Add #include <linux/module.h> to spi-tegra.c to fix a compilation error
after the removal of module.h from device.h (in patch: "include: replace
linux/module.h with "struct module" wherever possible").
Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
When creating flushed receive CQEs, set the QPID field in the t4_cqe
to the SQ QID and not the RQ QID. Otherwise the poll code will not
find the correct QP context.
Signed-off by: Jonathan Lallinger <jonathan@ogc.us>
Signed-off by: Steve Wise <swise@ogc.us>
Signed-off-by: Roland Dreier <roland@purestorage.com>
At the time when a peer closes the connection, iw_cxgb4 will not send
a cq event if ibqp.uobject exists. In that case, its possible for a
user application to get blocked in ibv_get_cq_event().
To resolve this, call the cq's comp_handler to unblock any read from
ibv_get_cq_event(). This will trigger userspace to poll the cq and
collect flush status completions for any pending work requests.
Signed-off-by: Kumar Sanghvi <kumaras@chelsio.com>
Acked-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Safety reasons. We shouldn't be trying to reconfigure a risc
processor instruction queue unless it's stopped.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Previous driver over-reported errors and stopped the video fifo
(causing video to stop) if 1) A risc error occured 2) The risc processor
detected that it has missing lines in the video frame 3) The risc fifo
is about to overflow.
The previous driver reported all three of these cases as risc errors
when this is technically not correct. So, the function was cleaned up.
1. Ensure that risc opcode related errors are correctly shown as such,
and not overly reported for non-risc-opcode cases.
2. Ensure that line sync and overflow errors are not reported as opcode errors.
3. Ensure that only the risc-op-code case can stop the fifo and dump channel
/ risc processor information.
The net result is that if video becomes unstable, cable disconnect, this will
not trigger a stop of the video firmware (due to missing lines errors).
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
1. Removed the verbose 'dump channel state to console', very noisy in
weak signal conditions.
2. No need for the video buffer timeout to stop the risc engine here.
Clearer and easier to maintain if start_video_dma() is the single
place that this is done.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Inform applications that multiple audio inputs are available on
non-tv inputs.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Ensure audio is established at driver start.
Ensure the correct defaults are established for the audio path if
the cards struct has nothing defined.
Allow the caller to select one of the two baseband input paths.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Add code to program the flatiron internal i2c ADC and pass the
appropriate audio mux enums to the cx25840 driver.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
The change effects cx23885 boards only and preserves support for existing
boards. Essentially, if we're using baseband audio into the cx23885 AV
core then we have to patch registers. The cx23885 driver will call
with either CX25840_AUDIO8 to signify tuner audio or AUDIO7 to
signify baseband audio. If/When we become more comfortable with this change
across a series of products then we may decide to relax the cx23885 only
restriction.
[liplianin@netup.ru: fix missing state declaration]
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Coding style and printk's cleaned up in following patches.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
This change is probably good for other boards also, but the change
has limited scope and is reasonably safe.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Coding style and printk's are cleaned up in subsequent patches
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
A handlful of coding style issue cleaned up in the following patches.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
vidioc_g_tuner should not do any memset for the parameters. Core
already does that. In particular, V4L2 core now does some handling
for the tuner type, and the tuner-core module relies on that.
So, doing any memset there is a very bad idea.
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
We really don't want this to work in the general case; device drivers
*shouldn't* care whether they are behind an IOMMU or not. But the
integrated graphics is a special case, because the IOMMU and the GTT are
all kind of smashed into one and generally horrifically buggy, so it's
reasonable for the graphics driver to want to know when the IOMMU is
active for the graphics hardware.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
There's no need for the API to modify it and having it const makes it
easier to use with random strings the board code has.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
To work around a hardware issue, we have to submit IOTLB flushes while
the graphics engine is idle. The graphics driver will (we hope) go to
great lengths to ensure that it gets that right on the affected
chipset(s)... so let's not screw it over by deferring the unmap and
doing it later. That wouldn't be very helpful.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
In the case where _regulator_enable returns an error it was not checked
if a supplying regulator exists before trying to disable it, leading
to a null pointer-dereference if no supplying regulator existed.
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
[stoth@kernellabs.com: add it to the makefile and fix snd_card binding]
[liplianin@netup.ru: videobuf: Remove the videobuf_sg_dma_map/unmap functions]
Signed-off-by: Mijhail Moreyra <mijhail.moreyra@gmail.com>
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
We need to have as less time between McPDM shutdown,
and power down of the DAC on the twl6040 codec as possible.
Request core to ignore the pmdown_time for the playback
stream.
Backround: with the McPDM protocol we are sendning not only
the pure audio stream, but OMAP McPDM also transmits
additional information (for example offset cancellation).
If McPDM is stopped prior to the DAC this information will
be not sent to the codec, which can result noise rendered
by the twl6040 codec.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
With this flag codec drivers can indicate that it is desired
to ignore the pmdown_time for DAPM shutdown sequence when
playback stream is stopped.
The DAPM sequence will be executed without delay in this case.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Current code defines AD193X_PLL_INPUT_MASK as (~0x6) which is quite
different from other MASK defines.
To make it consistent with other mask defines, define AD193X_PLL_INPUT_MASK
as 0x6 and change the code accordingly.
I think this change improves the readability.
Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Use snd_soc_update_bits for read-modify-write register access instead of
open-coding it using snd_soc_read and snd_soc_write
This patch also includes a comment fix in wm8990_set_dai_pll(),
if freq_in and freq_out are 0, what we do is to clear WM8990_PLL_ENA bit.
Thus the comment should be "Turn off PLL".
Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
If (fakepower & ((1 << WM8990_INMIXR_PWR_BIT) | (1 << WM8990_AINRMUX_PWR_BIT)))
is false, we should clear WM8990_AINR_ENA bits instead of WM8990_AINL_ENA.
Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
If (fakepower & ((1 << WM8400_INMIXR_PWR) | (1 << WM8400_AINRMUX_PWR)))
is false, we should clear WM8400_AINR_ENA bits instead of WM8400_AINL_ENA.
Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
