Daniel Mack [Fri, 24 Apr 2009 14:37:45 +0000 (16:37 +0200)]
ASoC: cs4270: add Master Playback Switch
This adds a new control named 'Master Playback Switch' for cs4270
codecs. It is implemented using the new SOC_DOUBLE_EXT macro to catch
the put function and store the information about manually set mute
controls from userspace. When a manual mute is set, we don't want the
soc core to un-mute the outputs.
Renamed cs4270_mute() to cs4270_dai_mute() to avoid confusion.
Signed-off-by: Daniel Mack <daniel@caiaq.de> Acked-by: Timur Tabi <timur@freescale.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Daniel Mack [Fri, 24 Apr 2009 13:00:25 +0000 (15:00 +0200)]
ASoC: cs4270: fix Master Capture Switch polarity
The control modifies the MUTE register, hence the polarity must be
inverted.
Signed-off-by: Daniel Mack <daniel@caiaq.de> Acked-By: Timur Tabi <timur@freescale.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Fri, 24 Apr 2009 15:33:10 +0000 (16:33 +0100)]
ASoC: Fix S3C64xx IIS device registration and support both ports
The S3C64xx IIS code had a number of problems with device registration.
The hardware has two IIS ports of which the driver supported only one
at once via a single exported DAI, attempting to identify the DAI to
use based on the dev->id of the ASoC platform device. As well as
limiting the driver to only supporting one IIS port at once this also
meant that the ID of the soc-audio device (or in future the card device)
had to match the IIS ID.
Fix both problems by converting the driver to register the DAIs based on
probing of platform devices registered by the arch/arm code, using those
platform devices to interact with the clock API.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Eric Miao [Thu, 23 Apr 2009 09:05:38 +0000 (17:05 +0800)]
ASoC: simplify the SSP DMA parameters settings by run-time generation
The SSP DMA parameters can actually be easily generated at run-time since
they are almost similar except for the FIFO width and direction. Another
benefit is the re-use of information from 'struct ssp_device', like SSDR
physical FIFO address and DRCMR register index for both directions.
Signed-off-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Reviewed-by: pHilipp Zabel <philipp.zabel@gmail.com>
Add DAPMs for VDL(Voice Down Link) path. To support VDL path, we have
to change DAPMs of outputs(Earpiece, PreDrive Left/Right, Headset
Left/Right, Carkit Left/Right) from mux to mixer.
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com> Acked-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Wed, 22 Apr 2009 17:24:55 +0000 (18:24 +0100)]
ASoC: Add power supply widget to DAPM
Many modern CODECs have shared resources on chip which must be enabled
for portions of the chip to work but which can be disabled at other times
in order to achieve power savings. Examples of such resources include
power supplies and some internal clocks.
Since these widgets are dependencies for the audio path but do not carry
audio signals they require slightly different handling to most widgets -
they do not contribute to the audio path and so should not be counted as
either inputs or outputs during path walks.
Cases where one supply provides a supply for another will require
additional work. There is also room for more optimisation of the graph
walking to avoid repeated checks for the same thing.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Mon, 20 Apr 2009 16:56:13 +0000 (17:56 +0100)]
ASoC: Make the DAPM power check an operation on the widget
Rather than having switch statements at point of use make the DAPM
power check a member of the widget structure and set it when we
instantiate the widget.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Mon, 20 Apr 2009 15:56:59 +0000 (16:56 +0100)]
ASoC: Factor out generic widget power checks
This will form a basis for further power check refactoring: the overall
goal of these changes is to allow us to check power separately to
applying it, allowing improvements in the power sequencing algorithms.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Add Voice DAI to support the PCM voice interface of the twl4030 codec.
The PCM voice interface can be used with 8-kHz(voice narrowband) or
16-kHz(voice wideband) sampling rates, and 16bits, and mono RX and mono
TX or stereo TX.
The PCM voice interface has two modes
- PCM mode1 : This uses the normal FS polarity and the rising edge of
the clock signal.
- PCM mode2 : This uses the FS polarity inverted and the falling edge
of the clock signal.
If the system master clock is not 26MHz or the twl4030 codec mode is not
option2, the voice PCM interface is not available.
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com> Acked-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Peter Ujfalusi [Fri, 17 Apr 2009 12:55:08 +0000 (15:55 +0300)]
ASoC: TWL4030: Fix for the constraint handling
The original implementation of the constraints were good against sane
applications.
If the opening sequence is:
stream1_open, stream1_hw_params, stream2_open, stream2_hw_params -> the
constraints are set correctly for stream2.
