Michael Schimek [Thu, 18 Jan 2007 19:17:39 +0000 (16:17 -0300)]
V4L/DVB (5077): Bttv cropping support
Adds the missing VIDIOC_CROPCAP, G_CROP and S_CROP ioctls, permitting
applications to capture or overlay a subsection of the picture or to
extend the capture window beyond active video, into the VBI area and the
horizontal blanking. VBI capturing can start and end on any line,
including the picture area, and apps can capture different lines of each
field and single fields.
For compatibility with existing applications, the open() function
resets the cropping and VBI capturing parameters and a VIDIOC_S_CROP
call is necessary to actually enable cropping.
Regrettably in PAL-M, PAL-N, PAL-Nc and NTSC-JP mode the maximum image
width will increase from 640 and 768 to 747 and 923 pixels respectively.
Like the VBI changes however, this should only affect applications which
depend on former driver limitations, such as never getting more than 640
pixels regardless of the requested width.
Also, new freedoms require additional checks for conflicts and some
applications may not expect an EBUSY error from the VIDIOC_QBUF and
VIDIOCMCAPTURE ioctls. These errors should be rare though.
So far, the patch has been tested on a UP machine with a bt878 in PAL-
BGHI and NTSC-M mode using xawtv, tvtime, mplayer/mencoder, zapping/
libzvbi and these tools: http://zapping.sf.net/bttv-crop-test.tar.bz2
I'd be grateful about comments or bug reports.
Signed-off-by: Michael H. Schimek <mschimek@gmx.at> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Sat, 30 Dec 2006 21:31:22 +0000 (18:31 -0300)]
V4L/DVB (5059): Pvrusb2: Be smarter about mode restoration
Switch back to the previous input selection when the radio device is
closed - but only do that if the current input selection is still the
radio (i.e. it appears that it hasn't been messed with).
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Clean up and tighten logic involving stream configuration. This
mainly involves changes to pvrusb2-v4l2.c, where we better clarify how
we use the stream configuration enum and implement a cleaner means to
control streaming for a given device node.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:36:33 +0000 (23:36 -0300)]
V4L/DVB (5053): Pvrusb2: Change default volume to something sane
The default volume of 65535 is too high. Make is something smaller.
Note that this _only_ changes the default value. Specifically, there
are no scaling or other more intrusive changes here. I'm just sick of
constantly having to reduce the volume every time I plug in and test
the device! (And unfortunately we can't do a better fix like scaling
the volume so that 65535 makes sense because doing so will screw up
any app - like MythTV - which expects the old scaling.) Too bad V4L
controls don't have better defined ranges.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:33:00 +0000 (23:33 -0300)]
V4L/DVB (5052): Pvrusb2: Remove stream claiming hack from /dev/radio
Trying to temporarily check that the stream is not claimed during open
of the radio device is at best a race condition. What's to stop
another app from claiming the stream anyway the instant after the
check is done? The implementation for this was dicey anyway. So it's
removed. The only "price" for this is that if /dev/radioX is opened
while streaming video, then the video stream is just going to switch
to radio mode anyway. If a user does this, he gets what he expects...
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:30:13 +0000 (23:30 -0300)]
V4L/DVB (5051): Pvrusb2: Better radio versus tv frequency handling
Separate track radio versus tv frequency so that when we switch modes
we can also switch to a sane frequency appropriate for the mode. Also
implement logic to automate mode switching in certain cases.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:23:22 +0000 (23:23 -0300)]
V4L/DVB (5047): Pvrusb2: Fix tuning calculation when in radio mode
Frequency units in V4L2 are apparently different when in radio mode
compared to tv mode. Why? Who knows. This change adapts the driver
appropriately - so that internally we always only deal in Hz and don't
have to muck with craziness like this.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:19:42 +0000 (23:19 -0300)]
V4L/DVB (5045): Pvrusb2: Fix heap corruption introduced by radio mods
We can't allocate v4l device structures in a block, since the v4l core
governs when each device actually gets freed. This bug was introduced
as part of the core radio implementation. Fix it.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Thu, 28 Dec 2006 02:12:28 +0000 (23:12 -0300)]
V4L/DVB (5041): Pvrusb2: Use separate enumeration for get/store of minor number
Use separate enum for get/store of minor number; we want pvr2_config
to go away eventually and since it really means something different,
don't use it here
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The "main" V4L2 interface patch. This is yet very incomplete, incorrect and
probably inappropriate for inclusion as-is, but at least with this I 'm able
to tune and play radio through a V4L2 program (pvr-radio.c, a "thumb" version
of ivtv-radio.c with just the essentials).
