platform/kernel/kernel-mfld-blackbay.git
17 years agoV4L/DVB (6316): Change list_for_each+list_entry to list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:43 +0000 (05:37 -0300)]
V4L/DVB (6316): Change list_for_each+list_entry to list_for_each_entry

The rest of V4L files.

There is one list_for_each+list_entry in cpia_pp.c that
wasn't changed because it expects the loop iterator to remain NULL if
the list is empty.

A bug in vivi is fixed; the 'safe' version needs to be used because the loop
deletes the list entries.

Simplify a second loop in vivi and get rid if an un-used variable in that loop.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6315): pvrusb2: Change list_for_each+list_entry to list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:42 +0000 (05:37 -0300)]
V4L/DVB (6315): pvrusb2: Change list_for_each+list_entry to list_for_each_entry

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Reviewed-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6314): saa7134: Replace list_for_each+list_entry with list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:42 +0000 (05:37 -0300)]
V4L/DVB (6314): saa7134: Replace list_for_each+list_entry with list_for_each_entry

A couple loops weren't changed because they expected the loop iterator
to be left as NULL if the list was empty.  Maybe the code should just
check for that first, then loop?

Adjust some of the loop logic to be simpler.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6313): ivtv: Replace list_for_each+list_entry with list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:41 +0000 (05:37 -0300)]
V4L/DVB (6313): ivtv: Replace list_for_each+list_entry with list_for_each_entry

Also fixed a few cases of cut&paste errors where 'buf' would be set to the
first entry in the list prior to be used as the loop iterator.  In one case
the value of buf was used before it was changed, but the rest were
unnecessary.

There was one list_for_each+list_entry loop that wasn't changed, since it
depending on the loop iterator being left as NULL if the list was empty.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
CC: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6312): cx88: Replace list_for_each+list_entry with list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:40 +0000 (05:37 -0300)]
V4L/DVB (6312): cx88: Replace list_for_each+list_entry with list_for_each_entry

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6311): dvb: Replace list_for_each+list_entry with list_for_each_entry
Trent Piepho [Wed, 10 Oct 2007 08:37:39 +0000 (05:37 -0300)]
V4L/DVB (6311): dvb: Replace list_for_each+list_entry with list_for_each_entry

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6308): V4L: zc0301, remove bad usage of ERESTARTSYS
Jiri Slaby [Wed, 10 Oct 2007 01:15:27 +0000 (22:15 -0300)]
V4L/DVB (6308): V4L: zc0301, remove bad usage of ERESTARTSYS

zc0301, remove bad usage of ERESTARTSYS

down_read_trylock can't be interrupted and so ERESTARTSYS would reach
userspace, which is not permitted. Change it to EAGAIN

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Luca Risolia <luca.risolia@studio.unibo.it>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6307): V4L: w9968cf, remove bad usage of ERESTARTSYS
Jiri Slaby [Wed, 10 Oct 2007 01:03:12 +0000 (22:03 -0300)]
V4L/DVB (6307): V4L: w9968cf, remove bad usage of ERESTARTSYS

w9968cf, remove bad usage of ERESTARTSYS

down_read_trylock can't be interrupted and so ERESTARTSYS would reach
userspace, which is not permitted. Change it to EAGAIN

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Luca Risolia <luca.risolia@studio.unibo.it>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6306): Few clenups for saa7134 resume code
Maxim Levitsky [Mon, 8 Oct 2007 15:07:35 +0000 (12:07 -0300)]
V4L/DVB (6306): Few clenups for saa7134 resume code

*Disable DMA explictly on suspend
*Enable DMA on resume, after all buffers were configured
*Disable overlay on resume - apps should enable it when X is resumed

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6305): V4L: videobuf-core.c avoid NULL dereferences in videobuf-core
Brandon Philips [Fri, 5 Oct 2007 19:26:27 +0000 (16:26 -0300)]
V4L/DVB (6305): V4L: videobuf-core.c avoid NULL dereferences in videobuf-core

The return value of videobuf_alloc() is unchecked but this function will
return NULL on an error.  Check for NULL and make videobuf_reqbufs()
return the number of successfully allocated buffers.

Also, fix saa7146_video.c and bttv-driver.c to use this returned
buffer count.

Tested against the vivi driver.  Not tested against saa7146 or bt8xx
devices.

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6301): pvrusb: Update DEBUGIFC sysfs to kernel 2.6.13+
Trent Piepho [Mon, 8 Oct 2007 22:05:28 +0000 (19:05 -0300)]
V4L/DVB (6301): pvrusb: Update DEBUGIFC sysfs to kernel 2.6.13+

The prototypes for the show and store methods of a device_attribute changed in
kernel 2.6.13, but the code in pvrusb2 was never updated.  I guess the
DEBUGIFC stuff isn't used much....

