Give non-highmem architectures access to the kmap API for the purposes of
overriding (this is what the attached patch does).
The proposal is that we should now require all architectures with coherence
issues to manage data coherence via the kmap/kunmap API. Thus driver
writers never have to write code like
kmap(page)
modify data in page
flush_kernel_dcache_page(page)
kunmap(page)
instead, kmap/kunmap will manage the coherence and driver (and filesystem)
writers don't need to worry about how to flush between kmap and kunmap.
For most architectures, the page only needs to be flushed if it was
actually written to *and* there are user mappings of it, so the best
implementation looks to be: clear the page dirty pte bit in the kernel page
tables on kmap and on kunmap, check page->mappings for user maps, and then
the dirty bit, and only flush if it both has user mappings and is dirty.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Original commit code assumes, that when a buffer on BJ_SyncData list is
locked, it is being written to disk. But this is not true and hence it can
lead to a potential data loss on crash. Also the code didn't count with
the fact that journal_dirty_data() can steal buffers from committing
transaction and hence could write buffers that no longer belong to the
committing transaction. Finally it could possibly happen that we tried
writing out one buffer several times.
The patch below tries to solve these problems by a complete rewrite of the
data commit code. We go through buffers on t_sync_datalist, lock buffers
needing write out and store them in an array. Buffers are also immediately
refiled to BJ_Locked list or unfiled (if the write out is completed). When
the array is full or we have to block on buffer lock, we submit all
accumulated buffers for IO.
[suitable for 2.6.18.x around the 2.6.19-rc2 timeframe]
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
get_cpu_var()/per_cpu()/__get_cpu_var() arguments must be simple
identifiers. Otherwise the arch dependent implementations might break.
This patch enforces the correct usage of the macros by producing a syntax
error if the variable is not a simple identifier.
Signed-off-by: Jan Blunck <jblunck@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The scheduler will stop load balancing if the most busy processor contains
processes pinned via processor affinity.
The scheduler currently only does one search for busiest cpu. If it cannot
pull any tasks away from the busiest cpu because they were pinned then the
scheduler goes into a corner and sulks leaving the idle processors idle.
F.e. If you have processor 0 busy running four tasks pinned via taskset,
there are none on processor 1 and one just started two processes on
processor 2 then the scheduler will not move one of the two processes away
from processor 2.
This patch fixes that issue by forcing the scheduler to come out of its
corner and retrying the load balancing by considering other processors for
load balancing.
This patch was originally developed by John Hawkes and discussed at
http://marc.theaimsgroup.com/?l=linux-kernel&m=113901368523205&w=2.
I have removed extraneous material and gone back to equipping struct rq
with the cpu the queue is associated with since this makes the patch much
easier and it is likely that others in the future will have the same
difficulty of figuring out which processor owns which runqueue.
The overhead added through these patches is a single word on the stack if
the kernel is configured to support 32 cpus or less (32 bit). For 32 bit
environments the maximum number of cpus that can be configued is 255 which
would result in the use of 32 bytes additional on the stack. On IA64 up to
1k cpus can be configured which will result in the use of 128 additional
bytes on the stack. The maximum additional cache footprint is one
cacheline. Typically memory use will be much less than a cacheline and the
additional cpumask will be placed on the stack in a cacheline that already
contains other local variable.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: John Hawkes <hawkes@sgi.com>
Cc: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Peter Williams <pwil3058@bigpond.net.au>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
OneNAND lock scheme depends on density and process of chip.
Some OneNAND chips support all block unlock
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Add autodetection for PCI subsystem ID 107d:6632, to detect as
a Leadtek PVR 2000
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This is the first in a series of patches to add full WinTV-HVR1300
support to Linux. This first patch will enable analog TV support
and DVB-T support. Later patches will add the hardware MPEG encoder
support.
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
For most drivers, VIDIOC_G_PARM will just return the current standard fps.
So, instead of failing, drivers based on video_ioctl2 will implement the
default method.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
We could go and work out if the target object is AGP or PCI but the
corner case of an Athlon 64 era box with PCI video is sufficiently
unusual it doesn't seem worth the extra work, at least until other cases
if any pop up.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Again depends on the PCIAGP_FAIL patch for a define. Someone with more
card knowledge should look at the ALIMAGIK case and whether latency can
be safely to set to 0xA or so.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
We should be doing this on all devices doing PCI<->PCI DMA. We only set
the _FAIL ones when the DMA will fail or may cause crashes. This relies
on the PCIAGP_FAIL patch I sent to Andrew already
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Implement fix suggested by Michael Hunold for a bug reported
by Philipp Matthias Hahn: Starting "kdetv" on a Siemens DVB-C 1.x
produced an oops because kdetv opened "/dev/vbi0".
Remove the V4L2_CAP_VBI_CAPTURE flag because it does not work with
this type of hardware anyway.
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Do not select FW_LOADER unless it is really required.
