffa0171e29 ffff8800791f7c58
ffffffffa0171ef8 ffff8800791f7c78 0000000000000246 ffff88001ea0ba00
ffff8800791f7c40 ffff8800791f7c40 00000000ff5d86a3 ffff8800791f7ca8
Call Trace:
[<ffffffff8127c5b0>] ? __posix_lock_file+0x40/0x760
[<ffffffffa0171e29>] ? rpc_make_runnable+0x99/0xa0 [sunrpc]
[<ffffffffa0171ef8>] ? rpc_wake_up_task_queue_locked.part.35+0xc8/0x250 [sunrpc]
[<ffffffff8127cd3a>] posix_lock_file_wait+0x4a/0x120
[<ffffffffa03e4f12>] ? nfs41_wake_and_assign_slot+0x32/0x40 [nfsv4]
[<ffffffffa03bf108>] ? nfs41_sequence_done+0xd8/0x2d0 [nfsv4]
[<ffffffffa03c116d>] do_vfs_lock+0x2d/0x30 [nfsv4]
[<ffffffffa03c251d>] nfs4_lock_done+0x1ad/0x210 [nfsv4]
[<ffffffffa0171a30>] ? __rpc_sleep_on_priority+0x390/0x390 [sunrpc]
[<ffffffffa0171a30>] ? __rpc_sleep_on_priority+0x390/0x390 [sunrpc]
[<ffffffffa0171a5c>] rpc_exit_task+0x2c/0xa0 [sunrpc]
[<ffffffffa0167450>] ? call_refreshresult+0x150/0x150 [sunrpc]
[<ffffffffa0172640>] __rpc_execute+0x90/0x460 [sunrpc]
[<ffffffffa0172a25>] rpc_async_schedule+0x15/0x20 [sunrpc]
[<ffffffff810baa1b>] process_one_work+0x1bb/0x410
[<ffffffff810bacc3>] worker_thread+0x53/0x480
[<ffffffff810bac70>] ? process_one_work+0x410/0x410
[<ffffffff810bac70>] ? process_one_work+0x410/0x410
[<ffffffff810c0b38>] kthread+0xd8/0xf0
[<ffffffff810c0a60>] ? kthread_worker_fn+0x180/0x180
[<ffffffff817a1aa2>] ret_from_fork+0x42/0x70
[<ffffffff810c0a60>] ? kthread_worker_fn+0x180/0x180
Jean says:
"Running locktests with a large number of iterations resulted in a
client crash. The test run took a while and hasn't finished after close
to 2 hours. The crash happened right after I gave up and killed the test
(after 107m) with Ctrl+C."
The crash happened because a NULL inode pointer got passed into
locks_get_lock_context. The call chain indicates that file_inode(filp)
returned NULL, which means that f_inode was NULL. Since that's zeroed
out in __fput, that suggests that this filp pointer outlived the last
reference.
Looking at the code, that seems possible. We copy the struct file_lock
that's passed in, but if the task is signalled at an inopportune time we
can end up trying to use that file_lock in rpciod context after the process
that requested it has already returned (and possibly put its filp
reference).
Fix this by taking an extra reference to the filp when we allocate the
lock info, and put it in nfs4_lock_release.
Reported-by: Jean Spector <jean@primarydata.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
When running the Connectathon basic tests against a Solaris NFS
server over NFSv4.0, test5 reports that stat(2) returns a file size
of zero instead of 1MB.
On success, nfs_commit_inode() can return a positive result; see
other call sites such as nfs_file_fsync_commit() and
nfs_commit_unstable_pages().
The call site recently added in nfs_wb_all() does not prevent that
positive return value from leaking to its callers. If it leaks
through nfs_sync_inode() back to nfs_getattr(), that causes stat(2)
to return a positive return value to user space while also not
filling in the passed-in struct stat.
Additional clean up: the new logic in nfs_wb_all() is rewritten in
bfields-normal form.
Fixes: 5bb89b4702 ("NFSv4.1/pnfs: Separate out metadata . . .")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Highlights include:
Stable patches:
- Fix a regression in /proc/self/mountstats
- Fix the pNFS flexfiles O_DIRECT support
- Fix high load average due to callback thread sleeping
Bugfixes:
- Various patches to fix the pNFS layoutcommit support
- Do not cache pNFS deviceids unless server notifications are enabled
- Fix a SUNRPC transport reconnection regression
- make debugfs file creation failure non-fatal in SUNRPC
- Another fix for circular directory warnings on NFSv4 "junctioned" mountpoints
- Fix locking around NFSv4.2 fallocate() support
- Truncating NFSv4 file opens should also sync O_DIRECT writes
- Prevent infinite loop in rpcrdma_ep_create()
Features:
- Various improvements to the RDMA transport code's handling of memory
registration
- Various code cleanups
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=j5dP
-----END PGP SIGNATURE-----
Merge tag 'nfs-for-4.1-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client updates from Trond Myklebust:
"Another set of mainly bugfixes and a couple of cleanups. No new
functionality in this round.
