Merge tag 'drm-habanalabs-next-2023-04-10' of https://git./linux/kernel/git/ogabbay/linux into drm-next

This tag contains additional habanalabs driver changes for v6.4:

- uAPI changes:
  - Add a definition of a new Gaudi2 server type. This is used by userspace
    to know what is the connectivity between the accelerators inside the
    server

- New features and improvements:
  - speedup h/w queues test in Gaudi2 to reduce device initialization times.

- Firmware related fixes:
  - Fixes to the handshake protocol during f/w initialization.
  - Sync f/w events interrupt in hard reset to avoid warning message.
  - Improvements to extraction of the firmware version.

- Misc bug fixes and code cleanups. Notable fixes are:
  - Multiple fixes for interrupt handling in Gaudi2.
  - Unmap mapped memory in case TLB invalidation fails.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Oded Gabbay <ogabbay@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230410124637.GA2441888@ogabbay-vm-u20.habana-labs.com

accel/habanalabs: add missing error flow in hl_sysfs_init()

hl_sysfs_fini() is called only if hl_sysfs_init() completes
successfully. Therefore if hl_sysfs_init() fails, need to remove any
sysfs group that was added until that point.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: speedup h/w queues test in Gaudi2

HW queues testing at driver load and after reset takes a substantial
amount of time.
This commit reduces the queues test time in Gaudi2 devices by running
all the tests in parallel instead of one after the other.
Time measurements on tests duration shows that the new method is almost
x100 faster than the serial approach.

Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: fix handling of arc farm sei event

There is only single eq entry for arc farm sei event which aggregates
events from the four arc farms.
Fix the code to handle this event according to this behavior.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: remove Gaudi1 multi MSI code

Multi MSI interrupts aren't working in Gaudi1 and because of that,
we are only using a single MSI interrupt. Therefore, let's remove this
dead code in order to avoid confusion.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs/uapi: new Gaudi2 server type

Add definition of a new Gaudi2 server type. This represents
the connectivity between the cards in that server type.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: fixes for unexpected error interrupt

Removing redundant asic prop variable as we don't need to expose this
to common code. In addition, fix some typos.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: don't wait for STS_OK after sending COMMS WFE

Sending COMMS_GOTO_WFE instructs the FW's CPU to halt (WFE state).
Once sent, FW's CPU isn't expected to continue communicating with LKD.
Therefore, the stage of waiting for COMMS_STS_OK should be skipped or
else waiting for COMMS_STS_OK will simply timeout, which will trigger
unexpected behavior.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: sync f/w events interrupt in hard reset

Receiving events from FW, while the device is in hard reset, causes
a warning message in Driver log. The message may point to a
problem in the Driver or FW. But It also can appear as a result
of events that have been sent from FW just before the hard reset.
In order to avoid receiving events from FW while the device is in reset
and is already in 'disabled' mode, sync the f/w events interrupt right
before setting the device to 'disabled'.

Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: fix wrong reset and event flags

During event handling, driver sets relevant reset and user event
notifier flags. Fix few wrong flags settings.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: fix events mask of decoder abnormal interrupts

The decoder IRQ status register may have several set bits upon an
abnormal interrupt. Therefore, when setting the events mask, need to
check all bits and not using if-else.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: remove completion from abnormal interrupt work name

Decoder abnormal interrupts are for errors and not for completion, so
rename the relevant work and work function to not include 'completion'.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: print raw binning masks in debug level

There are rare cases of failures when cards are initialized due to
wrong values in efuse mappings that are parsed by firmware.

To help debug those cases, print (in debug level) the raw binning masks
as fetched from the firmware during device initialization.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: fix HBM MMU interrupt handling

Current mapping between HMMU event and HMMU block is wrong.
In addition the captured address in case of a page fault or
an access error is scrambled, Hence we must call the descramble
function.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: improvements to FW ver extraction

1. Rename the func to hl_get_preboot_major_minor because we also set
   the extracted values in hdev fields.

2. Free the allocated string in the calling function which makes more
   sense

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: fix access error clear event

The register which needs to be cleared is the valid register instead
of the address.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

accel/habanalabs: send disable pci when compute ctx is active

Fix an issue in hard reset flow in which the driver didn't send a
disable pci message if there was an active compute context.
In hard reset, disable pci message should be sent no matter if
a compute context exists or not.

Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: remove duplicated disable pci msg

The disable pci message is sent in reset device. It informs the FW not
to raise more EQs. The Driver may ignore received EQs, when the device
is in disabled mode.
The duplication happens when hard reset is scheduled during compute
reset and also performs 'escalate_reset_flow'.

Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: change COMMS warning messages to error level

COMMS protocol is used for LKD <--> FW communication, and any
communication failure between the two might turn out to be
destructive, hence, it should be well emphasized.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: check return value of add_va_block_locked

since the function might fail and we should propagate the failure.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: print event type when device is disabled

When the device is in disabled state, the driver isn't suppose to
receive any events from FW. Printing the event type, as part of the
message that was already printed, shall help to get more info if this
unexpected message is received.

Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: unmap mapped memory when TLB inv fails

Once a memory mapping is added to the page tables, it's followed by
a TLB invalidation request which could potentially fail (HW failure).
Removing the mapping is simply a part of this failure handling routine.
TLB invalidation failure prints were updated to be more accurate.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>

accel/habanalabs: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
u16 pci_command;

pci_read_config_word(dev, PCI_COMMAND, &pci_command);
if (pci_command & PCI_COMMAND_MASTER) {
pci_command &= ~PCI_COMMAND_MASTER;
pci_write_config_word(dev, PCI_COMMAND, pci_command);
}

pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

Merge tag 'drm-intel-next-2023-04-06' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

- Fix DPT+shmem combo and add i915.enable_dpt modparam (Ville)
- i915.enable_sagv module parameter (Ville)
- Correction to QGV related register addresses (Vinod)
- IPS debugfs per-crtc and new file for false_color (Ville)
- More clean-up and reorganization of Display code (Jani)
- DP DSC related fixes and improvements (Stanislav, Ankit, Suraj, Swati)
- Make utility pin asserts more accurate (Ville)
- Meteor Lake enabling (Daniele)
- High refresh rate PSR fixes (Jouni)
- Cursor and Plane chicken register fixes (Ville)
- Align the ADL-P TypeC sequences with hardware specification (Imre)
- Documentation build fixes and improvements to catch bugs earlier (Lee, Jani)
- PL1 power limit hwmon entry changed to use 0 as disabled state (Ashutosh)
- DP aux sync fix and improvements (Ville)
- DP MST fixes and w/a (Stanislav)
- Limit PXP drm-errors or warning on firmware API failures (Alan)

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZC7RR3Laet8ywHRo@intel.com

Merge tag 'drm/tegra/for-6.4-rc1' of https://gitlab.freedesktop.org/drm/tegra into drm-next

drm/tegra: Changes for v6.4-rc1

The majority of this is minor cleanups and fixes. Other than those, this
contains Uwe's conversion to the new driver remove callback and Thomas'
fbdev DRM client conversion. The driver can now also be built on other
architectures to easy compile coverage.

Finally, this adds Mikko as a second maintainer for the driver. As a
next step we also want Tegra DRM to move into drm-misc to streamline the
maintenance process.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Thierry Reding <thierry.reding@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406121404.967704-1-thierry.reding@gmail.com

drm/i915/wakeref: fix kernel-doc comment

Fix the warning:

drivers/gpu/drm/i915/intel_wakeref.h:118: warning: expecting prototype
for intel_wakeref_get_if_in_use(). Prototype was for
intel_wakeref_get_if_active() instead

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405104142.766598-2-jani.nikula@intel.com

drm/i915/tc: demote a kernel-doc comment to a regular comment

There's not much point in a static work function having a kernel-doc
comment. Just clean it up and make it a regular comment.