Therefore, it kinda gives an idea of what is needed to support this, hm,
interface (partly used also by e.g., kradio). Please point out any mistakes
on this code. I 'm sure I 'm messing up some struct initialization somewhere
but currently I 'm too lazy to actually think this through until I complete
the functionality (e.g., handle the VIDIOC_S_STD, ENUMINPUT, etc ioctls
appropriately).
V4L/DVB (5038): Pvrusb2: Implement stream claim checking function
Add (and expose) a new function, pvr2_channel_check_stream_no_lock(), in
pvrusb2-context.c. This is hopefully the last V4L2 interface related patch
to change anything outside pvrusb2-v4l2.c.
We need this to implement the open() for the radio device. The reason is
that within the *enter_context() section of open() we need to ensure nobody
is streaming and if we cannot, we should cleanup after ourselves and return
-EBUSY. We cannot just use claim_stream() because:
1) That would cause a deadlock trying to re-acquire the context lock
2) We only need to ensure that nobody is streaming. We don't need to
actually acquire the stream.
Again, this is a kinda ugly patch. Feel free to improve.
V4L/DVB (5037): Pvrusb2: Implement multiple minor device number handling
This is the first patch in preparation of the V4L2/IVTV radio interface.
It does away with the assumption of only one minor per device. It also
adds a file to show the radio minor as well. This can be useful for a
program like pvr-radio.c (when it grows up), since this way it can search
for the minor of the /dev/radioX device it opened and use the video minor
of the same driver instance to get to the actual stream.
The implementation looks kinda ugly. Feel free to improve (that is the
reason behind separate patches anyway).
V4L/DVB (5036): Pvrusb2: Fix for min/max control value checking
In the previous patch we exploited the get_{min,max}_value facility to adjust
min/max allowable frequencies on the fly, depending on tuner mode.
Unfortunately, this facility was not used inside the *sym_to_val() function
that translates what we echo to sysfs, which means we got an -ERANGE despite
asking for a frequency between what we read to be min/max.
This patch corrects this small omission.
V4L/DVB (5035): Pvrusb2: Enable radio mode round #2
This is the logic that:
a) Ensures /sys/class/pvrusb2/sn-*/ctl_frequency/{max,min}_val are
"automagically" reset to sane values on each mode change.
b) Allows tuning to a radio frequency by something like:
echo `perl -e "print int(94.9*16000 + 0.5)"` \
> /sys/class/pvrusb2/sn-*/ctl_input/cur_val
The trick was to take advantage of the already existing .get_{min,max}_value
function pointers in pvr2_ctrl, to "dynamically override" the hardcoded values
for min/max frequency at runtime.
For a moment I thought to dispose of the hardcoded MIN/MAX_FREQ and use the
hirange/lowrange fields of the v4l2_tuner struct instead, but then I see that
tuner-core.c kinda hardcodes these as well, so I decided to not bother.
V4L/DVB (5034): Pvrusb2: Enable radio mode round #1
This is the logic that supports switching modes via e.g.,
echo radio > /sys/class/pvrusb2/sn-*/ctl_input/cur_val.
To do the mode switching we need to:
a) broadcast AUDC_SET_RADIO and
b) issue the CX2341X_ENC_MUTE_VIDEO command to the encoder.
