* Update fthd_ddr.c
Shipped in linux kernel 6.13, <linux/random.h> no longer includes <linux/prandom.h>. This file depended on that nested include, and this commit changes the include to use that header directly.
Without this, module compilation fails trying to use prandom_seed_state() which is in longer included.
5b3fdc9f2f
* Update fthd_ddr.c to include linux/random.h in kernels <= 6.12
When running Linux in a VM with the facetimehd module, the device must explicitly be removed.
If the device is not removed upon a shutdown of the VM, the device becomes inoperable until a full reboot of the host. Inoperable here means that the error message "Init failed! No wake signal" occurs.
This should help dracut to understand that the firmware is also needed
if the module is put into initrd.
Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Makes the driver automatically load when hardware is detected. This
should have been done a long time ago. Ooops.
Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Skype and other applications don't support stepwise frame sizes and
frame intervals. Instead they try to handle them as discrete devices
which fails. It seems they will never fix this so instead we reduce our
functionality to just support the discrete type.
Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
ioremap_nocache() got removed from upstream because ioremap() always
gave "nocache". Use ioremap() instead.
Fixes: 4bdc0d676a643140bdf17dbf7eafedee3d496a3c
remove ioremap_nocache and devm_ioremap_nocache
Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
In kernel 5.4 I otherwise get this warning
WARNING: CPU: 1 PID: 451 at drivers/media/v4l2-core/v4l2-dev.c:863 __video_register_device+0x64b/0xe90 [videodev]
The code there is:
/* the device_caps field MUST be set for all but subdevs */
if (WARN_ON(type != VFL_TYPE_SUBDEV && !vdev->device_caps))
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
My dev_priv->sensorid1 value is 0x248, which doesn't appear in the case statement. That said, the code below my change indicates that a set file itself is optional, so there should be no difference in behavior if the filename is not found or the filename is known but doesn't exist. The module and camera are functional on my machine with this change, with the caveat that no calibration is performed.
Kernel commit: 5a8361f7ecceaed64b4064000d16cb703462be49
ACPICA: Integrate package handling with module-level code
The above commit fixed a problem with ACPI that previously prevented
ASPM to be enabled. Unfortunately our driver broke because the link to
the device went down while streaming. Fix this for now by disabling ASPM
for the device entirely.
Note: There is an ACPI method we're calling to "power on/off" the device
so the real solution here might be to get it working correctly. I will
need to investigate that further so no ASPM will have to do for now.
Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