Highlights include:
Stable patches:
- Fix a regression in /proc/self/mountstats
- Fix the pNFS flexfiles O_DIRECT support
- Fix high load average due to callback thread sleeping
Bugfixes:
- Various patches to fix the pNFS layoutcommit support
- Do not cache pNFS deviceids unless server notifications are enabled
- Fix a SUNRPC transport reconnection regression
- make debugfs file creation failure non-fatal in SUNRPC
- Another fix for circular directory warnings on NFSv4 "junctioned"
mountpoints
- Fix locking around NFSv4.2 fallocate() support
- Truncating NFSv4 file opens should also sync O_DIRECT writes
- Prevent infinite loop in rpcrdma_ep_create()
Features:
- Various improvements to the RDMA transport code's handling of
memory registration
- Various code cleanups"
* tag 'nfs-for-4.1-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (55 commits)
fs/nfs: fix new compiler warning about boolean in switch
nfs: Remove unneeded casts in nfs
NFS: Don't attempt to decode missing directory entries
Revert "nfs: replace nfs_add_stats with nfs_inc_stats when add one"
NFS: Rename idmap.c to nfs4idmap.c
NFS: Move nfs_idmap.h into fs/nfs/
NFS: Remove CONFIG_NFS_V4 checks from nfs_idmap.h
NFS: Add a stub for GETDEVICELIST
nfs: remove WARN_ON_ONCE from nfs_direct_good_bytes
nfs: fix DIO good bytes calculation
nfs: Fetch MOUNTED_ON_FILEID when updating an inode
sunrpc: make debugfs file creation failure non-fatal
nfs: fix high load average due to callback thread sleeping
NFS: Reduce time spent holding the i_mutex during fallocate()
NFS: Don't zap caches on fallocate()
xprtrdma: Make rpcrdma_{un}map_one() into inline functions
xprtrdma: Handle non-SEND completions via a callout
xprtrdma: Add "open" memreg op
xprtrdma: Add "destroy MRs" memreg op
xprtrdma: Add "reset MRs" memreg op
...
Pull fourth vfs update from Al Viro:
"d_inode() annotations from David Howells (sat in for-next since before
the beginning of merge window) + four assorted fixes"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
RCU pathwalk breakage when running into a symlink overmounting something
fix I_DIO_WAKEUP definition
direct-io: only inc/dec inode->i_dio_count for file systems
fs/9p: fix readdir()
VFS: assorted d_backing_inode() annotations
VFS: fs/inode.c helpers: d_inode() annotations
VFS: fs/cachefiles: d_backing_inode() annotations
VFS: fs library helpers: d_inode() annotations
VFS: assorted weird filesystems: d_inode() annotations
VFS: normal filesystems (and lustre): d_inode() annotations
VFS: security/: d_inode() annotations
VFS: security/: d_backing_inode() annotations
VFS: net/: d_inode() annotations
VFS: net/unix: d_backing_inode() annotations
VFS: kernel/: d_inode() annotations
VFS: audit: d_backing_inode() annotations
VFS: Fix up some ->d_inode accesses in the chelsio driver
VFS: Cachefiles should perform fs modifications on the top layer only
VFS: AF_UNIX sockets should call mknod on the top layer only
do_blockdev_direct_IO() increments and decrements the inode
->i_dio_count for each IO operation. It does this to protect against
truncate of a file. Block devices don't need this sort of protection.
