| Age | Commit message (Collapse) | Author |
|---|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Add scripts/hg-pull-req.pl, which generates a nicely formatted pull
request. Mention it in README.patches.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
The first argument changed in 2.6.13, but it can be make backward compatible
with a macro.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
CONFIG_BIGPHYS_AREA is an out-of-tree patch. While it may be interesting
to have this configuration flag when compiling v4l-dvb out-of-tree, for
development purposes, there's no sense on keeping those unused code
inside kernel.
This patch do some cosmetic changes on zoran_driver.c, preserving both
supports there, and adds the defaults to be used by gentree.pl. This
way, gentree.pl will handle the removal of the dead code when submitting
code to mainstream, while keeping the complete code at v4l-dvb tree.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
The out of tree v4l-dvb build system uses slightly different compiler
flags than the kernel does. This caused problems when trying to compile
the out of tree version of the source symlinked into the kernel tree.
The workaround is to have the script that creates the symlinks also
create a new file called drivers/media/Kbuild, which the kernel build
system will use in preference to divers/media/Makefile. The new Kbuild
file exports the necessary EXTRA_CFLAGS to make building work and then
includes the real drivers/media/Makefile.
The dvb Makefiles were changed in a previous patch from 'EXTRA_CFLAGS='
to 'EXTRA_CFLAGS+=' so they wouldn't replace the new EXTRA_CFLAGS.
The symlink creating script is also re-written to make the symlinks
faster with less duplication of commands and avoid some unnecessary
links.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
V4L/DVB tree keeps backward compatibility with vanilla 2.6.x kernels.
However, if some API changes happens on 2.6.x.y, this wouldn't be
handled.
This patch improves the compatibility capabilities of the tree by
allowing customized scripts to be added on make_config_compat.pl.
So, when generating config-compat.h, some compat checks can be done
against the kernel tree.
The practical effect is allowing compilation on 2.6.17.x trees, used by
several distros, where some changes on netdevice.h affects dvb-net,
stopping its compilation.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hermann Pitton <hermann-pitton@arcor.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Installation logs were very prolific. This patch makes the installation message
cleaner, while preserving all information.
The generated log is now much more clear, especially if compiling just a few
device drivers.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Most of the v4l-dvb drivers now depend on the kernel Kconfig option
HAS_IOMEM, which isn't defined by kernels older than 2.6.22.
Assume HAS_IOMEM is true for older kernels to make the v4l-dvb Kconfig
files work.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
A number of drivers depend on "!VIDEO_HELPER_CHIPS_AUTO", so when VHCA is on,
they are disabled. Make the allyesconfig default be VHCA off so these drivers
will be enabled with allyesconfig.
Even though these drivers are disabled when VHCA is on, it is still possible
to enable them via a "select", at which point a driver will be on even though
its dependencies are not met. Which seems like it doesn't follow the
specification.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
A backported kernel patch added a "depends on" clause to a menu, which wasn't
yet handled.
Everything in a menu gains implicitly gains a copy of the menu's dependencies
We handle this the same way as if/endif is handled. It may not work properly
if someone makes a nonsense file like:
menu "foo"
if X
endmenu
endif
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
kernel-sync:
Backport the original patch from Mattia Dongili <malattia@linux.it>
meye: make meye use sony-laptop instead of sonypi
Change sonypi_camera_command() calls to sony_pic_camera_command() and use
the renamed macros.
Also, added compat stuff to allow compiling meye driver for kernels < 2.6.22
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Deal with string options and the default values.
There is only one string option, and it's disabled, but someone could
always add more.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Update make_myconfig.pl and make_kconfig.pl to handle menuconfig options the
same way as config options.
Update make_kconfig.pl to process if/endif directives. All config options
inside an if/endif block gain a dependency on the 'if' expression.
Make the script a little more efficient by not trying to match a line after it
has already found a match for it. This helped to enhance debug mode, so
that it will print lines which were not understood rather than silently
skipping them.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Changed usbvision cards table to allow:
1) Not repeat USB ID on two structs;
2) Not need to specify both usb and card description tables at
the same order, removing some magic;
Some cards had duplicated names. Fixed.
