Newer devices don't place the PMIC CLK line inside an EFI
var. Instead, those are found at the _PR0 table.

Add a parser for it, using info stored via the DSDT table.
Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Add support code for this driver to use ACPI power management.

Yet, at least with known devices, this won't work without
further changes.

The rationale is that the ACPI handling code checks for the _PR? tables
in order to know what is required to switch the device from power state
D0 (_PR0) up to D3COLD (_PR3).

The adev->flags.power_manageable is set to true if the device has a _PR0
table, which can be checked by calling acpi_device_power_manageable(adev).

However, this only says that the device can be set to power off mode.

At least on the DSDT tables we've seen so far, there's no _PR3 nor _PS3
(which would have a somewhat similar effect).

So, using ACPI for power management won't work, except if adding
an ACPI override logic somewhere.
Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The gmin_subdev_add() currently doesn't use ACPI device
power management. In order to prepare for adding support
for it, let's shift some things, placing the PM-related
stuff at the end of the probing logic.

Let's also store the current gs on a temporary var, in
order to simplify the source code.
Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Right now, the driver is not doing the right thing to detect
the clock like used by the sensor, at least on devices
without the gmin's EFI vars.

Add some notes at the code to explain why and skip the wrong
value provided by the _DSM table.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The Rockchip Video Decoder driver is expecting that the values in a
scaling list are in zig-zag order and applies the inverse scanning process
to get the values in matrix order.

Commit 0b0393d5 ("media: uapi: h264: clarify expected
scaling_list_4x4/8x8 order") clarified that the values in the scaling list
should already be in matrix order.

Fix this by removing the reordering and change to use two memcpy.

Fixes: cd33c830 ("media: rkvdec: Add the rkvdec driver")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Tested-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Reviewed-by: Ezequiel Garcia <ezequiel@collabora.com>
[hverkuil-cisco@xs4all.nl: rkvdec_scaling_matrix -> rkvdec_h264_scaling_list]
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Wrong loop filter flag is unset when tiles enabled flag is not set,
this cause HEVC decoding issues with Rockchip Video Decoder.

Fix this by unsetting the loop filter across tiles enabled flag instead of
the pps loop filter across slices enabled flag when tiles are disabled.

Fixes: 256fa392 ("media: v4l: Add definitions for HEVC stateless decoding")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Commit 9495b7e9 ("driver core: platform:
Initialize dma_parms for platform devices") in v5.7-rc5 causes
vb2_dma_contig_clear_max_seg_size() to kfree memory that was not
allocated by vb2_dma_contig_set_max_seg_size().

The assumption in vb2_dma_contig_set_max_seg_size() seems to be that
dev->dma_parms is always NULL when the driver is probed, and the case
where dev->dma_parms has bee initialized by someone else than the driver
(by calling vb2_dma_contig_set_max_seg_size) will cause a failure.

All the current users of these functions are platform devices, which now
always have dma_parms set by the driver core. To fix the issue for v5.7,
make vb2_dma_contig_set_max_seg_size() return an error if dma_parms is
NULL to be on the safe side, and remove the kfree code from
vb2_dma_contig_clear_max_seg_size().

For v5.8 we should remove the two functions and move the
dma_set_max_seg_size() calls into the drivers.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Fixes: 9495b7e9 ("driver core: platform: Initialize dma_parms for platform devices")
Cc: stable@vger.kernel.org
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Remove description of non-existing v4l2_subdev.nevents and replace the
undefined flag V4L2_SUBDEV_USES_EVENTS by the correct flag
V4L2_SUBDEV_FL_HAS_EVENTS, which is already documented in v4l2_subdev.flags

