- 25 Sep, 2009 11 commits
-
-
Wu Fengguang authored
Debug traces show that in per-bdi writeback, the inode under writeback almost always get redirtied by a busy dirtier. We used to call redirty_tail() in this case, which could delay inode for up to 30s. This is unacceptable because it now happens so frequently for plain cp/dd, that the accumulated delays could make writeback of big files very slow. So let's distinguish between data redirty and metadata only redirty. The first one is caused by a busy dirtier, while the latter one could happen in XFS, NFS, etc. when they are doing delalloc or updating isize. The inode being busy dirtied will now be requeued for next io, while the inode being redirtied by fs will continue to be delayed to avoid repeated IO. CC: Jan Kara <jack@suse.cz> CC: Theodore Ts'o <tytso@mit.edu> CC: Dave Chinner <david@fromorbit.com> CC: Chris Mason <chris.mason@oracle.com> CC: Christoph Hellwig <hch@infradead.org> Signed-off-by:
Wu Fengguang <fengguang.wu@intel.com> Signed-off-by:
Jens Axboe <jens.axboe@oracle.com>
-
Jens Axboe authored
Currently we pin the inode->i_sb for every single inode. This increases cache traffic on sb->s_umount sem. Lets instead cache the inode sb pin state and keep the super_block pinned for as long as keep writing out inodes from the same super_block. Signed-off-by: -
Jens Axboe authored
If we only moved inodes from a single super_block to the temporary list, there's no point in doing a resort for multiple super_blocks. Signed-off-by: -
Shaohua Li authored
__mark_inode_dirty adds inode to wb dirty list in random order. If a disk has several partitions, writeback might keep spindle moving between partitions. To reduce the move, better write big chunk of one partition and then move to another. Inodes from one fs usually are in one partion, so idealy move indoes from one fs together should reduce spindle move. This patch tries to address this. Before per-bdi writeback is added, the behavior is write indoes from one fs first and then another, so the patch restores previous behavior. The loop in the patch is a bit ugly, should we add a dirty list for each superblock in bdi_writeback? Test in a two partition disk with attached fio script shows about 3% ~ 6% improvement. Signed-off-by:
Shaohua Li <shaohua.li@intel.com> Reviewed-by:
-
Jens Axboe authored
Signed-off-by: -
Jens Axboe authored
And throw some comments in there, too. Reviewed-by:
-
Wu Fengguang authored
Make the if-else straight in writeback_single_inode(). No behavior change. Cc: Jan Kara <jack@suse.cz> Cc: Michael Rubin <mrubin@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by:
Fengguang Wu <wfg@mail.ustc.edu.cn> Signed-off-by:
-
Wu Fengguang authored
Fix the kupdate case, which disregards wbc.more_io and stop writeback prematurely even when there are more inodes to be synced. wbc.more_io should always be respected. Also remove the pages_skipped check. It will set when some page(s) of some inode(s) cannot be written for now. Such inodes will be delayed for a while. This variable has nothing to do with whether there are other writeable inodes. CC: Jan Kara <jack@suse.cz> CC: Dave Chinner <david@fromorbit.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
-
Wu Fengguang authored
Treat bdi_start_writeback(0) as a special request to do background write, and stop such work when we are below the background dirty threshold. Also simplify the (nr_pages <= 0) checks. Since we already pass in nr_pages=LONG_MAX for WB_SYNC_ALL and background writes, we don't need to worry about it being decreased to zero. Reported-by:
Richard Kennedy <richard@rsk.demon.co.uk> CC: Jan Kara <jack@suse.cz> Acked-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
-
Wu Fengguang authored
Some filesystem may choose to write much more than ratelimit_pages before calling balance_dirty_pages_ratelimited_nr(). So it is safer to determine number to write based on real number of dirtied pages. Otherwise it is possible that loop { btrfs_file_write(): dirty 1024 pages balance_dirty_pages(): write up to 48 pages (= ratelimit_pages * 1.5) } in which the writeback rate cannot keep up with dirty rate, and the dirty pages go all the way beyond dirty_thresh. The increased write_chunk may make the dirtier more bumpy. So filesystems shall be take care not to dirty too much at a time (eg. > 4MB) without checking the ratelimit. Signed-off-by: -