This fixes the kernel-doc warnings:

drivers/gpu/drm/i915/display/intel_tc.c:1370: warning: Function
parameter or member 'work' not described in
'intel_tc_port_disconnect_phy_work'

drivers/gpu/drm/i915/display/intel_tc.c:1370: warning: Excess function
parameter 'dig_port' description in 'intel_tc_port_disconnect_phy_work'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405104142.766598-1-jani.nikula@intel.com

Merge tag 'drm-misc-next-2023-04-06' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

drm-misc-next for v6.4-rc1:

UAPI Changes:

Cross-subsystem Changes:
- Document port and rotation dt bindings better.
- For panel timing DT bindings, document that vsync and hsync are
  first, rather than last in image.
- Fix video/aperture typos.

Core Changes:
- Reject prime DMA-Buf attachment if get_sg_table is missing.
  (For self-importing dma-buf only.)
- Add prime import/export to vram-helper.
- Fix oops in drm/vblank when init is not called.
- Fixup xres/yres_virtual and other fixes in fb helper.
- Improve SCDC debugs.
- Skip setting deadline on modesets.
- Assorted TTM fixes.

Driver Changes:
- Add lima usage stats.
- Assorted fixes to bridge/lt8192b, tc358767, ivpu,
  bridge/ti-sn65dsi83, ps8640.
- Use pci aperture helpers in drm/ast lynxfb, radeonfb.
- Revert some lima patches, as they required a commit that has been
  reverted upstream.
- Add AUO NE135FBM-N41 v8.1 eDP panel.
- Add QAIC accel driver.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/64bb9696-a76a-89d9-1866-bcdf7c69c284@linux.intel.com

Merge tag 'drm-intel-gt-next-2023-04-06' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

UAPI Changes:

- (Build-time only, should not have any impact)
  drm/i915/uapi: Replace fake flex-array with flexible-array member

  "Zero-length arrays as fake flexible arrays are deprecated and we are
  moving towards adopting C99 flexible-array members instead."

  This is on core kernel request moving towards GCC 13.

Driver Changes:

- Fix context runtime accounting on sysfs fdinfo for heavy workloads (Tvrtko)
- Add support for OA media units on MTL (Umesh)
- Add new workarounds for Meteorlake (Daniele, Radhakrishna, Haridhar)
- Fix sysfs to read actual frequency for MTL and Gen6 and earlier
  (Ashutosh)
- Synchronize i915/BIOS on C6 enabling on MTL (Vinay)
- Fix DMAR error noise due to GPU error capture (Andrej)
- Fix forcewake during BAR resize on discrete (Andrzej)
- Flush lmem contents after construction on discrete (Chris)
- Fix GuC loading timeout on systems where IFWI programs low boot
  frequency (John)
- Fix race condition UAF in i915_perf_add_config_ioctl (Min)

- Sanitycheck MMIO access early in driver load and during forcewake
  (Matt)
- Wakeref fixes for GuC RC error scenario and active VM tracking (Chris)
- Cancel HuC delayed load timer on reset (Daniele)
- Limit double GT reset to pre-MTL (Daniele)
- Use i915 instead of dev_priv insied the file_priv structure (Andi)
- Improve GuC load error reporting (John)
- Simplify VCS/BSD engine selection logic (Tvrtko)
- Perform uc late init after probe error injection (Andrzej)
- Fix format for perf_limit_reasons in debugfs (Vinay)
- Create per-gt debugfs files (Andi)

- Documentation and kerneldoc fixes (Nirmoy, Lee)
- Selftest improvements (Fei, Jonathan)

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZC6APj/feB+jBf2d@jlahtine-mobl.ger.corp.intel.com

drm/tegra: Avoid potential 32-bit integer overflow

In tegra_sor_compute_config(), the 32-bit value mode->clock is
multiplied by 1000, and assigned to the u64 variable pclk. We can avoid
a potential 32-bit integer overflow by casting mode->clock to u64 before
we do the arithmetic and assignment.

Signed-off-by: Nur Hussein <hussein@unixcat.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Implement fbdev emulation as in-kernel client

Move code from ad-hoc fbdev callbacks into DRM client functions
and remove the old callbacks. The functions instruct the client
to poll for changed output or restore the display. The DRM core
calls both, the old callbacks and the new client helpers, from
the same places. The new functions perform the same operation as
before, so there's no change in functionality.

