Adam Baker [Sun, 29 Mar 2009 22:17:10 +0000 (19:17 -0300)]
V4L/DVB (11387): Sensor orientation reporting
Add support to the SQ-905 driver to pass back to user space the
sensor orientation information obtained from the camera during init.
Modifies gspca and the videodev2.h header to create the necessary
API.
[mchehab@redhat.com: Changed "Output is" to "Frames are" at the comments, as suggested at LMML] Signed-off-by: Adam Baker <linux@baker-net.org.uk> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Remove support for the debug call VIDIOC_INT_S_AUDIO_ROUTING from cx18
and ivtv. These internal ioctls shouldn't be exposed. These were only
used through the cx18-ctl and ivtv-ctl utilities, and only when testing
a new card variant.
This cleanup allows the removal of this ioctl from v4l2-common.h.
Cc: Andy Walls <awalls@radix.net> Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
It is no longer needed to use a struct pointer as argument, since v4l2_subdev
doesn't require that ioctl-like approach anymore. Instead just pass the input,
output and config (new!) arguments directly.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 1 Apr 2009 06:15:52 +0000 (03:15 -0300)]
V4L/DVB (11377): v4l: increase version numbers of drivers converted to v4l2_subdev.
With all the v4l2_subdev changes that were made to these drivers it is a
good idea to increase the version number of each driver.
It's just the patch level that is increased, except for the zoran and saa7146
drivers where the minor number was increased due to the more substantial
changes that were made to those two drivers.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 1 Apr 2009 06:57:53 +0000 (03:57 -0300)]
V4L/DVB (11373): v4l2-common: add explicit v4l2_device pointer as first arg to new_(probed)_subdev
The functions v4l2_i2c_new_subdev and v4l2_i2c_new_probed_subdev relied on
i2c_get_adapdata to return the v4l2_device. However, this is not always
possible on embedded platforms. So modify the API to pass the v4l2_device
pointer explicitly.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Sun, 29 Mar 2009 23:04:44 +0000 (20:04 -0300)]
V4L/DVB (11372): v4l2: use old-style i2c API for kernels < 2.6.26 instead of < 2.6.22
Originally the intention was to switch to the new style i2c API starting with
the introduction of the API in 2.6.22. However, the i2c_new_probed_device()
function has a lethal bug that wasn't fixed until 2.6.25. Or more accurately,
it was only fixed in the stable series of 2.6.25 and 2.6.26.
Given the fact that the new i2c API also changed starting with 2.6.26 (the
addition of i2c_device_id), it is easiest to switch APIs starting with
2.6.26.
This patch updates all the legacy code accordingly.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 1 Apr 2009 06:52:39 +0000 (03:52 -0300)]
V4L/DVB (11370): v4l2-subdev: move s_std from tuner to core.
s_std didn't belong in the tuner ops. Stricly speaking it should be part of
the video ops, but it is used by audio and tuner devices as well, so it is
more efficient to make it part of the core ops.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Sun, 29 Mar 2009 22:20:26 +0000 (19:20 -0300)]
V4L/DVB (11369): v4l2-subdev: add load_fw and use that instead of abusing core->init.
The init callback was used in several places to load firmware. Make a separate
load_fw callback for that. This makes the code a lot more understandable.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 1 Apr 2009 06:41:09 +0000 (03:41 -0300)]
V4L/DVB (11367): v4l2-common: remove legacy code
Now that all drivers are converted to v4l2_subdev we can remove legacy code
in v4l2-common. Also move the documentation of the internal API to
v4l2-subdev.h where it really belongs.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Sun, 29 Mar 2009 21:00:07 +0000 (18:00 -0300)]
V4L/DVB (11366): v4l: remove obsolete header and source
v4l2-subdev.c and v4l2-i2c-drv-legacy.h were used to support the old i2c
API. All v4l drivers are now converted to v4l2_subdev, so these two files
can be removed.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 1 Apr 2009 06:32:22 +0000 (03:32 -0300)]
V4L/DVB (11364): tuner: remove i2c legacy code.
All drivers that use the tuner module now use v4l2_subdev, so we can remove the
legacy code from this module.