Fixes: commit 02adb1cc ("[media] v4l: subdev: Events support")
Signed-off-by: Michael Rodin <mrodin@de.adit-jv.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Commit 9495b7e9 ("driver core: platform: Initialize dma_parms for
platform devices") in v5.7-rc5 added allocation of dma_parms structure to
all platform devices. Then vb2_dma_contig_set_max_seg_size() have been
changed not to allocate dma_parms structure and rely on the one allocated
by the device core. Lets allocate the needed structure also for the
devices created for the 2 MFC device memory ports.
Reported-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Suggested-by: Ulf Hansson <ulf.hansson@linaro.org>
Fixes: 9495b7e9 ("driver core: platform: Initialize dma_parms for platform devices")
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Recently, MEDIA_CEC_SUPPORT became indepedent of MEDIA_SUPPORT.
However, if MEDIA_SUPPORT is not enabled, MEDIA_SUPPORT_FILTER is not
defined, and MEDIA_CEC_SUPPORT is thus enabled by default, which is not
desirable.

Fix this by adding a dependency on MEDIA_CEC_SUPPORT to the default
configuration.

Fixes: 46d2a3b9 ("media: place CEC menu before MEDIA_SUPPORT")
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

This allows the VE clocks and PLL_VE to be disabled most of the time.
A runtime PM reference is held while streaming.
Signed-off-by: Jernej Skrabec <jernej.skrabec@siol.net>
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Previously, the output format was programmed as part of the ioctl()
handler. However, this has two problems:

  1) If there are multiple active streams with different output
     formats, the hardware will use whichever format was set last
     for both streams. Similarly, an ioctl() done in an inactive
     context will wrongly affect other active contexts.
  2) The registers are written while the device is not actively
     streaming. To enable runtime PM tied to the streaming state,
     all hardware access needs to be moved inside cedrus_device_run().

The call to cedrus_dst_format_set() is now placed just before the
codec-specific callback that programs the hardware.

Cc: <stable@vger.kernel.org>
Fixes: 50e76151 ("media: platform: Add Cedrus VPU decoder driver")
Suggested-by: Jernej Skrabec <jernej.skrabec@siol.net>
Suggested-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Samuel Holland <samuel@sholland.org>
Tested-by: Jernej Skrabec <jernej.skrabec@siol.net>
Reviewed-by: Jernej Skrabec <jernej.skrabec@siol.net>
Reviewed-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

As there are several ways where the driver could possible
retrieve sensor data, make the prints clearer about what
was detected and from where.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Now that the EFI _DSM table is parsed by the driver, we don't
need a DMI match anymore for Asus Transform T101HA.

This reverts commit 0a76fd8e.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

This reverts commit 0d64e942.

As gmin_subdev_add() now takes the ACPI handle directly,
we can deprecate the code that were doing this inside each
I2C driver.

PS.: This also reverts commit c03496b3 ("media: atomisp: add a notice about possible leak resources")
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Instead of keep hardcoding device-specific tables, read them
directly from the ACPI BIOS, if available.

This method is know to work with Asus T101HA device. the
same table is also visible on EzPad devices. So, it seems
that at least some BIOSes use this method to pass data about
ISP2401-connected sensors.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The hive_isp_css_custom_host_hrt.h code, together
with atomisp_helper.h, provides an abstraction layer for
some functions inside atomisp_compat_css20.c and atomisp_cmd.c.

There's no good reason for that. In a matter of fact, after
removing the abstraction, the code looked a lot cleaner
and easier to understand.

So, get rid of them.

While here, get rid also of the udelay(1) abstraction code.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Some parts of the driver have their own implementation of
memcpy() & friends. Replace all of them by strscpy().
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Replace usages of strcpy(), strlcpy() and strncpy() in favor
of strscpy().
Suggested-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

If the sensor doesn't implement support for g_frame_interval,
it won't return the expected fps rate.

Instead of keeping DFS on its minimal value (which will likely
not work), set it to the max.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are several parts of the driver that could produce
a "dfs failed!" message. Change the texts, in order to help
identifying from where they're coming.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Instead of having a static var to detect it, let's use the
already-existing arch-specific bytes, as this is how other
parts of the code also checks when it needs to do something
different, depending on an specific chipset version.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Instead of hardcoding the intel family values there, use
the already defined ones from asm/intel-family.h.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The logic which sets the hpll_freq for BYT sets hpll_freq
to 1600MHz, but ignores it, and sets it again after reading
from-device-specific EFI vars (this time, using a default
of 2000MHz).