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6300): CodingStyle cleanup
Trent Piepho [Mon, 8 Oct 2007 21:36:21 +0000 (18:36 -0300)]
V4L/DVB (6300): CodingStyle cleanup

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6299): dvb: Add dependencies for VIDEOBUF_DVB
Trent Piepho [Mon, 8 Oct 2007 21:34:11 +0000 (18:34 -0300)]
V4L/DVB (6299): dvb: Add dependencies for VIDEOBUF_DVB

It needs to select VIDEOBUF_GEN and VIDEOBUF_DMA_SG, since it uses those
modules.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6297): cx23885: remove wrong Kconfig selection of VIDEOBUF
Michael Krufky [Mon, 8 Oct 2007 20:18:17 +0000 (17:18 -0300)]
V4L/DVB (6297): cx23885: remove wrong Kconfig selection of VIDEOBUF

fix bad Kconfig dependency of cx23885 module.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6296): dib0700: add support for AverMedia DVB-T Express card
Joachim Steiger [Mon, 8 Oct 2007 23:08:46 +0000 (20:08 -0300)]
V4L/DVB (6296): dib0700: add support for AverMedia DVB-T Express card

add Avermedia dvb-t express card 34 (usb2) id

Signed-off-by: Joachim Steiger <roh@hyte.de>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6295): saa7134: add autodetection for KWorld ATSC-115
Eric Sandeen [Mon, 8 Oct 2007 19:13:02 +0000 (16:13 -0300)]
V4L/DVB (6295): saa7134: add autodetection for KWorld ATSC-115

Recognize the KWorld ATSC115 PCI ID as a hardware clone of the ATSC110.

Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6293): V4L: convert struct class_device to struct device
Kay Sievers [Mon, 8 Oct 2007 19:26:13 +0000 (16:26 -0300)]
V4L/DVB (6293): V4L: convert struct class_device to struct device

The currently used "struct class_device" will be removed from the
kernel. Here is a patch that converts all users in drivers/media/video/
to struct device.

Reviewed-by: Thierry Merle <thierry.merle@free.fr>
Reviewed-by: Mike Isely <isely@pobox.com>
Reviewed-by: Luca Risolia <luca.risolia@studio.unibo.it>
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6292): videobuf_core init always require callback implementation
Mauro Carvalho Chehab [Mon, 8 Oct 2007 15:20:02 +0000 (12:20 -0300)]
V4L/DVB (6292): videobuf_core init always require callback implementation

In the past, videobuf_queue_init were used to initialize PCI DMA videobuffers.
This patch renames it, to avoid confusion with the previous kernel API, doing:
s/videobuf_queue_init/void videobuf_queue_core_init/

Also, the operations is now part of the function parameter. The function will
also add a test if this is defined, otherwise producing BUG.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6291): Fix: avoid oops on some SMP machines
Mauro Carvalho Chehab [Mon, 8 Oct 2007 14:48:57 +0000 (11:48 -0300)]
V4L/DVB (6291): Fix: avoid oops on some SMP machines

This workaround fix a bug that happens on some SMP machines. On those machines,
videobuf_iolock is called too soon, before file .mmap handler. This patch calls
the scheduler before iolocking, allowing it to properly call the pending mmap.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6290): remove videobuf_set_pci_ops
Mauro Carvalho Chehab [Mon, 8 Oct 2007 14:43:49 +0000 (11:43 -0300)]
V4L/DVB (6290): remove videobuf_set_pci_ops

Before the videobuf redesign, a procedure for re-using videobuf without PCI
scatter/gather where provided by changing the pci-dependent operations by
other operations.

With the newer approach, those methods are obsolete and can safelly be removed.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6289): Remove reference to dead CONFIG_UST and ust.h header
Robert P. J. Day [Mon, 8 Oct 2007 12:20:49 +0000 (09:20 -0300)]
V4L/DVB (6289): Remove reference to dead CONFIG_UST and ust.h header

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6287): Fix DMA Scatter/Gather constructor
Mauro Carvalho Chehab [Mon, 8 Oct 2007 12:04:09 +0000 (09:04 -0300)]
V4L/DVB (6287): Fix DMA Scatter/Gather constructor

cx23885 driver were converted to use the newer videobuf support. Unfortunately,
the constructor weren't changed. This causes an oops, since the abstract methods
(implemented as callbacks) aren't defined.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
17 years agoV4L/DVB (6286): Add support for MSI TV @nywhere A/D NB
Mauro Carvalho Chehab [Sun, 7 Oct 2007 05:17:09 +0000 (02:17 -0300)]
V4L/DVB (6286): Add support for MSI TV @nywhere A/D NB

This is a Lifeview hybrid OEM board.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6285): Remove pointless kmalloc() return value cast in Zoran PCI controller...
Jesper Juhl [Fri, 24 Aug 2007 00:22:15 +0000 (21:22 -0300)]
V4L/DVB (6285): Remove pointless kmalloc() return value cast in Zoran PCI controller driver

No need to cast the void pointer returned by kmalloc() in
drivers/media/video/zoran_driver.c::v4l_fbuffer_alloc().

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6284): cx23885: Update to new videobuf code
Trent Piepho [Thu, 4 Oct 2007 08:28:45 +0000 (05:28 -0300)]
V4L/DVB (6284): cx23885: Update to new videobuf code

cx23885 was still uses the old video-buf includes and code, which would only
`work' if one happened to be compiling against a kernel that had the old
headers.  Even then, it wouldn't actually work, it would just compile without
errors.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6283): videobuf: Remove references to old Kconfig option name
Trent Piepho [Thu, 4 Oct 2007 07:54:11 +0000 (04:54 -0300)]
V4L/DVB (6283): videobuf: Remove references to old Kconfig option name

CONFIG_VIDEO_BUF_DVB became CONFIG_VIDEOBUF_DVB.

But in these cases, it makes more sense to use CONFIG_VIDEO_SAA7134_DVB
or CONFIG_VIDEO_CX88_DVB_MODULE depending on the driver.

The reference in cx23885.h should just be removed, as the code there needs to
be included if DVB is on or off.  I do not think you can even compile the
cx23885 driver without DVB.  It's clearly just leftover from when the file was
obvious copied from the cx88 driver (which is not mentioned in the copyright
BTW).

