Samuel Ortiz [Thu, 18 Dec 2008 10:38:02 +0000 (11:38 +0100)]
mfd: Use irq_to_desc in twl4030 code
The global irq_desc array is soon going to be accessible only with
!CONFIG_SPARSE_IRQ. We should start using the generic irq_to_desc()
routines instead.
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Driver for battery charger integrated into Dialog Semiconductor DA9030 PMIC
Signed-off-by: Mike Rapoport <mike@compulab.co.il> Acked-by: Anton Vorontsov <cbouatmailru@gmail.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
David Brownell [Sun, 7 Dec 2008 18:10:58 +0000 (19:10 +0100)]
mfd: twl4030 regulator bug fixes
This contains two bugfixes to the initial twl4030 regulator
support patch related to USB:
(a) always overwrite the old list of consumers ... else
the regulator handles all use the same "usb1v5" name;
(b) don't set up the "usbcp" regulator, which turns out
to be managed through separate controls, usually ULPI
directly from the OTG controller.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
David Brownell [Sun, 30 Nov 2008 23:43:58 +0000 (00:43 +0100)]
mfd: twl4030: create some regulator devices
Initial code to create twl4030 voltage regulator devices, using
the new regulator framework. Note that this now starts to care
what name is used to declare the TWL chip:
- TWL4030 is the "old" chip; newer ones have a bigger variety
of VAUX2 voltages.
- TWL5030 is the core "new" chip; TPS65950 is its catalog version.
- The TPS65930 and TPS65920 are cost-reduced catalog versions of
TWL5030 parts ... fewer regulators, no battery charger, etc.
Board-specific regulator configuration should be provided, listing
which regulators are used and their constraints (e.g. 1.8V only).
Code that could ("should"?) leverage the regulator stuff includes
TWL4030 USB transceiver support and MMC glue, LCD support for the
3430SDP and Labrador boards, and S-Video output.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
David Brownell [Sun, 30 Nov 2008 23:35:33 +0000 (00:35 +0100)]
mfd: twl4030: cleanup symbols and OMAP dependency
Finish removing dependency of TWL driver stack on platform-specific
IRQ definitions ... and remove the build dependency on OMAP.
This lets the TWL4030 code be included in test builds for most
platforms, and will make it easier for non-OMAP folk to update
most of this code for new APIs etc.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
David Brownell [Sun, 30 Nov 2008 23:31:04 +0000 (00:31 +0100)]
mfd: twl4030: simplified child creation code
Minor cleanup to twl4030-core: define a helper function to populate
a single child node, and use it to replace six inconsistent versions
of the same logic. Both object and source code shrink.
As part of this, some devices now have more IRQ resources: battery
charger, keypad, ADC, and USB transceiver. That helps to remove some
irq #defines that block the children's drivers code from compiling on
non-OMAP platforms.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Mark Brown [Sun, 30 Nov 2008 21:45:14 +0000 (22:45 +0100)]
power_supply: Add battery health reporting for WM8350
Implement support for reporting battery health in the WM8350 battery
interface. Since we are now able to report this via the classs remove
the diagnostics from the interrupt handler.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Anton Vorontsov <cbouatmailru@gmail.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Mark Brown [Sun, 30 Nov 2008 21:43:21 +0000 (22:43 +0100)]
power_supply: Add cold to the POWER_SUPPLY_HEALTH report values
Some systems are able to report problems with batteries being under
temperature.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Anton Vorontsov <cbouatmailru@gmail.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Mark Brown [Mon, 24 Nov 2008 19:22:58 +0000 (20:22 +0100)]
mfd: Refactor WM8350 chip identification
Since the WM8350 driver was originally written the semantics for the
identification registers of the chip have been clarified, allowing
us to do an exact match on all the fields. This avoids mistakenly
running on unsupported hardware.
Also change to using the datasheet names more consistently for
legibility and fix a printk() that should be dev_err().