Replace all code that initializes or releases fbdev emulation
throughout the driver. Instead initialize the fbdev client by a
single call to tegra_fbdev_setup() after tegra has registered
its DRM device. As in most drivers, tegra's fbdev emulation now
acts like a regular DRM client.

The fbdev client setup consists of the initial preparation and the
hot-plugging of the display. The latter creates the fbdev device
and sets up the fbdev framebuffer. The setup performs display
hot-plugging once. If no display can be detected, DRM probe helpers
re-run the detection on each hotplug event.

A call to drm_dev_unregister() releases the client automatically.
No further action is required within tegra. If the fbdev
framebuffer has been fully set up, struct fb_ops.fb_destroy
implements the release. For partially initialized emulation, the
fbdev client reverts the initial setup.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Initialize fbdev DRM client

Initialize the fbdev client in the fbdev code with empty helper
functions. Also clean up the client. The helpers will later
implement various functionality of the DRM client. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Hide fbdev support behind config option

Only build tegra's fbdev emulation if CONFIG_DRM_FBDEV_EMULATION
has been enabled. As part of this change, move the code into its
own source file. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/ttm: Make the call to ttm_tt_populate() interruptible when faulting

When swapping in, or under memory pressure ttm_tt_populate() may sleep
for a substantiable amount of time. Allow interrupts during the sleep.
This will also allow us to inject -EINTR errors during swapin in upcoming
patches.

Also avoid returning VM_FAULT_OOM, since that will confuse the core
mm, making it print out a confused message and retrying the fault.
Return VM_FAULT_SIGBUS also under OOM conditions.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404200650.11043-4-thomas.hellstrom@linux.intel.com

drm/ttm: Reduce the number of used allocation orders for TTM pages

When swapping out, we will split multi-order pages both in order to
move them to the swap-cache and to be able to return memory to the
swap cache as soon as possible on a page-by-page basis.
Reduce the page max order to the system PMD size, as we can then be nicer
to the system and avoid splitting gigantic pages.

Looking forward to when we might be able to swap out PMD size folios
without splitting, this will also be a benefit.

v2:
- Include all orders up to the PMD size (Christian König)
v3:
- Avoid compilation errors for architectures with special PFN_SHIFTs

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404200650.11043-3-thomas.hellstrom@linux.intel.com

drm/ttm/pool: Fix ttm_pool_alloc error path

When hitting an error, the error path forgot to unmap dma mappings and
could call set_pages_wb() on already uncached pages.

Fix this by introducing a common ttm_pool_free_range() function that
does the right thing.

v2:
- Simplify that common function (Christian König)
v3:
- Rename that common function to ttm_pool_free_range() (Christian König)

Fixes: d099fc8f540a ("drm/ttm: new TT backend allocation pool v3")
Cc: Christian König <christian.koenig@amd.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Huang Rui <ray.huang@amd.com>
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404200650.11043-2-thomas.hellstrom@linux.intel.com

MAINTAINERS: Add entry for QAIC driver

Add MAINTAINERS entry for the Qualcomm Cloud AI 100 driver.

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-9-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add qaic driver to the build system

Now that we have all the components of a minimum QAIC which can boot and
run an AIC100 device, add the infrastructure that allows the QAIC driver
to be built.

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-8-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add mhi_qaic_cntl

Some of the MHI channels for an AIC100 device need to be routed to
userspace so that userspace can communicate directly with QSM. The MHI
bus does not support this, and while the WWAN subsystem does (for the same
reasons), AIC100 is not a WWAN device. Also, MHI is not something that
other accelerators are expected to share, thus an accel subsystem function
that meets this usecase is unlikely.

Create a QAIC specific MHI userspace shim that exposes these channels.