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6279): en_50221: convert to kthread API
Christoph Hellwig [Wed, 3 Oct 2007 14:23:01 +0000 (11:23 -0300)]
V4L/DVB (6279): en_50221: convert to kthread API

Here's an attempted update to the full kthread API + wake_up_process:

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
CC: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6278): Buf: fix typo that caused data loss when readng streams from device
Maxim Levitsky [Tue, 2 Oct 2007 12:03:39 +0000 (09:03 -0300)]
V4L/DVB (6278): Buf: fix typo that caused data loss when readng streams from device

If videobuf_read_stream reads two or more buffers it was overwriting the first one

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6277): vivi cleanup: remove the unused vars
Mauro Carvalho Chehab [Fri, 28 Sep 2007 00:00:50 +0000 (21:00 -0300)]
V4L/DVB (6277): vivi cleanup: remove the unused vars

Vivi driver is now simpler. This patch removes the now unused vars.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6276): V4L: videobuf-core.c lock before streaming check
Brandon Philips [Thu, 27 Sep 2007 23:55:28 +0000 (20:55 -0300)]
V4L/DVB (6276): V4L: videobuf-core.c lock before streaming check

The reading/streaming fields are used for mutual exclusion of the queue and
should be protected by the queue lock.

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6275): V4L: vivi.c remove the "resource" locking
Brandon Philips [Thu, 27 Sep 2007 23:55:17 +0000 (20:55 -0300)]
V4L/DVB (6275): V4L: vivi.c remove the "resource" locking

The "resource" locking in vivi isn't needed since
streamon/streamoff/read_stream do mutual exclusion using
q->reading/q->streaming.

Plus it is sort of broken:

a) res_locked() use in vivi_read() is racey.
b) res_free() calls mutex_lock twice causing streamoff to break

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6274): V4L: vivi.c replace logic in vivi_poll with videobuf_poll_stream
Brandon Philips [Thu, 27 Sep 2007 23:55:02 +0000 (20:55 -0300)]
V4L/DVB (6274): V4L: vivi.c replace logic in vivi_poll with videobuf_poll_stream

Since vivi is using videobuf_read_stream() it can use videobuf_poll_stream()
now.

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6273): V4L: vivi.c vidioc_try_fmt_cap() negotiate a valid field
Brandon Philips [Thu, 27 Sep 2007 23:54:52 +0000 (20:54 -0300)]
V4L/DVB (6273): V4L: vivi.c vidioc_try_fmt_cap() negotiate a valid field

If the client provides V4L2_FIELD_ANY vivi should return a valid field :)

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6272): V4L: properly fix support for capturing interlaced video in saa7134
Maxim Levitsky [Thu, 27 Sep 2007 23:44:39 +0000 (20:44 -0300)]
V4L/DVB (6272): V4L: properly fix support for capturing interlaced video in saa7134

By "capturing interlaced video" I mean that card ensures that top field
is really top and vice versa (I think it takes the filed ID from signal)
Properly turn on/off that support depending on signal state

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6271): V4L: Add basic support for suspend/resume for saa7134
Maxim Levitsky [Thu, 27 Sep 2007 23:34:25 +0000 (20:34 -0300)]
V4L/DVB (6271): V4L: Add basic support for suspend/resume for saa7134

This adds support for suspend/resume for core of saa7134

Should fix bug#7220

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6270): V4L: Honor dev->ctl_invert when setting up the decoder in saa7134
Maxim Levitsky [Thu, 27 Sep 2007 23:34:20 +0000 (20:34 -0300)]
V4L/DVB (6270): V4L: Honor dev->ctl_invert when setting up the decoder in saa7134

When user sets dev->ctl_invert, driver writes negative values to
SAA7134_DEC_LUMA_CONTRAST and SAA7134_DEC_CHROMA_SATURATION,
but general code that initializes decorder ignores that

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6269): V4L: Fix a "scheduling while atomic" bug in saa7134
Maxim Levitsky [Thu, 27 Sep 2007 23:34:15 +0000 (20:34 -0300)]
V4L/DVB (6269): V4L: Fix a "scheduling while atomic" bug in saa7134

set_tvnorm can sleep in saa7134_i2c_xfer
(it will be called through tuner code)
but code calls it under spinlock. Fix that

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6268): V4L: Fix a lock inversion in generic videobuf code
Maxim Levitsky [Thu, 27 Sep 2007 23:34:09 +0000 (20:34 -0300)]
V4L/DVB (6268): V4L: Fix a lock inversion in generic videobuf code

videobuf_qbuf takes q->lock, and then calls
q->ops->buf_prepare which by design in all drivers calls
videobuf_iolock which calls videobuf_dma_init_user and this
takes current->mm->mmap_sem

on the other hand if user calls mumap from other thread, sys_munmap
takes current->mm->mmap_sem and videobuf_vm_close takes q->lock

Since this can occur only for V4L2_MEMORY_MMAP buffers, take
current->mm->mmap_sem in qbuf, before q->lock, and don't take
current->mm->mmap_sem videobuf_dma_init_user for those buffers

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6266): videobuf cleanup: mmap check is common to all videobuf. Make it at...
Mauro Carvalho Chehab [Thu, 27 Sep 2007 21:25:44 +0000 (18:25 -0300)]
V4L/DVB (6266): videobuf cleanup: mmap check is common to all videobuf. Make it at core

