This patch speeds up lmbench lat_mmap test by about another 2% after the
first patch.
Before:
avg = 462.286
std = 5.46106
After:
avg = 453.12
std = 9.58257
(50 runs of each, stddev gives a reasonable confidence)
It does this by introducing mnt_clone_write, which avoids some heavyweight
operations of mnt_want_write if called on a vfsmount which we know already
has a write count; and mnt_want_write_file, which can call mnt_clone_write
if the file is open for write.
After these two patches, mnt_want_write and mnt_drop_write go from 7% on
the profile down to 1.3% (including mnt_clone_write).
[AV: mnt_want_write_file() should take file alone and derive mnt from it;
not only all callers have that form, but that's the only mnt about which
we know that it's already held for write if file is opened for write]
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This patch speeds up lmbench lat_mmap test by about 8%. lat_mmap is set up
basically to mmap a 64MB file on tmpfs, fault in its pages, then unmap it.
A microbenchmark yes, but it exercises some important paths in the mm.
Before:
avg = 501.9
std = 14.7773
After:
avg = 462.286
std = 5.46106
(50 runs of each, stddev gives a reasonable confidence, but there is quite
a bit of variation there still)
It does this by removing the complex per-cpu locking and counter-cache and
replaces it with a percpu counter in struct vfsmount. This makes the code
much simpler, and avoids spinlocks (although the msync is still pretty
costly, unfortunately). It results in about 900 bytes smaller code too. It
does increase the size of a vfsmount, however.
It should also give a speedup on large systems if CPUs are frequently operating
on different mounts (because the existing scheme has to operate on an atomic in
the struct vfsmount when switching between mounts). But I'm most interested in
the single threaded path performance for the moment.
[AV: minor cleanup]
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
These guys are what we add as submounts; checks for "is that attached in
our namespace" are simply irrelevant for those and counterproductive for
use of private vfsmount trees a-la what NFS folks want.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
New field: nd->root. When pathname resolution wants to know the root,
check if nd->root.mnt is non-NULL; use nd->root if it is, otherwise
copy current->fs->root there. After path_walk() is finished, we check
if we'd got a cached value in nd->root and drop it. Before calling
path_walk() we should either set nd->root.mnt to NULL *or* copy (and
pin down) some path to nd->root. In the latter case we won't be
looking at current->fs->root at all.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Split do_path_lookup(), opencode the call from do_filp_open()
do_filp_open() is the only caller of do_path_lookup() that
cares about root afterwards (it keeps resolving symlinks on
O_CREAT path after it'd done LOOKUP_PARENT walk). So when
we start caching fs->root in path_walk(), it'll need a different
treatment.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This patch adds an -oexpose_privroot option to allow access to the privroot.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: (23 commits)
Btrfs: fix extent_buffer leak during tree log replay
Btrfs: fix oops when btrfs_inherit_iflags called with a NULL dir
Btrfs: fix -o nodatasum printk spelling
Btrfs: check duplicate backrefs for both data and metadata
Btrfs: init worker struct fields before kthread-run
Btrfs: pin buffers during write_dev_supers
Btrfs: avoid races between super writeout and device list updates
Fix btrfs when ACLs are configured out
Btrfs: fdatasync should skip metadata writeout
Btrfs: remove crc32c.h and use libcrc32c directly.
Btrfs: implement FS_IOC_GETFLAGS/SETFLAGS/GETVERSION
Btrfs: autodetect SSD devices
Btrfs: add mount -o ssd_spread to spread allocations out
Btrfs: avoid allocation clusters that are too spread out
Btrfs: Add mount -o nossd
Btrfs: avoid IO stalls behind congested devices in a multi-device FS
Btrfs: don't allow WRITE_SYNC bios to starve out regular writes
Btrfs: fix metadata dirty throttling limits
Btrfs: reduce mount -o ssd CPU usage
Btrfs: balance btree more often
...
* 'for-linus' of git://git.infradead.org/users/eparis/notify:
fsnotify: allow groups to set freeing_mark to null
inotify/dnotify: should_send_event shouldn't match on FS_EVENT_ON_CHILD
dnotify: do not bother to lock entry->lock when reading mask
dnotify: do not use ?true:false when assigning to a bool
fsnotify: move events should indicate the event was on a child
inotify: reimplement inotify using fsnotify
fsnotify: handle filesystem unmounts with fsnotify marks
fsnotify: fsnotify marks on inodes pin them in core
fsnotify: allow groups to add private data to events
fsnotify: add correlations between events
fsnotify: include pathnames with entries when possible
fsnotify: generic notification queue and waitq
dnotify: reimplement dnotify using fsnotify
fsnotify: parent event notification
fsnotify: add marks to inodes so groups can interpret how to handle those inodes
fsnotify: unified filesystem notification backend
* 'for-linus' of git://linux-arm.org/linux-2.6:
kmemleak: Add the corresponding MAINTAINERS entry
kmemleak: Simple testing module for kmemleak
kmemleak: Enable the building of the memory leak detector
kmemleak: Remove some of the kmemleak false positives
kmemleak: Add modules support
kmemleak: Add kmemleak_alloc callback from alloc_large_system_hash
kmemleak: Add the vmalloc memory allocation/freeing hooks
kmemleak: Add the slub memory allocation/freeing hooks
kmemleak: Add the slob memory allocation/freeing hooks
kmemleak: Add the slab memory allocation/freeing hooks
kmemleak: Add documentation on the memory leak detector
kmemleak: Add the base support
Manual conflict resolution (with the slab/earlyboot changes) in:
drivers/char/vt.c
init/main.c
mm/slab.c
* 'perfcounters-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (574 commits)
perf_counter: Turn off by default
perf_counter: Add counter->id to the throttle event
perf_counter: Better align code
perf_counter: Rename L2 to LL cache
perf_counter: Standardize event names
perf_counter: Rename enums
perf_counter tools: Clean up u64 usage
perf_counter: Rename perf_counter_limit sysctl
perf_counter: More paranoia settings
perf_counter: powerpc: Implement generalized cache events for POWER processors
perf_counters: powerpc: Add support for POWER7 processors
perf_counter: Accurate period data
perf_counter: Introduce struct for sample data
perf_counter tools: Normalize data using per sample period data
perf_counter: Annotate exit ctx recursion
perf_counter tools: Propagate signals properly
perf_counter tools: Small frequency related fixes
perf_counter: More aggressive frequency adjustment
perf_counter/x86: Fix the model number of Intel Core2 processors
perf_counter, x86: Correct some event and umask values for Intel processors
...
Most fsnotify listeners (all but inotify) do not care about marks being
freed. Allow groups to set freeing_mark to null and do not call any
function if it is set that way.
Signed-off-by: Eric Paris <eparis@redhat.com>
inotify and dnotify will both indicate that they want any event which came
from a child inode. The fix is to mask off FS_EVENT_ON_CHILD when deciding
if inotify or dnotify is interested in a given event.
Signed-off-by: Eric Paris <eparis@redhat.com>
entry->lock is needed to make sure entry->mask does not change while
manipulating it. In dnotify_should_send_event() we don't care if we get an
old or a new mask value out of this entry so there is no point it taking
the lock.
Signed-off-by: Eric Paris <eparis@redhat.com>
dnotify_should send event assigned a bool using ?true:false when computing
a bit operation. This is poitless and the bool type does this for us.
