It's not uncommon for some workloads to do a bunch of I/O to a file and
delete it just afterward. If knfsd has a cached open file however, then
the file may still be open when the dentry is unlinked. If the
underlying filesystem is nfs, then that could trigger it to do a
sillyrename.
On a REMOVE or RENAME scan the nfsd_file cache for open files that
correspond to the inode, and proactively unhash and put their
references. This should prevent any delete-on-last-close activity from
occurring, solely due to knfsd's open file cache.
This must be done synchronously though so we use the variants that call
flush_delayed_fput. There are deadlock possibilities if you call
flush_delayed_fput while holding locks, however. In the case of
nfsd_rename, we don't even do the lookups of the dentries to be renamed
until we've locked for rename.
Once we've figured out what the target dentry is for a rename, check to
see whether there are cached open files associated with it. If there
are, then unwind all of the locking, close them all, and then reattempt
the rename.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The raparms cache was set up in order to ensure that we carry readahead
information forward from one RPC call to the next. In other words, it
was set up because each RPC call was forced to open a struct file, then
close it, causing the loss of readahead information that is normally
cached in that struct file, and used to keep the page cache filled when
a user calls read() multiple times on the same file descriptor.
Now that we cache the struct file, and reuse it for all the I/O calls
to a given file by a given user, we no longer have to keep a separate
readahead cache.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp.
Since we now presume that the struct file will be persistent in most
cases, we can stop fiddling with the raparms in the read code. This
also means that we don't really care about the rd_tmp_file field
anymore.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Have them keep an nfsd_file reference instead of a struct file.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Use cached filps if possible instead of opening a new one every time.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Currently, NFSv2/3 reads and writes have to open a file, do the read or
write and then close it again for each RPC. This is highly inefficient,
especially when the underlying filesystem has a relatively slow open
routine.
This patch adds a new open file cache to knfsd. Rather than doing an
open for each RPC, the read/write handlers can call into this cache to
see if there is one already there for the correct filehandle and
NFS_MAY_READ/WRITE flags.
If there isn't an entry, then we create a new one and attempt to
perform the open. If there is, then we wait until the entry is fully
instantiated and return it if it is at the end of the wait. If it's
not, then we attempt to take over construction.
Since the main goal is to speed up NFSv2/3 I/O, we don't want to
close these files on last put of these objects. We need to keep them
around for a little while since we never know when the next READ/WRITE
will come in.
Cache entries have a hardcoded 1s timeout, and we have a recurring
workqueue job that walks the cache and purges any entries that have
expired.
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Richard Sharpe <richard.sharpe@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
"cb" is never actually NULL in these functions.
On a quick skim of the history, they seem to have been there from the
beginning. I'm not sure if they originally served a purpose.
Reported-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
A process could race in an open and attempt to read one of these files
before i_private is initialized, and get a spurious error.
Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
synchronize_rcu() gets called multiple times each time a client is
destroyed. If the laundromat thread has a lot of clients to destroy,
the delay can be noticeable. This was causing pynfs test RENEW3 to
fail.
We could embed an rcu_head in each inode and do the kref_put in an rcu
callback. But simplest is just to take a lock here.
(I also wonder if the laundromat thread would be better replaced by a
bunch of scheduled work or timers or something.)
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
syzbot is reporting that nfsd_mkdir() forgot to remove dentry created by
d_alloc_name() when __nfsd_mkdir() failed (due to memory allocation fault
injection) [1].
[1] https://syzkaller.appspot.com/bug?id=ce41a1f769ea4637ebffedf004a803e8405b4674
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reported-by: syzbot <syzbot+2c95195d5d433f6ed6cb@syzkaller.appspotmail.com>
Fixes: e8a79fb14f ("nfsd: add nfsd/clients directory")
[bfields: clean up in nfsd_mkdir instead of __nfsd_mkdir]
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Pull vfs mount updates from Al Viro:
"The first part of mount updates.
Convert filesystems to use the new mount API"
* 'work.mount0' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
mnt_init(): call shmem_init() unconditionally
constify ksys_mount() string arguments
don't bother with registering rootfs
init_rootfs(): don't bother with init_ramfs_fs()
vfs: Convert smackfs to use the new mount API
vfs: Convert selinuxfs to use the new mount API
vfs: Convert securityfs to use the new mount API
vfs: Convert apparmorfs to use the new mount API
vfs: Convert openpromfs to use the new mount API
vfs: Convert xenfs to use the new mount API
vfs: Convert gadgetfs to use the new mount API
vfs: Convert oprofilefs to use the new mount API
vfs: Convert ibmasmfs to use the new mount API
vfs: Convert qib_fs/ipathfs to use the new mount API
vfs: Convert efivarfs to use the new mount API
vfs: Convert configfs to use the new mount API
vfs: Convert binfmt_misc to use the new mount API
convenience helper: get_tree_single()
convenience helper get_tree_nodev()
vfs: Kill sget_userns()
...
Here is the "big" driver core and debugfs changes for 5.3-rc1
It's a lot of different patches, all across the tree due to some api
changes and lots of debugfs cleanups. Because of this, there is going
to be some merge issues with your tree at the moment, I'll follow up
with the expected resolutions to make it easier for you.