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6265): Prevent for calling mmap_free without an allocated buffer
Mauro Carvalho Chehab [Tue, 25 Sep 2007 14:53:24 +0000 (11:53 -0300)]
V4L/DVB (6265): Prevent for calling mmap_free without an allocated buffer

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6264): Make the vertical lines to move
Mauro Carvalho Chehab [Sat, 22 Sep 2007 05:01:33 +0000 (02:01 -0300)]
V4L/DVB (6264): Make the vertical lines to move

While this is not the standard color bar behaviour, having some movement
there allows to check if buffers are being properly handled.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6263): Fix buffer release code
Mauro Carvalho Chehab [Thu, 6 Sep 2007 23:11:35 +0000 (20:11 -0300)]
V4L/DVB (6263): Fix buffer release code

Release code should happen before the cleaning of map variable.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6262): An allocation error message were being printed as a debug msg
Mauro Carvalho Chehab [Thu, 6 Sep 2007 22:08:24 +0000 (19:08 -0300)]
V4L/DVB (6262): An allocation error message were being printed as a debug msg

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6261): Cleans mem->vmalloc after vfree
Mauro Carvalho Chehab [Thu, 6 Sep 2007 21:55:07 +0000 (18:55 -0300)]
V4L/DVB (6261): Cleans mem->vmalloc after vfree

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6260): Fix Kconfig dependency
Mauro Carvalho Chehab [Tue, 2 Oct 2007 14:16:16 +0000 (11:16 -0300)]
V4L/DVB (6260): Fix Kconfig dependency

Thanks to Michael Krufky for pointing this to me.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6259): Fix vivi poll() method
Mauro Carvalho Chehab [Mon, 27 Aug 2007 10:37:34 +0000 (07:37 -0300)]
V4L/DVB (6259): Fix vivi poll() method

Due to the replace of videobuf_read_one to videobuf_read_stream, poll()
method implementation is wrong. This fixes poll() implementation, making
read of /dev/video? to work again.

With this method, an USB driver can use video-buf, without needing to
request memory from the DMA-safe area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6257): Rename video-buf-dvb to videobuf-dvb to be consistent with the other...
Mauro Carvalho Chehab [Thu, 23 Aug 2007 21:12:08 +0000 (18:12 -0300)]
V4L/DVB (6257): Rename video-buf-dvb to videobuf-dvb to be consistent with the other patches

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6256): Remove the obsolete video-buf module
Mauro Carvalho Chehab [Thu, 23 Aug 2007 19:31:15 +0000 (16:31 -0300)]
V4L/DVB (6256): Remove the obsolete video-buf module

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6255): Convert vivi to use videobuf-vmalloc
Mauro Carvalho Chehab [Fri, 3 Aug 2007 02:31:54 +0000 (23:31 -0300)]
V4L/DVB (6255): Convert vivi to use videobuf-vmalloc

This patch removes the usage of videobuf-dma-sg from vivi driver, using
instead videobuf-vmalloc. This way, vivi will be useful for testing the
newer method. Reverting this patch won't hurt vivi, since both methods
work fine.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6254): Add videobuf-vmalloc
Mauro Carvalho Chehab [Fri, 3 Aug 2007 02:31:33 +0000 (23:31 -0300)]
V4L/DVB (6254): Add videobuf-vmalloc

Adds a newer videobuf-vmalloc module. This module uses the same
videobuf controls, but implements memory allocation based on vmalloc
methods.

With this method, an USB driver can use video-buf, without needing to
request memory from the DMA-safe area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6253): Convert vivi to use the newer videobuf-dma-sg module
Mauro Carvalho Chehab [Thu, 23 Aug 2007 19:41:14 +0000 (16:41 -0300)]
V4L/DVB (6253): Convert vivi to use the newer videobuf-dma-sg module

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6252): Adapt drivers to use the newer videobuf modules
Mauro Carvalho Chehab [Thu, 23 Aug 2007 19:37:49 +0000 (16:37 -0300)]
V4L/DVB (6252): Adapt drivers to use the newer videobuf modules

PCI-dependent videobuf_foo methods were renamed as videobuf_pci_foo.

Also, videobuf_dmabuf is now part of videobuf-dma-sg private struct.
So, to access it, a subroutine call is needed.

This patch renames all occurences of those function calls to be
consistent with the video-buf split.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6251): Replace video-buf to a more generic approach
Mauro Carvalho Chehab [Thu, 23 Aug 2007 19:26:14 +0000 (16:26 -0300)]
V4L/DVB (6251): Replace video-buf to a more generic approach

video-buf currently does two different tasks:

- Manages video buffers with a common code that allows
  implementing all the V4L2 different modes of buffering;

- Controls memory allocations

While the first task is generic, the second were written to support PCI DMA
Scatter/Gather needs. The original approach can't even work for those
video capture hardware that don't support scatter/gather.

I did one approach to make it more generic. While the approach worked
fine for vivi driver, it were not generic enough to handle USB needs.