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Mark Brown [Sun, 16 Nov 2008 19:16:47 +0000 (20:16 +0100)]
power_supply: Add support for WM8350 PMU
This patch adds support for the PMU provided by the WM8350 which
implements battery, line and USB supplies including a battery charger.
The hardware functions largely autonomously, with minimal software
control required to initiate fast charging.
Support for configuration of the USB supply is not yet implemented.
This means that the hardware will remain in the mode configured at
startup, by default limiting the current drawn from USB to 100mA.
This driver was originally written by Liam Girdwood with subsequent
updates for submission by Mark Brown.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Anton Vorontsov <cbouatmailru@gmail.com> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
David Brownell [Sat, 8 Nov 2008 00:13:16 +0000 (01:13 +0100)]
mfd: allow reading entire register banks on twl4030
Minor change to the TWL4030 utility interface: support reads
of all 256 bytes in each register bank (vs just 255). This
can help when debugging, but is otherwise a NOP.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Julia Lawall [Sat, 3 Jan 2009 16:54:53 +0000 (17:54 +0100)]
ALSA: sound/usb: Use negated usb_endpoint_xfer_control, etc
This patch extends 42a6e66f1e40a930d093c33ba0bb9d8d8e4555ed by using
usb_endpoint_xfer_control, usb_endpoint_xfer_isoc, usb_endpoint_xfer_bulk,
and usb_endpoint_xfer_int in the negated case as well.
This patch also rewrites some calls to usb_endpoint_dir_in as negated calls
to !usb_endpoint_dir_out, and vice versa, to better correspond to the
intent of the original code.
The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)
Ingo Molnar [Sun, 4 Jan 2009 09:55:02 +0000 (10:55 +0100)]
intel-iommu: fix build error with INTR_REMAP=y and DMAR=n
dmar.o can be built in the CONFIG_INTR_REMAP=y case but
iommu_calculate_agaw() is only available if VT-d is built as well.
So create an inline version of iommu_calculate_agaw() for the
!CONFIG_DMAR case. The iommu->agaw value wont be used in this
case, but the code is cleaner (has less #ifdefs) if we have it around
unconditionally.
Theodore Ts'o [Tue, 6 Jan 2009 16:14:25 +0000 (11:14 -0500)]
ext4: Remove code to create the journal inode
This code has been obsolete in quite some time, since the supported
method for adding a journal inode is to use tune2fs (or to creating
new filesystem with a journal via mke2fs or mkfs.ext4).
ext4: provide function to release metadata pages under memory pressure
Pages in the page cache belonging to ext4 data files are released via
the ext4_releasepage() function specified in the ext4 inode's
address_space_ops. However, metadata blocks (such as indirect blocks,
directory blocks, etc) are managed via the block device
address_space_ops, and they can not be released by
try_to_free_buffers() if they have a journal head attached to them.
To address this, we supply a release_metadata function which calls
jbd2_journal_try_to_free_buffers() function to free the metadata, and
which is called by the block device's blkdev_releasepage() function.
ext3: provide function to release metadata pages under memory pressure
Pages in the page cache belonging to ext3 data files are released via
the ext3_releasepage() function specified in the ext3 inode's
address_space_ops. However, metadata blocks (such as indirect blocks,
directory blocks, etc) are managed via the block device
address_space_ops, and they can not be released by
try_to_free_buffers() if they have a journal head attached to them.
To address this, we supply a try_to_free_pages() function which calls
journal_try_to_free_buffers() function to free the metadata, and which
is called by the block device's blkdev_releasepage() function.
Oliver Neukum [Mon, 15 Dec 2008 12:12:08 +0000 (13:12 +0100)]
HID: fix reference count leak hidraw
The hidraw subsystem has a bug that prevents the close syscall from ever
reaching the low level driver, leading to a resource leak. Fix by replacing
postdecrement with predecrement.