Jan Kara authored
If all inodes are under writeback (e.g. in case when there's only one inode with dirty pages), wb_writeback() with WB_SYNC_NONE work basically degrades to busylooping until I_SYNC flags of the inode is cleared. Fix the problem by waiting on I_SYNC flags of an inode on b_more_io list in case we failed to write anything. Tested-by:
Jan Kara <jack@suse.cz> Signed-off-by:
-
- 22 Sep, 2009 29 commits
-
-
Linus Torvalds authored
Merge branch 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: perf_event, powerpc: Fix compilation after big perf_counter rename
-
git://git.kernel.org/pub/scm/linux/kernel/git/vegard/kmemcheckLinus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/vegard/kmemcheck: kmemcheck: add missing braces to do-while in kmemcheck_annotate_bitfield kmemcheck: update documentation kmemcheck: depend on HAVE_ARCH_KMEMCHECK kmemcheck: remove useless check kmemcheck: remove duplicated #include
-
git://linux-nfs.org/~bfields/linuxLinus Torvalds authored
* 'for-2.6.32' of git://linux-nfs.org/~bfields/linux: (68 commits) nfsd4: nfsv4 clients should cross mountpoints nfsd: revise 4.1 status documentation sunrpc/cache: avoid variable over-loading in cache_defer_req sunrpc/cache: use list_del_init for the list_head entries in cache_deferred_req nfsd: return success for non-NFS4 nfs4_state_start nfsd41: Refactor create_client() nfsd41: modify nfsd4.1 backchannel to use new xprt class nfsd41: Backchannel: Implement cb_recall over NFSv4.1 nfsd41: Backchannel: cb_sequence callback nfsd41: Backchannel: Setup sequence information nfsd41: Backchannel: Server backchannel RPC wait queue nfsd41: Backchannel: Add sequence arguments to callback RPC arguments nfsd41: Backchannel: callback infrastructure nfsd4: use common rpc_cred for all callbacks nfsd4: allow nfs4 state startup to fail SUNRPC: Defer the auth_gss upcall when the RPC call is asynchronous nfsd4: fix null dereference creating nfsv4 callback client nfsd4: fix whitespace in NFSPROC4_CLNT_CB_NULL definition nfsd41: sunrpc: add new xprt class for nfsv4.1 backchannel sunrpc/cache: simplify cache_fresh_locked and cache_fresh_unlocked. ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivialLinus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (34 commits) trivial: fix typo in aic7xxx comment trivial: fix comment typo in drivers/ata/pata_hpt37x.c trivial: typo in kernel-parameters.txt trivial: fix typo in tracing documentation trivial: add __init/__exit macros in drivers/gpio/bt8xxgpio.c trivial: add __init macro/ fix of __exit macro location in ipmi_poweroff.c trivial: remove unnecessary semicolons trivial: Fix duplicated word "options" in comment trivial: kbuild: remove extraneous blank line after declaration of usage() trivial: improve help text for mm debug config options trivial: doc: hpfall: accept disk device to unload as argument trivial: doc: hpfall: reduce risk that hpfall can do harm trivial: SubmittingPatches: Fix reference to renumbered step trivial: fix typos "man[ae]g?ment" -> "management" trivial: media/video/cx88: add __init/__exit macros to cx88 drivers trivial: fix typo in CONFIG_DEBUG_FS in gcov doc trivial: fix missing printk space in amd_k7_smp_check trivial: fix typo s/ketymap/keymap/ in comment trivial: fix typo "to to" in multiple files trivial: fix typos in comments s/DGBU/DBGU/ ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hidLinus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: HID: Remove duplicate Kconfig entry HID: consolidate connect and disconnect into core code HID: fix non-atomic allocation in hid_input_report
-
David Härdeman authored
Add a driver for the the Consumer IR (CIR) functionality of the Winbond WPCD376I chipset (found on e.g. Intel DG45FC motherboards). Signed-off-by:
David Härdeman <david@hardeman.nu> Reviewed-by:
