overlay needs underlying fs to support d_type. Recently I put in a
patch in to detect this condition and started failing mount if
underlying fs did not support d_type.
But this breaks existing configurations over kernel upgrade. Those who
are running docker (partially broken configuration) with xfs not
supporting d_type, are surprised that after kernel upgrade docker does
not run anymore.
https://github.com/docker/docker/issues/22937#issuecomment-229881315
So instead of erroring out, detect broken configuration and warn
about it. This should allow existing docker setups to continue
working after kernel upgrade.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 45aebeaf4f ("ovl: Ensure upper filesystem supports d_type")
Cc: <stable@vger.kernel.org> 4.6
Pull vfs fixes from Al Viro:
"Tmpfs readdir throughput regression fix (this cycle) + some -stable
fodder all over the place.
One missing bit is Miklos' tonight locks.c fix - NFS folks had already
grabbed that one by the time I woke up ;-)"
[ The locks.c fix came through the nfsd tree just moments ago ]
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
namespace: update event counter when umounting a deleted dentry
9p: use file_dentry()
ceph: fix d_obtain_alias() misuses
lockless next_positive()
libfs.c: new helper - next_positive()
dcache_{readdir,dir_lseek}(): don't bother with nested ->d_lock
Pull lockd/locks fixes from Bruce Fields:
"One fix for lockd soft lookups in an error path, and one fix for file
leases on overlayfs"
* tag 'nfsd-4.7-3' of git://linux-nfs.org/~bfields/linux:
locks: use file_inode()
lockd: unregister notifier blocks if the service fails to come up completely
Pull libnvdimm fixes from Dan Williams:
"1/ Two regression fixes since v4.6: one for the byte order of a sysfs
attribute (bz121161) and another for QEMU 2.6's NVDIMM _DSM (ACPI
Device Specific Method) implementation that gets tripped up by new
auto-probing behavior in the NFIT driver.
2/ A fix tagged for -stable that stops the kernel from
clobbering/ignoring changes to the configuration of a 'pfn'
instance ("struct page" driver). For example changing the
alignment from 2M to 1G may silently revert to 2M if that value is
currently stored on media.
3/ A fix from Eric for an xfstests failure in dax. It is not
currently tagged for -stable since it requires an 8-exabyte file
system to trigger, and there appear to be no user visible side
effects"
* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
nfit: fix format interface code byte order
dax: fix offset overflow in dax_io
acpi, nfit: fix acpi_check_dsm() vs zero functions implemented
libnvdimm, pfn, dax: fix initialization vs autodetect for mode + alignment
(Another one for the f_path debacle.)
ltp fcntl33 testcase caused an Oops in selinux_file_send_sigiotask.
The reason is that generic_add_lease() used filp->f_path.dentry->inode
while all the others use file_inode(). This makes a difference for files
opened on overlayfs since the former will point to the overlay inode the
latter to the underlying inode.
So generic_add_lease() added the lease to the overlay inode and
generic_delete_lease() removed it from the underlying inode. When the file
was released the lease remained on the overlay inode's lock list, resulting
in use after free.
Reported-by: Eryu Guan <eguan@redhat.com>
Fixes: 4bacc9c923 ("overlayfs: Make f_path always point to the overlay and f_inode to the underlay")
Cc: <stable@vger.kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
lockless_dereference() was added which can be used in place of
hard-coding smp_read_barrier_depends().
Signed-off-by: He Kuang <hekuang@huawei.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
- m_start() in fs/namespace.c expects that ns->event is incremented each
time a mount added or removed from ns->list.
- umount_tree() removes items from the list but does not increment event
counter, expecting that it's done before the function is called.
- There are some codepaths that call umount_tree() without updating
"event" counter. e.g. from __detach_mounts().
- When this happens m_start may reuse a cached mount structure that no
longer belongs to ns->list (i.e. use after free which usually leads
to infinite loop).
This change fixes the above problem by incrementing global event counter
before invoking umount_tree().
