When a read to the preferred mirror returns an error, the flexfiles
driver records the error in the inode list and currently marks the
layout for return before failing over the attempted read to the next
mirror.
What we actually want to do is fire off a LAYOUTERROR to notify the
MDS that there is an issue with the preferred mirror, then we fail
over. Only once we've failed to read from all mirrors should we
return the layout.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If a layout segment gets invalidated while a pNFS I/O operation
is queued for transmission, then we ideally want to abort
immediately. This is particularly the case when there is a large
number of I/O related RPCs queued in the RPC layer, and the layout
segment gets invalidated due to an ENOSPC error, or an EACCES (because
the client was fenced). We may end up forced to spam the MDS with a
lot of otherwise unnecessary LAYOUTERRORs after that I/O fails.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix the memory barriers in nfs4_mark_deviceid_unavailable() and
nfs4_test_deviceid_unavailable().
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the attempt to instantiate the mirror's layout DS pointer failed,
then that pointer may hold a value of type ERR_PTR(), so we need
to check that before we dereference it.
Fixes: 65990d1afb ("pNFS/flexfiles: Fix a deadlock on LAYOUTGET")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
These really should have been there from the beginning, but we never
noticed because there was enough slack in the RPC request for the extra
bytes. Chuck's recent patch to use au_cslack and au_rslack to compute
buffer size shrunk the buffer enough that this was now a problem for
SEEK operations on my test client.
Fixes: f4ac1674f5 ("nfs: Add ALLOCATE support")
Fixes: 2e72448b07 ("NFS: Add COPY nfs operation")
Fixes: cb95deea0b ("NFS OFFLOAD_CANCEL xdr")
Fixes: 624bd5b7b6 ("nfs: Add DEALLOCATE support")
Fixes: 1c6dcbe5ce ("NFS: Implement SEEK")
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Ensure that if we call nfs41_sequence_process() a second time for the
same rpc_task, then we only process the results once.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If we have to retransmit a request, we should ensure that we reinitialise
the sequence results structure, since in the event of a signal
we need to treat the request as if it had not been sent.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: stable@vger.kernel.org
udp_poll() checks the struct file for the O_NONBLOCK flag, so we must not
call it with a NULL file pointer.
Fixes: 0ffe86f480 ("SUNRPC: Use poll() to fix up the socket requeue races")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If a bulk layout recall or a metadata server reboot coincides with a
umount, then holding a reference to an inode is unsafe unless we
also hold a reference to the super block.
Fixes: fd9a8d7160 ("NFSv4.1: Fix bulk recall and destroy of layouts")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix a soft lockup when NFS client delegation recovery is attempted
but the inode is in the process of being freed. When the
igrab(inode) call fails, and we have to restart the recovery process,
we need to ensure that we won't attempt to recover the same delegation
again.
Fixes: 45870d6909 ("NFSv4.1: Test delegation stateids when server...")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
A 'false retry' in NFSv4.1 occurs when the client attempts to transmit a
new RPC call using a slot+sequence number combination that references an
already cached one. Currently, the Linux NFS client will do this if a
user process interrupts an RPC call that is in progress.
The problem with doing so is that we defeat the main mechanism used by
the server to differentiate between a new call and a replayed one. Even
if the server is able to perfectly cache the arguments of the old call,
it cannot know if the client intended to replay or send a new call.
The obvious fix is to bump the sequence number pre-emptively if an
RPC call is interrupted, but in order to deal with the corner cases
where the interrupted call is not actually received and processed by
the server, we need to interpret the error NFS4ERR_SEQ_MISORDERED
as a sign that we need to either wait or locate a correct sequence
number that lies between the value we sent, and the last value that
was acked by a SEQUENCE call on that slot.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Tested-by: Jason Tibbitts <tibbs@math.uh.edu>
Now that we send the pages using a struct msghdr, instead of
using sendpage(), we no longer need to 'prime the socket' with
an address for unconnected UDP messages.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the client stream receive code receives an ESHUTDOWN error either
because the server closed the connection, or because it sent a
callback which cannot be processed, then we should shut down
the connection.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the message read completes, but the socket returned an error
condition, we should ensure to propagate that error.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
A zero length fragment is really a bug, but let's ensure we don't
go nuts when one turns up.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
To ensure that the receive worker has exclusive access to the stream record
info, we must not reset the contents other than when holding the
transport->recv_mutex.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
After setxattr, the nfsv3 cached the acl which set by user.
But at the backend, the shared file system (eg. ext4) will check
the acl, if it can merged with mode, it won't add acl to the file.
So, the nfsv3 cached acl is redundant.
Don't 'set_cached_acl' when setxattr.
Signed-off-by: ZhangXiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
As the block and SCSI layouts can only read/write fixed-length
blocks, we must perform read-modify-write when data to be written is
not aligned to a block boundary or smaller than the block size.
(612aa983a0 pnfs: add flag to force read-modify-write in ->write_begin)
The current code tries to see if we have to do read-modify-write
on block-oriented pNFS layouts by just checking !PageUptodate(page),
but the same condition also applies for overwriting of any uncached
potions of existing files, making such operations excessively slow
even it is block-aligned.