Note that TUNER_SET_CONFIG is still called by tuner-simple.c, so we have
to handle it via a .command callback. There must be a better way to do this,
but for now this will work.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (11135): cx231xx: use usb_make_path() for bus_info
VIDIOC_QUERYCAP should return the proper bus info. In the case of USB, this
should be generated by usb_make_path(), being something like:
usb-0000:00:1d.7-2
Sri Deevi [Sat, 21 Mar 2009 02:33:48 +0000 (23:33 -0300)]
V4L/DVB (11128): cx231xx: convert the calls to subdev format
This patch converts cx231xx to the new v4l2 dev/subdev, doing:
- Conversion of i2c calls to subdev calls;
- all subdev calls to call_all();
- Corrected the header file order in cx231xx.h;
- Added tuner frequency setting.
Sri Deevi [Fri, 13 Mar 2009 21:35:14 +0000 (18:35 -0300)]
V4L/DVB (11038): Fix the issue with audio module & correction of Names
The audio module requested in driver differs with module
created by Makefile. Makefile is corrected to create the same module name
required by driver. Also, corrected the strings that shows wrong name.
Janne Grunau [Wed, 1 Apr 2009 11:30:06 +0000 (08:30 -0300)]
V4L/DVB (11351): v4l: use usb_interface for v4l2_device_register
If usb_interface.dev is used as dev parameter for v4l2_device_register
v4l2_dev.name contains the v4l driver/module name and usb device and
interface instead of a simple "usb x-y".
It also matches the recommendation to set the parent devices for usb
drivers.
Signed-off-by: Janne Grunau <j@jannau.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (11349): mx3-camera: adapt the clock definition and the driver to the new clock naming
With the i.MX31 transition to clkdev clock names have changed, but mistakenly
the "mx3-camera.0" has been registered with a non-NULL connection ID, which is
not necessary, since this is the only clock, used by the capture interface
driver. Fix the clock definition and the driver to use NULL as a connection ID.
Stuart Hall [Thu, 2 Apr 2009 08:25:09 +0000 (05:25 -0300)]
V4L/DVB (11345): af9015: support for DigitalNow TinyTwin remote
Patch to provide basic support for DigitalNow TinyTwin Remote. It uses
same remote as TwinHan AzureWave AD-TU700(704J).
Signed-off-by: Stuart Hall <mailing-lists@enginuities.com> Signed-off-by: Antti Palosaari <crope@iki.fi> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (11343): au0828: make i2c clock speed per-board configurable
Setup the i2c clock speed to be definable on a per-board basis. This allows
us to explicitly set the clock speed to 30 KHz on the 950q, and also gets rid
of code which sets it on a basis of what chip the i2c master is talking to
at any given time (which could have caused issues because i2c slaves should
never receive commands at a clock higher than their supported clock speed).
Change the #define entries for the i2c registers to be more meaningful, and
document the valid values for the registers. Note that this changeset makes
*no* functional changes to the code.
Any time a struct (especially one not defined by this driver) is
allocated, we MUST zero its underlying storage. This makes our usage
of the struct predictable and robust against future changes where
fields might be added that we don't know about. Failing to do this
with tuner_setup left the config field uninitialized which then caused
trouble with the tuner type used for HVR-1950 devices.
Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Russell King [Sun, 29 Mar 2009 11:12:27 +0000 (08:12 -0300)]
V4L/DVB (11329): Fix buglets in v4l1 compatibility layer
The following patch fixes a few bugs I've noticed in the V4L1
compatibility layer:
- VIDEO_MODE_AUTO for get/set input ioctls was not being handled
- wrong V4L2 ioctl being used in v4l1_compat_select_tuner
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (11328): Add AVerMedia A310 USB IDs to CE6230 driver.
The CE6230 DVB USB driver works correctly for the AVerMedia A310 USB2.0
DVB-T tuner. Add the required USB ID's and hardware names so that the
driver will handle it.
Signed-off-by: Juan Jesús García de Soria Lucena <skandalfo@gmail.com> Acked-by: Antti Palosaari <crope@iki.fi> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
MT9M001 datasheet says, that the data is ready on the falling edge of the pixel
clock, but the driver wrongly sets the SOCAM_PCLK_SAMPLE_RISING flag. Changing
this doesn't seem to produce any visible difference, still, it is better to
comply to the datasheet.
soc-camera re-uses struct devices multiple times in calls to device_register(),
therefore it has to reset the embedded struct kobject to avoid the "tried to
init an initialized object" error, which then also erases its name. Now with
the transition to kobject's name for device names, we have to re-initialise the
name before each call to device_register().