But if the sequence is:
stream1_open, stream2_open, stream2_hw_params, stream1_hw_params -> than stream2
would receive constraint rate = 0, sample_bits = 0, since the stream1 has not
yet called hw_params...
The command to trigger this event:
gst-launch-0.10 alsasrc device=hw:0 ! alsasink device=hw:0 sync=false
This patch does some 'black magic' in order to always set the correct
constraints and sets it only when it is needed for the other stream.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Jarkko Nikula [Fri, 17 Apr 2009 11:42:26 +0000 (14:42 +0300)]
ASoC: OMAP: Update contact addresses
My email address is going to expire soon so update it. Adding also
Peter Ujfalusi <peter.ujfalusi@nokia.com> as a second contact to OMAP core
drivers since I won't have anymore access to non-public OMAP documentation
in the future and Peter is working with these drivers as well.
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com> Cc: Peter Ujfalusi <peter.ujfalusi@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Peter Ujfalusi [Wed, 15 Apr 2009 12:38:56 +0000 (15:38 +0300)]
ASoC: OMAP: Add DSP_A mode support for mcbsp
DSP_A mode is similar to the DSP_B, but the MSB is delayed with
one bclk (appears after the FS pulse and not under it).
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Acked-by: Jarkko Nikula <jarkko.nikula@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Peter Ujfalusi [Wed, 15 Apr 2009 12:38:55 +0000 (15:38 +0300)]
ASoC: OMAP: Use single-phase for DSP mode
Use single-phase mode for the DSP mode and keep the dual phase
mode for the I2S mode.
The mono (1 channel) mode already used single phase mode,
now it is more cleaner. There is no need to configure the
second phase, when the single phase is used.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Acked-by: Jarkko Nikula <jarkko.nikula@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Jarkko Nikula [Wed, 15 Apr 2009 10:48:17 +0000 (13:48 +0300)]
ASoC: OMAP: Fix FS polarity in OSK5912 machine driver
Using inverted FS polarity in OSK5912 must be an error since TLV320AIC23
do not have support for inverted polarities. This is mostly due the hassle
with the DSP formats in OMAP McBSP DAI and inversion on OMAP side probably
just made this configuration working at some point.
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com> Acked-by: Arun KS <arunks@mistralsolutions.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Thanks to Peter Ujfalusi <peter.ujfalusi@nokia.com> for noticing and being
part of the fix. Now the FS length definition is more clear by defining
it with FWID(0).
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com> Acked-by: Peter Ujfalusi <peter.ujfalusi@nokia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Daniel Mack [Wed, 15 Apr 2009 18:24:45 +0000 (20:24 +0200)]
ASoC: pxa-ssp: allow setting of dai format 0
pxa_ssp_set_dai_fmt() currently has an early exit if the desired format
equals the current configuration. This is correct behaviour unless this
function is called with a zero value parameter for the first time.
Zero is a valid value for this function, but the early exit is bogus in
this case.
Hence, set priv->dai_fmt to -1 in the beginning so we can configure the
port.
Signed-off-by: Daniel Mack <daniel@caiaq.de> Cc: pHilipp Zabel <philipp.zabel@gmail.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Wed, 15 Apr 2009 20:37:46 +0000 (21:37 +0100)]
ASoC: Volume controls are never of boolean type
Some limited volume controls (mostly simple attenuations) have only two
settings so the ASoC info functions misreport them as booleans. Since
we currently have no better information check for " Volume" in the
control name and always report any controls matching as being integer.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Tue, 7 Apr 2009 18:20:14 +0000 (19:20 +0100)]
ASoC: Add WM8960 CODEC driver
The WM8960 is a low power, high quality stereo codec designed for
portable digital audio applications.
Stereo class D speaker drivers provide 1W per channel into 8W loads.
Guaranteed low leakage, excellent PSRR and pop/click suppression
mechanisms enable direct battery connection for the speaker supply.
The device also integrates a complete microphone interface and a stereo
headphone driver. External component requirements are drastically
reduced as no separate microphone, speaker or headphone amplifiers are
required. Advanced on-chip digital signal processing performs automatic
level control for the microphone or line input.
Stereo 24-bit sigma-delta ADCs and DACs are used with low power
over-sampling digital interpolation and decimation filters and a
flexible digital audio interface.
The master clock can be input directly or generated internally by an
onboard PLL, supporting most commonly-used clocking schemes.