Change-Id: I622c8e84dcb9fb63542372c5dbf0178ee86bb589
Cc: stable@vger.kernel.org
Signed-off-by: Andrey Ulanov <andreyu@google.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
v9fs may be used as lower layer of overlayfs and accessing f_path.dentry
can lead to a crash. In this case it's a NULL pointer dereference in
p9_fid_create().
Fix by replacing direct access of file->f_path.dentry with the
file_dentry() accessor, which will always return a native object.
Reported-by: Alessio Igor Bogani <alessioigorbogani@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Tested-by: Alessio Igor Bogani <alessioigorbogani@gmail.com>
Fixes: 4bacc9c923 ("overlayfs: Make f_path always point to the overlay and f_inode to the underlay")
Cc: <stable@vger.kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Filesystem uids which don't map into a user namespace may result
in inode->i_uid being INVALID_UID. A symlink and its parent
could have different owners in the filesystem can both get
mapped to INVALID_UID, which may result in following a symlink
when this would not have otherwise been permitted when protected
symlinks are enabled.
Signed-off-by: Seth Forshee <seth.forshee@canonical.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Update posix_acl_valid to verify that an acl is within a user namespace.
Update the callers of posix_acl_valid to pass in an appropriate
user namespace. For posix_acl_xattr_set and v9fs_xattr_set_acl pass in
inode->i_sb->s_user_ns to posix_acl_valid. For md_unpack_acl pass in
&init_user_ns as no inode or superblock is in sight.
Acked-by: Seth Forshee <seth.forshee@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
If the lockd service fails to start up then we need to be sure that the
notifier blocks are not registered, otherwise a subsequent start of the
service could cause the same notifier to be registered twice, leading to
soft lockups.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Cc: stable@vger.kernel.org
Fixes: 0751ddf77b "lockd: Register callbacks on the inetaddr_chain..."
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Fix up nfs4_do_handle_exception() so that it can check if the operation
that received the NFS4ERR_BAD_STATEID was using a defunct delegation.
Apply that to the case of SETATTR, which will currently return EIO
in some cases where this happens.
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
If a pNFS client sets hdr->pgio_done_cb, then we should not overwrite that
in nfs4_proc_read_setup()
Fixes: 75bf47ebf6 ("pNFS/flexfile: Fix erroneous fall back to...")
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
An extent with lblock = 4294967295 and len = 1 will pass the
ext4_valid_extent() test:
ext4_lblk_t last = lblock + len - 1;
if (len == 0 || lblock > last)
return 0;
since last = 4294967295 + 1 - 1 = 4294967295. This would later trigger
the BUG_ON(es->es_lblk + es->es_len < es->es_lblk) in ext4_es_end().
We can simplify it by removing the - 1 altogether and changing the test
to use lblock + len <= lblock, since now if len = 0, then lblock + 0 ==
lblock and it fails, and if len > 0 then lblock + len > lblock in order
to pass (i.e. it doesn't overflow).
Fixes: 5946d0893 ("ext4: check for overlapping extents in ext4_valid_extent_entries()")
Fixes: 2f974865f ("ext4: check for zero length extent explicitly")
Cc: Eryu Guan <guaneryu@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Phil Turnbull <phil.turnbull@oracle.com>
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The jbd2 journal stores the commit time in 64-bit seconds and 32-bit
nanoseconds, which avoids an overflow in 2038, but it gets the numbers
from current_kernel_time(), which uses 'long' seconds on 32-bit
architectures.
This simply changes the code to call current_kernel_time64() so
we use 64-bit seconds consistently.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org
So far we were tracking only dependency on transaction commit due to
starting a new handle (which may require commit to start a new
transaction). Now add tracking also for other cases where we wait for
transaction commit. This way lockdep can catch deadlocks e. g. because we
call jbd2_journal_stop() for a synchronous handle with some locks held
which rank below transaction start.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Currently lockdep map is tracked in each journal handle. To be able to
expand lockdep support to cover also other cases where we depend on
transaction commit and where handle is not available, move lockdep map
into struct journal_s. Since this makes the lockdep map shared for all
handles, we have to use rwsem_acquire_read() for acquisitions now.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The transaction the handle references is free to commit once we've
decremented t_updates counter. Move the lockdep instrumentation to that
place. Currently it was a bit later which did not really matter but
subsequent improvements to lockdep instrumentation would cause false
positives with it.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
While sending the blocking directIO in fuse, the write request is broken
into sub-requests, each of default size 128k and all the requests are sent
in non-blocking background mode if async_dio mode is supported by libfuse.