This patch creates two different modules, one containing the generic
video buffer handling (videobuf-core) and another with PCI DMA S/G.
After this patch, it would be simpler to write an USB video-buf and a
non-SG DMA module.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981
Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
17 years agoV4L/DVB (6250): bw-qcam use data_reverse instead of manually poking the control regis...
Brett Warden [Tue, 2 Oct 2007 20:37:21 +0000 (17:37 -0300)]
V4L/DVB (6250): bw-qcam use data_reverse instead of manually poking the control register fix

coding-style repairs

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6249): Add Typhoon Tv-Tuner PCI to bttv-cards.c
Sascha Sommer [Tue, 2 Oct 2007 15:23:39 +0000 (12:23 -0300)]
V4L/DVB (6249): Add Typhoon Tv-Tuner PCI to bttv-cards.c

Adds an entry for the Typhoon Tv-Tuner PCI to bttv-cards.c

Signed-off-by: Sascha Sommer <saschasommer@freenet.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6247): Fix bug #8689: Fixes IR stop/start during suspend/resume
Mauro Carvalho Chehab [Mon, 1 Oct 2007 11:51:39 +0000 (08:51 -0300)]
V4L/DVB (6247): Fix bug #8689: Fixes IR stop/start during suspend/resume

IR workqueue should be disabled during suspend. This avoids some troubles, like
the one reported on bug #8689:

"The Hauppauge HVR 1100 ir-remote control does not work after resume from
suspend to ram or disk."

This patch disables IR before suspending, re-enabling it after resume.

Thanks to Peter Poklop <Peter.Poklop@gmx.at> for reporting it and helping with
the fix.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Reviewed-by: Peter Poklop <Peter.Poklop@gmx.at>
17 years agoV4L/DVB (6246): GemTek Radio card - Control Word
Trent Piepho [Mon, 1 Oct 2007 03:38:30 +0000 (00:38 -0300)]
V4L/DVB (6246):  GemTek Radio card - Control Word

Redoes the way the control word is stored and set.
The existing code was a lot more complicated than it needed to be.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Douglas Schilling Landgraf <dougsland@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Reviewed-by: Pekka Seppänen <pexu@kapsi.fi>
17 years agoV4L/DVB (6245): GemTek Radio card - frequency calculation
Trent Piepho [Mon, 1 Oct 2007 03:32:25 +0000 (00:32 -0300)]
V4L/DVB (6245):  GemTek Radio card - frequency calculation

Frequency calculation to use better math.  It's still the same
IF offset and step size (which are not the same as the datasheet says) as
the code was before.  It's just more efficient and accurate.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Reviewed-by: Pekka Seppänen <pexu@kapsi.fi>
Signed-off-by: Douglas Schilling Landgraf <dougsland@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6244): [PATCH 1/2] GemTek Radio card
Pekka Seppanen [Mon, 1 Oct 2007 03:27:55 +0000 (00:27 -0300)]
V4L/DVB (6244): [PATCH 1/2] GemTek Radio card

Code cleanup for GemTek Radio card driver. Removed unnecessary / invalid
I/O commands and rewrote code for tuning on-board BU2614FS chip. Adds
several new module params for power users. Includes automatic device
probing.

Signed-off-by: Pekka Seppanen <pexu@kapsi.fi>
Signed-off-by: Douglas Schilling Landgraf <dougsland@gmail.com>
Reviewed-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6243): [PATCH 2/2] GemTek Radio card
Pekka Seppanen [Mon, 1 Oct 2007 00:49:01 +0000 (21:49 -0300)]
V4L/DVB (6243): [PATCH 2/2] GemTek Radio card

Details now match with radio-gemtek.c, eg. no more different ports.
Included a short note about cards that should be compatible with
radio-gemtek module.

Signed-off-by: Pekka Seppanen <pexu@kapsi.fi>
Signed-off-by: Douglas Schilling Landgraf <dougsland@gmail.com>
Reviewed-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6241): fix typo in DVB_PLL Kconfig help text
Michael Krufky [Thu, 27 Sep 2007 14:37:19 +0000 (11:37 -0300)]
V4L/DVB (6241): fix typo in DVB_PLL Kconfig help text

s/driver/drives/1

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6238): bw-qcam: use data_reverse instead of manually poking the control...
Brett Warden [Fri, 28 Sep 2007 06:19:04 +0000 (03:19 -0300)]
V4L/DVB (6238): bw-qcam: use data_reverse instead of manually poking the control register

Fixes use of parport_write_control() to match the newer interface that
requires explicit parport_data_reverse() and parport_data_forward() calls.
This eliminates the following error message and restores the original
intended behavior:

parport0 (bw-qcam): use data_reverse for this!

Also increases threshold in qc_detect() from 300 to 400, as my camera often
results in a count of approx 330.  Added a kernel error message to indicate
detection failure.

Thanks Ray and Randy for your comments, and for pointing out that I
needed to reset the port to forward mode!

Signed-off-by: Brett T. Warden <brett.warden@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6237): Oops in pwc v4l driver
Oliver Neukum [Wed, 26 Sep 2007 13:19:01 +0000 (10:19 -0300)]
V4L/DVB (6237): Oops in pwc v4l driver

The pwc driver is defficient in locking, which can trigger an oops
when disconnecting.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
CC: Luc Saillard <luc@saillard.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6235): cafe_ccic: default to allocating DMA buffers at probe time
Andres Salomon [Wed, 19 Sep 2007 05:44:18 +0000 (02:44 -0300)]
V4L/DVB (6235): cafe_ccic: default to allocating DMA buffers at probe time

By default, we allocate DMA buffers when actually reading from the video
capture device.  On a system with 128MB or 256MB of ram, it's very easy
for that memory to quickly become fragmented.  We've had users report
having 30+MB of memory free, but the cafe_ccic driver is still unable to
allocate DMA buffers.