Signed-off-by: Eric Paris <eparis@redhat.com>
Reimplement inotify_user using fsnotify. This should be feature for feature
exactly the same as the original inotify_user. This does not make any changes
to the in kernel inotify feature used by audit. Those patches (and the eventual
removal of in kernel inotify) will come after the new inotify_user proves to be
working correctly.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
When an fs is unmounted with an fsnotify mark entry attached to one of its
inodes we need to destroy that mark entry and we also (like inotify) send
an unmount event.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
This patch pins any inodes with an fsnotify mark in core. The idea is that
as soon as the mark is removed from the inode->fsnotify_mark_entries list
the inode will be iput. In reality is doesn't quite work exactly this way.
The igrab will happen when the mark is added to an inode, but the iput will
happen when the inode pointer is NULL'd inside the mark.
It's possible that 2 racing things will try to remove the mark from
different directions. One may try to remove the mark because of an
explicit request and one might try to remove it because the inode was
deleted. It's possible that the removal because of inode deletion will
remove the mark from the inode's list, but the removal by explicit request
will actually set entry->inode == NULL; and call the iput. This is safe.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
inotify needs per group information attached to events. This patch allows
groups to attach private information and implements a callback so that
information can be freed when an event is being destroyed.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
As part of the standard inotify events it includes a correlation cookie
between two dentry move operations. This patch includes the same behaviour
in fsnotify events. It is needed so that inotify userspace can be
implemented on top of fsnotify.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
When inotify wants to send events to a directory about a child it includes
the name of the original file. This patch collects that filename and makes
it available for notification.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
inotify needs to do asyc notification in which event information is stored
on a queue until the listener is ready to receive it. This patch
implements a generic notification queue for inotify (and later fanotify) to
store events to be sent at a later time.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Reimplement dnotify using fsnotify.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
inotify and dnotify both use a similar parent notification mechanism. We
add a generic parent notification mechanism to fsnotify for both of these
to use. This new machanism also adds the dentry flag optimization which
exists for inotify to dnotify.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
This patch creates a way for fsnotify groups to attach marks to inodes.
These marks have little meaning to the generic fsnotify infrastructure
and thus their meaning should be interpreted by the group that attached
them to the inode's list.
dnotify and inotify will make use of these markings to indicate which
inodes are of interest to their respective groups. But this implementation
has the useful property that in the future other listeners could actually
use the marks for the exact opposite reason, aka to indicate which inodes
it had NO interest in.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
fsnotify is a backend for filesystem notification. fsnotify does
not provide any userspace interface but does provide the basis
needed for other notification schemes such as dnotify. fsnotify
can be extended to be the backend for inotify or the upcoming
fanotify. fsnotify provides a mechanism for "groups" to register for
some set of filesystem events and to then deliver those events to
those groups for processing.
fsnotify has a number of benefits, the first being actually shrinking the size
of an inode. Before fsnotify to support both dnotify and inotify an inode had
unsigned long i_dnotify_mask; /* Directory notify events */
struct dnotify_struct *i_dnotify; /* for directory notifications */
struct list_head inotify_watches; /* watches on this inode */
struct mutex inotify_mutex; /* protects the watches list
But with fsnotify this same functionallity (and more) is done with just
__u32 i_fsnotify_mask; /* all events for this inode */
struct hlist_head i_fsnotify_mark_entries; /* marks on this inode */
That's right, inotify, dnotify, and fanotify all in 64 bits. We used that
much space just in inotify_watches alone, before this patch set.
fsnotify object lifetime and locking is MUCH better than what we have today.
inotify locking is incredibly complex. See 8f7b0ba1c8 as an example of
what's been busted since inception. inotify needs to know internal semantics
of superblock destruction and unmounting to function. The inode pinning and
vfs contortions are horrible.
no fsnotify implementers do allocation under locks. This means things like
f04b30de3 which (due to an overabundance of caution) changes GFP_KERNEL to
GFP_NOFS can be reverted. There are no longer any allocation rules when using
or implementing your own fsnotify listener.
fsnotify paves the way for fanotify. In brief fanotify is a notification
mechanism that delivers the lisener both an 'event' and an open file descriptor
to the object in question. This means that fanotify is pathname agnostic.
Some on lkml may not care for the original companies or users that pushed for
TALPA, but fanotify was designed with flexibility and input for other users in
mind. The readahead group expressed interest in fanotify as it could be used
to profile disk access on boot without breaking the audit system. The desktop
search groups have also expressed interest in fanotify as it solves a number
of the race conditions and problems present with managing inotify when more
than a limited number of specific files are of interest. fanotify can provide
for a userspace access control system which makes it a clean interface for AV
vendors to hook without trying to do binary patching on the syscall table,
LSM, and everywhere else they do their things today. With this patch series
fanotify can be implemented in less than 1200 lines of easy to review code.
Almost all of which is the socket based user interface.
This patch series builds fsnotify to the point that it can implement
dnotify and inotify_user. Patches exist and will be sent soon after
acceptance to finish the in kernel inotify conversion (audit) and implement
fanotify.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
* 'for-2.6.31' of git://git.kernel.dk/linux-2.6-block: (153 commits)
block: add request clone interface (v2)
floppy: fix hibernation
ramdisk: remove long-deprecated "ramdisk=" boot-time parameter
fs/bio.c: add missing __user annotation
block: prevent possible io_context->refcount overflow
Add serial number support for virtio_blk, V4a
block: Add missing bounce_pfn stacking and fix comments
Revert "block: Fix bounce limit setting in DM"
cciss: decode unit attention in SCSI error handling code
cciss: Remove no longer needed sendcmd reject processing code
cciss: change SCSI error handling routines to work with interrupts enabled.
cciss: separate error processing and command retrying code in sendcmd_withirq_core()
cciss: factor out fix target status processing code from sendcmd functions
cciss: simplify interface of sendcmd() and sendcmd_withirq()
cciss: factor out core of sendcmd_withirq() for use by SCSI error handling code
cciss: Use schedule_timeout_uninterruptible in SCSI error handling code
block: needs to set the residual length of a bidi request
Revert "block: implement blkdev_readpages"
block: Fix bounce limit setting in DM
Removed reference to non-existing file Documentation/PCI/PCI-DMA-mapping.txt
...
Manually fix conflicts with tracing updates in:
block/blk-sysfs.c
drivers/ide/ide-atapi.c
drivers/ide/ide-cd.c
drivers/ide/ide-floppy.c
drivers/ide/ide-tape.c
include/trace/events/block.h
kernel/trace/blktrace.c
* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
cifs: remove never-used in6_addr option
cifs: add addr= mount option alias for ip=
[CIFS] Add mention of new mount parm (forceuid) to cifs readme
cifs: make overriding of ownership conditional on new mount options
cifs: fix IPv6 address length check
cifs: clean up set_cifs_acl interfaces
cifs: reorganize get_cifs_acl
[CIFS] Update readme to indicate change to default mount (serverino)
cifs: make serverino the default when mounting
cifs: rename cifs_iget to cifs_root_iget
cifs: make cnvrtDosUnixTm take a little-endian args and an offset
cifs: have cifs_NTtimeToUnix take a little-endian arg
cifs: tighten up default file_mode/dir_mode
cifs: fix artificial limit on reading symlinks
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (49 commits)
ext4: Avoid corrupting the uninitialized bit in the extent during truncate
ext4: Don't treat a truncation of a zero-length file as replace-via-truncate
ext4: fix dx_map_entry to support 256k directory blocks
ext4: truncate the file properly if we fail to copy data from userspace
ext4: Avoid leaking blocks after a block allocation failure
ext4: Change all super.c messages to print the device
ext4: Get rid of EXTEND_DISKSIZE flag of ext4_get_blocks_handle()
ext4: super.c whitespace cleanup
jbd2: Fix minor typos in comments in fs/jbd2/journal.c
ext4: Clean up calls to ext4_get_group_desc()
ext4: remove unused function __ext4_write_dirty_metadata
ext2: Fix memory leak in ext2_fill_super() in case of a failed mount
ext3: Fix memory leak in ext3_fill_super() in case of a failed mount
ext4: Fix memory leak in ext4_fill_super() in case of a failed mount
ext4: down i_data_sem only for read when walking tree for fiemap
ext4: Add a comprehensive block validity check to ext4_get_blocks()
ext4: Clean up ext4_get_blocks() so it does not depend on bh_result->b_state
ext4: Merge ext4_da_get_block_write() into mpage_da_map_blocks()
ext4: Add BUG_ON debugging checks to noalloc_get_block_write()
ext4: Add documentation to the ext4_*get_block* functions
...