Other than the debugfs cleanups, in this set of changes we have:
- bus iteration function cleanups (will cause build warnings
with s390 and coresight drivers in your tree)
- scripts/get_abi.pl tool to display and parse Documentation/ABI
entries in a simple way
- cleanups to Documenatation/ABI/ entries to make them parse
easier due to typos and other minor things
- default_attrs use for some ktype users
- driver model documentation file conversions to .rst
- compressed firmware file loading
- deferred probe fixes
All of these have been in linux-next for a while, with a bunch of merge
issues that Stephen has been patient with me for. Other than the merge
issues, functionality is working properly in linux-next :)
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
iG0EABECAC0WIQT0tgzFv3jCIUoxPcsxR9QN2y37KQUCXSgpnQ8cZ3JlZ0Brcm9h
aC5jb20ACgkQMUfUDdst+ykcwgCfS30OR4JmwZydWGJ7zK/cHqk+KjsAnjOxjC1K
LpRyb3zX29oChFaZkc5a
=XrEZ
-----END PGP SIGNATURE-----
Merge tag 'driver-core-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core and debugfs updates from Greg KH:
"Here is the "big" driver core and debugfs changes for 5.3-rc1
It's a lot of different patches, all across the tree due to some api
changes and lots of debugfs cleanups.
Other than the debugfs cleanups, in this set of changes we have:
- bus iteration function cleanups
- scripts/get_abi.pl tool to display and parse Documentation/ABI
entries in a simple way
- cleanups to Documenatation/ABI/ entries to make them parse easier
due to typos and other minor things
- default_attrs use for some ktype users
- driver model documentation file conversions to .rst
- compressed firmware file loading
- deferred probe fixes
All of these have been in linux-next for a while, with a bunch of
merge issues that Stephen has been patient with me for"
* tag 'driver-core-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (102 commits)
debugfs: make error message a bit more verbose
orangefs: fix build warning from debugfs cleanup patch
ubifs: fix build warning after debugfs cleanup patch
driver: core: Allow subsystems to continue deferring probe
drivers: base: cacheinfo: Ensure cpu hotplug work is done before Intel RDT
arch_topology: Remove error messages on out-of-memory conditions
lib: notifier-error-inject: no need to check return value of debugfs_create functions
swiotlb: no need to check return value of debugfs_create functions
ceph: no need to check return value of debugfs_create functions
sunrpc: no need to check return value of debugfs_create functions
ubifs: no need to check return value of debugfs_create functions
orangefs: no need to check return value of debugfs_create functions
nfsd: no need to check return value of debugfs_create functions
lib: 842: no need to check return value of debugfs_create functions
debugfs: provide pr_fmt() macro
debugfs: log errors when something goes wrong
drivers: s390/cio: Fix compilation warning about const qualifiers
drivers: Add generic helper to match by of_node
driver_find_device: Unify the match function with class_find_device()
bus_find_device: Unify the match callback with class_find_device
...
Fix sparse warning:
fs/nfsd/nfsctl.c:1221:22: warning:
symbol '__get_nfsdfs_client' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Fix sparse warnings:
fs/nfsd/nfs4state.c:1908:6: warning: symbol 'drop_client' was not declared. Should it be static?
fs/nfsd/nfs4state.c:2518:6: warning: symbol 'force_expire_client' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Probable cut&paste typo - use the correct field size.
(Not currently a practical problem since these two fields have the same
size, but we should fix it anyway.)
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Decode the implementation ID and display in nfsd/clients/#/info. It may
be help identify the client. It won't be used otherwise.
(When this went into the protocol, I thought the implementation ID would
be a slippery slope towards implementation-specific workarounds as with
the http user-agent. But I guess I was wrong, the risk seems pretty low
now.)
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
NFSv4 clients are automatically expired and all their locks removed if
they don't contact the server for a certain amount of time (the lease
period, 90 seconds by default).
There can still be situations where that's not enough, so allow
userspace to force expiry by writing "expire\n" to the new
nfsd/client/#/ctl file.
(The generic "ctl" name is because I expect we may want to allow other
operations on clients in the future.)
The write will not return until the client is expired and all of its
locks and other state removed.
The fault injection code also provides a way of expiring clients, but it
fails if there are any in-progress RPC's referencing the client. Also,
its method of selecting a client to expire is a little more
primitive--it uses an IP address, which can't always uniquely specify an
NFSv4 client.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Add a nfsd/clients/#/opens file to list some information about all the
opens held by the given client, including open modes, device numbers,
inode numbers, and open owners.
Open owners are totally opaque but seem to sometimes have some useful
ascii strings included, so passing through printable ascii characters
and escaping the rest seems useful while still being machine-readable.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Add ip address, full client-provided identifier, and minor version.
There's much more that could possibly be useful but this is a start.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
rpc_copy_addr() copies only the IP address and misses any port numbers.
It seems potentially useful to keep the port number around too.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
We want clientid's on the wire to be randomized for reasons explained in
ebd7c72c63 "nfsd: randomize SETCLIENTID reply to help distinguish
servers". But I'd rather have mostly small integers for the clients/
directory.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Keep a second reference count which is what is really used to decide
when to free the client's memory.