Our workaround has been to make use of the 'alloc_bufs_at_load' parameter
to allocate DMA buffers during device probing.  This patch makes DMA
buffer allocation happen during device probe by default, and changes
the parameter to 'alloc_bufs_at_read'.  The camera hardware is there,
if the cafe_ccic driver is enabled/loaded it should do its best to ensure
that the camera is actually usable; delaying DMA buffer allocation
saves an insignicant amount of memory, and causes the driver to be much
less useful.

Signed-off-by: Andres Salomon <dilinger@debian.org>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6232): dvb-pll: report whether input rf will be autoselected or set via...
Michael Krufky [Sun, 9 Sep 2007 08:16:34 +0000 (05:16 -0300)]
V4L/DVB (6232): dvb-pll: report whether input rf will be autoselected or set via insmod option

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6231): dvb-pll: always show pll name if forced via insmod option
Michael Krufky [Sun, 9 Sep 2007 08:08:30 +0000 (05:08 -0300)]
V4L/DVB (6231): dvb-pll: always show pll name if forced via insmod option

The name of the pll will be shown if forced via insmod option,
or if debug is enabled.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6230): dvb-pll: add module option to force dvb-pll desc id (for debug use...
Michael Krufky [Fri, 7 Sep 2007 21:27:43 +0000 (18:27 -0300)]
V4L/DVB (6230): dvb-pll: add module option to force dvb-pll desc id (for debug use only)

Add a module option to force the dvb-pll module to use an alternate dvb-pll
description without having to recompile the kernel.

Having a module option like this is useful in some cases, where the vendor
may release an alternate revision of the hardware using a different tuner,
but without changing the pci subsystem / usb device ids.

This option is intended for debugging purposes _only_.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6229): dvb-pll: increase DVB_PLL_MAX to 64
Michael Krufky [Sun, 9 Sep 2007 16:00:45 +0000 (13:00 -0300)]
V4L/DVB (6229): dvb-pll: increase DVB_PLL_MAX to 64

Increased DVB_PLL_MAX from 16 to a figure that would never be reached in a
practical sense.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6228): dvb-pll: add module option to specify rf input
Michael Krufky [Fri, 7 Sep 2007 21:19:57 +0000 (18:19 -0300)]
V4L/DVB (6228): dvb-pll: add module option to specify rf input

Add a module option to dvb-pll, called "input" to specify which rf
input to use on devices with multiple rf inputs.  If the module option
is not specified, then the driver will autoselect the rf input, as per
previous behavior.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6227): dvb-pll: store instance ID in dvb_pll_priv structure
Michael Krufky [Fri, 7 Sep 2007 21:11:15 +0000 (18:11 -0300)]
V4L/DVB (6227): dvb-pll: store instance ID in dvb_pll_priv structure

Store an instance ID in the dvb_pll_priv structure, so that module options
specific to a given pll may be used by the functions within the driver.

When debug is turned on, print a message indicating which pll was attached
and it's instance id.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6226): dvb-pll: pass fe pointer into dvb_pll_configure() and set() functions
Michael Krufky [Fri, 7 Sep 2007 21:03:58 +0000 (18:03 -0300)]
V4L/DVB (6226): dvb-pll: pass fe pointer into dvb_pll_configure() and set() functions

The pll-specific set() function will need access to the dvb_pll_priv
structure for new functionality.  This patch gives access to this
structure to the required functions.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6225): AverTV Studio 307 has only one composite input
Stas Sergeev [Mon, 17 Sep 2007 18:34:09 +0000 (15:34 -0300)]
V4L/DVB (6225): AverTV Studio 307 has only one composite input

AverTV Studio 307 has only one composite input.

Signed-off-by: Stas Sergeev <stsp@aknet.ru>
Acked-by: Nickolay V. Shmyrev <nshmyrev@yandex.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6224): VIDEO_CX23885 depends on DVB_CORE
Michael Krufky [Tue, 18 Sep 2007 16:25:45 +0000 (13:25 -0300)]
V4L/DVB (6224): VIDEO_CX23885 depends on DVB_CORE

Fix the following dependency issue:

ERROR: "dvb_dmx_init" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_unregister_adapter" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_register_frontend" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_unregister_frontend" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_net_release" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_frontend_detach" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_dmxdev_release" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_dmx_swfilter" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_net_init" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_dmx_release" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_register_adapter" [drivers/media/video/video-buf-dvb.ko] undefined!
ERROR: "dvb_dmxdev_init" [drivers/media/video/video-buf-dvb.ko] undefined!

ERROR: "mt2131_attach" [drivers/media/video/cx23885/cx23885.ko] undefined!
ERROR: "s5h1409_attach" [drivers/media/video/cx23885/cx23885.ko] undefined!

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6222): select DVB_PLL if !DVB_FE_CUSTOMISE for FusionHDTV5 Express
Michael Krufky [Sat, 15 Sep 2007 21:45:10 +0000 (18:45 -0300)]
V4L/DVB (6222): select DVB_PLL if !DVB_FE_CUSTOMISE for FusionHDTV5 Express

VIDEO_CX23885 must select DVB_PLL if !DVB_FE_CUSTOMISE for FusionHDTV5 Express

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6221): budget-ci: select TT keymap for DVB-S TT 1500
Oliver Endriss [Tue, 18 Sep 2007 01:17:12 +0000 (22:17 -0300)]
V4L/DVB (6221): budget-ci: select TT keymap for DVB-S TT 1500

DVB-S TT 1500 must use the TT keymap.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6220): TCM825x: Add support for vidioc_int_g_ifparm call, small cleanups
Sakari Ailus [Thu, 30 Aug 2007 12:20:42 +0000 (09:20 -0300)]
V4L/DVB (6220): TCM825x: Add support for vidioc_int_g_ifparm call, small cleanups

vidioc_int_g_ifparm returns platform-specific information about the
interface settings used by the sensor. Support for [gs]_ext_clk has
been removed.