The process which issue the write wait for the completion of all the
sub-requests. Sending multiple requests parallely gives a chance to perform
parallel writes in the user space fuse implementation if it is
multi-threaded and hence improves the performance.
When there is a size extending aio dio write, we switch to blocking mode so
that we can properly update the size of the file after completion of the
writes. However, in this situation all the sub-requests are sent in
serialized manner where the next request is sent only after receiving the
reply of the current request. Hence the multi-threaded user space
implementation is not utilized properly.
This patch changes the size extending aio dio behavior to exactly follow
blocking dio. For multi threaded fuse implementation having 10 threads and
using buffer size of 64MB to perform async directIO, we are getting double
the speed.
Signed-off-by: Ashish Sangwan <ashishsangwan2@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Negotiate with userspace filesystems whether they support parallel readdir
and lookup. Disable parallelism by default for fear of breaking fuse
filesystems.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 9902af79c0 ("parallel lookups: actual switch to rwsem")
Fixes: d9b3dbdcfd ("fuse: switch to ->iterate_shared()")
The simple_write_to_buffer() already increments the @ppos on success,
see fs/libfs.c simple_write_to_buffer() comment:
"
On success, the number of bytes written is returned and the offset @ppos
advanced by this number, or negative value is returned on error.
"
If the configfs_write_bin_file() is invoked with @count smaller than the
total length of the written binary file, it will be invoked multiple times.
Since configfs_write_bin_file() increments @ppos on success, after calling
simple_write_to_buffer(), the @ppos is incremented twice.
Subsequent invocation of configfs_write_bin_file() will result in the next
piece of data being written to the offset twice as long as the length of
the previous write, thus creating buffer with "holes" in it.
The simple testcase using DTO follows:
$ mkdir /sys/kernel/config/device-tree/overlays/1
$ dd bs=1 if=foo.dtbo of=/sys/kernel/config/device-tree/overlays/1/dtbo
Without this patch, the testcase will result in twice as big buffer in the
kernel, which is then passed to the cfs_overlay_item_dtbo_write() .
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Several cases of overlapping changes, except the packet scheduler
conflicts which deal with the addition of the free list parameter
to qdisc_enqueue().
Signed-off-by: David S. Miller <davem@davemloft.net>
The two methods essentially do the same: find the real dentry/inode
belonging to an overlay dentry. The difference is in the usage:
vfs_open() uses ->d_select_inode() and expects the function to perform
copy-up if necessary based on the open flags argument.
file_dentry() uses ->d_real() passing in the overlay dentry as well as the
underlying inode.
vfs_rename() uses ->d_select_inode() but passes zero flags. ->d_real()
with a zero inode would have worked just as well here.
This patch merges the functionality of ->d_select_inode() into ->d_real()
by adding an 'open_flags' argument to the latter.
[Al Viro] Make the signature of d_real() match that of ->d_real() again.
And constify the inode argument, while we are at it.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Pull NFS client bugfixes from Anna Schumaker:
"Stable bugfixes:
- Fix _cancel_empty_pagelist
- Fix a double page unlock
- Make nfs_atomic_open() call d_drop() on all ->open_context() errors.
- Fix another OPEN_DOWNGRADE bug
Other bugfixes:
- Ensure we handle delegation errors in nfs4_proc_layoutget()
- Layout stateids start out as being invalid
- Add sparse lock annotations for pnfs_find_alloc_layout
- Handle bad delegation stateids in nfs4_layoutget_handle_exception
- Fix up O_DIRECT results
- Fix potential use after free of state in nfs4_do_reclaim.