The first is done by adding a new pvr2_i2c_op and having it trigger on
input change, the second by adding this command in pvr2_encoder_start()
and requesting an encoder restart on input change by setting
stale_subsys_mask appropriately.
The clues about AUDC_SET_RADIO and CX2341X_ENC_MUTE_VIDEO were kindly
provided by Hans Verkuil on the pvrusb2 mailing list. The idea to
implement mode switching this way (on input change) is due to Mike Isely.
Why AUDC_SET_RADIO/VIDIOC_S_STD are used for switching? I can 't be sure,
but I think this can be traced to a cornell student being the first to
implement radio support in ivtv "as a different standard". I think the
rest just evolved from there (it 's in the ivtv ML archives).
V4L/DVB (5074): Some fixes at stream waitqueue on vivi
There are several potential troubles on vivi waitqueue code:
- Watchdog timer should be reset at every received frame;
- Watchdog timer should be reset at the beginning of vivi_thread();
- Checks for errors when creating a newer thread with kernel_thread();
- Wake up vivi_thread() after creating it.
Bill Dirks asked me to update his entries at kernel files, since
he change his e-mail.
I've also updated a few web broken links or obsolete info to the curent
sites where V4L drivers and API are being discussed currently.
CC: Bill Dirks <bill@thedirks.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Luca Risolia [Mon, 8 Jan 2007 13:43:56 +0000 (10:43 -0300)]
V4L/DVB (5062): SN9C102 driver updates
- Add support for SN9C105 and SN9C120
- Add some more USB device identifiers
- Add support for OV7660
- Implement audio ioctl's and VIDIOC_ENUM_FRAMESIZES
- Add preliminary support for 0x0c45/0x6007
- Documentation updates
- Generic improvements Signed-off-by: Luca Risolia <luca.risolia@studio.unibo.it> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Dwaine Garden [Mon, 8 Jan 2007 00:13:55 +0000 (21:13 -0300)]
V4L/DVB (5032): Improves some USBVision info messages
Replaces the info statements with printk(KERN_INFO statements.
This will cut down on the useless information which is showing up
in the kernel messages log file.
Signed-off-by: Dwaine P. Garden <DwaineGarden@rogers.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Mike Isely [Wed, 3 Jan 2007 21:08:06 +0000 (18:08 -0300)]
V4L/DVB: MAINTAINERS: tag pvrusb2 list as subscribers-only
Posting to the pvrusb2 mailing list is for subscribers only. Anyone can
subscribe of course. This is done purely to keep spammers and similar pond
scum from bothering the subscribers of the list. This patch marks the
pvrusb2 list tag in MAINTAINERS to reflect this situation.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Adrian Bunk [Thu, 28 Dec 2006 15:47:47 +0000 (12:47 -0300)]
V4L/DVB (5017): DVB: fix compile error
This patch fixes the following compile error:
<-- snip -->
...
LD drivers/media/video/built-in.o
drivers/media/video/saa7134/built-in.o:(.data+0x85ec): multiple definition of `ir_rc5_remote_gap'
drivers/media/video/bt8xx/built-in.o:(.data+0x734c): first defined here
drivers/media/video/saa7134/built-in.o:(.data+0x85f0): multiple definition of `ir_rc5_key_timeout'
drivers/media/video/bt8xx/built-in.o:(.data+0x7350): first defined here
make[4]: *** [drivers/media/video/built-in.o] Error 1
<-- snip -->
Since this variables were needlessly global, this patch implements the
trivial fix of making them static.
Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (5012a): Remove some unused code from kernel mainstream
There are some long time unused code under some media driver source files.
There's no need of keeping it at mainstream.
Those unused code will remain available at V4L/DVB master tree and also at
kernel history.
Tobias Klauser [Tue, 26 Dec 2006 10:33:48 +0000 (07:33 -0300)]
V4L/DVB (5011): DVB: Remove unneeded void * casts in ttpci/av7110
The patch removes unneeded void * casts for the following (void *) pointers:
- struct file: private_data
- struct dvb_device: priv
- struct dvb_demux: priv
- struct dvb_adapter: priv
The patch also contains some whitespace and coding style cleanups in the
relevant areas.