This driver was originally written by Liam Girdwood, with substantial
subsequent additions and updates for feature completeness and changes in
the ASoC framework from me.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Daniel Ribeiro [Wed, 8 Apr 2009 13:51:24 +0000 (10:51 -0300)]
ASoC: pxa-ssp.c fix clock/frame invert
SCMODE(0): Data Driven (Falling), Data Sampled (Rising), Idle State (Low)
SCMODE(1): Data Driven (Rising), Data Sampled (Falling), Idle State (Low)
SCMODE(2): Data Driven (Rising), Data Sampled (Falling), Idle State (High)
SCMODE(3): Data Driven (Falling), Data Sampled (Rising), Idle State (High)
SCMODE(3) does not invert the clock polarity compared to the default SCMODE(0).
This patch also adds all possible NF/IF, NB/IB combinations to the DSP_A and
DSP_B modes.
Signed-off-by: Daniel Ribeiro <drwyrm@gmail.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Mon, 13 Apr 2009 10:09:18 +0000 (11:09 +0100)]
ASoC: Factor out application of power for generic widgets
This is simple code motion, intended to support future refactoring of
the DAPM algorithms and (more immediately) the additon of events for
DACs and ADCs.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Thu, 9 Apr 2009 15:40:41 +0000 (16:40 +0100)]
ASoC: Disable S3C64xx support in Kconfig
Due to the process and communications issues with the 2.6.30 S3C
platform merges none of the underlying arch/arm code for S3C64xx audio
support made it into mainline, rendering the drivers useless. Disable
them in Kconfig to avoid user confusion - users patching in the required
support can always reenable this too.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Eric Miao [Thu, 9 Apr 2009 06:13:07 +0000 (14:13 +0800)]
ASoC: magician: remove un-necessary #include of pxa-regs.h and hardware.h
Signed-off-by: Eric Miao <eric.miao@marvell.com> Cc: Philipp Zabel <philipp.zabel@gmail.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Merge branch 'core/softlockup' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'core/softlockup' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
softlockup: make DETECT_HUNG_TASK default depend on DETECT_SOFTLOCKUP
softlockup: move 'one' to the softlockup section in sysctl.c
softlockup: ensure the task has been switched out once
softlockup: remove timestamp checking from hung_task
softlockup: convert read_lock in hung_task to rcu_read_lock
softlockup: check all tasks in hung_task
softlockup: remove unused definition for spawn_softlockup_task
softlockup: fix potential race in hung_task when resetting timeout
softlockup: fix to allow compiling with !DETECT_HUNG_TASK
softlockup: decouple hung tasks check from softlockup detection
Merge branch 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
branch tracer, intel-iommu: fix build with CONFIG_BRANCH_TRACER=y
branch tracer: Fix for enabling branch profiling makes sparse unusable
ftrace: Correct a text align for event format output
Update /debug/tracing/README
tracing/ftrace: alloc the started cpumask for the trace file
tracing, x86: remove duplicated #include
ftrace: Add check of sched_stopped for probe_sched_wakeup
function-graph: add proper initialization for init task
tracing/ftrace: fix missing include string.h
tracing: fix incorrect return type of ns2usecs()
tracing: remove CALLER_ADDR2 from wakeup tracer
blktrace: fix pdu_len when tracing packet command requests
blktrace: small cleanup in blk_msg_write()
blktrace: NUL-terminate user space messages
tracing: move scripts/trace/power.pl to scripts/tracing/power.pl
Merge branch 'irq/threaded' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'irq/threaded' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
genirq: fix devres.o build for GENERIC_HARDIRQS=n
genirq: provide old request_irq() for CONFIG_GENERIC_HARDIRQ=n
genirq: threaded irq handlers review fixups
genirq: add support for threaded interrupts to devres
genirq: add threaded interrupt handler support
Commit c2ec175c39f62949438354f603f4aa170846aabb ("mm: page_mkwrite
change prototype to match fault") exposed a bug in the NFS
implementation of page_mkwrite. We should be returning 0 on success...