Signed-off-by: Oliver Neukum <oneukum@suse.de> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Matt Helsley [Sat, 13 Dec 2008 13:28:54 +0000 (14:28 +0100)]
HID: add proper support for pensketch 12x9 tablet
The Genius PenSketch 12x9 tablet has a puck (labeled a
"Tablet Mouse") in addition to a pen. Without registering a quirk
the tablet appears to be a single input device that reports the
wrong axis information in /proc/bus/input/devices, and sends
incorrect events (e.g. ABS_Z instead of ABS_Y). This information
confuses the X evdev driver and makes the device impossible to
use.
The quirk fixes events and splits the device into multiple input
event devices so that at least the puck is useful.
Signed-off-by: Matt Helsley <matt.helsley@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Alexey Klimov [Mon, 8 Dec 2008 11:40:14 +0000 (12:40 +0100)]
HID: don't allow DealExtreme usb-radio be handled by usb hid driver
This device is already handled by radio-si470x driver, and we therefore
want usbhid to ignore it. Patch places usb ids of that device in
ignore section of hid-core.c
Lev Babiev [Sat, 3 Jan 2009 23:36:56 +0000 (00:36 +0100)]
HID: driver for TopSeed Cyberlink quirky remote
I recently picked up a Cyberlink branded remote control produced
by TopSeed Tech Corp. Alas, it appears that this device is using
non-standard mappings for some of it's keys (Usage page 0xffbc).
Signed-off-by: Lev Babiev <harley@hosers.org> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Parag Warudkar [Mon, 22 Dec 2008 21:50:52 +0000 (22:50 +0100)]
HID: make boot protocol drivers depend on EMBEDDED
The usbmouse and usbkbd modules are not supposed to be used with regular USB
mice and keyboards. Make them depend on EMBEDDED to prevent them from being
built and loaded on non-EMBEDDED configs.
Oliver Neukum [Tue, 16 Dec 2008 09:55:15 +0000 (10:55 +0100)]
HID: hiddev cleanup -- handle all error conditions properly
This is a cleanup of hiddev and fixes the following issues:
- thread safety by locking in read & ioctl, introducing a per device mutex
- race between ioctl and disconnect, introducing a flag and locking
in form of a per low level device mutex
- race between open and other methods, making sure only successfully
opened devices are put on the list, changing order of events
- range checking both upper and lower limits of the minor range
- make sure further calls to open fail for unplugged devices even if
the device still has opened files
- error checking for low level open
- possible loss of wakeup events, using standard waiting macros
- race in initialisation by moving registration after full initialisation
Signed-off-by: Oliver Neukum <oneukum@suse.de> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
I have implemented Force Feedback driver for another "GreeAsia" based device
(0e8f:0012 "GreenAsia Inc. USB Joystick"). The functionality was tested with
MANTA Warior MM816 and SpeedLink Strike2 SL-6635 and fftest software -
everything seems to work right.
David Brownell [Thu, 11 Dec 2008 13:54:07 +0000 (14:54 +0100)]
HID: switch specialized drivers from "default y" to !EMBEDDED
Fix the obnoxious "default y" for all the "special" HID code, which forces folk
with EMBEDDED defined to manually override that inappropriate default for
almost 20 choices. The general policy is against "default y"; it should apply
here too.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Jiri Kosina [Thu, 4 Dec 2008 15:16:46 +0000 (16:16 +0100)]
HID: set proper dev.parent in hidraw
We need to properly set parent of the hidraw device (which is the
corresponding physical device itself) in order to hidraw devices not
end up under virtual device tree.
Reported-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Kay Sievers [Thu, 30 Oct 2008 23:12:32 +0000 (00:12 +0100)]
HID: struct device - replace bus_id with dev_name(), dev_set_name()
This patch is part of a larger patch series which will remove
the "char bus_id[20]" name string from struct device. The device
name is managed in the kobject anyway, and without any size
limitation, and just needlessly copied into "struct device".
To set and read the device name dev_name(dev) and dev_set_name(dev)
must be used. If your code uses static kobjects, which it shouldn't
do, "const char *init_name" can be used to statically provide the
name the registered device should have. At registration time, the
init_name field is cleared, to enforce the use of dev_name(dev) to
access the device name at a later time.