There are allocations for which the main pointer cannot be found but
they are not memory leaks. This patch fixes some of them. For more
information on false positives, see Documentation/kmemleak.txt.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
* serial-from-alan: (79 commits)
moxa: prevent opening unavailable ports
imx: serial: use tty_encode_baud_rate to set true rate
imx: serial: add IrDA support to serial driver
imx: serial: use rational library function
lib: isolate rational fractions helper function
imx: serial: handle initialisation failure correctly
imx: serial: be sure to stop xmit upon shutdown
imx: serial: notify higher layers in case xmit IRQ was not called
imx: serial: fix one bit field type
imx: serial: fix whitespaces (no changes in functionality)
tty: use prepare/finish_wait
tty: remove sleep_on
sierra: driver interface blacklisting
sierra: driver urb handling improvements
tty: resolve some sierra breakage
timbuart: Fix the termios logic
serial: Added Timberdale UART driver
tty: Add URL for ttydev queue
devpts: unregister the file system on error
tty: Untangle termios and mm mutex dependencies
...
During tree log replay, we read in the tree log roots,
process them and then free them. A recent change
takes an extra reference on the root node of the tree
when the root is read in, and stores that reference
in root->commit_root.
This reference was not being freed, leaving us with
one buffer pinned in ram for each subvol with
a tree log root after a crash.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
lookup_inline_extent_backref only checks for duplicate backref for data
extents. It assumes backrefs for tree block never conflict.
This patch makes lookup_inline_extent_backref check for duplicate backrefs
for both data and tree block, so that we can detect potential bug earlier.
This is a safety check, strictly speaking it is not required.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* 'tracing-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (244 commits)
Revert "x86, bts: reenable ptrace branch trace support"
tracing: do not translate event helper macros in print format
ftrace/documentation: fix typo in function grapher name
tracing/events: convert block trace points to TRACE_EVENT(), fix !CONFIG_BLOCK
tracing: add protection around module events unload
tracing: add trace_seq_vprint interface
tracing: fix the block trace points print size
tracing/events: convert block trace points to TRACE_EVENT()
ring-buffer: fix ret in rb_add_time_stamp
ring-buffer: pass in lockdep class key for reader_lock
tracing: add annotation to what type of stack trace is recorded
tracing: fix multiple use of __print_flags and __print_symbolic
tracing/events: fix output format of user stack
tracing/events: fix output format of kernel stack
tracing/trace_stack: fix the number of entries in the header
ring-buffer: discard timestamps that are at the start of the buffer
ring-buffer: try to discard unneeded timestamps
ring-buffer: fix bug in ring_buffer_discard_commit
ftrace: do not profile functions when disabled
tracing: make trace pipe recognize latency format flag
...
This patch fixes a bug which may result race condition
between btrfs_start_workers() and worker_loop().
btrfs_start_workers() executed in a parent thread writes
on workers->worker and worker_loop() in a child thread
reads workers->worker. However, there is no synchronization
enforcing the order of two operations.
This patch makes btrfs_start_workers() fill workers->worker
before it starts a child thread with worker_loop()
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sched: fix typo in sched-rt-group.txt file
ftrace: fix typo about map of kernel priority in ftrace.txt file.
sched: properly define the sched_group::cpumask and sched_domain::span fields
sched, timers: cleanup avenrun users
sched, timers: move calc_load() to scheduler
sched: Don't export sched_mc_power_savings on multi-socket single core system
sched: emit thread info flags with stack trace
sched: rt: document the risk of small values in the bandwidth settings
sched: Replace first_cpu() with cpumask_first() in ILB nomination code
sched: remove extra call overhead for schedule()
sched: use group_first_cpu() instead of cpumask_first(sched_group_cpus())
wait: don't use __wake_up_common()
sched: Nominate a power-efficient ilb in select_nohz_balancer()
sched: Nominate idle load balancer from a semi-idle package.
sched: remove redundant hierarchy walk in check_preempt_wakeup
write_dev_supers is called in sequence. First is it called with wait == 0,
which starts IO on all of the super blocks for a given device. Then it is
called with wait == 1 to make sure they all reach the disk.
It doesn't currently pin the buffers between the two calls, and it also
assumes the buffers won't go away between the two calls, leading to
an oops if the VM manages to free the buffers in the middle of the sync.
This fixes that assumption and updates the code to return an error if things
are not up to date when the wait == 1 run is done.
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
On multi-device filesystems, btrfs writes supers to all of the devices
before considering a sync complete. There wasn't any additional
locking between super writeout and the device list management code
because device management was done inside a transaction and
super writeout only happened with no transation writers running.
With the btrfs fsync log and other async transaction updates, this
has been racey for some time. This adds a mutex to protect
the device list. The existing volume mutex could not be reused due to
transaction lock ordering requirements.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This option was never used to my knowledge. Remove it before someone
does...
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
The unitialized bit was not properly getting preserved in in an extent
which is partially truncated because the it was geting set to the
value of the first extent to be removed or truncated as part of the
truncate operation, and if there are multiple extents are getting
removed or modified as part of the truncate operation, it is only the
last extent which will might be partially truncated, and its
uninitalized bit is not necessarily the same as the first extent to be
truncated.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
When you look in /proc/mounts, the address of the server gets displayed
as "addr=". That's really a better option to use anyway since it's more
generic. What if we eventually want to support non-IP transports? It
also makes CIFS option consistent with the NFS option of the same name.
Begin the migration to that option name by adding an alias for ip=
called addr=.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
... otherwise generic_permission() will allow *anything* for all
files you don't own and that have some group permissions.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
In btrfs, fdatasync and fsync are identical, but
fdatasync should skip committing transaction when
inode->i_state is set just I_DIRTY_SYNC and this indicates
only atime or/and mtime updates.
Following patch improves fdatasync throughput.
--file-block-size=4K --file-total-size=16G --file-test-mode=rndwr
--file-fsync-mode=fdatasync run
Results:
-2.6.30-rc8
Test execution summary:
total time: 1980.6540s
total number of events: 10001
total time taken by event execution: 1192.9804
per-request statistics:
min: 0.0000s
avg: 0.1193s
max: 15.3720s
approx. 95 percentile: 0.7257s
Threads fairness:
events (avg/stddev): 625.0625/151.32
execution time (avg/stddev): 74.5613/9.46
-2.6.30-rc8-patched
Test execution summary:
total time: 1695.9118s
total number of events: 10000
total time taken by event execution: 871.3214
per-request statistics:
min: 0.0000s
avg: 0.0871s
max: 10.4644s
approx. 95 percentile: 0.4787s
Threads fairness:
events (avg/stddev): 625.0000/131.86
execution time (avg/stddev): 54.4576/8.98
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There's no need to preserve this abstraction; it used to let us use
hardware crc32c support directly, but libcrc32c is already doing that for us
through the crypto API -- so we're already using the Intel crc32c
acceleration where appropriate.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Add support for the standard attributes set via chattr and read via
lsattr. Currently we store the attributes in the flags value in
the btrfs inode, but I wonder whether we should split it into two so
that we don't have to keep converting between the two formats.