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
PCI: pci_slot: grab refcount on slot's bus
PCI Hotplug: acpiphp: grab refcount on p2p subordinate bus
PCI: allow PCI core hotplug to remove PCI root bus
PCI: Fix oops in pci_vpd_truncate
PCI: don't corrupt enable_cnt when doing manual resource alignment
PCI: annotate pci_rescan_bus as __ref, not __devinit
PCI-IOV: fix missing kernel-doc
PCI: Setup disabled bridges even if buses are added
PCI: SR-IOV quirk for Intel 82576 NIC
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
loop: mutex already unlocked in loop_clr_fd()
cfq-iosched: don't let idling interfere with plugging
block: remove unused REQ_UNPLUG
cfq-iosched: kill two unused cfqq flags
cfq-iosched: change dispatch logic to deal with single requests at the time
mflash: initial support
cciss: change to discover first memory BAR
cciss: kernel scan thread for MSA2012
cciss: fix residual count for block pc requests
block: fix inconsistency in I/O stat accounting code
block: elevator quiescing helpers
Mark Brown [Tue, 7 Apr 2009 17:45:21 +0000 (18:45 +0100)]
ASoC: Add WM8988 CODEC driver
The WM8988 is a low power, high quality stereo CODEC designed for
portable digital audio applications.
The device integrates complete interfaces to 2 stereo headphone or line
out ports. External component requirements are drastically reduced as no
separate headphone amplifiers are required. Advanced on-chip digital
signal processing performs graphic equaliser, 3-D sound enhancement and
automatic level control for the microphone or line input.
The WM8988 can operate as a master or a slave, with various master clock
frequencies including 12 or 24MHz for USB devices, or standard 256fs
rates like 12.288MHz and 24.576MHz. Different audio sample rates such as
96kHz, 48kHz, 44.1kHz are generated directly from the master clock
without the need for an external PLL.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Mark Brown [Tue, 7 Apr 2009 17:10:13 +0000 (18:10 +0100)]
ASoC: Provide core support for symmetric sample rates
Many devices require symmetric configurations of capture and playback
data formats, often due to shared clocking but sometimes also due to
other shared playback and record configuration in the device. Start
providing core support for this by allowing the DAIs or the machine
to specify that the sample rates used should be kept symmetric.
A flag symmetric_rates is provided in the snd_soc_dai and
snd_soc_dai_link structures. If this is set in either of the DAIs or in
the machine then a constraint will be applied when a stream is already
open preventing any changes in sample rate.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
The code that enables branch tracing for all (non-constant) branches
plays games with the preprocessor and #define's the C 'if ()' construct
to do tracing.
That's all fine, but it fails for some unusual but valid C code that is
sometimes used in macros, notably by the intel-iommu code:
if (i=drhd->iommu, drhd->ignored) ..
because now the preprocessor complains about multiple arguments to the
'if' macro.
So make the macro expansion of this particularly horrid trick use
varargs, and handle the case of comma-expressions in if-statements. Use
another macro to do it cleanly in just one place.
This replaces a patch by David (and acked by Steven) that did this all
inside that one already-too-horrid macro.
Tested-by: Ingo Molnar <mingo@elte.hu> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6: (36 commits)
ALSA: hda - Add VREF powerdown sequence for another board
ALSA: oss - volume control for CSWITCH and CROUTE
ALSA: hda - add missing comma in ad1884_slave_vols
sound: usb-audio: allow period sizes less than 1 ms
sound: usb-audio: save data packet interval in audioformat structure
sound: usb-audio: remove check_hw_params_convention()
sound: usb-audio: show sample format width in proc file
ASoC: fsl_dma: Pass the proper device for dma mapping routines
ASoC: Fix null dereference in ak4535_remove()
ALSA: hda - enable SPDIF output for Intel DX58SO board
ALSA: snd-atmel-abdac: increase periods_min to 6 instead of 4
ALSA: snd-atmel-abdac: replace bus_id with dev_name()
ALSA: snd-atmel-ac97c: replace bus_id with dev_name()
ALSA: snd-atmel-ac97c: cleanup registers when removing driver
ALSA: snd-atmel-ac97c: do a proper reset of the external codec
ALSA: snd-atmel-ac97c: enable interrupts to catch events for error reporting
ALSA: snd-atmel-ac97c: set correct size for buffer hardware parameter
ALSA: snd-atmel-ac97c: do not overwrite OCA and ICA when assigning channels
ALSA: snd-atmel-ac97c: remove dead break statements after return in switch case
ALSA: snd-atmel-ac97c: cleanup register definitions
...