We need to get rid of all occurrences of bus_id in the entire tree
to be able to enable the new interface. Please apply this patch,
and possibly convert any remaining remaining occurrences of bus_id.
We want to submit a patch to -next, which will remove bus_id from
"struct device", to find the remaining pieces to convert, and finally
switch over to the new api, which will remove the 20 bytes array
and does no longer have a size limitation.
Alan Stern [Thu, 30 Oct 2008 22:58:51 +0000 (23:58 +0100)]
HID: automatically call usbhid_set_leds in usbhid driver
This patch (as1146c) makes usbhid automatically call usbhid_set_leds()
for any device that supports the keyboard boot protocol.
In theory this should be perfectly safe. BIOSes send the LED output
report as part of their normal device initialization, so any keyboard
device supporting the boot protocol has to be able to handle it.
As a side effect, the hid-dell and hid-bright drivers are no longer
needed, and the Logitech keyboard driver can be removed from hid-lg.
Linus Torvalds [Sat, 3 Jan 2009 20:04:39 +0000 (12:04 -0800)]
Merge branch 'cpus4096-for-linus-3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'cpus4096-for-linus-3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (77 commits)
x86: setup_per_cpu_areas() cleanup
cpumask: fix compile error when CONFIG_NR_CPUS is not defined
cpumask: use alloc_cpumask_var_node where appropriate
cpumask: convert shared_cpu_map in acpi_processor* structs to cpumask_var_t
x86: use cpumask_var_t in acpi/boot.c
x86: cleanup some remaining usages of NR_CPUS where s/b nr_cpu_ids
sched: put back some stack hog changes that were undone in kernel/sched.c
x86: enable cpus display of kernel_max and offlined cpus
ia64: cpumask fix for is_affinity_mask_valid()
cpumask: convert RCU implementations, fix
xtensa: define __fls
mn10300: define __fls
m32r: define __fls
h8300: define __fls
frv: define __fls
cris: define __fls
cpumask: CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
cpumask: zero extra bits in alloc_cpumask_var_node
cpumask: replace for_each_cpu_mask_nr with for_each_cpu in kernel/time/
cpumask: convert mm/
...
Linus Torvalds [Sat, 3 Jan 2009 20:02:18 +0000 (12:02 -0800)]
Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6: (34 commits)
V4L/DVB (10173): Missing v4l2_prio_close in radio_release
V4L/DVB (10172): add DVB_DEVICE_TYPE= to uevent
V4L/DVB (10171): Use usb_set_intfdata
V4L/DVB (10170): tuner-simple: prevent possible OOPS caused by divide by zero error
V4L/DVB (10168): sms1xxx: fix inverted gpio for lna control on tiger r2
V4L/DVB (10167): sms1xxx: add support for inverted gpio
V4L/DVB (10166): dvb frontend: stop using non-C99 compliant comments
V4L/DVB (10165): Add FE_CAN_2G_MODULATION flag to frontends that support DVB-S2
V4L/DVB (10164): Add missing S2 caps flag to S2API
V4L/DVB (10163): em28xx: allocate adev together with struct em28xx dev
V4L/DVB (10162): tuner-simple: Fix tuner type set message
V4L/DVB (10161): saa7134: fix autodetection for AVer TV GO 007 FM Plus
V4L/DVB (10160): em28xx: update chip id for em2710
V4L/DVB (10157): Add USB ID for the Sil4701 radio from DealExtreme
V4L/DVB (10156): saa7134: Add support for Avermedia AVer TV GO 007 FM Plus
V4L/DVB (10155): Add TEA5764 radio driver
V4L/DVB (10154): saa7134: fix a merge conflict on Behold H6 board
V4L/DVB (10153): Add the Beholder H6 card to DVB-T part of sources.
V4L/DVB (10152): Change configuration of the Beholder H6 card
V4L/DVB (10151): Fix I2C bridge error in zl10353
...