Remove the btrfs_clear_flag/btrfs_set_flag/btrfs_test_flag macros
as they were confusing the existing code and got in the way of the
new additions.
Also add the FS_IOC_GETVERSION ioctl for getting i_generation as it's
trivial.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
During mount, btrfs will check the queue nonrot flag
for all the devices found in the FS. If they are all
non-rotating, SSD mode is enabled by default.
If the FS was mounted with -o nossd, the non-rotating
flag is ignored.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Some SSDs perform best when reusing block numbers often, while
others perform much better when clustering strictly allocates
big chunks of unused space.
The default mount -o ssd will find rough groupings of blocks
where there are a bunch of free blocks that might have some
allocated blocks mixed in.
mount -o ssd_spread will make sure there are no allocated blocks
mixed in. It should perform better on lower end SSDs.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
In SSD mode for data, and all the time for metadata the allocator
will try to find a cluster of nearby blocks for allocations. This
commit adds extra checks to make sure that each free block in the
cluster is close to the last one.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The btrfs IO submission threads try to service a bunch of devices with a small
number of threads. They do a congestion check to try and avoid waiting
on requests for a busy device.
The checks make sure we've sent a few requests down to a given device just so
that we aren't bouncing between busy devices without actually sending down
any IO. The counter used to decide if we can switch to the next device
is somewhat overloaded. It is also being used to decide if we've done
a good batch of requests between the WRITE_SYNC or regular priority lists.
It may get reset to zero often, leaving us hammering on a busy device
instead of moving on to another disk.
This commit adds a new counter for the number of bios sent while
servicing a device. It doesn't get reset or fiddled with. On
multi-device filesystems, this fixes IO stalls in streaming
write workloads.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs uses dedicated threads to submit bios when checksumming is on,
which allows us to make sure the threads dedicated to checksumming don't get
stuck waiting for requests. For each btrfs device, there are
two lists of bios. One list is for WRITE_SYNC bios and the other
is for regular priority bios.
The IO submission threads used to process all of the WRITE_SYNC bios first and
then switch to the regular bios. This commit makes sure we don't completely
starve the regular bios by rotating between the two lists.
WRITE_SYNC bios are still favored 2:1 over the regular bios, and this tries
to run in batches to avoid seeking. Benchmarking shows this eliminates
stalls during streaming buffered writes on both multi-device and
single device filesystems.
If the regular bios starve, the system can end up with a large amount of ram
pinned down in writeback pages. If we are a little more fair between the two
classes, we're able to keep throughput up and make progress on the bulk of
our dirty ram.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Once a metadata block has been written, it must be recowed, so the
btrfs dirty balancing call has a check to make sure a fair amount of metadata
was actually dirty before it started writing it back to disk.
A previous commit had changed the dirty tracking for metadata without
updating the btrfs dirty balancing checks. This commit switches it
to use the correct counter.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The block allocator in SSD mode will try to find groups of free blocks
that are close together. This commit makes it loop less on a given
group size before bumping it.
The end result is that we are less likely to fill small holes in the
available free space, but we don't waste as much CPU building the
large cluster used by ssd mode.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
With the new back reference code, the cost of a balance has gone down
in terms of the number of back reference updates done. This commit
makes us more aggressively balance leaves and nodes as they become
less full.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When the delayed reference code was added, some checks were added
to avoid extra balancing while the delayed references were being flushed.
This made for less efficient btrees, but it reduced the chances of
loops where no forward progress was made because the balances made
more delayed ref updates.
With the new dead root removal code and the mixed back references,
the extent allocation tree is no longer using precise back refs, and
the delayed reference updates don't carry the risk of looping forever
anymore. So, the balance avoidance is no longer required.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit introduces a new kind of back reference for btrfs metadata.
Once a filesystem has been mounted with this commit, IT WILL NO LONGER
BE MOUNTABLE BY OLDER KERNELS.
When a tree block in subvolume tree is cow'd, the reference counts of all
extents it points to are increased by one. At transaction commit time,
the old root of the subvolume is recorded in a "dead root" data structure,
and the btree it points to is later walked, dropping reference counts
and freeing any blocks where the reference count goes to 0.
The increments done during cow and decrements done after commit cancel out,
and the walk is a very expensive way to go about freeing the blocks that
are no longer referenced by the new btree root. This commit reduces the
transaction overhead by avoiding the need for dead root records.
When a non-shared tree block is cow'd, we free the old block at once, and the
new block inherits old block's references. When a tree block with reference
count > 1 is cow'd, we increase the reference counts of all extents
the new block points to by one, and decrease the old block's reference count by
one.
This dead tree avoidance code removes the need to modify the reference
counts of lower level extents when a non-shared tree block is cow'd.
But we still need to update back ref for all pointers in the block.
This is because the location of the block is recorded in the back ref
item.
We can solve this by introducing a new type of back ref. The new
back ref provides information about pointer's key, level and in which
tree the pointer lives. This information allow us to find the pointer
by searching the tree. The shortcoming of the new back ref is that it
only works for pointers in tree blocks referenced by their owner trees.
This is mostly a problem for snapshots, where resolving one of these
fuzzy back references would be O(number_of_snapshots) and quite slow.
The solution used here is to use the fuzzy back references in the common
case where a given tree block is only referenced by one root,
and use the full back references when multiple roots have a reference
on a given block.
This commit adds per subvolume red-black tree to keep trace of cached
inodes. The red-black tree helps the balancing code to find cached
inodes whose inode numbers within a given range.
This commit improves the balancing code by introducing several data
structures to keep the state of balancing. The most important one
is the back ref cache. It caches how the upper level tree blocks are
referenced. This greatly reduce the overhead of checking back ref.
The improved balancing code scales significantly better with a large
number of snapshots.
This is a very large commit and was written in a number of
pieces. But, they depend heavily on the disk format change and were
squashed together to make sure git bisect didn't end up in a
bad state wrt space balancing or the format change.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There are some 'start = state->end + 1;' like code in set_extent_bit
and clear_extent_bit. They overflow when end == (u64)-1.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch rips out the XFS ACL handling code and uses the generic
fs/posix_acl.c code instead. The ondisk format is of course left
unchanged.
This also introduces the same ACL caching all other Linux filesystems do
by adding pointers to the acl and default acl in struct xfs_inode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Although get_block() callback function can return extent of contiguous
blocks with bh->b_size, nilfs_get_block() function did not support
this feature.
This adds contiguous lookup feature to the block mapping codes of
nilfs, and allows the nilfs_get_blocks() function to return the extent
information by applying the feature.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This applies block_sync_page() function to the sync_page method of
page caches for meta data files, gc page caches, and btree node
buffers. This is a companion patch of ("nilfs2: enable sync_page
mothod") which applied the function for data pages.
This allows lock_page() for those meta data to unplug pending bio
requests.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Previously, default_backing_dev_info was used for the mapping of btree
node caches. This uses device dependent backing_dev_info to allow
detailed control of the device for the btree node pages.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This helps userland programs like the rmcp command to distinguish
error codes returned against a checkpoint removal request.
