- 26 Sep, 2006 40 commits
-
-
Trent Piepho authored
The release callback was installed in ops.release, it should have been ops.release_sec. This prevented to front-end release from getting called. Signed-off-by:
Trent Piepho <xyzzy@speakeasy.org> Signed-off-by:
Mauro Carvalho Chehab <mchehab@infradead.org>
-
Patrick Boettcher authored
Still the pid_parse bit was erased. Signed-off-by:
Patrick Boettcher <pb@linuxtv.org> Signed-off-by:
-
Patrick Boettcher authored
Matthieu Castet found that with the rewritten dib3000mc-driver the PID-parsing was handled correctly. This changeset fixes it. Thanks Matthieu Signed-off-by:
-
Jochen Issing authored
After some structural changes, here comes the updated patch for the Club 3D zap250mini USB dvb device. Adding support for a WideView-based device with a different "COLD" ID. Signed-off-by:
Jochen Issing <jochen.issing@isign-softart.de> Signed-off-by:
-
Adrian Bunk authored
The Coverity checker spotted the following "we dereference d->fe only when we know it's NULL" bug: <-- snip --> ... static int dibusb_dib3000mb_frontend_attach(struct dvb_usb_device *d) { ... if ((d->fe = dib3000mb_attach(&demod_cfg,&d->i2c_adap,&st->ops)) == NULL) { d->fe->ops.tuner_ops.init = dvb_usb_tuner_init_i2c; d->fe->ops.tuner_ops.set_params = dvb_usb_tuner_set_params_i2c; return -ENODEV; } ... <-- snip --> Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
There are several boards on V4L that can work with several different helper modules for audio/video encoding/decoding and similar stuff. This patch adds the capability of choicing between autoselecting the pertinent helper modules for each driver, or to manually selecting them. Acked-by:
Michael Krufky <mkrufky@linuxtv.org> Acked-by:
Mike Isely <isely@pobox.com> Signed-off-by:
-
Mauro Carvalho Chehab authored
Several audio/video encoders/decoders were just bound as if they were part of the core driver. In fact, they are generic enough to be used by other drivers as well, since they use I2C event bus to communicate. Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
Several audio decoders were just bound as if they were part of bttv driver. In fact, they are generic enough to be used by other drivers as well, since they use I2C event bus to communicate. Signed-off-by: -
Alexey Dobriyan authored
Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
-
Eric Sesterhenn authored
when checking the -Wextra signedness warnings issued by gcc 4.1 I came across this one: drivers/media/video/vivi.c:1001: warning: comparison of unsigned expression < 0 is always false Since videobuf_reqbufs() returns negative values on errors the current code does no real error checking since gcc removes the comparison. This patch fixes this issue by making ret a normal, signed integer. Signed-off-by:
Eric Sesterhenn <snakebyte@gmx.de> Signed-off-by:
-
Mauro Carvalho Chehab authored
Several changes at register stuff were done since commit b45009b0, but I've forgot to add the copyright line for this. Those changes allowed cx88 to have multiple tuners, allowing it to support hardware with separated radio and video tuners. It also allows binding a tuner to an specific i2c address. Signed-off-by:
-
Mauro Carvalho Chehab authored
The Coverity checker spotted the following two array overflows. Registers 0xcc and 0xdc were cached on reg[] array, with only 128 elements, instead of 256. Signed-off-by: -
Mauro Carvalho Chehab authored
Signed-off-by: -
Mauro Carvalho Chehab authored
As pointed by The Coverity checker, commit c7c0b34c broked input selection. Signed-off-by:
-
Ricardo Cerqueira authored
cx88-alsa can be compiled in (in distro built kernels, for example), but not used. In those cases, the audio DMA restart is needed for proper sound on NICAM channels. This patch enables the DMA restart even with ALSA, but the functions now check if cx88-alsa is really active. Signed-off-by:
Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by:
-
Christoph Pfister authored
Apparently there are two types of knc1 dvb-s plus cards with different tuners (as for the knc1 dvb-s ones). This patch adds support for those cards with subsystem id 1894:0011. Successfully tested by: Benjamin Weber <benny1990@gmail.com> Signed-off-by:
Christoph Pfister <christophpfister@gmail.com> Signed-off-by:
Andrew de Quincey <adq_dvb@lidskialf.net> Signed-off-by:
-
Andrew de Quincey authored
This actually needs the same configuration as the other DVBC cards; simply no one had managed to test before to find out. Signed-off-by:
-
Alexey Dobriyan authored
Signed-off-by:
-
Mauro Carvalho Chehab authored
Although not dangerous, it may allow a normal user with access to a machine to untune a TV channel. Signed-off-by: -
Chris Pascoe authored
If the FX2 does not reset properly during reboot, it can present an invalid USB device ID and fail to attach. Prevent this situation from occuring by resetting the USB part of the card when the PCI part probes. Also fix the GPIO configurations so that analog capture will not inadvertantly reset the USB part. Signed-off-by:
Chris Pascoe <c.pascoe@itee.uq.edu.au> Signed-off-by:
-
Chris Pascoe authored
More register setup which improves the ZL10353's tuning characteristics. Signed-off-by:
-
Chris Pascoe authored
Force parallel transport stream output on the ZL10353 attached to a bluebird device. Addresses the problem where a frontend lock was observed, but no MPEG transport data was received. Signed-off-by:
-
Oliver Endriss authored
lnbp21 module must use fe->ops.release_sec, not fe->ops.release. The release function of the frontend driver was never called for cards with lnbp21... Signed-off-by:
Oliver Endriss <o.endriss@gmx.de> Signed-off-by:
-
Oliver Endriss authored
dvb_frontend_detach() used invalid config option CONFIG_DVB_DETACH, so dvb_frontend_detach() did not call symbol_put_addr(). Replaced CONFIG_DVB_DETACH by CONFIG_DVB_CORE_ATTACH. Signed-off-by:
-
Trent Piepho authored
from whitespace cleaning script Signed-off-by:
-
Trent Piepho authored
If an eeprom defined two tuners and they supported more than eight standards combined (as opposed to each), it would overflow an array. Signed-off-by:
-
Mauro Carvalho Chehab authored
drivers/media/video/pvrusb2/pvrusb2-i2c-cmd-v4l2.c: In function 'set_standard': drivers/media/video/pvrusb2/pvrusb2-i2c-cmd-v4l2.c:33: warning: format '%llx' expects type 'long long unsigned int', but argument 2 has type 'v4l2_std_id' Signed-off-by: -
Andrew Morton authored
drivers/media/radio/radio-cadet.c: In function 'cadet_do_ioctl': drivers/media/radio/radio-cadet.c:362: warning: implicit declaration of function 'KERNEL_VERSION' Signed-off-by:
-
Patrick Boettcher authored
Forget to put the MODULE-define and thus to give the license. Now GPL Signed-off-by:
-
Dave Jones authored
DVB null terminates its device names, which seems odd, and should be unnecessary. Acked-by:
Alan Cox <alan@redhat.com> Signed-off-by:
Dave Jones <davej@redhat.com> Signed-off-by:
-
Yeasah Pell authored
1) It sets LNBDCPol differently based on the card type. Now it should work properly for both the kworld and geniatech cards. 2) It stops returning an error for the SEC_VOLTAGE_OFF voltage command (the cx88-dvb level handles the actual voltage on/off, but it still passes the ioctl down to the cx24123 level, which previously rejected the OFF as invalid so the ioctl would report failure) Acked-by:
Manu Abraham <manu@linuxtv.org> Signed-off-by:
Yeasah Pell <yeasah@schwide.net> Signed-off-by:
-
Oliver Endriss authored
Some DVB-T cards with Grundig 29504-401 tuner use subsystem id 13c2:0001 (reported by Stuart Butterfield). Signed-off-by:
-
Mauro Carvalho Chehab authored
Hardware algorithm needs to be configured to: 1) Increase timeout constants, to detect weaker signals; 2) do a wider zigzag search. Acked-by:
-
Mauro Carvalho Chehab authored
Tested on x64 with a bttv board. Most Get ioctls are fixed. The only non-completely working one is VIDIOCGAUDIO. All other IOR ioctls give the same results on x86_64 and i386 architectures. Thanks to Alastair Poole <netstar@gatheringofgray.com> for part of this patch and tests on ppc64. Signed-off-by:
Alastair Poole <netstar@gatheringofgray.com> Signed-off-by:
-
Mauro Carvalho Chehab authored
Signed-off-by:
-