The change does not affect the optimization for modify-write-read
cases (38c73044f5 NFS: read-modify-write page updating),
because partial update of !PageUptodate() pages can only happen
in layouts that can do arbitrary length read/write and never
in block-based ones.
Testing results:
We ran fio on one of the pNFS clients running 4.20 kernel
(vanilla and patched) in this configuration to read/write/overwrite
files on the storage array, exported as pnfs share by the server.
pNFS clients ---1G Ethernet--- pNFS server
(HP DL360 G8) (HP DL360 G8)
| |
| |
+------8G Fiber Channel--------+
|
Storage Array
(HP P6350)
Throughput of overwrite (both buffered and O_SYNC) is noticeably
improved.
Ops. |block size| Throughput |
| (KiB) | (MiB/s) |
| | 4.20 | patched|
---------+----------+----------------+
buffered | 4| 21.3 | 232 |
overwrite| 32| 22.2 | 256 |
| 512| 22.4 | 260 |
---------+----------+----------------+
O_SYNC | 4| 3.84| 4.77|
overwrite| 32| 12.2 | 32.0 |
| 512| 18.5 | 152 |
---------+----------+----------------+
Read and write (buffered and O_SYNC) by the same client remain unchanged
by the patch either negatively or positively, as they should do.
Ops. |block size| Throughput |
| (KiB) | (MiB/s) |
| | 4.20 | patched|
---------+----------+----------------+
read | 4| 548 | 550 |
| 32| 547 | 551 |
| 512| 548 | 551 |
---------+----------+----------------+
buffered | 4| 237 | 244 |
write | 32| 261 | 268 |
| 512| 265 | 272 |
---------+----------+----------------+
O_SYNC | 4| 0.46| 0.46|
write | 32| 3.60| 3.57|
| 512| 105 | 106 |
---------+----------+----------------+
Signed-off-by: Kazuo Ito <ito_kazuo_g3@lab.ntt.co.jp>
Tested-by: Hiroyuki Watanabe <watanabe.hiroyuki@lab.ntt.co.jp>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
nfs_want_read_modify_write() didn't check for !PagePrivate when pNFS
block or SCSI layout was in use, therefore we could lose data forever
if the page being written was filled by a read before completion.
Signed-off-by: Kazuo Ito <ito_kazuo_g3@lab.ntt.co.jp>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
This fixes the typo in comments of nfs_readdir_alloc_pages().
Because nfs_readdir_large_page and nfs_readdir_free_pagearray had been
renamed.
Signed-off-by: Liguang Zhang <zhangliguang@linux.alibaba.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When listing very large directories via NFS, clients may take a long
time to complete. There are about three factors involved:
First of all, ls and practically every other method of listing a
directory including python os.listdir and find rely on libc readdir().
However readdir() only reads 32K of directory entries at a time, which
means that if you have a lot of files in the same directory, it is going
to take an insanely long time to read all the directory entries.
Secondly, libc readdir() reads 32K of directory entries at a time, in
kernel space 32K buffer split into 8 pages. One NFS readdirplus rpc will
be called for one page, which introduces many readdirplus rpc calls.
Lastly, one NFS readdirplus rpc asks for 32K data (filled by nfs_dentry)
to fill one page (filled by dentry), we found that nearly one third of
data was wasted.
To solve above problems, pagecache mechanism was introduced. One NFS
readdirplus rpc will ask for a large data (more than 32k), the data can
fill more than one page, the cached pages can be used for next readdir
call. This can reduce many readdirplus rpc calls and improve readdirplus
performance.
TESTING:
When listing very large directories(include 300 thousand files) via NFS
time ls -l /nfs_mount | wc -l
without the patch:
300001
real 1m53.524s
user 0m2.314s
sys 0m2.599s
with the patch:
300001
real 0m23.487s
user 0m2.305s
sys 0m2.558s
Improved performance: 79.6%
readdirplus rpc calls decrease: 85%
Signed-off-by: Liguang Zhang <zhangliguang@linux.alibaba.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
In the rare and unsupported case of a hostname list nfs_parse_devname
will modify dev_name. There is no need to modify dev_name as the all
that is being computed is the length of the hostname, so the computed
length can just be shorted.
Fixes: dc04589827 ("NFS: Use common device name parsing logic for NFSv4 and NFSv2/v3")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
As reported by Dan Carpenter, this test for acred->cred being set is
inconsistent with the dereference of the pointer a few lines earlier.
An 'auth_cred' *always* has ->cred set - every place that creates one
initializes this field, often as the first thing done.
So remove this test.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Drop LIST_HEAD where the variable it declares has never
been used.
The semantic patch that fixes this problem is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@@
identifier x;
@@
- LIST_HEAD(x);
... when != x
// </smpl>
Fixes: 0e20162ed1 ("NFSv4.1 Use MDS auth flavor for data server connection")
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Because we clear XPRT_SOCK_DATA_READY before reading, we can end up
with a situation where new data arrives, causing xs_data_ready() to
queue up a second receive worker job for the same socket, which then
immediately gets stuck waiting on the transport receive mutex.