Previously -EPERM was returned, and not discriminable from real
permission errors. This also allows removal of the latest checkpoint
because the deletion leads to create a new checkpoint, and thus it's
harmless for the filesystem.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This adds a missing sync_page method which unplugs bio requests when
waiting for page locks. This will improve read performance of nilfs.
Here is a measurement result using dd command.
Without this patch:
# mount -t nilfs2 /dev/sde1 /test
# dd if=/test/aaa of=/dev/null bs=512k
1024+0 records in
1024+0 records out
536870912 bytes (537 MB) copied, 6.00688 seconds, 89.4 MB/s
With this patch:
# mount -t nilfs2 /dev/sde1 /test
# dd if=/test/aaa of=/dev/null bs=512k
1024+0 records in
1024+0 records out
536870912 bytes (537 MB) copied, 3.54998 seconds, 151 MB/s
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This sets BIO_RW_UNPLUG flag on the last bio of each segment during
write. The last bio should be unplugged immediately because the
caller waits for the completion after the submission.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Nilfs has some ioctl commands to read out metadata from meta data
files:
- NILFS_IOCTL_GET_CPINFO for checkpoint file,
- NILFS_IOCTL_GET_SUINFO for segment usage file, and
- NILFS_IOCTL_GET_VINFO for Disk Address Transalation (DAT) file,
respectively.
Every routine on these metadata files is implemented so that it allows
future expansion of on-disk format. But, the above ioctl commands do
not support expansion even though nilfs_argv structure can handle
arbitrary size for data exchanged via ioctl.
This allows future expansion of the following structures which give
basic format of the "get information" ioctls:
- struct nilfs_cpinfo
- struct nilfs_suinfo
- struct nilfs_vinfo
So, this introduces forward compatility of such ioctl commands.
In this patch, a sanity check in nilfs_ioctl_get_info() function is
changed to accept larger data structure [1], and metadata read
routines are rewritten so that they become compatible for larger
structures; the routines will just ignore the remaining fields which
the current version of nilfs doesn't know.
[1] The ioctl function already has another upper limit (PAGE_SIZE
against a structure, which appears in nilfs_ioctl_wrap_copy
function), and this will not cause security problem.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Hi,
I introduced "is_partially_uptodate" aops for NILFS2.
A page can have multiple buffers and even if a page is not uptodate, some buffers
can be uptodate on pagesize != blocksize environment.
This aops checks that all buffers which correspond to a part of a file
that we want to read are uptodate. If so, we do not have to issue actual
read IO to HDD even if a page is not uptodate because the portion we
want to read are uptodate.
"block_is_partially_uptodate" function is already used by ext2/3/4.
With the following patch random read/write mixed workloads or random read after
random write workloads can be optimized and we can get performance improvement.
I did a performance test using the sysbench.
1 --file-block-size=8K --file-total-size=2G --file-test-mode=rndrw --file-fsync-freq=0 --fil
e-rw-ratio=1 run
-2.6.30-rc5
Test execution summary:
total time: 151.2907s
total number of events: 200000
total time taken by event execution: 2409.8387
per-request statistics:
min: 0.0000s
avg: 0.0120s
max: 0.9306s
approx. 95 percentile: 0.0439s
Threads fairness:
events (avg/stddev): 12500.0000/238.52
execution time (avg/stddev): 150.6149/0.01
-2.6.30-rc5-patched
Test execution summary:
total time: 140.8828s
total number of events: 200000
total time taken by event execution: 2240.8577
per-request statistics:
min: 0.0000s
avg: 0.0112s
max: 0.8750s
approx. 95 percentile: 0.0418s
Threads fairness:
events (avg/stddev): 12500.0000/218.43
execution time (avg/stddev): 140.0536/0.01
arch: ia64
pagesize: 16k
Thanks.
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Previously, the bmap codes of nilfs used three types of function
tables. The abuse of indirect function calls decreased source
readability and suffered many indirect jumps which would confuse
branch prediction of processors.
This eliminates one type of the function tables,
nilfs_bmap_ptr_operations, which was used to dispatch low level
pointer operations of the nilfs bmap.
This adds a new integer variable "b_ptr_type" to nilfs_bmap struct,
and uses the value to select the pointer operations.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This will cut off 16 bytes from the nilfs_bmap struct which is
embedded in the on-memory inode of nilfs.
The b_high field was never used, and the b_low field stores a constant
value which can be determined by whether the inode uses btree for
block mapping or not.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This indirect function is set to NULL only for gc cache inodes, but
the gc cache inodes never call this function.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Two get block function for btree nodes, nilfs_bmap_get_block() and
nilfs_bmap_get_new_block(), are called only from the btree codes.
This relocation will increase opportunities of compiler optimization.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
nilfs_bmap_delete_block() is a wrapper function calling
nilfs_btnode_delete(). This removes it for simplicity.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
nilfs_bmap_put_block() is a wrapper function calling brelse(). This
eliminates the wrapper for simplicity.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This will eliminate obsolete list operations of nilfs_segment_entry
structure which has been used to handle mutiple segment numbers.
The patch ("nilfs2: remove list of freeing segments") removed use of
the structure from the segment constructor code, and this patch
simplifies the remaining code by integrating it into recovery.c.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This will clean up the removal list of segments and the related
functions from segment.c and ioctl.c, which have hurt code
readability.
This elimination is applied by using nilfs_sufile_updatev() previously
introduced in the patch ("nilfs2: add sufile function that can modify
multiple segment usages").
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This is a preparation for the later cleanup patch ("nilfs2: remove
list of freeing segments").
This adds nilfs_sufile_updatev() to sufile, which can modify multiple
segment usages at a time.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
This simplifies some low level functions of bmap.
Three bmap pointer operations, nilfs_bmap_start_v(),
nilfs_bmap_commit_v(), and nilfs_bmap_abort_v(), are unified into one
nilfs_bmap_start_v() function. And the related indirect function calls
are replaced with it.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
libosd users that need to work with bios, must sometime use
the request_queue associated with the osd_dev. Make a wrapper for
that, and convert all in-tree users.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
For supporting of chained-bios we can not inspect the first
bio only, as before. Caller shall pass the total length of the
request, ie. sum_bytes(bio-chain).
Also since the bio might be a chain we don't set it's direction
on behalf of it's callers. The bio direction should be properly
set prior to this call. So fix a couple of write users that now
need to set the bio direction properly
[In this patch I change both library code and user sites at
exofs, to make it easy on integration. It should be submitted
via James's scsi-misc tree.]
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
CC: Jeff Garzik <jeff@garzik.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
By popular demand, define usefull wrappers for osd_req_read/write
that recieve kernel pointers. All users had their own.
Also remove these from exofs
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
If a page was partially zeroed as the result of a truncate, then it was
not being correctly marked dirty. This resulted in the deleted data
reappearing if the file was read back via direct I/O.
Reported-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
In commit code, we scan buffers attached to a transaction. During this
scan, we sometimes have to drop j_list_lock and then we recheck whether
the journal buffer head didn't get freed by journal_try_to_free_buffers().
But checking for buffer_jbd(bh) isn't enough because a new journal head
could get attached to our buffer head. So add a check whether the journal
head remained the same and whether it's still at the same transaction and
list.
This is a nasty bug and can cause problems like memory corruption (use after
free) or trigger various assertions in JBD code (observed).
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: <stable@kernel.org>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The recent ->lookup() deadlock correction required the directory inode
mutex to be dropped while waiting for expire completion. We were
concerned about side effects from this change and one has been identified.
I saw several error messages.
They cause autofs to become quite confused and don't really point to the
actual problem.