Linus Torvalds [Sat, 3 Jan 2009 20:00:07 +0000 (12:00 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc:
mmc: warn about voltage mismatches
mmc_spi: Add support for OpenFirmware bindings
pxamci: fix dma_unmap_sg length
mmc_block: ensure all sectors that do not have errors are read
drivers/mmc: Move a dereference below a NULL test
sdhci: handle built-in sdhci with modular leds class
mmc: balanc pci_iomap with pci_iounmap
mmc_block: print better error messages
mmc: Add mmc_vddrange_to_ocrmask() helper function
ricoh_mmc: Handle newer models of Ricoh controllers
mmc: Add 8-bit bus width support
sdhci: activate led support also when module
mmc: trivial annotation of 'blocks'
pci: use pci_ioremap_bar() in drivers/mmc
sdricoh_cs: Add support for Bay Controller devices
mmc: at91_mci: reorder timer setup and mmc_add_host() call
Linus Torvalds [Sat, 3 Jan 2009 19:46:17 +0000 (11:46 -0800)]
Make %p print '(null)' for NULL pointers
Before, when we only ever printed out the pointer value itself, a NULL
pointer would never cause issues and might as well be printed out as
just its numeric value.
However, with the extended %p formats, especially %pR, we might validly
want to print out resources for debugging. And sometimes they don't
even exist, and the resource pointer is just NULL. Print it out as
such, rather than oopsing.
This is a more generic version of a patch done by Trent Piepho (catching
all %p cases rather than just %pR, and using "(null)" instead of
"[NULL]" to match glibc).
Al Viro [Sat, 3 Jan 2009 07:16:23 +0000 (07:16 +0000)]
sanitize ifdefs in binfmt_aout
They are actually alpha vs. i386/arm/m68k i.e. ecoff vs. aout.
In the only place where we actually tried to handle arm and i386/m68k in
different ways (START_DATA() in coredump handling), the arm variant
works for all of them (i386 and m68k have u.start_code set to 0).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mike Travis [Thu, 1 Jan 2009 02:08:47 +0000 (18:08 -0800)]
cpumask: use alloc_cpumask_var_node where appropriate
Impact: Reduce inter-node memory traffic.
Reduces inter-node memory traffic (offloading the global system bus)
by allocating referenced struct cpumasks on the same node as the
referring struct.
Signed-off-by: Mike Travis <travis@sgi.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rusty Russell [Thu, 1 Jan 2009 02:08:47 +0000 (18:08 -0800)]
cpumask: convert shared_cpu_map in acpi_processor* structs to cpumask_var_t
Impact: Reduce memory usage, use new API.
This is part of an effort to reduce structure sizes for machines
configured with large NR_CPUS. cpumask_t gets replaced by
cpumask_var_t, which is either struct cpumask[1] (small NR_CPUS) or
struct cpumask * (large NR_CPUS).
(Changes to powernow-k* by <travis>.)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Mike Travis <travis@sgi.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Mike Travis [Thu, 1 Jan 2009 02:08:45 +0000 (18:08 -0800)]
sched: put back some stack hog changes that were undone in kernel/sched.c
Impact: prevents panic from stack overflow on numa-capable machines.
Some of the "removal of stack hogs" changes in kernel/sched.c by using
node_to_cpumask_ptr were undone by the early cpumask API updates, and
causes a panic due to stack overflow. This patch undoes those changes
by using cpumask_of_node() which returns a 'const struct cpumask *'.
In addition, cpu_coregoup_map is replaced with cpu_coregroup_mask further
reducing stack usage. (Both of these updates removed 9 FIXME's!)
Also:
Pick up some remaining changes from the old 'cpumask_t' functions to
the new 'struct cpumask *' functions.
Optimize memory traffic by allocating each percpu local_cpu_mask on the
same node as the referring cpu.
Signed-off-by: Mike Travis <travis@sgi.com> Acked-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ingo Molnar [Sat, 3 Jan 2009 11:50:46 +0000 (12:50 +0100)]
ia64: cpumask fix for is_affinity_mask_valid()
Impact: build fix on ia64
ia64's default_affinity_write() still had old cpumask_t usage:
/home/mingo/tip/kernel/irq/proc.c: In function `default_affinity_write':
/home/mingo/tip/kernel/irq/proc.c:114: error: incompatible type for argument 1 of `is_affinity_mask_valid'
make[3]: *** [kernel/irq/proc.o] Error 1
make[3]: *** Waiting for unfinished jobs....
