Jesper Juhl [Fri, 23 Jun 2006 16:27:36 +0000 (13:27 -0300)]
V4L/DVB (4154): Fix use-after-free bug in cpia2 driver
The coverity checker detected a use-after-free error in
drivers/media/video/cpia2/cpia2_v4l.c::cpia2_close() (coverity
error #1281).
What happens is that we lock cam->busy_lock, then proceed to free
resources, and in the case of (--cam->open_count == 0) we finish off by
doing a kfree(cam) and then at the end of the function we do a
mutex_unlock(&cam->busy_lock) which will explode since it'll dereference
the free'd `cam' :
...
mutex_lock(&cam->busy_lock);
...
if (--cam->open_count == 0) {
...
if (!cam->present) {
video_unregister_device(dev);
kfree(cam);
}
}
mutex_unlock(&cam->busy_lock); <--- PROBLEM, cam no longer around.
...
Since this only happens in the case of open_count going down to zero I
don't see a problem with just releasing the mutex after unregistering the
device and just before the kfree(). In this case there is nothing around
that we can race against; we are in the release method, open_count is zero,
(!cam->present) and the device has just been unregistered, so letting go of
the mutex at this point looks safe to me.
Patch below to implement that solution. Acked-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (4132): Don't kill cx88 DMA sound on channel change
There's a call at cx88-tvaudio to reset DMA sound on channel change; that
call was killing cx88-alsa whenever the channel was changed, resulting
in no sound. Remove those calls if cx88-alsa is selected to be compiled.
V4L/DVB (4131): Fix cx88-alsa vs IRQ remote conflict
cx88-alsa was resetting the card on load, causing the IRQ IR handler
to go away (maybe others, too). There's no actual need to reset the
card, though, so that line was removed
Michael Krufky [Tue, 20 Jun 2006 18:56:02 +0000 (15:56 -0300)]
V4L/DVB (4124): Kconfig: fix description for saa7115 module
This Kconfig description is incorrect, due to a previous merge a while back.
CONFIG_SAA711X builds module saa7115, which is the newer module, and is
not obsoleted.
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Zac Bowling [Mon, 19 Jun 2006 22:49:20 +0000 (19:49 -0300)]
V4L/DVB (4115): Saa7134 card (LifeView3000 NTSC)
Add detection for the LifeView3000 NTSC model (right now the PAL version is
the only one in there, which is sort of annoying for me living in the US
and all.. sigh...)
This is one of those things that I like to classify "as a works for me so
take it or leave it" :-) . Not that worried because its an $18 video
capture card that didn't look it was selling to well in the US, but then
again who knows.
The documentation needs updating too I bet but its behind anyways usually
to what is in the code.
Michael Krufky [Mon, 12 Jun 2006 17:18:26 +0000 (14:18 -0300)]
V4L/DVB (4099): Cx88-dvb: clean up long linewraps
This patch splits up dvb_pll_attach and isl6421_attach function calls
such that the lines do not exceed 80 columns. This patch is purely cosmetic,
with no functional change at all.
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Dyks, Axel (XL) [Sun, 11 Jun 2006 20:14:35 +0000 (17:14 -0300)]
V4L/DVB (4090): Fix cinergyt2_poll() to allow non-blocking IO on frontend
cinergyt2_poll()" shouldn't return (POLLIN | POLLRDNORM | POLLPRI) when
there are no pending events. User space programs that do non-bocking IO
using "select()" and/or "poll()" would otherwise produce high system load. Acked-by: Andreas Oberritter <obi@linuxtv.org> Signed-off-by: Axel Dyks <xl@xlsigned.net> Acked-by: Andreas Oberritter <obi@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The description of the card has been updated to it's full name/model.
The tuner has also been switched to a more compatible one (radio
wasn't working, now it is)
Eric Sesterhenn [Tue, 6 Jun 2006 14:20:08 +0000 (11:20 -0300)]
V4L/DVB (4070): Zoran strncpy() fix
The zoran driver uses strncpy() in an unsafe way. This patch uses the proper
sizeof()-1 size parameter. Since all strncpy() targets are initialised with
memset() the trailing '\0' is already set. Where std->name was the target for
the strncpy() we overwrote 8 Bytes of the std structure with zeros.
Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (4068): Removed all references to kernel stuff from videodev.h and videodev2.h
The videodev.h and videodev2.h describe the public API for V4L and V4L2.
It shouldn't have there any kernel-specific stuff. Those were moved to
v4l2-dev.h.
This patch removes some uneeded headers and include v4l2-common.h on all
V4L driver. This header includes device implementation of V4L2 API provided
on v4l2-dev.h as well as V4L2 internal ioctls that provides connections
between master driver and its i2c devices.
Sub-carrier frequency for PAL/M is slicely different than NTSC/M.
Without this patch, colors don't work on PAL/M.
Setting method also improved to allow other video standards to be included.
V4L/DVB (4065): Several improvements at videodev.c
Videodev now is capable of better handling V4L2 api, by
processing V4L2 ioctls and using callbacks to the driver.
The drivers should be migrated to the newer way and the older
one will be obsoleted soon.
V4L/DVB (4057): Cx88: IR remote support for DTV2000H
Adds support for the Y0400052 remote supplied with this card.
In addition to adding a number of buttons to the current winfast definition,
it enables all the keys currently masked out with #ifdef 0.
It is supplied separately as it remaps two keys from the current definition,
The teletext button now sends KEY_TEXT instead of KEY_SUBTITLE, as that
keycode is used by the subtitle button.
KEY_BACK was changed to KEY_LAST to group it with KEY_NEXT.
Other then that the keys don't overlap, so this should support several
different versions of the Leadtek remotes.
Randy Dunlap [Fri, 26 May 2006 13:32:13 +0000 (10:32 -0300)]
V4L/DVB (4047): Doc. sources: expose video4linux/
Documentation/video4linux/:
Expose example and tool source files in the Documentation/ directory in
their own files instead of being buried (almost hidden) in readme/txt files.
This will make them more visible/usable to users who may need
to use them, to developers who may need to test with them, and
to janitors who would update them if they were more visible.
Also, if any of these possibly should not be in the kernel tree at
all, it will be clearer that they are here and we can discuss if
they should be removed.
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
linux/videodev2.h uses types such as __u8 but it fails to include
<linux/types.h>. Within the kernel, that's not a problem because
<linux/time.h> already includes <linux/types.h>. However, there are
user apps that try to include videodev2.h (e.g., ekiga) and at least
on ia64, it causes compilation failures since <linux/types.h> doesn't
get included for any other reason, leaving __u8 etc. undefined. The
attached patch fixes the problem for me.
Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (4028): Change dvb_frontend_ops to be a real field instead of a pointer field inside dvb_frontend
The dvb_frontend_ops is a pointer inside dvb_frontend. That's why every demod-driver
is having a field of dvb_frontend_ops in its private-state-struct and
using the reference for filling the pointer-field in dvb_frontend.
- It saves at least two lines of code per demod-driver,
- reduces object size (one less dereference per frontend_ops-access),
- be coherent with dvb_tuner_ops,
- makes it a little bit easier for newbies to understand how it works and
- avoids stupid mistakes because you would have to copy the dvb_frontend_ops
always, before you could assign the static pointer directly, which was
dangerous.
Signed-off-by: Patrick Boettcher <pb@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This is a patch which includes support for the GENPIX 8PSK->USB module. The
board supports QPSK, BPSK and 8PSK decoding (though I don't think it will be
DVB-S2 compliant) With the following patch, the boad is equivalent to a budget
card (no CA Module)
The patch which adds 8psk suppot will follow, but is seperate, as it requires
DVB-S2 support
More info on the board can be found at www.genpix-electronics.com (and they
host the requisite firmwares there as well)
Signed off by: Alan Nisota <alannisota@gmail.com>
Signed-off-by: Patrick Boettcher <pb@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (4026): Bttv-gpio-irq is no longer used, remove it
This functionality was used when remote control input for BTTV was handled
by a standalone module, to hook some functions of that module into the main
bttv core.