Things like:
handle_packet_missing_direct:1376: can't find map entry for (43,1827932)
which is usually totally fatal (although in this case it wouldn't be
except that I treat is as such because it normally is).
do_mount_direct: direct trigger not valid or already mounted
/test/nested/g3c/s1/ss1
which is recoverable, however if this problem is at play it can cause
autofs to become quite confused as to the dependencies in the mount tree
because mount triggers end up mounted multiple times. It's hard to
accurately check for this over mounting case and automount shouldn't need
to if the kernel module is doing its job.
There was one other message, similar in consequence of this last one but I
can't locate a log example just now.
When checking if a mount has already completed prior to adding a new mount
request to the wait queue we check if the dentry is hashed and, if so, if
it is a mount point. But, if a mount successfully completed while we
slept on the wait queue mutex the dentry must exist for the mount to have
completed so the test is not really needed.
Mounts can also be done on top of a global root dentry, so for the above
case, where a mount request completes and the wait queue entry has already
been removed, the hashed test returning false can cause an incorrect
callback to the daemon. Also, d_mountpoint() is not sufficient to check
if a mount has completed for the multi-mount case when we don't have a
real mount at the base of the tree.
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In ocfs2, fdatasync and fsync are identical.
I think fdatasync should skip committing transaction when
inode->i_state is set just I_DIRTY_SYNC and this indicates
only atime or/and mtime updates.
Following patch improves fdatasync throughput.
#sysbench --num-threads=16 --max-requests=300000 --test=fileio
--file-block-size=4K --file-total-size=16G --file-test-mode=rndwr
--file-fsync-mode=fdatasync run
Results:
-2.6.30-rc8
Test execution summary:
total time: 107.1445s
total number of events: 119559
total time taken by event execution: 116.1050
per-request statistics:
min: 0.0000s
avg: 0.0010s
max: 0.1220s
approx. 95 percentile: 0.0016s
Threads fairness:
events (avg/stddev): 7472.4375/303.60
execution time (avg/stddev): 7.2566/0.64
-2.6.30-rc8-patched
Test execution summary:
total time: 86.8529s
total number of events: 300016
total time taken by event execution: 24.3077
per-request statistics:
min: 0.0000s
avg: 0.0001s
max: 0.0336s
approx. 95 percentile: 0.0001s
Threads fairness:
events (avg/stddev): 18751.0000/718.75
execution time (avg/stddev): 1.5192/0.05
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Acked-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
TRACE_EVENT is a more generic way to define tracepoints. Doing so adds
these new capabilities to this tracepoint:
- zero-copy and per-cpu splice() tracing
- binary tracing without printf overhead
- structured logging records exposed under /debug/tracing/events
- trace events embedded in function tracer output and other plugins
- user-defined, per tracepoint filter expressions
...
Cons:
- no dev_t info for the output of plug, unplug_timer and unplug_io events.
no dev_t info for getrq and sleeprq events if bio == NULL.
no dev_t info for rq_abort,...,rq_requeue events if rq->rq_disk == NULL.
This is mainly because we can't get the deivce from a request queue.
But this may change in the future.
- A packet command is converted to a string in TP_assign, not TP_print.
While blktrace do the convertion just before output.
Since pc requests should be rather rare, this is not a big issue.
- In blktrace, an event can have 2 different print formats, but a TRACE_EVENT
has a unique format, which means we have some unused data in a trace entry.
The overhead is minimized by using __dynamic_array() instead of __array().
I've benchmarked the ioctl blktrace vs the splice based TRACE_EVENT tracing:
dd dd + ioctl blktrace dd + TRACE_EVENT (splice)
1 7.36s, 42.7 MB/s 7.50s, 42.0 MB/s 7.41s, 42.5 MB/s
2 7.43s, 42.3 MB/s 7.48s, 42.1 MB/s 7.43s, 42.4 MB/s
3 7.38s, 42.6 MB/s 7.45s, 42.2 MB/s 7.41s, 42.5 MB/s
So the overhead of tracing is very small, and no regression when using
those trace events vs blktrace.
And the binary output of TRACE_EVENT is much smaller than blktrace:
# ls -l -h
-rw-r--r-- 1 root root 8.8M 06-09 13:24 sda.blktrace.0
-rw-r--r-- 1 root root 195K 06-09 13:24 sda.blktrace.1
-rw-r--r-- 1 root root 2.7M 06-09 13:25 trace_splice.out
Following are some comparisons between TRACE_EVENT and blktrace:
plug:
kjournald-480 [000] 303.084981: block_plug: [kjournald]
kjournald-480 [000] 303.084981: 8,0 P N [kjournald]
unplug_io:
kblockd/0-118 [000] 300.052973: block_unplug_io: [kblockd/0] 1
kblockd/0-118 [000] 300.052974: 8,0 U N [kblockd/0] 1
remap:
kjournald-480 [000] 303.085042: block_remap: 8,0 W 102736992 + 8 <- (8,8) 33384
kjournald-480 [000] 303.085043: 8,0 A W 102736992 + 8 <- (8,8) 33384
bio_backmerge:
kjournald-480 [000] 303.085086: block_bio_backmerge: 8,0 W 102737032 + 8 [kjournald]
kjournald-480 [000] 303.085086: 8,0 M W 102737032 + 8 [kjournald]
getrq:
kjournald-480 [000] 303.084974: block_getrq: 8,0 W 102736984 + 8 [kjournald]
kjournald-480 [000] 303.084975: 8,0 G W 102736984 + 8 [kjournald]
bash-2066 [001] 1072.953770: 8,0 G N [bash]
bash-2066 [001] 1072.953773: block_getrq: 0,0 N 0 + 0 [bash]
rq_complete:
konsole-2065 [001] 300.053184: block_rq_complete: 8,0 W () 103669040 + 16 [0]
konsole-2065 [001] 300.053191: 8,0 C W 103669040 + 16 [0]
ksoftirqd/1-7 [001] 1072.953811: 8,0 C N (5a 00 08 00 00 00 00 00 24 00) [0]
ksoftirqd/1-7 [001] 1072.953813: block_rq_complete: 0,0 N (5a 00 08 00 00 00 00 00 24 00) 0 + 0 [0]
rq_insert:
kjournald-480 [000] 303.084985: block_rq_insert: 8,0 W 0 () 102736984 + 8 [kjournald]
kjournald-480 [000] 303.084986: 8,0 I W 102736984 + 8 [kjournald]
Changelog from v2 -> v3:
- use the newly introduced __dynamic_array().
Changelog from v1 -> v2:
- use __string() instead of __array() to minimize the memory required
to store hex dump of rq->cmd().
- support large pc requests.
- add missing blk_fill_rwbs_rq() in block_rq_requeue TRACE_EVENT.
- some cleanups.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <4A2DF669.5070905@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If a non-existent file is opened via O_WRONLY|O_CREAT|O_TRUNC, there's
no need to treat this as a true file truncation, so we shouldn't
activate the replace-via-truncate hueristic.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
CUSE enables implementing character devices in userspace. With recent
additions of ioctl and poll support, FUSE already has most of what's
necessary to implement character devices. All CUSE has to do is
bonding all those components - FUSE, chardev and the driver model -
nicely.
When client opens /dev/cuse, kernel starts conversation with
CUSE_INIT. The client tells CUSE which device it wants to create. As
the previous patch made fuse_file usable without associated
fuse_inode, CUSE doesn't create super block or inodes. It attaches
fuse_file to cdev file->private_data during open and set ff->fi to
NULL. The rest of the operation is almost identical to FUSE direct IO
case.