Jesse Barnes <jbarnes@virtuousgeek.org> Cc: Dmitry Torokhov <dtor@mail.ru> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
David Härdeman authored
The shutdown method is used by the winbond cir driver to setup the hardware for wake-from-S5. Signed-off-by:
Bjorn Helgaas <bjorn.helgaas@hp.com> Signed-off-by:
-
Henrik Rydberg authored
On resume from suspend, the driver currently resets the logical state as if it was brought up from halt. This patch uses the dev_pm_ops.resume/restore methods to synchronize the hardware with the memorized logical state, in effect bringing back the accelerometer and backlight to the state prior to suspend. Works for both suspend to ram and hibernation. The patch has zero effect on the running state. Signed-off-by:
Henrik Rydberg <rydberg@euromail.se> Cc: Nicolas Boichat <nicolas@boichat.ch> Signed-off-by:
-
Roel Kluin authored
If already requested, gpio_data and irq should be freed in the case of an error. Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Acked-by:
Jonathan Cameron <jic23@cam.ac.uk> Cc: David Brownell <david-b@pacbell.net> Signed-off-by:
-
Michael Abbott authored
Occasionally it is helpful to be able to turn a temperature sensor off (for example if it's making unwanted electrical noise). This patch adds a sysfs node to put any adm1021 compatible device into low power mode. Signed-off-by:
Michael Abbott <michael.abbott@diamond.ac.uk> Cc: Jean Delvare <khali@linux-fr.org> Signed-off-by:
-
Michael Abbott authored
The ADM1023 temperature sensor supports higher resolution for its external sensor (sensitivity of 1/8 deg C). This patch makes this higher resolution available through the appropriate temperature sysfs nodes. Curiously, this functionality was available in the 2.4 kernel driver (but formatted in a less helpful manner). Cc: Jean Delvare <khali@linux-fr.org> Signed-off-by:
-
Daniel Mack authored
Signed-off-by:
Daniel Mack <daniel@caiaq.de> Acked-by:
Pavel Machek <pavel@ucw.cz> Cc: Eric Piel <eric.piel@tremplin-utc.net> Signed-off-by:
-
Daniel Mack authored
This enabled power management functions for the SPI transport layer of the lis3 devices. The device's suspend mode is only entered in case no wakeup threshold has been given. In this case, the device is supposed to wake up the system and must thus not be put to deep sleep. [randy.dunlap@oracle.com: fix lis3-spi for CONFIG_PM=n] Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
Daniel Mack authored
This offers a way for platforms to define flags and thresholds for the free-fall/wakeup functions of the lis302d chips. More registers needed to be seperated as they are specific to the Signed-off-by:
-
Daniel Mack authored
Bit 0x80 in CTRL_REG3 is an ACTIVE_LOW rather than an ACTIVE_HIGH function, I got that wrong during my last change. Signed-off-by:
-
Michael Riepe authored
Enable the coretemp driver on an Intel Atom. I'm not sure if the readings are correct, however - on my 330, the driver reports values between 27 and 41 °C (with core1 being about 8°C hotter than core0, given the same load). Maybe the maximum temperature of 100 °C is wrong for Atom CPUs. Cc: Arjan van de Ven <arjan@infradead.org> Cc: Rudolf Marek <r.marek@assembler.cz> Signed-off-by:
-
Mike Frysinger authored
Add checks for Blackfin-specific issues that seem to crop up from time to time. In particular, we have helper macros to break a 32bit address into the hi/lo parts, and we want to make sure people use the csync/ssync variant that includes fun anomaly workarounds. Signed-off-by:
Mike Frysinger <vapier@gentoo.org> Signed-off-by:
Bryan Wu <cooloney@kernel.org> Cc: Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Andy Whitcroft authored
Signed-off-by:
Andy Whitcroft <apw@canonical.com> Signed-off-by:
-
Andy Whitcroft authored
Limit our type matcher to the s/u/le/be etc sizes that actually exist to prevent miss categorising s2 as a type. Fix up the spelling of the error also. Signed-off-by:
-
Andy Whitcroft authored
We should not recommend braces for the following: #define pr_fmt(fmt) "%s: " fmt, __func__ allow things with double quotes round them to avoid this check. Signed-off-by: -
Hannes Eder authored