Remove the first set, as this will be overriden anyway.

While here, do minor adjustments on comments and on a
printk message.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There's a hack at the driver that selects a different table
for a BYT tablet, which sets the maximum frequency to 320 MHz,
instead of 400 MHz.

After looking at the Intel Aero Yocto's version from:

	https://download.01.org/aero/deb/pool/main/l/linux-4.4.76-aero-1.3/

It was noticed that this depends on an Android-specific modprobe parameter,
which uses a macro (INTEL_MID_BOARD) from this file:

	arch/x86/include/asm/spid.h

>From the comments there, it looks like this macro parses a
variable passed at boot time:

	cmdline : androidboot.spid=vend:cust:manu:plat:prod:hard

The devices in question are identified there as:

	INTEL_BYT_TABLET_BLK_PRO = 0x0000
	INTEL_BYT_TABLET_BLK_ENG = 0x8000

Well, this is something that we don't have upstream. So,
without further details about that, we can't really parse it.

If we ever end supporting those devices with the upstream driver,
this patch can be reverted and the device can be detected
via DMI (or maybe via PCI ID?).
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

As both isp2400 and isp2401 files are identical, remove one
of them and remove the test for ISP variant.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are some parameters that are different between
isp2400 and isp2401. None of those are actually used.

So, get rid of them.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are lots of mess with IRQ ifdef settings. As the
*_global.h will already detect the type of IRQ system at
compile time, we can get rid of them, replacing by just
one ifdef for ISP2401.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are some ifdefs there that end doing the same thing.
Get rid of them.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The contents of hive_isp_css_2401_irq_types_hrt.h and
hive_isp_css_common/irq_global.h are identical, except for
one unused enum:

On isp2401, this IRQ line has this name:

	hrt_isp_css_irq_is2401 = HIVE_GP_DEV_IRQ_ISP_PMEM_ERROR_BIT_ID,

While the same bit is named as:
	hrt_isp_css_irq_isp_pmem_error = HIVE_GP_DEV_IRQ_ISP_PMEM_ERROR_BIT_ID,

At the isp2400 version.

Remove one of them, in order to reduce the code differences
between those two versions.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

- Get rid of typedefs;
- Get rid of a duplicated enum type with different names for
  ISP2400 and ISP2401;
- adjust indentation on the touched code.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

When an IRQ is not handled, it is nice to know what's the
reason.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There's a different table for some BYT variants that depend
on something inside a FIXME ifdef.

Place this also inside it, just to shut up a clang-11 warning.
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The abstraction layer for kvfree() was removed, but there
is still a left-over code there.
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are two names for the same IRQ, but just one is used.
Remove the unused one.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There's nothing there, just comments.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

When building with clang, multiple copies of the structures to be
initialized are passed around on the stack and copied locally, using an
insane amount of stack space:

drivers/staging/media/atomisp/pci/sh_css.c:2371:1: error: stack frame size of 26864 bytes in function 'create_pipe' [-Werror,-Wframe-larger-than=]

Use constantly-allocated variables plus an explicit memcpy()
to avoid that.
Co-authored-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Fixes: 6dc9a256 ("media: atomisp: convert default struct values to use compound-literals with designated initializers")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

For most warnings, the current code is OK. There are still
some issues with implicit-fallthough warnings.

Solve those and re-enable all warnings for this driver.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

This table used to be used also to translate between ia_css
abstraction and V4L2 fourcc codes.

This was removed on a past patch, but the table now contains
two fields with identical values.

Get rid of one of them.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Without that driver, there is a link failure in

ERROR: modpost: "intel_soc_pmic_exec_mipi_pmic_seq_element"
[drivers/staging/media/atomisp/pci/atomisp_gmin_platform.ko] undefined!

Add an explicit Kconfig dependency.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>