Each CUSE device has a corresponding directory /sys/class/cuse/DEVNAME
(which is symlink to /sys/devices/virtual/class/DEVNAME if
SYSFS_DEPRECATED is turned off) which hosts "waiting" and "abort"
among other things. Those two files have the same meaning as the FUSE
control files.
The only notable lacking feature compared to in-kernel implementation
is mmap support.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
The dx_map_entry structure doesn't support over 64KB block size by
current usage of its member("offs"). Because "offs" treats an offset
of copies of the ext4_dir_entry_2 structure as is. This member size is
16 bits. But real offset for over 64KB(256KB) block size needs 18
bits. However, real offset keeps 4 byte boundary, so lower 2 bits is
not used.
Therefore, we do the following to fix this limitation:
For "store":
we divide the real offset by 4 and then store this result to "offs"
member.
For "use":
we multiply "offs" member by 4 and then use this result
as real offset.
Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
SYNC_BDFLUSH is a leftover from IRIX and rather misnamed for todays
code. Make xfs_sync_fsdata and xfs_dq_sync use the SYNC_TRYLOCK flag
for not blocking on logs just as the inode sync code already does.
For xfs_sync_fsdata it's a trivial 1:1 replacement, but for xfs_qm_sync
I use the opportunity to decouple the non-blocking lock case from the
different flushing modes, similar to the inode sync code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
We want to wait for all I/O to finish when we do data integrity syncs. So
there is no reason to keep SYNC_WAIT separate from SYNC_IOWAIT. This
causes a little change in behaviour for the ENOSPC flushing code which now
does a second submission and wait of buffered I/O, but that should finish
ASAP as we already did an asynchronous writeout earlier.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
xfs_sync_inodes is used to write back either file data or inode metadata.
In general we always do these separately, except for one fishy case in
xfs_fs_put_super that does both. So separate xfs_sync_inodes into
separate xfs_sync_data and xfs_sync_attr functions. In xfs_fs_put_super
we first call the data sync and then the attr sync as that was the previous
order. The moved log force in that path doesn't make a difference because
we will force the log again as part of the real unmount process.
The filesystem readonly checks are not performed by the new function but
instead moved into the callers, given that most callers alredy have it
further up in the stack. Also add debug checks that we do not pass in
incorrect flags in the new xfs_sync_data and xfs_sync_attr function and
fix the one place that did pass in a wrong flag.
Also remove a comment mentioning xfs_sync_inodes that has been incorrect
for a while because we always take either the iolock or ilock in the
sync path these days.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Use xfs_inode_ag_iterator instead of opencoding the inode walk in the
quota code. Mark xfs_inode_ag_iterator and xfs_sync_inode_valid non-static
to allow using them from the quota code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Given that we walk across the per-ag inode lists so often, it makes sense to
introduce an iterator for this.
Convert the sync and reclaim code to use this new iterator, quota code will
follow in the next patch.
Also change xfs_reclaim_inode to return -EGAIN instead of 1 for an inode
already under reclaim. This simplifies the AG iterator and doesn't
matter for the only other caller.
[hch: merged the lookup and execute callbacks back into one to get the
pag_ici_lock locking correct and simplify the code flow]
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
The noblock parameter of xfs_reclaim_inodes is only ever set to zero. Remove
it and all the conditional code that is never executed.
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Separate the validation of inodes found by the radix
tree walk from the radix tree lookup.
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
In many cases we only want to sync inode metadata. Split out the inode
flushing into a separate helper to prepare factoring the inode sync code.
Based on a patch from Dave Chinner, but redone to keep the current behaviour
exactly and leave changes to the flushing logic to another patch.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
In many cases we only want to sync inode data. Start spliting the inode sync
into data sync and inode sync by factoring out the inode data flush.
[hch: minor cleanups]
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Kill the quota ops function vector and replace it with direct calls or
stubs in the CONFIG_XFS_QUOTA=n case.
Make sure we check XFS_IS_QUOTA_RUNNING in the right spots. We can remove
the number of those checks because the XFS_TRANS_DQ_DIRTY flag can't be set
otherwise.
This brings us back closer to the way this code worked in IRIX and earlier
Linux versions, but we keep a lot of the more useful factoring of common
code.
Eventually we should also kill xfs_qm_bhv.c, but that's left for a later
patch.
Reduces the size of the source code by about 250 lines and the size of
XFS module by about 1.5 kilobytes with quotas enabled:
text data bss dec hex filename
615957 2960 3848 622765 980ad fs/xfs/xfs.o
617231 3152 3848 624231 98667 fs/xfs/xfs.o.old
Fallout:
- xfs_qm_dqattach is split into xfs_qm_dqattach_locked which expects
the inode locked and xfs_qm_dqattach which does the locking around it,
thus removing XFS_QMOPT_ILOCKED.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Arkadiusz has seen really strange crashes in xfs_qm_dqcheck that
I can only explain by a log item being too smal to actually fit the
xfs_dqblk_t we're dereferencing all over xfs_qm_dqcheck. So add
graceful checks for NULL or too small quota items to the log recovery
code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Commit a6634fba3dec4a92f0a2c4e30c80b634c0576ad5 in xfsprogs increased the
maximum log size supported by mkfs. Merged back the changes to xfs_fs.h
so the growfs enforced the same limit and the headers are in sync.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Split up fat_generic_ioctl and add separate functions for the two
implemented ioctls.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
* Add ->set_capacity block device method and use it in rescan_partitions()
to attempt enabling native capacity of the device upon detecting the
partition which exceeds device capacity.
* Add GENHD_FL_NATIVE_CAPACITY flag to try limit attempts of enabling
native capacity during partition scan.
Together with the consecutive patch implementing ->set_capacity method in
ide-gd device driver this allows automatic disabling of Host Protected Area
(HPA) if any partitions overlapping HPA are detected.
Cc: Robert Hancock <hancockrwd@gmail.com>
Cc: Frans Pop <elendil@planet.nl>
Cc: "Andries E. Brouwer" <Andries.Brouwer@cwi.nl>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Emphatically-Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
The current warning message says only about the kernel's action taken
without mentioning the underlying reason behind it.
Noticed-by: Robert Hancock <hancockrwd@gmail.com>
Cc: Frans Pop <elendil@planet.nl>
Cc: "Andries E. Brouwer" <Andries.Brouwer@cwi.nl>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Emphatically-Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
CONFIG_IMA=y inode activity leaks iint_cache and radix_tree_node objects
until the system runs out of memory. Nowhere is calling ima_inode_free()
a.k.a. ima_iint_delete(). Fix that by calling it from destroy_inode().
Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have a bit of a problem with the uid= option. The basic issue is that
it means too many things and has too many side-effects.
It's possible to allow an unprivileged user to mount a filesystem if the
user owns the mountpoint, /bin/mount is setuid root, and the mount is
set up in /etc/fstab with the "user" option.
When doing this though, /bin/mount automatically adds the "uid=" and
"gid=" options to the share. This is fortunate since the correct uid=
option is needed in order to tell the upcall what user's credcache to
use when generating the SPNEGO blob.
On a mount without unix extensions this is fine -- you generally will
want the files to be owned by the "owner" of the mount. The problem
comes in on a mount with unix extensions. With those enabled, the
uid/gid options cause the ownership of files to be overriden even though
the server is sending along the ownership info.
This means that it's not possible to have a mount by an unprivileged
user that shows the server's file ownership info. The result is also
inode permissions that have no reflection at all on the server. You
simply cannot separate ownership from the mode in this fashion.