For a capable multiqueue setup, this atomic int is the only shared
state between applications accessing the device for O_DIRECT, and it
presents a scaling wall for that. In my testing, as much as 30% of
system time is spent incrementing and decrementing this value. A mixed
read/write workload improved from ~2.5M IOPS to ~9.6M IOPS, with
better latencies too. Before:
clat percentiles (usec):
| 1.00th=[ 33], 5.00th=[ 34], 10.00th=[ 34], 20.00th=[ 34],
| 30.00th=[ 34], 40.00th=[ 34], 50.00th=[ 35], 60.00th=[ 35],
| 70.00th=[ 35], 80.00th=[ 35], 90.00th=[ 37], 95.00th=[ 80],
| 99.00th=[ 98], 99.50th=[ 151], 99.90th=[ 155], 99.95th=[ 155],
| 99.99th=[ 165]
After:
clat percentiles (usec):
| 1.00th=[ 95], 5.00th=[ 108], 10.00th=[ 129], 20.00th=[ 149],
| 30.00th=[ 155], 40.00th=[ 161], 50.00th=[ 167], 60.00th=[ 171],
| 70.00th=[ 177], 80.00th=[ 185], 90.00th=[ 201], 95.00th=[ 270],
| 99.00th=[ 390], 99.50th=[ 398], 99.90th=[ 418], 99.95th=[ 422],
| 99.99th=[ 438]
In other setups, Robert Elliott reported seeing good performance
improvements:
https://lkml.org/lkml/2015/4/3/557
The more applications accessing the device, the worse it gets.
Add a new direct-io flags, DIO_SKIP_DIO_COUNT, which tells
do_blockdev_direct_IO() that it need not worry about incrementing
or decrementing the inode i_dio_count for this caller.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Elliott, Robert (Server Storage) <elliott@hp.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* bugfixes:
NFSv4: Return delegations synchronously in evict_inode
SUNRPC: Fix a regression when reconnecting
NFS: remount with security change should return EINVAL
nfs: do not export discarded symbols
NFSv4.1: don't export static symbol
The brand new GCC 5.1.0 warns by default on using a boolean in the
switch condition. This results in the following warning:
fs/nfs/nfs4proc.c: In function 'nfs4_proc_get_rootfh':
fs/nfs/nfs4proc.c:3100:10: warning: switch condition has boolean value [-Wswitch-bool]
switch (auth_probe) {
^
This code was obviously using switch to make use of the fall-through
semantics (without the usual comment, though).
Rewrite that code using if statements to avoid the warning and make
the code a bit more readable on the way.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This reverts commit 5a254d08b0.
Since commit 5a254d08b0 ("nfs: replace nfs_add_stats with
nfs_inc_stats when add one"), nfs_readpage and nfs_do_writepage use
nfs_inc_stats to increment NFSIOS_READPAGES and NFSIOS_WRITEPAGES
instead of nfs_add_stats.
However nfs_inc_stats does not do the same thing as nfs_add_stats with
value 1 because these functions work on distinct stats:
nfs_inc_stats increments stats from "enum nfs_stat_eventcounters" (in
server->io_stats->events) and nfs_add_stats those from "enum
nfs_stat_bytecounters" (in server->io_stats->bytes).
Signed-off-by: Nicolas Iooss <nicolas.iooss_linux@m4x.org>
Fixes: 5a254d08b0 ("nfs: replace nfs_add_stats with nfs_inc_stats...")
Cc: stable@vger.kernel.org # 3.19+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
I added the nfs4 prefix to make it obvious that this file is built into
the NFS v4 module, and not the generic client.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This file is only used internally to the NFS v4 module, so it doesn't
need to be in the global include path. I also renamed it from
nfs_idmap.h to nfs4idmap.h to emphasize that it's an NFSv4-only include
file.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
The idmapper is completely internal to the NFS v4 module, so this macro
will always evaluate to true. This patch also removes unnecessary
includes of this file from the generic NFS client.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
d4b18c3e (pnfs: remove GETDEVICELIST implementation) removed the
GETDEVICELIST operation from the NFS client, but left a "hole" in the
nfs4_procedures array. This caused /proc/self/mountstats to report an
operation named "51" where GETDEVICELIST used to be. This patch adds a
stub to fix mountstats.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Fixes: d4b18c3e (pnfs: remove GETDEVICELIST implementation)
Cc: stable@vger.kernel.org # 3.17+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
For flexfiles driver, we might choose to read from mirror index other
than 0 while mirror_count is always 1 for read.
Reported-by: Jean Spector <jean@primarydata.com>
Cc: <stable@vger.kernel.org> # v3.19+
Cc: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
For direct read that has IO size larger than rsize, we'll split
it into several READ requests and nfs_direct_good_bytes() would
count completed bytes incorrectly by eating last zero count reply.
Fix it by handling mirror and non-mirror cases differently such that
we only count mirrored writes differently.
This fixes 5fadeb47("nfs: count DIO good bytes correctly with mirroring").
Reported-by: Jean Spector <jean@primarydata.com>
Cc: <stable@vger.kernel.org> # v3.19+
Signed-off-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2ef47eb1 (NFS: Fix use of nfs_attr_use_mounted_on_fileid()) was a good
start to fixing a circular directory structure warning for NFS v4
"junctioned" mountpoints. Unfortunately, further testing continued to
generate this error.