- Mark the layout stateid invalid when all segments are removed
- Don't let readdirplus revalidate an inode that was marked as stale
- Fix potential race in nfs_fhget()
- Fix an unused variable warning"
* tag 'nfs-for-4.7-2' of git://git.linux-nfs.org/projects/anna/linux-nfs:
NFS: Fix another OPEN_DOWNGRADE bug
make nfs_atomic_open() call d_drop() on all ->open_context() errors.
NFS: Fix an unused variable warning
NFS: Fix potential race in nfs_fhget()
NFS: Don't let readdirplus revalidate an inode that was marked as stale
NFSv4.1/pnfs: Mark the layout stateid invalid when all segments are removed
NFS: Fix a double page unlock
pnfs_nfs: fix _cancel_empty_pagelist
nfs4: Fix potential use after free of state in nfs4_do_reclaim.
NFS: Fix up O_DIRECT results
NFS/pnfs: handle bad delegation stateids in nfs4_layoutget_handle_exception
NFSv4.1/pnfs: Add sparse lock annotations for pnfs_find_alloc_layout
NFSv4.1/pnfs: Layout stateids start out as being invalid
NFSv4.1/pnfs: Ensure we handle delegation errors in nfs4_proc_layoutget()
When truncating a file we should check write access on the underlying
inode. And we should do so on the lower file as well (before copy-up) for
consistency.
Original patch and test case by Aihua Zhang.
- - >o >o - - test.c - - >o >o - -
#include <stdio.h>
#include <errno.h>
#include <unistd.h>
int main(int argc, char *argv[])
{
int ret;
ret = truncate(argv[0], 4096);
if (ret != -1) {
fprintf(stderr, "truncate(argv[0]) should have failed\n");
return 1;
}
if (errno != ETXTBSY) {
perror("truncate(argv[0])");
return 1;
}
return 0;
}
- - >o >o - - >o >o - - >o >o - -
Reported-by: Aihua Zhang <zhangaihua1@huawei.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Cc: <stable@vger.kernel.org>
When using the 'default_permissions' mount option, ovl_permission() on
non-directories was missing a dput(alias), resulting in "BUG Dentry still
in use".
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 8d3095f4ad ("ovl: default permissions")
Cc: <stable@vger.kernel.org> # v4.5+
Olga Kornievskaia reports that the following test fails to trigger
an OPEN_DOWNGRADE on the wire, and only triggers the final CLOSE.
fd0 = open(foo, RDRW) -- should be open on the wire for "both"
fd1 = open(foo, RDONLY) -- should be open on the wire for "read"
close(fd0) -- should trigger an open_downgrade
read(fd1)
close(fd1)
The issue is that we're missing a check for whether or not the current
state transitioned from an O_RDWR state as opposed to having transitioned
from a combination of O_RDONLY and O_WRONLY.
Reported-by: Olga Kornievskaia <aglo@umich.edu>
Fixes: cd9288ffae ("NFSv4: Fix another bug in the close/open_downgrade code")
Cc: stable@vger.kernel.org # 2.6.33+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The CEC ioctls didn't have compat32 support, so they returned -ENOTTY
when used in a 32 bit application on a 64 bit kernel.
Since all the CEC ioctls are 32-bit compatible adding support for this
API is trivial.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Add checks to notify_change to verify that uid and gid changes
will map into the superblock's user namespace. If they do not
fail with -EOVERFLOW.
This is mandatory so that fileystems don't have to even think
of dealing with ia_uid and ia_gid that
--EWB Moved the test from inode_change_ok to notify_change
Signed-off-by: Seth Forshee <seth.forshee@canonical.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
This isn't functionally apparent for some reason, but
when we test io at extreme offsets at the end of the loff_t
rang, such as in fstests xfs/071, the calculation of
"max" in dax_io() can be wrong due to pos + size overflowing.
For example,
# xfs_io -c "pwrite 9223372036854771712 512" /mnt/test/file
enters dax_io with:
start 0x7ffffffffffff000
end 0x7ffffffffffff200
and the rounded up "size" variable is 0x1000. This yields:
pos + size 0x8000000000000000 (overflows loff_t)
end 0x7ffffffffffff200
Due to the overflow, the min() function picks the wrong
value for the "max" variable, and when we send (max - pos)
into i.e. copy_from_iter_pmem() it is also the wrong value.