This behavior also makes MultiuserMount option less usable. Once you
pass in the uid= option for a mount, then you can't use unix ownership
info and allow someone to share the mount.
While I'm not thrilled with it, the only solution I can see is to stop
making uid=/gid= force the overriding of ownership on mounts, and to add
new mount options that turn this behavior on.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
OK, that's probably the easiest way to do that, as much as I don't like it...
Since iget() et.al. will not accept I_FREEING (will wait to go away
and restart), and since we'd better have serialization between new/free
on fs data structures anyway, we can afford simply skipping I_FREEING
et.al. in insert_inode_locked().
We do that from new_inode, so it won't race with free_inode in any interesting
ways and it won't race with iget (of any origin; nfsd or in case of fs
corruption a lookup) since both still will wait for I_LOCK.
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Acked-by: Jan Kara <jack@suse.cz>
Tested-by: David Watson <dbwatson@ukfsn.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The nobh_truncate_page() function is used by ext2, exofs, and jfs. Of
these three, only ext2 and jfs's get_block() function pays attention
to bh->b_size --- which is normally always the filesystem blocksize
except when the get_block() function is called by either
mpage_readpage(), mpage_readpages(), or the direct I/O routines in
fs/direct_io.c.
Unfortunately, nobh_truncate_page() does not initialize map_bh before
calling the filesystem-supplied get_block() function. So ext2 and jfs
will try to calculate the number of blocks to map by taking stack
garbage and shifting it left by inode->i_blkbits. This should be
*mostly* harmless (except the filesystem will do some unnneeded work)
unless the stack garbage is less than filesystem's blocksize, in which
case maxblocks will be zero, and the attempt to find out whether or
not the filesystem has a hole at a given logical block will fail, and
the page cache entry might not get zero'ed out.
Also if the stack garbage in in map_bh->state happens to have the
BH_Mapped bit set, there could be an attempt to call readpage() on a
non-existent page, which could cause nobh_truncate_page() to return an
error when it should not.
Fix this by initializing map_bh->state and map_bh->size.
Fortunately, it's probably fairly unlikely that ext2 and jfs users
mount with nobh these days.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: Fix oops and use after free during space balancing
Btrfs: set device->total_disk_bytes when adding new device
This patch uses sget() to get a reference to the
existing gfs2 sb when mouting the gfs2meta filesystem
(in fact thats just another mount of the gfs2
filesystem with a different root and this interface
is for backward compatibility).
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Reported-by: Benjamin Marzinski <bmarzins@redhat.com>
Tested-by: Benjamin Marzinski <bmarzins@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
In generic_perform_write if we fail to copy the user data we don't
update the inode->i_size. We should truncate the file in the above
case so that we don't have blocks allocated outside inode->i_size. Add
the inode to orphan list in the same transaction as block allocation
This ensures that if we crash in between the recovery would do the
truncate.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
We should add inode to the orphan list in the same transaction
as block allocation. This ensures that if we crash after a failed
block allocation and before we do a vmtruncate we don't leak block
(ie block marked as used in bitmap but not claimed by the inode).
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch changes ext4 super.c to include the device name with all
warning/error messages, by using a new utility function ext4_msg.
It's a rather large patch, but very mechanic. I left debug printks
alone.
This is a straightforward port of a patch which Andi Kleen did for
ext3.
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Get rid of EXTEND_DISKSIZE flag of ext4_get_blocks_handle(). This
seems to be a relict from some old days and setting disksize in this
function does not make much sense. Currently it was set only by
ext4_getblk(). Since the parameter has some effect only if create ==
1, it is easy to check by grepping through the sources that the three
callers which end up calling ext4_getblk() with create == 1
(ext4_append, ext4_quota_write, ext4_mkdir) do the right thing and set
disksize themselves.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This reverts commit db2dbb12dc.
It apparently causes problems with partition table read-ahead
on archs with large page sizes. Until that problem is diagnosed
further, just drop the readpages support on block devices.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
The btrfs allocator uses list_for_each to walk the available block
groups when searching for free blocks. It starts off with a hint
to help find the best block group for a given allocation.
The hint is resolved into a block group, but we don't properly check
to make sure the block group we find isn't in the middle of being
freed due to filesystem shrinking or balancing. If it is being
freed, the list pointers in it are bogus and can't be trusted. But,
the code happily goes along and uses them in the list_for_each loop,
leading to all kinds of fun.
The fix used here is to check to make sure the block group we find really
is on the list before we use it. list_del_init is used when removing
it from the list, so we can do a proper check.
The allocation clustering code has a similar bug where it will trust
the block group in the current free space cluster. If our allocation
flags have changed (going from single spindle dup to raid1 for example)
because the drives in the FS have changed, we're not allowed to use
the old block group any more.
The fix used here is to check the current cluster against the
current allocation flags.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
It would be nice to know how often we get checksum failures. Even
better, how many of them we can fix with the single bit ecc. So, we add
a statistics structure. The structure can be installed into debugfs
wherever the user wants.
For ocfs2, we'll put it in the superblock-specific debugfs directory and
pass it down from our higher-level functions. The stats are only
registered with debugfs when the filesystem supports metadata ecc.
Signed-off-by: Joel Becker <joel.becker@oracle.com>
When a dentry is unlinked, the unlinking node takes an EX on the dentry lock
before moving the dentry to the orphan directory. Other nodes that have
this dentry in cache have a PR on the same dentry lock. When the EX is
requested, the other nodes flag the corresponding inode as MAYBE_ORPHANED
during downconvert. The inode is finally deleted when the last node to iput
the inode sees that i_nlink==0 and the MAYBE_ORPHANED flag is set.
A problem arises if a node is forced to free dentry locks because of memory
pressure. If this happens, the node will no longer get downconvert
notifications for the dentries that have been unlinked on another node.
If it also happens that node is actively using the corresponding inode and
happens to be the one performing the last iput on that inode, it will fail
to delete the inode as it will not have the MAYBE_ORPHANED flag set.
This patch fixes this shortcoming by introducing a periodic scan of the
orphan directories to delete such inodes. Care has been taken to distribute
the workload across the cluster so that no one node has to perform the task
all the time.
Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
We use ordering ip_alloc_sem -> local alloc locks in ocfs2_write_begin().
So change lock ordering in ocfs2_extend_dir() and ocfs2_expand_inline_dir()
to also use this lock ordering.
Signed-off-by: Jan Kara <jack@suse.cz>
Acked-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
In ocfs2_finish_quota_recovery() we acquired global quota file lock and started
recovering local quota file. During this process we need to get quota
structures, which calls ocfs2_dquot_acquire() which gets global quota file lock
again. This second lock can block in case some other node has requested the
quota file lock in the mean time. Fix the problem by moving quota file locking
down into the function where it is really needed. Then dqget() or dqput()
won't be called with the lock held.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
We called vfs_dq_transfer() with global quota file lock held. This can lead
to deadlocks as if vfs_dq_transfer() has to allocate new quota structure,
it calls ocfs2_dquot_acquire() which tries to get quota file lock again and
this can block if another node requested the lock in the mean time.
Since we have to call vfs_dq_transfer() with transaction already started
and quota file lock ranks above the transaction start, we cannot just rely
on ocfs2_dquot_acquire() or ocfs2_dquot_release() on getting the lock
if they need it. We fix the problem by acquiring pointers to all quota
structures needed by vfs_dq_transfer() already before calling the function.
By this we are sure that all quota structures are properly allocated and
they can be freed only after we drop references to them. Thus we don't need
quota file lock anywhere inside vfs_dq_transfer().
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