A test for an specific board were doing by using a string comparation.
The comparation were wrong. Also, it is not a good practice to recognize
a board based on his string name.
Acked-by: Thierry MERLE <thierry.merle@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Annoyingly, the Hg mq extension will strip lines starting with "From: "
from the patch header. Since the v4l-dvb format commit message contains
a line starting with "From: ", it will mess them up.
This changes prep_commit_msg to create a decoy "from" line to keep mq from
messing up commit messages.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
Adds a qrefresh target to the top-level Makefile. It's like commit, but
for use with the Hg mq extension. It will execute hg qrefresh, so you
should have a current mq patch created with hg qnew already!
It will run the whitespace and cardlist scripts. If you do not have a
commit message from the current mq patch, it will create a default one,
as make commit does. If one already exists, then it will put a comment
at the beginning of the existing message with the current diffstat and a
note of any changes the whitespace script made. The diffstat is done
with qdiff, which will show the total diffstat of the patch instead of
just the diffstat of from the last version of the patch to the current
version.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
In one case a Kbuild file would do something like:
foo-objs := a.o b.o
foo-objs += c.o
analyze_build didn't understand that the second lines _adds_ c.o to the
objects that make up foo.ko. It would think that foo.ko was made up of
only c.o. Should be fixed now.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Ludwig Nussel <ludwig.nussel@suse.de>
If defined, use $(DESTDIR) as a prefix for installing and removing
modules. depmod will also use $(DESTDIR) as a prefix when it is run as
part of module installation.
Signed-off-by: Ludwig Nussel <ludwig.nussel@suse.de>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Michael Krufky <mkrufky@linuxtv.org>
The lgh06xf module was removed, so we should remove it from versions.txt
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
When parsing hgrc files, don't get confused by 'username' appearing in
other sections, like [smtp].
Check the repo's hgrc file first, then ~/.hgrc
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Everytime a build is started, the build system creates links to every
single source file from the v4l directory. This patch changes the
command used, to one that is about 15x faster. When building just one
file, the creating the links could take the majority of the total build
time.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
tea5761 is a driver wrote back on 2005, to support Philips tea5671
radio, meanting to be used on some devices that would be lauched with
those chips.
Recently, some people seems to be interested on having support for those
chips at Linux Kernel.
Let's enable compilation of this driver and ask they to test if this
works fine.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
The #ifndef true test didn't work, as true is an enum, not a macro.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
If Module.symvers is kept around, modpost will think symbols from
previous builds are still defined when in the current build they are not.
This leads to undefined symbol warnings getting masked.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
From: Hans Verkuil <hverkuil@xs4all.nl>
It took three core maintainers, over four years of work, eight new i2c
modules, eleven new V4L2 ioctls, three new DVB video ioctls, a Sliced
VBI API, a new MPEG encoder API, an enhanced DVB video MPEG decoding
API, major YUV/OSD contributions from Ian and John, web/wiki/svn/trac
support from Axel Thimm, (hardware) support from Hauppauge, support and
assistance from the v4l-dvb people and the many, many users of ivtv to
finally make it possible to merge this driver into the kernel.
Thank you all!
Signed-off-by: Kevin Thayer <nufan_wfk@yahoo.com>
Signed-off-by: Chris Kennedy <c@groovy.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: John P Harvey <john.p.harvey@btinternet.com>
Signed-off-by: Ian Armstrong <ian@iarmst.demon.co.uk>
|
|
From: Oliver Endriss <o.endriss@gmx.de>
Fix compilation error introduced by changeset 6d08d71cbcfb.
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
gentree.pl were part of Gerd Knorr scripts, used to generate patches
from his tree against kernel. Several newer capabilities were added to
the original script, including the ability to handle with multiple
levels of #if/#else/#endif and support for #elif.