Merge branch 'i2c-for-2630-v2' of git://aeryn.fluff.org.uk/bjdooks/linux
* 'i2c-for-2630-v2' of git://aeryn.fluff.org.uk/bjdooks/linux:
i2c: imx: Make disable_delay a per-device variable
i2c: xtensa s6000 i2c driver
powerpc/85xx: i2c-mpc: use new I2C bindings for the Socates board
i2c: i2c-mpc: make I2C bus speed configurable
i2c: i2c-mpc: use dev based printout function
i2c: i2c-mpc: various coding style fixes
i2c: imx: Add missing request_mem_region in probe()
i2c: i2c-s3c2410: Initialise Samsung I2C controller early
i2c-s3c2410: Simplify bus frequency calculation
i2c-s3c2410: sda_delay should be in ns, not clock ticks
i2c: iMX/MXC support
Alan Cox [Tue, 7 Apr 2009 14:30:57 +0000 (15:30 +0100)]
parport: Use the PCI IRQ if offered
PCI parallel port devices can IRQ share so we should stop them hogging
the line and making a mess on modern PC systems. We know the sharing
side works as the PCMCIA driver has shared the parallel port IRQ for
some time.
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sonic Zhang [Tue, 7 Apr 2009 15:52:26 +0000 (16:52 +0100)]
tty: Blackin CTS/RTS
Both software emulated and hardware based CTS and RTS are enabled in
serial driver.
The CTS RTS PIN connection on BF548 UART port is defined as a modem
device not as a host device. In order to test it under Linux, please
nake a cross UART cable to exchange CTS and RTS signal.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use the new general RS485 Linux data structure (introduced by Alan with
commit number c26c56c0f40e200e61d1390629c806f6adaffbcc) in the Cris
architecture too (currently, Cris still uses the old private data
structure instead of the new one).
Signed-off-by: Claudio Scordino <claudio@evidence.eu.com> Tested-by: Hinko Kocevar <hinko.kocevar@cetrtapot.si> Tested-by: Janez Cufer <janez.cufer@cetrtapot.si> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There's a possible deadlock in generic_file_splice_write(),
splice_from_pipe() and ocfs2_file_splice_write():
- task A calls generic_file_splice_write()
- this calls inode_double_lock(), which locks i_mutex on both
pipe->inode and target inode
- ordering depends on inode pointers, can happen that pipe->inode is
locked first
- __splice_from_pipe() needs more data, calls pipe_wait()
- this releases lock on pipe->inode, goes to interruptible sleep
- task B calls generic_file_splice_write(), similarly to the first
- this locks pipe->inode, then tries to lock inode, but that is
already held by task A
- task A is interrupted, it tries to lock pipe->inode, but fails, as
it is already held by task B
- ABBA deadlock
Fix this by explicitly ordering locks: the outer lock must be on
target inode and the inner lock (which is later unlocked and relocked)
must be on pipe->inode. This is OK, pipe inodes and target inodes
form two nonoverlapping sets, generic_file_splice_write() and friends
are not called with a target which is a pipe.
After a review of user's feedback for finding out other compatibility
issues, I found nilfs improperly initializes timestamps in inode;
CURRENT_TIME was used there instead of CURRENT_TIME_SEC even though nilfs
didn't have nanosecond timestamps on disk. A few users gave us the report
that the tar program sometimes failed to expand symbolic links on nilfs,
and it turned out to be the cause.
Instead of applying the above displacement, I've decided to support
nanosecond timestamps on this occation. Fortunetaly, a needless 64-bit
field was in the nilfs_inode struct, and I found it's available for this
purpose without impact for the users.
So, this will do the enhancement and resolve the tar problem.
The former versions didn't have extra super blocks. This improves the
weak point by introducing another super block at unused region in tail of
the partition.
This doesn't break disk format compatibility; older versions just ingore
the secondary super block, and new versions just recover it if it doesn't
exist. The partition created by an old mkfs may not have unused region,
but in that case, the secondary super block will not be added.
This doesn't make more redundant copies of the super block; it is a future
work.
nilfs2: simplify handling of active state of segments
will reduce some lines of segment constructor. Previously, the state was
complexly controlled through a list of segments in order to keep
consistency in meta data of usage state of segments. Instead, this
presents ``calculated'' active flags to userland cleaner program and stop
maintaining its real flag on disk.
Only by this fake flag, the cleaner cannot exactly know if each segment is
reclaimable or not. However, the recent extension of nilfs_sustat ioctl
struct (nilfs2-extend-nilfs_sustat-ioctl-struct.patch) can prevent the
cleaner from reclaiming in-use segment wrongly.
nilfs2: mark minor flag for checkpoint created by internal operation
Nilfs creates checkpoints even for garbage collection or metadata updates
such as checkpoint mode change. So, user often sees checkpoints created
only by such internal operations.
This is inconvenient in some situations. For example, application that
monitors checkpoints and changes them to snapshots, will fall into an
infinite loop because it cannot distinguish internally created
checkpoints.
This patch solves this sort of problem by adding a flag to checkpoint for
identification.