Fix indentation and remove useless & characters.

Remove experiment for typechecking slave callback function arguments.

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6219): V4L: Int if: Remove [gs]_ext_clk
Sakari Ailus [Thu, 30 Aug 2007 12:20:41 +0000 (09:20 -0300)]
V4L/DVB (6219): V4L: Int if: Remove [gs]_ext_clk

Remove support for g_ext_clk and s_ext_clk. The same functionality is
now handled by g_ifparm.

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6218): V4L: Int if: Use -ENOIOCTLCMD for nonexistent ioctls
Sakari Ailus [Thu, 30 Aug 2007 12:20:40 +0000 (09:20 -0300)]
V4L/DVB (6218): V4L: Int if: Use -ENOIOCTLCMD for nonexistent ioctls

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6217): V4L: Int if: Add BT.656 interface support
Sakari Ailus [Thu, 30 Aug 2007 12:20:39 +0000 (09:20 -0300)]
V4L/DVB (6217): V4L: Int if: Add BT.656 interface support

This patch adds BT.656 interface settings for [gs]_ifparm.

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6216): V4L: Int if: add vidioc_int_g_ifparm, other updates
Sakari Ailus [Thu, 30 Aug 2007 12:20:38 +0000 (09:20 -0300)]
V4L/DVB (6216): V4L: Int if: add vidioc_int_g_ifparm, other updates

vidioc_int_g_ifparm can be used to obtain hardware-specific information
about the interface used by the slave.

Rearrange v4l2-int-device.h as well.

Also remove useless & characters.

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6215): Bugfix for saa6588.c, add forgotten spin_lock_init()
Hans-Jürgen Koch [Wed, 12 Sep 2007 19:20:43 +0000 (16:20 -0300)]
V4L/DVB (6215): Bugfix for saa6588.c, add forgotten spin_lock_init()

There's a serious bug in saa6588.c, it uses a non-initialized spin_lock.
Funny thing is that it works fine with bttv, but completly freezes the
machine if e.g. saa7134 is loaded.

Thanks to Derek Philip for reporting this bug on the rdsd-devel list.

This patch adds the missing spin_lock_init().

Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6214): usbvision: Don't support I2C_M_REV_DIR_ADDR
Jean Delvare [Sun, 9 Sep 2007 09:17:44 +0000 (06:17 -0300)]
V4L/DVB (6214): usbvision: Don't support I2C_M_REV_DIR_ADDR

I2C adapters should only support I2C_M_REV_DIR_ADDR if they really have
to (i.e. if they are connected to a broken I2C device which needs this
deviation from the standard I2C protocol.) As no media chip driver
uses I2C_M_REV_DIR_ADDR, I don't think that the usbvision driver needs
to support it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Thierry Merle <thierry.merle@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6212): pvrusb2: I2C adapter tweaks from Jean Delvare
Jean Delvare [Sun, 9 Sep 2007 02:19:32 +0000 (23:19 -0300)]
V4L/DVB (6212): pvrusb2: I2C adapter tweaks from Jean Delvare

* I2C adapters aren't expected to handle I2C_M_NOSTART unless they
  really have to. As the pvrusb2 driver doesn't support it, I take it
  that it doesn't need it so it shouldn't mention it at all.
* I2C_FUNC_SMBUS_EMUL includes I2C_FUNC_SMBUS_BYTE_DATA so listing
  both is redundant.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6211): pvrusb2: Allocate a debug mask bit for reporting video standard things
Mike Isely [Sun, 9 Sep 2007 01:32:12 +0000 (22:32 -0300)]
V4L/DVB (6211): pvrusb2: Allocate a debug mask bit for reporting video standard things

It's useful to see specific details for how the pvrusb2 driver is
figuring out things related to the video standard, independent of
other initialization activities.  So let's set up a separate debug
mask bit for this and turn it on.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6210): pvrusb2: Do a far better job at setting the default initial video...
Mike Isely [Sun, 9 Sep 2007 01:28:51 +0000 (22:28 -0300)]
V4L/DVB (6210): pvrusb2: Do a far better job at setting the default initial video standard

The v4l tveeprom logic tells us what video standards are supported by
the hardware, however it doesn't directly tell us what should be the
preferred initial standard.  For example "NTSC/NTSC-J" devices are
reported by tveeprom as support NTSC-M and PAL-M, and while that might
be true, in the vast majority of cases NTSC-M is really what the user
is going to want.  However the driver previously just arbitrarily
picked the "lowest numbered" standard as the initial default, which in
that case would have been PAL-M.  (And making matters more confusing -
this only caused real problems on 24xxx devices because the saa7115 on
29xxx seems to autodetect the right answer anyway.)  This change
implements an algorithm that uses the set of "supported" standards as
a hint to decide on the initial standard.  This algorithm ONLY comes
into play if the driver isn't specifically told what to do; said
another way - the user can always still change the standard via the
sysfs interface, via the usual V4L methods, or even specified as a
module parameter.  The idea here is only to pick a better starting
point if the user (or app) doesn't otherwise do something to set the
standard; otherwise this change has no real impact.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6209): pvrusb2: Better discriminate among device types
Mike Isely [Sun, 9 Sep 2007 01:18:50 +0000 (22:18 -0300)]
V4L/DVB (6209): pvrusb2: Better discriminate among device types