Antti Seppälä [Wed, 20 Dec 2006 14:10:35 +0000 (11:10 -0300)]
V4L/DVB (4999): [PATCH] Cablestar2 support
This patch changes the initialization of alps tdee4 tuner in
flexcop-fe-tuner.c to match what is used in the old driver that was
written specifically for Cablestar cards by Patrick Boettcher. This
patch should make Cablestar2 work again with recent dvb drivers without
breaking other stv0297 based cards.
Signed-off-by: Antti Seppala <ajhseppa@niksula.hut.fi> Signed-off-by: Patrick Boettcher <pb@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Hans Verkuil [Wed, 20 Dec 2006 09:50:18 +0000 (06:50 -0300)]
V4L/DVB (4987): Improve cx2341x documentation
Document the program index table format, removed unused interrupt documentation
and improve the documentation regarding the audio mode (stereo/joint/dual/mono).
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Hans Verkuil [Tue, 19 Dec 2006 01:52:21 +0000 (22:52 -0300)]
V4L/DVB (4986): Removed unimplemented cx2341x API commands
The commands CX2341X_DEC_SET_AUDIO_OUTPUT, CX2341X_DEC_SET_AV_DELAY and
CX2341X_ENC_SET_3_2_PULLDOWN are not implemented in the Conexant firmware.
So these commands are removed. This also means that the V4L2_CID_MPEG_VIDEO_PULLDOWN
control in cx2341x.c and pvrusb2-hdw.c is removed.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Hans Verkuil [Tue, 19 Dec 2006 01:43:36 +0000 (22:43 -0300)]
V4L/DVB (4985): Update cx2341x documentation.
Removed a few unimplemented commands. Added a note for a few fields that are
not implemented in the firmware, and clarified several issues around reverse
playback.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
WARNING: drivers/video/i810/i810fb.o - Section mismatch: reference
to .init.data: from .text between 'i810_check_params' (at offset
0x1123) and 'encode_fix'
yres cannot be declared __devinitdata as it is used in
i810_check_params(), which isn't __devinit.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ben Dooks [Tue, 20 Feb 2007 21:58:21 +0000 (13:58 -0800)]
[PATCH] fb: SM501 framebuffer driver
Driver for the Silicon Motion SM501 multifunction device framebuffer
subsystem.
This driver supports both the CRT and LCD panel heads, with some simple
acceleration for the cursor plotting and support for screen panning. There
is no current support for bitblt/drawing engines, which should be added at
a later date.
This has been tested on a number of configurations, including PCI and
generic-bus, on PPC, ARM and SH4
[akpm@linux-foundation.org: fix warnings] Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Vincent Sanders <vince@arm.linux.org.u.> Acked-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Philipp Zabel [Tue, 20 Feb 2007 21:58:20 +0000 (13:58 -0800)]
[PATCH] GPIO API: SA1100 wrapper cleanup
Based on the discussion last december (http://lkml.org/lkml/2006/12/20/241),
this patch
- adds gpio_direction_input/output functions to
generic.c instead of making them inline,
- fixes comment and includes and uses inline functions
instead of macros in gpio.h
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Philipp Zabel [Tue, 20 Feb 2007 21:58:20 +0000 (13:58 -0800)]
[PATCH] GPIO API: S3C2410 wrapper cleanup
this one adds an #include <asm/arch/regs-gpio.h>.
Tested by Roman Moravcik on s3c2440.
Based on the discussion last december
(http://lkml.org/lkml/2006/12/20/243), this patch
- fixes comment and includes in gpio.h
- adds the gpio_to_irq definition for S3C2400
- includes asm/arch/regs-gpio.h for pin direction
definitions
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Harald Welte [Tue, 20 Feb 2007 21:58:19 +0000 (13:58 -0800)]
[PATCH] spi_s3c2410_gpio.c spi mode 2 and 3 support
Add transfer modes 2 and 3 to the S3C24XX gpio SPI driver
Signed-off-by: Harald Welte <laforge@openmoko.org> Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Tue, 20 Feb 2007 21:58:19 +0000 (13:58 -0800)]
[PATCH] SPI controller build/warning fixes
The signature of the per-device cleanup() routine changed to remove its
const-ness. Three new SPI controller drivers now need that change, to
eliminate build warnings.