Start with QAIC_SAHARA which is required to boot AIC100 and is consumed by
the kickstart application as documented in aic100.rst

Each AIC100 instance (currently, up to 16) in a system will create a
chardev for QAIC_SAHARA. This chardev will be found as
/dev/<mhi instance>_QAIC_SAHARA
For example - /dev/mhi0_QAIC_SAHARA

Signed-off-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-7-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add datapath

Add the datapath component that manages BOs and submits them to running
workloads on the qaic device via the dma_bridge hardware. This allows
QAIC clients to interact with their workloads (run inferences) via the
following ioctls along with mmap():

DRM_IOCTL_QAIC_CREATE_BO
DRM_IOCTL_QAIC_MMAP_BO
DRM_IOCTL_QAIC_ATTACH_SLICE_BO
DRM_IOCTL_QAIC_EXECUTE_BO
DRM_IOCTL_QAIC_PARTIAL_EXECUTE_BO
DRM_IOCTL_QAIC_WAIT_BO
DRM_IOCTL_QAIC_PERF_STATS_BO

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-6-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add control path

Add the control path component that talks to the management processor (QSM)
to load workloads onto the AIC100 device. This implements the KMD portion
of the NNC protocol over the QAIC_CONTROL MHI channel and the
DRM_IOCTL_QAIC_MANAGE IOCTL to userspace. With this functionality, QAIC
clients are able to load, run, and cleanup their workloads on the device
but not interact with the workloads (run inferences).

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-5-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add MHI controller

An AIC100 device contains a MHI interface with a number of different
channels for controlling different aspects of the device. The MHI
controller works with the MHI bus to enable and drive that interface.

AIC100 uses the BHI protocol in PBL to load SBL. The MHI controller
expects the SBL to be located at /lib/firmware/qcom/aic100/sbl.bin and
expects the MHI bus to manage the process of loading and sending SBL to
the device.

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-4-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add uapi and core driver file

Add the QAIC driver uapi file and core driver file that binds to the PCIe
device. The core driver file also creates the accel device and manages
all the interconnections between the different parts of the driver.

The driver can be built as a module. If so, it will be called "qaic.ko".

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-3-git-send-email-quic_jhugo@quicinc.com

accel/qaic: Add documentation for AIC100 accelerator driver

The Qualcomm Cloud AI 100 (AIC100) device is an Artificial Intelligence
accelerator PCIe card. It contains a number of components both in the
SoC and on the card which facilitate running workloads:

QSM: management processor
NSPs: workload compute units
DMA Bridge: dedicated data mover for the workloads
MHI: multiplexed communication channels
DDR: workload storage and memory

The Linux kernel driver for AIC100 is called "QAIC" and is located in the
accel subsystem.

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Acked-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1679932497-30277-2-git-send-email-quic_jhugo@quicinc.com

drm/atomic-helper: Don't set deadline for modesets

If the crtc is being switched on or off then the semantics of
computing the timestampe of the next vblank is somewhat ill-defined.
And indeed, the code splats with a warning in the timestamp
computation code. Specifically it hits the check to make sure that
atomic drivers have full set up the timing constants in the drm_vblank
structure, and that's just not the case before the crtc is actually
on.

For robustness it seems best to just not set deadlines for modesets.

v2: Also skip on inactive crtc (Ville)

Link: https://lore.kernel.org/dri-devel/dfc21f18-7e1e-48f0-c05a-d659b9c90b91@linaro.org/
Fixes: d39e48ca80c0 ("drm/atomic-helper: Set fence deadline for vblank")
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Rob Clark <robdclark@chromium.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> # test patch only
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405133105.947834-1-daniel.vetter@ffwll.ch

drm/scdc-helper: Pimp SCDC debugs

Include the device and connector information in the SCDC
debugs. Makes it easier to figure out who did what.

v2: Rely on connector->ddc (Maxime)

Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Robert Foss <rfoss@kernel.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Emma Anholt <emma@anholt.net>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: intel-gfx@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403223652.18848-1-ville.syrjala@linux.intel.com
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Thierry Reding <treding@nvidia.com>

drm/fb-helper: fix input validation gaps in check_var

Apparently drivers need to check all this stuff themselves, which for
most things makes sense I guess. And for everything else we luck out,
because modern distros stopped supporting any other fbdev drivers than
drm ones and I really don't want to argue anymore about who needs to
check stuff. Therefore fixing all this just for drm fbdev emulation is
good enough.