My server is configured like this:
anna@nfsd ~ % df
Filesystem Size Used Avail Use% Mounted on
/dev/vda1 9.1G 2.0G 6.5G 24% /
/dev/vdc1 1014M 33M 982M 4% /exports
/dev/vdc2 1014M 33M 982M 4% /exports/vol1
/dev/vdc3 1014M 33M 982M 4% /exports/vol1/vol2
anna@nfsd ~ % cat /etc/exports
/exports/ *(rw,async,no_subtree_check,no_root_squash)
/exports/vol1/ *(rw,async,no_subtree_check,no_root_squash)
/exports/vol1/vol2 *(rw,async,no_subtree_check,no_root_squash)
I've been running chown across the entire mountpoint twice in a row to
hit this problem. The first run succeeds, but the second one fails with
the circular directory warning along with:
anna@client ~ % dmesg
[Apr 3 14:28] NFS: server 192.168.100.204 error: fileid changed
fsid 0:39: expected fileid 0x100080, got 0x80
WHere 0x80 is the mountpoint's fileid and 0x100080 is the mounted-on
fileid.
This patch fixes the issue by requesting an updated mounted-on fileid
from the server during nfs_update_inode(), and then checking that the
fileid stored in the nfs_inode matches either the fileid or mounted-on
fileid returned by the server.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Chuck pointed out a problem that crept in with commit 6ffa30d3f7 (nfs:
don't call blocking operations while !TASK_RUNNING). Linux counts tasks
in uninterruptible sleep against the load average, so this caused the
system's load average to be pinned at at least 1 when there was a
NFSv4.1+ mount active.
Not a huge problem, but it's probably worth fixing before we get too
many complaints about it. This patch converts the code back to use
TASK_INTERRUPTIBLE sleep, simply has it flush any signals on each loop
iteration. In practice no one should really be signalling this thread at
all, so I think this is reasonably safe.
With this change, there's also no need to game the hung task watchdog so
we can also convert the schedule_timeout call back to a normal schedule.
Cc: <stable@vger.kernel.org>
Reported-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Tested-by: Chuck Lever <chuck.lever@oracle.com>
Fixes: commit 6ffa30d3f7 (“nfs: don't call blocking . . .”)
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
At the very least, we should not be taking the i_mutex until after
checking if the server even supports ALLOCATE or DEALLOCATE, allowing
v4.0 or v4.1 to exit without potentially waiting on a lock.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This patch adds a GETATTR to the end of ALLOCATE and DEALLOCATE
operations so we can set the updated inode size and change attribute
directly. DEALLOCATE will still need to release pagecache pages, so
nfs42_proc_deallocate() now calls truncate_pagecache_range() before
contacting the server.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Pull third hunk of vfs changes from Al Viro:
"This contains the ->direct_IO() changes from Omar + saner
generic_write_checks() + dealing with fcntl()/{read,write}() races
(mirroring O_APPEND/O_DIRECT into iocb->ki_flags and instead of
repeatedly looking at ->f_flags, which can be changed by fcntl(2),
check ->ki_flags - which cannot) + infrastructure bits for dhowells'
d_inode annotations + Christophs switch of /dev/loop to
vfs_iter_write()"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (30 commits)
block: loop: switch to VFS ITER_BVEC
configfs: Fix inconsistent use of file_inode() vs file->f_path.dentry->d_inode
VFS: Make pathwalk use d_is_reg() rather than S_ISREG()
VFS: Fix up debugfs to use d_is_dir() in place of S_ISDIR()
VFS: Combine inode checks with d_is_negative() and d_is_positive() in pathwalk
NFS: Don't use d_inode as a variable name
VFS: Impose ordering on accesses of d_inode and d_flags
VFS: Add owner-filesystem positive/negative dentry checks
nfs: generic_write_checks() shouldn't be done on swapout...
ocfs2: use __generic_file_write_iter()
mirror O_APPEND and O_DIRECT into iocb->ki_flags
switch generic_write_checks() to iocb and iter
ocfs2: move generic_write_checks() before the alignment checks
ocfs2_file_write_iter: stop messing with ppos
udf_file_write_iter: reorder and simplify
fuse: ->direct_IO() doesn't need generic_write_checks()
ext4_file_write_iter: move generic_write_checks() up
xfs_file_aio_write_checks: switch to iocb/iov_iter
generic_write_checks(): drop isblk argument
blkdev_write_iter: expand generic_file_checks() call in there
...