This also fixes a build bug with atmel_spi on AT91 systems.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
WARNING: drivers/parport/parport_pc.o - Section mismatch: reference
to .init.text: from .text between 'parport_pc_probe_port' (at offset
0x14f7) and 'parport_pc_unregister_port'
parport_dma_probe() cannot be declared __devinit as it is called
from parport_pc_probe_port() which isn't.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Peter Zijlstra [Tue, 20 Feb 2007 21:58:18 +0000 (13:58 -0800)]
[PATCH] lockdep: annotate BLKPG_DEL_PARTITION
>=============================================
>[ INFO: possible recursive locking detected ]
>2.6.19-1.2909.fc7 #1
>---------------------------------------------
>anaconda/587 is trying to acquire lock:
> (&bdev->bd_mutex){--..}, at: [<c05fb380>] mutex_lock+0x21/0x24
>
>but task is already holding lock:
> (&bdev->bd_mutex){--..}, at: [<c05fb380>] mutex_lock+0x21/0x24
>
>other info that might help us debug this:
>1 lock held by anaconda/587:
> #0: (&bdev->bd_mutex){--..}, at: [<c05fb380>] mutex_lock+0x21/0x24
>
>stack backtrace:
> [<c0405812>] show_trace_log_lvl+0x1a/0x2f
> [<c0405db2>] show_trace+0x12/0x14
> [<c0405e36>] dump_stack+0x16/0x18
> [<c043bd84>] __lock_acquire+0x116/0xa09
> [<c043c960>] lock_acquire+0x56/0x6f
> [<c05fb1fa>] __mutex_lock_slowpath+0xe5/0x24a
> [<c05fb380>] mutex_lock+0x21/0x24
> [<c04d82fb>] blkdev_ioctl+0x600/0x76d
> [<c04946b1>] block_ioctl+0x1b/0x1f
> [<c047ed5a>] do_ioctl+0x22/0x68
> [<c047eff2>] vfs_ioctl+0x252/0x265
> [<c047f04e>] sys_ioctl+0x49/0x63
> [<c0404070>] syscall_call+0x7/0xb
Annotate BLKPG_DEL_PARTITION's bd_mutex locking and add a little comment
clarifying the bd_mutex locking, because I confused myself and initially
thought the lock order was wrong too.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Adrian Bunk [Tue, 20 Feb 2007 21:58:17 +0000 (13:58 -0800)]
[PATCH] drivers/isdn/gigaset/: build asyncdata.o into the gigaset module
LD drivers/isdn/gigaset/built-in.o
drivers/isdn/gigaset/ser_gigaset.o: In function `gigaset_m10x_send_skb':
(.text+0xe50): multiple definition of `gigaset_m10x_send_skb'
drivers/isdn/gigaset/usb_gigaset.o:(.text+0x0): first defined here
drivers/isdn/gigaset/ser_gigaset.o: In function `gigaset_m10x_input':
(.text+0x1121): multiple definition of `gigaset_m10x_input'
drivers/isdn/gigaset/usb_gigaset.o:(.text+0x2d1): first defined here
make[4]: *** [drivers/isdn/gigaset/built-in.o] Error 1
Signed-off-by: Adrian Bunk <bunk@stusta.de> Cc: Tilman Schmidt <tilman@imap.cc> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Tue, 20 Feb 2007 21:58:16 +0000 (13:58 -0800)]
[PATCH] remove modpost false warnings on ARM
This patch stops "modpost" from issuing erroneous modpost warnings on ARM
builds, which it's been doing since since maybe last summer. A canonical
example would be driver method table entries:
WARNING: <path> - Section mismatch: reference to .exit.text:<name>_remove
from .data after '$d' (at offset 0x4)
That "$d" symbol is generated by tools conformant with ARM ABI specs; in
this case it's a symbol **in the middle of** a "<name>_driver" struct.