This is a bunch of cleanup in various places to improve behavior based
on actual device type being driven.  While this doesn't actually
affect operation with existing devices, it cleans things up so that it
will be easier / more deterministic when other devices are added.
Ideally we should make stuff like this table-driven, but for now this
is just a series of small incremental (read: safe) improvements.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6208): pvrusb2: Implement programmatic means to extract prom contents
Mike Isely [Sun, 9 Sep 2007 01:16:27 +0000 (22:16 -0300)]
V4L/DVB (6208): pvrusb2: Implement programmatic means to extract prom contents

The pvrusb2 driver already has a method for extracting the FX2's
program memory back out to a user application; this ability is used to
facilitate manual firmware extraction as per the procedure documented
on the pvrusb2 web site.  This change follows that pattern and
implements a corresponding method to grab the binary contents of the
PVR USB2 prom (which for PVR USB2 devices can contain information in
addition to the usual Hauppauge metadata).

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6207): pvrusb2: Fix a potential oops in an error leg cleanup
Mike Isely [Sun, 9 Sep 2007 01:11:46 +0000 (22:11 -0300)]
V4L/DVB (6207): pvrusb2: Fix a potential oops in an error leg cleanup

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6205): pvrusb2: Fix oops in error leg cleanup
Mike Isely [Sun, 9 Sep 2007 01:10:11 +0000 (22:10 -0300)]
V4L/DVB (6205): pvrusb2: Fix oops in error leg cleanup

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6203): Fix SVideo input on KWorld DVB-T 220 boards
Simon Farnsworth [Mon, 10 Sep 2007 16:37:26 +0000 (13:37 -0300)]
V4L/DVB (6203): Fix SVideo input on KWorld DVB-T 220 boards

Fix SVideo input on KWorld DVB-T 220 boards. Without this patch, the
luma pin on the SVideo input is treated as a composite in, and the
chroma pin is ignored.

Also, fix the radio, and provide a second composite input for people who
are used to the existing composite on SVideo connector behaviour.

Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
Reviewed-by: Hermann Pitton <hermann-pitton@arcor.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6200): cx23885: use a pointer to the required i2c bus in dvb_register function
Michael Krufky [Sun, 9 Sep 2007 06:55:34 +0000 (03:55 -0300)]
V4L/DVB (6200): cx23885: use a pointer to the required i2c bus in dvb_register function

Store a pointer to the required i2c_bus so that we do not put the wrong
analog demod into standby.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6199): cx23885: Changes to allow demodulators on each transport bus
Steven Toth [Sun, 9 Sep 2007 00:31:56 +0000 (21:31 -0300)]
V4L/DVB (6199): cx23885: Changes to allow demodulators on each transport bus

cx23885: Changes to allow demodulators on each transport bus.

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6198): add CARDLIST.cx23885 to Documentation/
Michael Krufky [Sat, 8 Sep 2007 18:58:45 +0000 (15:58 -0300)]
V4L/DVB (6198): add CARDLIST.cx23885 to Documentation/

Cardlist generated by cx23885.pl

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6196): cx23885: add support for DViCO FusionHDTV 5 Express
Michael Krufky [Sat, 8 Sep 2007 18:17:13 +0000 (15:17 -0300)]
V4L/DVB (6196): cx23885: add support for DViCO FusionHDTV 5 Express

This patch adds digital ATSC / QAM support for the DViCO FusionHDTV5 Express.
Remote control is supported by ir-kbd-i2c, RTC is supported by rtc-isl1208.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6195): Changes to support MPEG TS on VIDB
Steven Toth [Sat, 8 Sep 2007 18:07:02 +0000 (15:07 -0300)]
V4L/DVB (6195): Changes to support MPEG TS on VIDB

Changes to support MPEG TS on VIDB

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6194): Changes to support interrupts on VIDB
Steven Toth [Sat, 8 Sep 2007 17:21:03 +0000 (14:21 -0300)]
V4L/DVB (6194): Changes to support interrupts on VIDB

Changes to support interrupts on VIDB

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6193): cx23885: define Video B Interrupt Status register bit values
Michael Krufky [Sat, 8 Sep 2007 14:25:37 +0000 (11:25 -0300)]
V4L/DVB (6193): cx23885: define Video B Interrupt Status register bit values

- define missing register bit values for VID_B
- corrected VID/VBI_B_GPCNT_CTL register addresses

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6192): Ensure start_dma() is capable of starting dma on port VIDB
Steven Toth [Thu, 6 Sep 2007 19:07:49 +0000 (16:07 -0300)]
V4L/DVB (6192): Ensure start_dma() is capable of starting dma on port VIDB

start_dma() would fail to start dma if a device used VIDB (portb).

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6191): Removed a redundant switch()
Steven Toth [Thu, 6 Sep 2007 18:50:31 +0000 (15:50 -0300)]
V4L/DVB (6191): Removed a redundant switch()

The switch() statement is no longer required.

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
17 years agoV4L/DVB (6190): cx23885: GPIO fix for non HVR1800lp boards
Steven Toth [Sat, 8 Sep 2007 22:08:17 +0000 (19:08 -0300)]
V4L/DVB (6190): cx23885: GPIO fix for non HVR1800lp boards

The HVR1250 and HVR1800 boards need the s5h1409 demod GPIO enabled.

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>