Next I'm going to add an nfsd/clients/ directory with a subdirectory for
each NFSv4 client. File objects under nfsd/clients/ will hold these
references.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Rename this to a more descriptive name: it counts the number of
in-progress rpc's referencing this client.
Next I'm going to add a second refcount with a slightly different use.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Keep around one internal mount of the nfsd filesystem so that we can add
stuff to it when clients come and go, regardless of whether anyone has
it mounted.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The failure to unregister the shrinker results will result in corruption
when the nfsd_net is freed.
Also clean up the drc_slab while we're here.
Reported-by: syzbot+83a43746cebef3508b49@syzkaller.appspotmail.com
Fixes: db17b61765c2 ("nfsd4: drc containerization")
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Commit bf8d909705 "nfsd: Decode and send 64bit time values" fixed the
code without updating the comment.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
After commit 95582b0083 "vfs: change inode times to use struct
timespec64" there are spots in the NFSv4 decoding where we decode the
protocol into a struct timeval and then convert that into a timeval64.
That's unnecesary in the NFSv4 case since the on-the-wire protocol also
uses 64-bit values. So just fix up our code to use timeval64 everywhere.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
After 89a26b3d29 "nfsd: split DRC global spinlock into per-bucket
locks", there is no longer a single global spinlock to protect these
stats.
So, really we need to fix that. For now, at least fix the comment.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The nfsd duplicate reply cache should not be shared between network
namespaces.
The most straightforward way to fix this is just to move every global in
the code to per-net-namespace memory, so that's what we do.
Still todo: sort out which members of nfsd_stats should be global and
which per-net-namespace.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Since commit 10a68cdf10 (nfsd: fix performance-limiting session
calculation) (Linux 5.1-rc1 and 4.19.31), shares from NFS servers with
1 TB of memory cannot be mounted anymore. The mount just hangs on the
client.
The gist of commit 10a68cdf10 is the change below.
-avail = clamp_t(int, avail, slotsize, avail/3);
+avail = clamp_t(int, avail, slotsize, total_avail/3);
Here are the macros.
#define min_t(type, x, y) __careful_cmp((type)(x), (type)(y), <)
#define clamp_t(type, val, lo, hi) min_t(type, max_t(type, val, lo), hi)
`total_avail` is 8,434,659,328 on the 1 TB machine. `clamp_t()` casts
the values to `int`, which for 32-bit integers can only hold values
−2,147,483,648 (−2^31) through 2,147,483,647 (2^31 − 1).
`avail` (in the function signature) is just 65536, so that no overflow
was happening. Before the commit the assignment would result in 21845,
and `num = 4`.
When using `total_avail`, it is causing the assignment to be
18446744072226137429 (printed as %lu), and `num` is then 4164608182.
My next guess is, that `nfsd_drc_mem_used` is then exceeded, and the
server thinks there is no memory available any more for this client.
Updating the arguments of `clamp_t()` and `min_t()` to `unsigned long`
fixes the issue.
Now, `avail = 65536` (before commit 10a68cdf10 `avail = 21845`), but
`num = 4` remains the same.
Fixes: c54f24e338 (nfsd: fix performance-limiting session calculation)
Cc: stable@vger.kernel.org
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
When calling debugfs functions, there is no need to ever check the
return value. The function can work or not, but the code logic should
never do something different based on this.
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Jeff Layton <jlayton@kernel.org>
Cc: linux-nfs@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20190612152603.GB18440@kroah.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert the nfsctl filesystem to the new internal mount API as the old
one will be obsoleted and removed. This allows greater flexibility in
communication of mount parameters between userspace, the VFS and the
filesystem.
See Documentation/filesystems/mount_api.txt for more information.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: "J. Bruce Fields" <bfields@fieldses.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: linux-nfs@vger.kernel.org
Based on 2 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license as published by
the free software foundation either version 2 of the license or at
your option any later version this program is distributed in the
hope that it will be useful but without any warranty without even
the implied warranty of merchantability or fitness for a particular
purpose see the gnu general public license for more details you
should have received a copy of the gnu general public license along
with this program if not write to the free software foundation inc
51 franklin street fifth floor boston ma 02110 1301 usa
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license as published by
the free software foundation either version 2 of the license or at
your option [no]_[pad]_[ctrl] any later version this program is
distributed in the hope that it will be useful but without any
warranty without even the implied warranty of merchantability or
fitness for a particular purpose see the gnu general public license
for more details you should have received a copy of the gnu general
public license along with this program if not write to the free
software foundation inc 51 franklin street fifth floor boston ma
02110 1301 usa
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-or-later
has been chosen to replace the boilerplate/reference in 176 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154040.652910950@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add SPDX license identifiers to all Make/Kconfig files which:
- Have no license information of any form
These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:
GPL-2.0-only
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add SPDX license identifiers to all files which:
- Have no license information of any form
- Have MODULE_LICENCE("GPL*") inside which was used in the initial
scan/conversion to ignore the file
These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:
GPL-2.0-only
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