The erroneous warnings appear to be issued because "modpost" whitelists
references from "<name>_driver" data into init and exit sections ... but
doesn't know should also include those "$d" mapping symbols, which are not
otherwise associated with "<name>_driver" symbols.
This patch prevents the modpost symbol lookup code from ever returning
those mapping symbols, so it will return a whitelisted symbol instead.
Then things work as expected.
Now to revert various code-bloating "fixes" that got merged because of this
modpost bug....
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Roman Zippel <zippel@linux-m68k.org> Acked-by: Sam Ravnborg <sam@ravnborg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Philipp Zabel [Tue, 20 Feb 2007 21:58:15 +0000 (13:58 -0800)]
[PATCH] GPIO API: PXA wrapper cleanup
Based on the discussion last december (http://lkml.org/lkml/2006/12/20/242),
this patch:
- moves the PXA_LAST_GPIO check into pxa_gpio_mode
- fixes comment and includes in gpio.h
- replaces the gpio_set/get_value macros with inline
functions and adds a non-inline version to avoid
code explosion when gpio is not a constant.
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Nicolas Pitre <nico@cam.org> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Tue, 20 Feb 2007 21:58:14 +0000 (13:58 -0800)]
[PATCH] at91_rtc updates
Various bug fixes to the at91rm9200 RTC:
- alarm: setalarm() should pay attention to the "enabled" flag
- init: cleaner handling of the wakeup flags, which cpu init should
really have set up. Doing it here is just a workaround.
- linkage: since the at91_rtc driver probe() routine is in the init
section, it should use platform_driver_probe() instead of leaving
that pointer around in the driver struct after init section removal.
- linkage: likewise, remove() belongs in the exit section.
Among other things, the init and alarm changes ensure that this driver
handles the new sysfs "wakealarm" attribute properly.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Tue, 20 Feb 2007 21:58:13 +0000 (13:58 -0800)]
[PATCH] rtc-sa1100 rtc_wklarm.enabled bugfixes
Some rtc-sa1100 bugfixes:
- The read_alarm() method reports the rtc_wkalrm.enabled field properly.
This patch is already in the handhelds.org tree.
- And the set_alarm() method now handles that flag correctly, rather than
making mismatched {en,dis}able_irq_wake() calls, which trigger runtime
warning messages. (Those calls are best made in suspend/resume methods.)
Note that while this SA1100/PXA RTC is fully capable of waking those ARM
processors from sleep states, that mechanism isn't properly supported on
either processor family, or in this driver. Some boards have board-specific
PM glue providing partial workarounds for the weak generic PM support.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[PATCH] Missing __user in pointer referenced within copy_from_user
Pointers to user data should be marked with a __user hint. This one is
missing.
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Morton [Tue, 20 Feb 2007 21:58:12 +0000 (13:58 -0800)]
[PATCH] genalloc warning fixes
lib/genalloc.c: In function 'gen_pool_alloc':
lib/genalloc.c:151: warning: passing argument 2 of '__set_bit' from incompatible pointer type
lib/genalloc.c: In function 'gen_pool_free':
lib/genalloc.c:190: warning: passing argument 2 of '__clear_bit' from incompatible pointer type
Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
affs wants to truncate the inode when the last user goes away, currently it
does that through a potentially racy i_count check in ->put_inode. But we
already have a method that's called just after the we dropped the last
reference, ->drop_inode. This patch implements affs_drop_inode to take
advantage of this.
Signed-off-by: Christoph Hellwig <hch@lst.de> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Tue, 20 Feb 2007 21:58:10 +0000 (13:58 -0800)]
[PATCH] autofs4: check for directory re-create in lookup
This problem was identified and fixed some time ago by Jeff Moyer but it fell
through the cracks somehow.