This somehow(tm) gets magically absorbed without incident,
probably because iter->count is correct. But it seems best
to fix it up properly by comparing the two values as
unsigned.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Pull cifs fixes from Steve French:
"Various small cifs/smb3 fixes, include some for stable, and some from
the recent SMB3 test event"
* 'for-next' of git://git.samba.org/sfrench/cifs-2.6:
File names with trailing period or space need special case conversion
Fix reconnect to not defer smb3 session reconnect long after socket reconnect
cifs: check hash calculating succeeded
cifs: dynamic allocation of ntlmssp blob
cifs: use CIFS_MAX_DOMAINNAME_LEN when converting the domain name
cifs: stuff the fl_owner into "pid" field in the lock request
When gfs2 attempts to write a page to a file that is being truncated,
and notices that the page is completely outside of the file size, it
tries to invalidate it. However, this may require a transaction for
journaled data files to revoke any buffers from the page on the active
items list. Unfortunately, this can happen inside a log flush, where a
transaction cannot be started. Also, gfs2 may need to be able to remove
the buffer from the ail1 list before it can finish the log flush.
To deal with this, when writing a page of a file with data journalling
enabled gfs2 now skips the check to see if the write is outside the file
size, and simply writes it anyway. This situation can only occur when
the truncate code still has the file locked exclusively, and hasn't
marked this block as free in the metadata (which happens later in
truc_dealloc). After gfs2 writes this page out, the truncation code
will shortly invalidate it and write out any revokes if necessary.
To do this, gfs2 now implements its own version of block_write_full_page
without the check, and calls the newly exported __block_write_full_page.
It also no longer calls gfs2_writepage_common from gfs2_jdata_writepage.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
gfs2 needs to be able to skip the check to see if a page is outside of
the file size when writing it out. gfs2 can get into a situation where
it needs to flush its in-memory log to disk while a truncate is in
progress. If the file being trucated has data journaling enabled, it is
possible that there are data blocks in the log that are past the end of
the file. gfs can't finish the log flush without either writing these
blocks out or revoking them. Otherwise, if the node crashed, it could
overwrite subsequent changes made by other nodes in the cluster when
it's journal was replayed.
Unfortunately, there is no way to add log entries to the log during a
flush. So gfs2 simply writes out the page instead. This situation can
only occur when the truncate code still has the file locked exclusively,
and hasn't marked this block as free in the metadata (which happens
later in truc_dealloc). After gfs2 writes this page out, the truncation
code will shortly invalidate it and write out any revokes if necessary.
In order to make this work, gfs2 needs to be able to skip the check for
writes outside the file size. Since the check exists in
block_write_full_page, this patch exports __block_write_full_page, which
doesn't have the check.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Make the code more readable by cleaning up the different ways of
initializing lock holders and checking for initialized lock holders:
mark lock holders as uninitialized by setting the holder's glock to NULL
(gfs2_holder_mark_uninitialized) instead of zeroing out the entire
object or using a separate flag. Recognize initialized holders by their
non-NULL glock (gfs2_holder_initialized). Don't zero out holder objects
which are immeditiately initialized via gfs2_holder_init or
gfs2_glock_nq_init.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Commit ff34245d switched from iget5_locked to iget_locked among other
things, but iget_locked doesn't work for filesystems larger than 2^32
blocks on 32-bit systems. Switch back to iget5_locked. Filesystems
larger than 2^32 blocks are unrealistic to work well on 32-bit systems,
so this is mostly a code cleanliness fix.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Now that gfs2_lookup_by_inum only takes the inode glock for new inodes
(and not for cached inodes anymore), there no longer is a need to
optimize the cached-inode case in gfs2_get_dentry or delete_work_func,
and gfs2_ilookup can be removed.