Impact: - More verbose help/usage message. - Make the option -f an alias for --file. - On -h, --help, and --version display help message and exit(0). - With no FILE(s) given, exit(1) with "no input files". - On invalid options display help/usage and exit(1). Based on a patch by Pavel Machek. Signed-off-by:
Hannes Eder <hannes@hanneseder.net> Acked-by:
Pavel Machek <pavel@suse.cz> Signed-off-by:
-
Andy Whitcroft authored
Ensure we terminate when there are no futher continuation lines when trying to determine relative indent of conditionals and their blocks. Reported-by:
John Daiker <daikerjohn@gmail.com> Signed-off-by:
-
Daniel Walker authored
This fixes the sanitation process in checkpatch.pl so that it blocks out the text after a C99 style comment the same way it does with block style comments. This prevents the text from getting processed as regular code. Signed-off-by:
Daniel Walker <dwalker@fifo99.com> Signed-off-by:
-
Andy Whitcroft authored
An else cannot start a type, it would have to be within a block after the else. This can trigger false modifier matching. Signed-off-by:
-
David Rientjes authored
Add kerneldoc annotations for function formals of type struct flex_array and gfp_t which are currently lacking. Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Dave Hansen <dave@linux.vnet.ibm.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
David Rientjes authored
FLEX_ARRAY_INIT(element_size, total_nr_elements) cannot determine if either parameter is valid, so flex arrays which are statically allocated with this interface can easily become corrupted or reference beyond its allocated memory. This removes FLEX_ARRAY_INIT() as a struct flex_array initializer since no initializer may perform the required checking. Instead, the array is now defined with a new interface: DEFINE_FLEX_ARRAY(name, element_size, total_nr_elements) This may be prefixed with `static' for file scope. This interface includes compile-time checking of the parameters to ensure they are valid. Since the validity of both element_size and total_nr_elements depend on FLEX_ARRAY_BASE_SIZE and FLEX_ARRAY_PART_SIZE, the kernel build will fail if either of these predefined values changes such that the array parameters are no longer valid. Since BUILD_BUG_ON() requires compile time constants, several of the static inline functions that were once local to lib/flex_array.c had to be moved to include/linux/flex_array.h. Signed-off-by:
Dave Hansen <dave@linux.vnet.ibm.com> Signed-off-by:
-
David Rientjes authored
Add a new function to the flex_array API: int flex_array_shrink(struct flex_array *fa) This function will free all unused second-level pages. Since elements are now poisoned if they are not allocated with __GFP_ZERO, it's possible to identify parts that consist solely of unused elements. flex_array_shrink() returns the number of pages freed. Signed-off-by:
-
David Rientjes authored
Newly initialized flex_array's and/or flex_array_part's are now poisoned with a new poison value, FLEX_ARRAY_FREE. It's value is similar to POISON_FREE used in the various slab allocators, but is different to distinguish between flex array's poisoned kmem and slab allocator poisoned kmem. This will allow us to identify flex_array_part's that only contain free elements (and free them with an addition to the flex_array API). This could also be extended in the future to identify `get' uses on elements that have not been `put'. If __GFP_ZERO is passed for a part's gfp mask, the poisoning is avoided. These elements are considered to be in-use since they have been initialized. Signed-off-by:
-
David Rientjes authored
Add a new function to the flex_array API: int flex_array_clear(struct flex_array *fa, unsigned int element_nr) This function will zero the element at element_nr in the flex_array. Although this is equivalent to using flex_array_put() and passing a pointer to zero'd memory, flex_array_clear() does not require such a pointer to memory that would most likely need to be allocated on the caller's stack which could be significantly large depending on element_size. Signed-off-by:
-