It is possible that a user space application could remove and re-create a
directory during a request. To avoid returning a failure from lookup
incorrectly when our current dentry is unhashed we need to check if another
positive, hashed dentry matching this one exists and if so return it instead
of a fail.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com> 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>
Ian Kent [Tue, 20 Feb 2007 21:58:10 +0000 (13:58 -0800)]
[PATCH] autofs4: fix another race between mount and expire
Jeff Moyer has identified a race between mount and expire.
What happens is that during an expire the situation can arise that a directory
is removed and another lookup is done before the expire issues a completion
status to the kernel module. In this case, since the the lookup gets a new
dentry, it doesn't know that there is an expire in progress and when it posts
its mount request, matches the existing expire request and waits for its
completion. ENOENT is then returned to user space from lookup (as the dentry
passed in is now unhashed) without having performed the mount request.
The solution used here is to keep track of dentrys in this unhashed state and
reuse them, if possible, in order to preserve the flags. Additionally, this
infrastructure will provide the framework for the reintroduction of caching of
mount fails removed earlier in development.
Signed-off-by: Ian Kent <raven@themaw.net> Acked-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Tue, 20 Feb 2007 21:58:09 +0000 (13:58 -0800)]
[PATCH] autofs4: header file update
The current header file definitions for autofs version 5 have caused a couple
of problems for application builds downstream.
This fixes the problem by separating the definitions.
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>
Nick Piggin [Tue, 20 Feb 2007 21:58:09 +0000 (13:58 -0800)]
[PATCH] fs: fix nobh data leak
nobh_prepare_write leaks data similarly to how simple_prepare_write did. Fix
by not marking the page uptodate until nobh_commit_write time. Again, this
could break weird use-cases, but none appear to exist in the tree.
We can safely remove the set_page_dirty, because as the comment says,
nobh_commit_write does set_page_dirty. If a filesystem wants to allocate
backing store for a page dirtied via mmap, page_mkwrite is the suggested
approach.
Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nick Piggin [Tue, 20 Feb 2007 21:58:08 +0000 (13:58 -0800)]
[PATCH] fs: fix libfs data leak
simple_prepare_write leaks uninitialised kernel data. This happens because
the it leaves an uninitialised "hole" over the part of the page that the
write is expected to go to. This is fine, but it then marks the page
uptodate, which means a concurrent read can come in and copy the
uninitialised memory into userspace before it written to.
Fix it by simply marking it uptodate in simple_commit_write instead, after
the hole has been filled in. This could theoretically break an fs that
uses simple_prepare_write and not simple_commit_write, and that relies on
the incorrect simple_prepare_write behaviour. Luckily, none of those
exists in the tree.
Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Dave Jones [Tue, 20 Feb 2007 21:58:07 +0000 (13:58 -0800)]
[PATCH] loosen dependancy on rtc cmos
This option is useful for all of the X86 subarchs afaik (and especially
X86_GENERICARCH).
Signed-off-by: Dave Jones <davej@redhat.com> Acked-by: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ben Dooks [Tue, 20 Feb 2007 21:58:01 +0000 (13:58 -0800)]
[PATCH] mfd: SM501 core driver
This driver provides the core functionality of the SM501, which is a
multi-function chip including two framebuffers, video acceleration, USB,
and many other peripheral blocks.
The driver exports a number of entries for the peripheral drivers to use.
Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Vincent Sanders <vince@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Miguel Ojeda [Tue, 20 Feb 2007 21:58:00 +0000 (13:58 -0800)]
[PATCH] cfag12864b: fix crash when built-in and no parport present
The problem comes when ks0108/cfag12864b are built-in and no parallel port is
present. ks0108_init() is called first, as it should be, but fails to load
(as there is no parallel port to use).
After that, cfag12864b_init() gets called, without knowing anything about
ks0108 failed, and calls ks0108_writecontrol(), which dereferences an
uninitialized pointer.