In addition, gfs2_get_dentry wasn't checking the GFS2_DIF_SYSTEM flag in
i_diskflags in the gfs2_ilookup case (see gfs2_lookup_by_inum); this
inconsistency goes away as well.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
The current gfs2_lookup_by_inum takes the glock of a presumed inode
identified by block number, verifies that the block is indeed an inode,
and then instantiates and reads the new inode via gfs2_inode_lookup.
However, instantiating a new inode may block on freeing a previous
instance of that inode (__wait_on_freeing_inode), and freeing an inode
requires to take the glock already held, leading to lock inversion and
deadlock.
Fix this by first instantiating the new inode, then verifying that the
block is an inode (if required), and then reading in the new inode, all
in gfs2_inode_lookup.
If the block we are looking for is not an inode, we discard the new
inode via iget_failed, which marks inodes as bad and unhashes them.
Other tasks waiting on that inode will get back a bad inode back from
ilookup or iget_locked; in that case, retry the lookup.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
In "NFSv4: Move dentry instantiation into the NFSv4-specific atomic open code"
unconditional d_drop() after the ->open_context() had been removed. It had
been correct for success cases (there ->open_context() itself had been doing
dcache manipulations), but not for error ones. Only one of those (ENOENT)
got a compensatory d_drop() added in that commit, but in fact it should've
been done for all errors. As it is, the case of O_CREAT non-exclusive open
on a hashed negative dentry racing with e.g. symlink creation from another
client ended up with ->open_context() getting an error and proceeding to
call nfs_lookup(). On a hashed dentry, which would've instantly triggered
BUG_ON() in d_materialise_unique() (or, these days, its equivalent in
d_splice_alias()).
Cc: stable@vger.kernel.org # v3.10+
Tested-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Also, if we are going to issue the barrier, we should do this after we
write out the parent directories if necessary.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
If there are no pending blocks to be released after a commit, forcing
a journal commit has no hope of helping. It's possible that a commit
had just completed, so if there are now free blocks available for
allocation, it's worth retrying the commit.
Reported-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Pull btrfs fixes part 2 from Chris Mason:
"This has one patch from Omar to bring iterate_shared back to btrfs.
We have a tree of work we queue up for directory items and it doesn't
lend itself well to shared access. While we're cleaning it up, Omar
has changed things to use an exclusive lock when there are delayed
items"
* 'for-linus-4.7-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix ->iterate_shared() by upgrading i_rwsem for delayed nodes
Pull btrfs fixes from Chris Mason:
"I have a two part pull this time because one of the patches Dave
Sterba collected needed to be against v4.7-rc2 or higher (we used
rc4). I try to make my for-linus-xx branch testable on top of the
last major so we can hand fixes to people on the list more easily, so
I've split this pull in two.
This first part has some fixes and two performance improvements that
we've been testing for some time.
Josef's two performance fixes are most notable. The transid tracking
patch makes a big improvement on pretty much every workload"
* 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: Force stripesize to the value of sectorsize
btrfs: fix disk_i_size update bug when fallocate() fails
Btrfs: fix error handling in map_private_extent_buffer
Btrfs: fix error return code in btrfs_init_test_fs()
Btrfs: don't do nocow check unless we have to
btrfs: fix deadlock in delayed_ref_async_start
Btrfs: track transid for delayed ref flushing
Commit fe742fd4f9 ("Revert "btrfs: switch to ->iterate_shared()"")
backed out the conversion to ->iterate_shared() for Btrfs because the
delayed inode handling in btrfs_real_readdir() is racy. However, we can
still do readdir in parallel if there are no delayed nodes.
This is a temporary fix which upgrades the shared inode lock to an
exclusive lock only when we have delayed items until we come up with a
more complete solution. While we're here, rename the
btrfs_{get,put}_delayed_items functions to make it very clear that
they're just for readdir.
Tested with xfstests and by doing a parallel kernel build:
while make tinyconfig && make -j4 && git clean dqfx; do
:
done
along with a bunch of parallel finds in another shell:
while true; do
for ((i=0; i<4; i++)); do
find . >/dev/null &
done
wait
done
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