V4L/DVB (11323): pxa-camera: simplify the .buf_queue path by merging two loops
pxa_dma_update_sg_tail() is called only once, runs exactly the same loop as the
caller and has to recalculate the last element in an sg-list, that the caller
has already calculated. Eliminate redundancy by merging the two loops and
re-using the calculated pointer. This also saves a bit of performance which is
always good during video-capture.
Robert Jarzmik [Tue, 31 Mar 2009 06:44:22 +0000 (03:44 -0300)]
V4L/DVB (11322): pxa_camera: Fix overrun condition on last buffer
The last buffer queued will often overrun, as the DMA chain
is finished, and the time the dma irq handler is activated,
the QCI fifos are filled by the sensor.
The fix is to ignore the overrun condition on the last
queued buffer, and restart the capture only on intermediate
buffers of the chain.
Moreover, a fix was added to the very unlikely condition
where in YUV422P mode, one channel overruns while another
completes at the very same time. The capture is restarted
after the overrun as before, but the other channel
completion is now ignored.
The DMA transfers in pxa_camera showed some weaknesses in
multiple queued buffers context :
- poll/select problem
The bug shows up with capture_example tool from v4l2 hg
tree. The process just "stalls" on a "select timeout".
- multiple buffers DMA starting
When multiple buffers were queued, the DMA channels were
always started right away. This is not optimal, as a
special case appears when the first EOF was not yet
reached, and the DMA channels were prematurely started.
- Maintainability
DMA code was a bit obfuscated. Rationalize the code to be
easily maintainable by anyone.
- DMA hot chaining
DMA is not stopped anymore to queue a buffer, the buffer
is queued with DMA running. As a tribute, a corner case
exists where chaining happens while DMA finishes the
chain, and the capture is restarted to deal with the
missed link buffer.
This patch attemps to address these issues / improvements.
All planes were PAGE aligned (ie. 4096 bytes aligned). This
is not consistent with YUV422 format, which requires Y, U
and V planes glued together. The new implementation forces
the alignement on 8 bytes (DMA requirement), which is almost
always the case (granted by width x height being a multiple
of 8).
The test cases include tests in both YUV422 and RGB565 :
- a picture of size 111 x 111 (cross RAM pages example)
- a picture of size 1023 x 4 in (under 1 RAM page)
- a picture of size 1024 x 4 in (exactly 1 RAM page)
- a picture of size 1025 x 4 in (over 1 RAM page)
- a picture of size 1280 x 1024 (many RAM pages)
Robert Jarzmik [Tue, 31 Mar 2009 06:44:21 +0000 (03:44 -0300)]
V4L/DVB (11319): pxa_camera: Enforce YUV422P frame sizes to be 16 multiples
Due to DMA constraints, the DMA chain always transfers bytes
from the QCI fifos to memory in 8 bytes units. In planar
formats, that could mean 0 padding between Y and U plane
(and between U and V plane), which is against YUV422P
standard.
Therefore, a frame size is required to be a multiple of 16
(so U plane size is a multiple of 8). It is enforced in
try_fmt() and set_fmt() primitives, be aligning height then
width on 4 multiples as need be, to reach a 16 multiple.
NFS: Fix a double free in nfs_parse_mount_options()
Due to an apparent typo, commit a67d18f89f5782806135aad4ee012ff78d45aae7
(NFS: load the rpc/rdma transport module automatically) lead to the
'proto=' mount option doing a double free, while Opt_mountproto leaks a
string.
ext3: make default data ordering mode configurable
This makes the defautl ext3 data ordering mode (when no explicit
ordering is set) configurable, so as to allow people to default to
'data=writeback' and get the resulting latency improvements.
This is a non-issue if a filesystem has been explicitly set to some
ordering (with 'tune2fs').
Mark Lord [Mon, 6 Apr 2009 19:26:24 +0000 (15:26 -0400)]
sata_mv: workaround errata SATA#13
Add remainder of workaround for errata SATA#13.
This prevents writes of certain adjacent 32-bit registers
from being combined into single 64-bit writes, which might
fail for the affected registers.
Most of sata_mv is already safe from this issue,
but adding this code to mv_write_cached_reg() will
catch the remaining cases and hopefully prevent future ones.
Signed-off-by: Mark Lord <mlord@pobox.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Mark Lord [Mon, 6 Apr 2009 16:30:43 +0000 (12:30 -0400)]
sata_mv: fix irq mask races
Prevent racing on the main interrupt mask during port_start and port_stop.