Init order is OK, I think. The problem is how to stop cfag12864b_init() being
called if ks0108 failed to load. modprobe does it for us, but, how when
built-in?
Signed-off-by: Miguel Ojeda Sandonis <maxextreme@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Akinobu Mita [Tue, 20 Feb 2007 21:57:56 +0000 (13:57 -0800)]
[PATCH] fault injection: split up stacktrace filter Kconfig option
There is no prompt for CONFIG_STACKTRACE, so FAULT_INJECTION cannot be
selected without LOCKDEP enabled. (found by Paolo 'Blaisorblade'
Giarrusso)
In order to fix such broken Kconfig dependency, this patch splits up the
stacktrace filter support for fault injection by new Kconfig option, which
enables to use fault injection on the architecture which doesn't have
general stacktrace support.
Cc: "Paolo 'Blaisorblade' Giarrusso" <blaisorblade@yahoo.it> Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Srinivasa Ds [Tue, 20 Feb 2007 21:57:54 +0000 (13:57 -0800)]
[PATCH] kprobes: list all active probes in the system
This patch lists all active probes in the system by scanning through
kprobe_table[]. It takes care of aggregate handlers and prints the type of
the probe. Letter "k" for kprobes, "j" for jprobes, "r" for kretprobes.
It also lists address of the instruction,its symbolic name(function name +
offset) and the module name. One can access this file through
/sys/kernel/debug/kprobes/list.
Output looks like this
=====================
llm40:~/a # cat /sys/kernel/debug/kprobes/list c0169ae3 r sys_read+0x0 c0169ae3 k sys_read+0x0 c01694c8 k vfs_write+0x0 c0167d20 r sys_open+0x0 f8e658a6 k reiserfs_delete_inode+0x0 reiserfs c0120f4a k do_fork+0x0 c0120f4a j do_fork+0x0 c0169b4a r sys_write+0x0 c0169b4a k sys_write+0x0 c0169622 r vfs_read+0x0
=================================
[akpm@linux-foundation.org: cleanup]
[ananth@in.ibm.com: sparc build fix] Signed-off-by: Srinivasa DS <srinivasa@in.ibm.com> Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nick Piggin [Tue, 20 Feb 2007 21:57:54 +0000 (13:57 -0800)]
[PATCH] fs: fix __block_write_full_page error case buffer submission
Andrew noticed that unlocking the page before submitting all buffers for
writeout could cause problems if the IO completes before we've finished
messing around with the page buffers, and they subsequently get freed.
Even if there were no bug, it is a good idea to bring the error case
into line with the common case here.
Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[PATCH] shm: make sysv ipc shared memory use stacked files
The current ipc shared memory code runs into several problems because it
does not quite use files like the rest of the kernel. With the option of
backing ipc shared memory with either hugetlbfs or ordinary shared memory
the problems got worse. With the added support for ipc namespaces things
behaved so unexpected that we now have several bad namespace reference
counting bugs when using what appears at first glance to be a reasonable
idiom.
So to attack these problems and hopefully make the code more maintainable
this patch simply uses the files provided by other parts of the kernel and
builds it's own files out of them. The shm files are allocated in do_shmat
and freed when their reference count drops to zero with their last unmap.
The file and vm operations that we don't want to implement or we don't
implement completely we just delegate to the operations of our backing
file.
This means that we now get an accurate shm_nattch count for we have a
hugetlbfs inode for backing store, and the shm accounting of last attach
and last detach time work as well.
This means that getting a reference to the ipc namespace when we create the
file and dropping the referenece in the release method is now safe and
correct.
This means we no longer need a special case for clearing VM_MAYWRITE
as our file descriptor now only has write permissions when we have
requested write access when calling shmat. Although VM_SHARED is now
cleared as well which I believe is harmless and is mostly likely a
minor bug fix.
By using the same set of operations for both the hugetlb case and regular
shared memory case shmdt is not simplified and made slightly more correct
as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting
all shared memory regions generated from sysvipc shared memory.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / linux-2.6 / log