The fix is to only clear XPRT_SOCK_DATA_READY once we're done reading,
and then to use poll() to check if we might need to queue up a new
job in order to deal with any new data.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Set memalloc_nofs_save() on all the rpciod/xprtiod jobs so that we
ensure memory allocations for asynchronous rpc calls don't ever end
up recursing back to the NFS layer for memory reclaim.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix up some compiler warnings about function parameters, etc not being
correctly described or formatted.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
All the allocations that we can hit in the NFS layer and sunrpc layers
themselves are already marked as GFP_NOFS, but we need to ensure that
any calls to generic kernel functionality do the right thing as well.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Allow the caller to pass error information when cleaning up a failed
I/O request so that we can conditionally take action to cancel the
request altogether if the error turned out to be fatal.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
In several places we're just moving the struct nfs_page from one list to
another by first removing from the existing list, then adding to the new
one.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the I/O completion failed with a fatal error, then we should just
exit nfs_pageio_complete_mirror() rather than try to recoalesce.
Fixes: a7d42ddb30 ("nfs: add mirroring support to pgio layer")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: stable@vger.kernel.org # v4.0+
Whether we need to exit early, or just reprocess the list, we
must not lost track of the request which failed to get recoalesced.
Fixes: 03d5eb65b5 ("NFS: Fix a memory leak in nfs_do_recoalesce")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: stable@vger.kernel.org # v4.0+
When we fail to add the request to the I/O queue, we currently leave it
to the caller to free the failed request. However since some of the
requests that fail are actually created by nfs_pageio_add_request()
itself, and are not passed back the caller, this leads to a leakage
issue, which can again cause page locks to leak.
This commit addresses the leakage by freeing the created requests on
error, using desc->pg_completion_ops->error_cleanup()
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fixes: a7d42ddb30 ("nfs: add mirroring support to pgio layer")
Cc: stable@vger.kernel.org # v4.0: c18b96a1b8: nfs: clean up rest of reqs
Cc: stable@vger.kernel.org # v4.0: d600ad1f2b: NFS41: pop some layoutget
Cc: stable@vger.kernel.org # v4.0+
Here are a few last-minute fixes for 5.0. The most significant one
is the OF-node refcount fix for ASoC simple-card, which could be
triggered on many boards. Another fix for ASoC core is for the
error handling in topology, while others are device-specific fixes
for Samsung and HD-audio.
-----BEGIN PGP SIGNATURE-----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=xyOI
-----END PGP SIGNATURE-----
Merge tag 'sound-5.0' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"Here are a few last-minute fixes for 5.0.
The most significant one is the OF-node refcount fix for ASoC
simple-card, which could be triggered on many boards. Another fix for
ASoC core is for the error handling in topology, while others are
device-specific fixes for Samsung and HD-audio"
* tag 'sound-5.0' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ASoC: simple-card: fixup refcount_t underflow
ASoC: topology: free created components in tplg load error
ALSA: hda/realtek: Disable PC beep in passthrough on alc285
ALSA: hda/realtek - Headset microphone and internal speaker support for System76 oryp5
ASoC: samsung: i2s: Fix prescaler setting for the secondary DAI
- Meson8B fixup for the sdc pins
- Fix SDC tile position for Qualcomm QCS404
-----BEGIN PGP SIGNATURE-----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=OC91
-----END PGP SIGNATURE-----
Merge tag 'pinctrl-v5.0-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
Pull pin control fixes from Linus Walleij:
"Some final pin control fixes (I hope) to round off the v5.0 pin
control development cycle.
Only driver fixes, one for stable:
- Meson8B fixup for the sdc pins
- Fix SDC tile position for Qualcomm QCS404"
* tag 'pinctrl-v5.0-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: meson: meson8b: fix the sdxc_a data 1..3 pins
pinctrl: qcom: qcs404: Correct SDC tile
- Per-instance irqchip on the MT7621
- Avoid direction setting using pin control on MMP2
-----BEGIN PGP SIGNATURE-----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=zceM
-----END PGP SIGNATURE-----
Merge tag 'gpio-v5.0-4' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
"Two GPIO fixes for the v5.0 series:
- Per-instance irqchip on the MT7621
- Avoid direction setting using pin control on MMP2"
* tag 'gpio-v5.0-4' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: pxa: avoid attempting to set pin direction via pinctrl on MMP2
gpio: MT7621: use a per instance irq_chip structure
Pull keys fixes from James Morris:
- Handle quotas better, allowing full quota to be reached.
- Fix the creation of shortcuts in the assoc_array internal
representation when the index key needs to be an exact multiple of
the machine word size.
- Fix a dependency loop between the request_key contruction record and
the request_key authentication key. The construction record isn't
really necessary and can be dispensed with.
- Set the timestamp on a new key rather than leaving it as 0. This
would ordinarily be fine - provided the system clock is never set to
a time before 1970
* 'fixes-v5.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
keys: Timestamp new keys
keys: Fix dependency loop between construction record and auth key
assoc_array: Fix shortcut creation
KEYS: allow reaching the keys quotas exactly