Otherwise, we end up with IRQs masked on inactive ports,
and hotplug insertions then get missed later on.
Found while debugging (out of tree) target mode operations,
but the bug is present and impacting mainline as well.
This patch should also be considered for -stable.
Signed-off-by: Mark Lord <mlord@pobox.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Mark Lord [Mon, 6 Apr 2009 16:29:49 +0000 (12:29 -0400)]
sata_mv: revert SoC irq breakage
Revert most of commit 6be96ac1d5e4d913e1f48299db083ada5321803b2,
originally from Lennert Buijtenheck (Marvell) and Saeed Bishara (Marvell),
since that commit causes sata_mv to oops at startup on SOC "Kirkwood".
The SOC variants do not have the hpriv->irq_{cause,mask}_ofs registers,
so don't try to write to them!
This patch should also be considered for -stable.
Reported-by: Maxime Bizon <mbizon@freebox.fr> Signed-off-by: Mark Lord <mlord@pobox.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
If an ipv4 packet (not locally generated with IP_DF flag not set) bigger
than mtu size is supposed to go via a xfrm ipv6 tunnel, the packetsize
check in xfrm4_tunnel_check_size() is omited and ipv6 drops the packet
without sending a notice to the original sender of the ipv4 packet.
Another issue is that ipv4 connection tracking does reassembling of
incomming fragmented packets. If such a reassembled packet is supposed to
go via a xfrm ipv6 tunnel it will be droped, even if the original sender
did proper fragmentation.
According to RFC 2473 (section 7) tunnel ipv6 packets resulting from the
encapsulation of an original packet are considered as locally generated
packets. If such a packet passed the checks in xfrm{4,6}_tunnel_check_size()
fragmentation is allowed according to RFC 2473 (section 7.1/7.2).
This patch sets skb->local_df in xfrm6_prepare_output() to achieve
fragmentation in this case.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jean Delvare [Mon, 6 Apr 2009 02:27:10 +0000 (02:27 +0000)]
ibm_newemac: Fix dangerous struct assumption
The ibm_newemac driver includes code which assumes that the
work_struct which is included in every delayed_work is the first
member of that structure. This is currently the case but might change
in the future, so use to_delayed_work() instead, which doesn't make
such an assumption.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Cc: Josh Boyer <jwboyer@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
gigaset: in file ops, check for device disconnect before anything else
When the device is disconnected, the dev structure goes away, so
trying to report another error via dev_printk is bound to oops.
To avoid that, first check whether the device is still connected
and return quietly if it isn't.
Impact: error handling
Signed-off-by: Tilman Schmidt <tilman@imap.cc> Reported-by: Paul Bolle <pebolle@tiscali.nl> Tested-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: David S. Miller <davem@davemloft.net>
bas_gigaset: use tasklet_hi_schedule for timing critical tasklets
The tasklets for isochronous data transfer need to run within 8 msec
to avoid over/underruns, so schedule them with high priority to fix
reported issues with occasional over/underruns.
Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
David Milburn [Fri, 3 Apr 2009 20:36:41 +0000 (15:36 -0500)]
libata: ahci enclosure management bios workaround
During driver initialization ahci_start_port may not be able
to turn LEDs off because the hardware may still be transmitting
a message. And since the BIOS may not be setting the LEDs to
off the drive LEDs may end up in a fault state. This has
been seen on ICH9r and ICH10r when configured in AHCI mode
instead of RAID mode, this patch doesn't key off a specific
set of device IDs but will give the EM transmit bit a chance
to clear if busy.
Signed-off-by: David Milburn <dmilburn@redhat.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Matthew Wilcox [Thu, 2 Apr 2009 14:37:25 +0000 (10:37 -0400)]
ata: Add TRIM infrastructure
This is common code shared between the IDE and libata implementations
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com> Signed-off-by: Matthew Wilcox <willy@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Tejun Heo [Tue, 31 Mar 2009 01:44:34 +0000 (10:44 +0900)]
ata_piix: VGN-BX297XP wants the controller power up on suspend
Sony VGN-BX297XP fails suspend if the controller is powered down when
calling into ACPI suspend. Add the machine to piix_broken_suspend
list.
This problem was reported by GNUtoo@no-log.org on bko#10293.
Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rafael J. Wysocki <rjw@sisk.pl> Reported-by: GNUtoo@no-log.org Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
projects / linux-2.6-omap-h63xx.git / log