Ingo Molnar [Sat, 3 Jan 2009 12:16:09 +0000 (13:16 +0100)]
cpumask: convert RCU implementations, fix
Impact: cleanup
This warning:
kernel/rcuclassic.c: In function ‘rcu_start_batch’:
kernel/rcuclassic.c:397: warning: passing argument 1 of ‘cpumask_andnot’ from incompatible pointer type
triggers because one usage site of rcp->cpumask was not converted
to to_cpumask(rcp->cpumask). There's no ill effects of this bug.
ext4: Fix s_dirty_blocks_counter if block allocation failed with nodelalloc
With nodelalloc option we need to update the dirty block counter on
block allocation failure. This is needed because we increment the
dirty block counter early in the block allocation phase. Without
the patch s_dirty_blocks_counter goes wrong so that filesystem's
free blocks decreases incorrectly.
ext4: Init the complete page while building buddy cache
We need to init the complete page during buddy cache init
by setting the contents to '1'. Otherwise we can see the
following errors after doing an online resize of the
filesystem:
EXT4-fs error (device sdb1): ext4_mb_mark_diskspace_used:
Allocating block 1040385 in system zone of 127 group
ext4: Don't allow new groups to be added during block allocation
After we mark the blocks in the buddy cache as allocated,
we need to ensure that we don't reinit the buddy cache until
the block bitmap is updated. This commit achieves this by holding
the group_info alloc_semaphore till ext4_mb_release_context
ext4: Use new buffer_head flag to check uninit group bitmaps initialization
For uninit block group, the on-disk bitmap is not initialized. That
implies we cannot depend on the uptodate flag on the bitmap
buffer_head to find bitmap validity. Use a new buffer_head flag which
would be set after we properly initialize the bitmap. This also
prevents (re-)initializing the uninit group bitmap every time we call
ext4_read_block_bitmap().
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode()
We need to make sure we update the inode bitmap and clear
EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since
ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide
whether to initialize the inode bitmap each time it is called.
(introduced by commit c806e68f.)
and ext4_new_inode does
if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group),
ino, inode_bitmap_bh->b_data))
......
...
spin_lock(sb_bgl_lock(sbi, group));
gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT);
i.e., on allocation we update the bitmap then we take the sb_bgl_lock
and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a
parallel ext4_read_inode_bitmap can zero out the bitmap in between
the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..)
The race results in below user visible errors
EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449
EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ...
EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ...
# ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71
ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle
ext4: Fix race between read_block_bitmap() and mark_diskspace_used()
We need to make sure we update the block bitmap and clear
EXT4_BG_BLOCK_UNINIT flag with sb_bgl_lock held, since
ext4_read_block_bitmap() looks at EXT4_BG_BLOCK_UNINIT to decide
whether to initialize the block bitmap each time it is called
(introduced by commit c806e68f), and this can race with block
allocations in ext4_mb_mark_diskspace_used().
ie on allocation we update the bitmap then we take the sb_bgl_lock
and clear the EXT4_BG_BLOCK_UNINIT flag. What can happen is a
parallel ext4_read_block_bitmap can zero out the bitmap in between
the above mb_set_bits and spin_lock(sb_bg_lock..)
The race results in below user visible errors
EXT4-fs error (device sdb1): ext4_mb_release_inode_pa: free 100, pa_free 105
EXT4-fs error (device sdb1): mb_free_blocks: double-free of inode 0's block ..
ext4: fix BUG when calling ext4_error with locked block group
The mballoc code likes to call ext4_error while it is holding locked
block groups. This can causes a scheduling in atomic context BUG. We
can't just unlock the block group and relock it after/if ext4_error
returns since that might result in race conditions in the case where
the filesystem is set to continue after finding errors.
projects / linux-2.6-omap-h63xx.git / log