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Add card support for Proteus Pro 2309, based on saa7130 bridge
Signed-off-by: Michal Majchrowicz <mmajchrowicz@gmail.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Do not return -EINVAL for index=0 in VIDIOC_ENUMSTD, because it is a
valid index
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The vivi.current_norm field is not initialized in vivi.c, so a
VIDIOC_G_STD ioctl without a prior call to VIDIOC_S_STD gives
unpredictable results. mplayer does exactly this.
Signed-off-by Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Replaced kernel_thread() with kthread_run() since kernel_thread() is
deprecated in drivers/modules.
Removed the completion and the wait queue which are now useless with
kthread. Also removed the allow_signal() call as signals don't apply
to kernel threads.
Fixed a small race condition when thread is stopped.
Please check if the timer vs. thread still works fine without the wait
queue.
Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This patch fixes a copy-paste bug in videodev.c where the vidioc_qbuf()
function gets called for the dqbuf ioctl.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
While converting the configuration for the old DiB3000MC-module to the new one
a wrong AGC configuration was introduced.
This is using the old one again.
Signed-off-by: Jose Alberto Reguero <jareguero@telefonica.net>
Signed-off-by: Patrick Boettcher <pb@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
There seems to be an off by one error in the dibusb-mb.c which causes
the "Artec T1 with AN2235" box to be detected as a totally different
box - but it only happens if the Artec is one with the correct USB
IDs. A patch is attached to the second post. However, even with this
patch, the box still won't tune. It will tune using a 2.6.12 kernel
though.
Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com>
Signed-off-by: Patrick Boettcher <pb@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Attached is a patch against latest HG which adds remote control support
to the DigiTV driver. It works for me;
Signed-off-by: Allan Third <allan.third@gmail.com>
Signed-off-by: Patrick Boettcher <pb@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
saa7134-alsa now unmutes/mutes the line when opening/closing the capture
device, and needs this symbol
Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This patch should enable unmuting the audio device when opening it (and
posterior muting when closing it), doing away with the need for unmute ioctls
or v4lctl usage.
Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
I've just determined that module-init-tools >=3.2 is needed for dvb_attach
to work. This adds a comment to Kconfig about it.
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
I had forgotten to add the wrapper round the tua6100_attach function if its
disabled in Kconfig
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Software I2C were using a very conservative value of udelay=16, meaning about
20Kbps. According with Philips I2C datasheet, the i2c should answer well for
times at the order of 4.7 us. So, using udelay=5 should work for all devices.
After this patch, the speed should be close to 66,67 Kbps, with the current
kernel software bitbang, with 30/60 duty cycle.
Anyway, added a new parameter (i2c_udelay) that would allow using conservative
values, if eventually a hardware doesn't support the datasheet values.
Thanks to Jean Delvare <khali@linux-fr.org> for pointing this improvement.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Updates feature-removal-schedule.txt to reflect the current scheduled date
to convert all V4L1 drivers to V4L2.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Port budget-av to use the new tua6100 driver
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This driver was originally in budget-av.c, however I have extracted it into
a seperate file to permit reuse.
I also reworked the code to make it maintainable. I then examined the KNC1
windows drivers and rewrote the code in order to configure the PLL as they
do. This solves a lot of reported tuning problems.
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Fix broken build when 24XXX support is not selected. This is required
due to the requirement of removing 24XXX ifdef's from the driver
source.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The cx22702 returns an 8 bit unshifted value for signal strength; this is
inconsistent with most other frontends
Signed-off-by: Bradley Derek Kite <bradley.kite@gmail.com>
Signed-off-by: Andrew de Quincey <adq_dvb@lidskialf.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Trivial patch to make Compro PS39U WebCam work with linux by using the
vicam driver.
The camera is just a vicam with another USB ID, so I added that ID to the
driver, and it works now.
Signed-off-by: Bas Bloemsaat <bas.bloemsaat@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The pvrusb2 driver resolution limits are now programmed the following
way, based on empirical measurements of the hardware:
Vertical max: 480 for NTSC, 576 otherwise
Vertical min: 75 for 24xxx, 17 otherwise
Horizontal max: 720
Horizontal min: 720 for 24xxx, 19 otherwise
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This repairs a problem introduced by a commit earlier applied.
Thanks-to: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Support for 24xxx devices was previously explicitly bracketed with
CONFIG_VIDEO_PVRUSB2_24XXX inside the code because we didn't trust the
stability of these changes. We trust it now; so there's no reason to
leave this out of the driver anymore.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Scaling were not working fine;
Some reserved registers were wrong;
On some situations, saa7115 were not properly being initializated.
Removed some duplicated code.
Thanks-to: Hans Verkuil <hverkuil@xs4all.nl> for co-working on this
patch.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
The VIDIOC_G_SLICED_VBI_CAP needs to receive the v4l2_buf_type field before
it can return a result. Hence this ioctl must be IOWR, not IOR. Since this
ioctl is still marked experimental we can make this change.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
This patch fixes the physical address and takes into account recent
changes in the USB input subsystem.
Acked-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Unai Uribarri <unaiur@gmail.com>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
saa7115 module is capable of calculating the proper FIR tables.
However, there were 4 hardcoded tables for the standard resolution.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