This patch corrects the credits for this valuable piece of code.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Mostly re-write gentree.pl to make a few improvements:
About three times faster
Correctly processes BTTV_VERSION_CODE checks
Properly handles #elif
Doesn't need absolute path names specified in command line arguments
Improved debug output, which now shows input line number
Can handle almost any general expression for a conditional compilation
directive. Can understand all these:
#if 0
#ifdef BROKEN_XAWTV
#ifndef MM_KERNEL
#if defined(STV0297_CS2)
#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,20)
#elif defined(MM_KERNEL) || !BROKEN_XAWTV
#if BTTV_VERSION_CODE > KERNEL_VERSION(0,8,0) || defined(BROKEN_XAWTV)
#if (1 > 2) || (I2C_PEC && defined(I2C_CLASS_TV_ANALOG))
#ifndef CONFIG_XC3028 /* Keep */ <- knows to keep this one in the code
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Relaxed some regular expressions to better handle whitespaces at the
middle of cpp macros;
Added #elif processing capability to the script.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Add an inline versin of vm_insert_page() that uses remap_pfn_range() to
compat.h.
Remove compat code from em28xx-video.c, sn9c102_core.c, and
usbvision-video.c since compat.h is now doing the same thing.
USB_ZC0301, USB_ET61X251, and USB_ZR364XX will now compile under 2.6.12.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Adjust konicawc and quickcam_messenger to use backward compat code
relating to changes in the input layer in 2.6.15. They should now work
in 2.6.12.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
compat.h was missing from usbvideo-i2c.c and zr364xx.c
zr364xx.c needed the struct semaphore -> struct mutex compat check.
Added an old i2c id to compat.h. These could be added to the Hg copy of
linux/i2c-id.h instead? There's just this one so far.
zr364xx needs 2.6.15 because of vm_insert_page(), so add to versions.txt
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Add -include linux/version.h to the cflags. Now code can have backward
compatibility test without including compat.h first.
Linux headers included from compat.h are removed, so that code will get
the same headers when compiling in v4l-dvb as it does in the kernel.
Many drivers have compat.h moved to the end of their include list, as
this lets compat.h do things it can't do at the beginning. Such as test
of something is defined to include compat code, or to put a wrapper
around a function without changing the function's name.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Trent Piepho <xyzzy@speakeasy.org>
Compat.h used to have an argument, now it doesn't. There was also some code:
if (current->flags & PF_FREEZE) { refrigerator(PF_FREEZE); }
which is the same as try_to_freeze(), and so can be replaced by it.
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Antti Palosaari <crope@iki.fi>
This patch adds driver for Sigmatek DVB-110 USB DVB-T stick. Stick has based
on hardware of Qtuantek QT1010 tuner, Zarlink ZL10353 (Intel CE 6353)
demodulator and Alcor Micro AU6610 DVB-T USB controller. HW is rather similar
as used in MSI Megasky GL861.
Currently, the driver works only in USB 2.0. In my understanding USB 1.1 is
also supported by hw but I cannot test it due to lack of USB 1.1 port. Device
supports only isochronous mode transfers. There is also eeprom in usb
controller(at least in address range 0x80 - 0xbf) for storing data, eg.
firmware. Anyway, firmware loading is not used / required by the device.
There seems to be at least one unknown I2C device in address 0xa0, probably
remote control or GPIO. Windows drivers reads registers from 0x00 to 0x07
from this unknown address.
Driver is based on gl861 module. Tuner has a lot of problems to lock with
megasky qt1010 module with this hardware with some broadcasting standards.
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
|
|
From: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
|
|
From: Michael Krufky <mkrufky@linuxtv.org>
- removed extra newline
- removed NULL entry
- fixed versions.txt
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
|
|
From: Aapo Tahkola <aet@rasterburn.org>
Currently, the driver works in bulk mode supporting both USB 2.0 and
1.0 with and without hardware pid filters.
The ULi m9205 also supports isochronous transfer mode, but I have
dropped support for it because it depends on firmware and does not work
on all USB host chips. Further, I have no firmware with remote
controller support for this mode.
Signed-off-by: Aapo Tahkola <aet@rasterburn.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
The recent updates on some webcams replaced vmalloc_32 by vmallo_32_user.
However, it seems that this doesn't exist before kernel 2.6.19
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