Since that module is now extinct, this can go away
Sylvain Pasche [Sun, 26 Mar 2006 02:14:42 +0000 (23:14 -0300)]
V4L/DVB (4023): Subject: Pinnacle PCTV grey remote control support
This adds support for the older (?) Pinnacle PCTV remotes (with all buttons
colored in grey). There's no autodetection for the type of remote, though;
saa7134 defaults to the colored one, to use the grey remote the
"pinnacle_remote=1" option must be passed to the saa7134 module
Angelo Marconi [Tue, 9 May 2006 21:27:48 +0000 (18:27 -0300)]
V4L/DVB (4016): Cx88-blackbird: add support for ProLink Pixelview Playtv@P7000
- Add support for ProLink Pixelview Playtv@P7000
Raw video and MPEG encoded video confirmed to work properly.
SVideo, Composite and FM inputs are untested - disabled for now.
Trent Piepho [Wed, 24 May 2006 02:54:44 +0000 (23:54 -0300)]
V4L/DVB (4002): Adjust VBI offset to match value reported in VBI format
The actual offset of the start of VBI data is incorrectly documented for both
the cx2388x and bt8x8. For the cx2388x, it appears to be about 220 to 224 +
VBI_V_DEL*2. The driver has been incorrectly reporting an offset of 244.
This patch adjusts VBI_V_DEL to 10, so that the offset of 244 is correct.
This is the same offset as the bttv driver, and expected by some software.
Signed-off-by: Christoph Pfister <christophpfister@bluemail.ch> Signed-off-by: Manu Abraham <manu@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (3988): Add math routines required by DVB demods
Signed-off-by: Christoph Pfister <christophpfister@bluemail.ch> Signed-off-by: Manu Abraham <manu@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (3984): Fix CI interface on KNC1 DVBT and DVBC cards
These cards need special handling for CI - reinitialising the frontend
device when the CI module is reset. Additionally the tda10021 needs to be set
into a different transport stream mode when a CI module is present.
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Michael Krufky [Mon, 22 May 2006 13:32:00 +0000 (10:32 -0300)]
V4L/DVB (3982): Cx88-blackbird: use encoder firmware filename defined in cx2341x.h
Since the filename of the cx23416 encoder firmware image
is defined in cx2341x.h, we don't need to explicitly define
it in cx88-blackbird.c anymore.
This changeset removes BLACKBIRD_FIRM_ENC_FILENAME,
replacing all occurrances with CX2341X_FIRM_ENC_FILENAME.
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Valentin Zagura [Mon, 22 May 2006 13:31:59 +0000 (10:31 -0300)]
V4L/DVB (3973): Cx88-blackbird: pause the encoder during frequency change
Added code to cx88-blackbird.c to treat the VIDIOC_S_FREQUENCY ioctl
to stop mpeg stream before changing the frequency and restart it after.
It seems that the mpeg stream needs to be paused while changing the
channel frequency, otherwise the mpeg encoder enters into a bad state,
producing artifacts.
Michael Krufky [Mon, 22 May 2006 13:31:55 +0000 (10:31 -0300)]
V4L/DVB (3971): Cx88-blackbird: use standard filename for cx23416 firmware
cx88-blackbird uses the same encoder firmware for the cx23416 chip that is
used by ivtv and pvrusb2.
This patch alters the driver to expect the cx23416 encoder firmware to be
located in the file, "v4l-cx2341x-enc.fw", instead of using
"blackbird-fw-enc.bin"
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
akpm@osdl.org [Mon, 22 May 2006 13:31:49 +0000 (10:31 -0300)]
V4L/DVB (3967): Pwc-dec23 oops fix
drivers/media/video/pwc/pwc-dec23.c: In function `pwc_dec23_decompress':
drivers/media/video/pwc/pwc-dec23.c:885: warning: null argument where non-null required (arg 2)
Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB (3966): Core: ULE fixes and RFC4326 additions
Fix some problems regarding support for Unidirectional Lightweight
Encapsulation (ULE) in dvbnet.c.
The original ULE code was based on a draft. In the meantime, ULE has been
published in RFC 4326 (ftp://ftp.rfc-editor.org/in-notes/rfc4326.txt).
With these fixes, and some additions (which are included in the patch), the
decaps code should now be complient to RFC4326.
Signed-off-by: Christian Praehauser <cpraehaus@cosy.sbg.ac.at> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
projects / linux-2.6 / log