Note that var->active is not set or validated. This is just control
flow for fbmem.c and needs to be validated in there as needed.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404194038.472803-3-daniel.vetter@ffwll.ch

drm/fb-helper: drop redundant pixclock check from drm_fb_helper_set_par()

The fb_check_var hook is supposed to validate all this stuff. Any
errors from fb_set_par are considered driver/hw issues and resulting
in dmesg warnings.

Luckily we do fix up the pixclock already, so this is all fine.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404194038.472803-2-daniel.vetter@ffwll.ch

drm/fb-helper: set x/yres_virtual in drm_fb_helper_check_var

Drivers are supposed to fix this up if needed if they don't outright
reject it. Uncovered by 6c11df58fd1a ("fbmem: Check virtual screen
sizes in fb_set_var()").

Reported-by: syzbot+20dcf81733d43ddff661@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?id=c5faf983bfa4a607de530cd3bb008888bf06cefc
Cc: stable@vger.kernel.org # v5.4+
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404194038.472803-1-daniel.vetter@ffwll.ch

drm/i915/clock: mass rename dev_priv to i915

Follow the contemporary naming style. Include some indentation fixes
while at it on the affected statements.

One function needs to keep using dev_priv due to implicit dev_priv usage
in a macro.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403122428.3526263-2-jani.nikula@intel.com

drm/i915: rename intel_pm.[ch] to intel_clock_gating.[ch]

Observe that intel_pm.[ch] is now purely about clock gating, so rename
them to intel_clock_gating.[ch]. Rename the functions to
intel_clock_gating_*() to follow coding conventions.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403122428.3526263-1-jani.nikula@intel.com

drm/i915/pxp: limit drm-errors or warning on firmware API failures

MESA driver is creating protected context on every driver handle
creation to query caps bits for app. So when running CI tests,
they are observing hundreds of drm_errors when enabling PXP
in .config but using SOC fusing or BIOS configuration that cannot
support PXP sessions.

The fixes tag referenced below was to resolve a related issue
where we wanted to silence error messages, but that case was due
to outdated IFWI (firmware) that definitely needed an upgrade and
was, at that point, considered a one-off case as opposed to today's
realization that default CI was enabling PXP in kernel config for
all testing.

So with this patch, let's strike a balance between issues that is
critical but are root-caused from HW/platform gaps (louder drm-warn
but just ONCE) vs other cases where it could also come from session
state machine (which cannot be a WARN_ONCE since it can be triggered
due to runtime operation events).

Let's use helpers for these so as more functions are added in future
features / HW (or as FW designers continue to bless upstreaming of
the error codes and meanings), we only need to update the helpers.

NOTE: Don't completely remove FW errors (via drm_debug) or else cusomer
apps that really needs to know that content protection failed won't
be aware of it.

v2: - Add fixes tag (Trvtko)
v3: - Break multi-line drm_dbg strings into separate drm_dbg (Daniele)
    - Fix couple of typecasting nits (Daniele)
v4: - Unsuccessful PXP FW cmd due to platform configuration shouldn't
      use drm_WARN_once (Tvrtko), Switched to use drm_info_once.
v5: - Added "reported-and-tested" by Eero.

Reported-and-tested-by: Eero Tamminen <eero.t.tamminen@intel.com>
Fixes: b762787bf767 ("drm/i915/pxp: Use drm_dbg if arb session failed due to fw version")
Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230323184156.4140659-1-alan.previn.teres.alexis@intel.com

drm/bridge: ti-sn65dsi86: Allow GPIO operations to sleep

There is no need to require non-sleeping GPIO access. Silence the
WARN_ON() if GPIO is using e.g. I2C expanders.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405135127.769665-1-alexander.stein@ew.tq-group.com

drm/panel-edp: Add AUO NE135FBM-N41 v8.1 panel entry

Add a panel entry with delay_200_500_e50 for the AUO NE135FBM-N41
version 8.1, found on a number of ACER laptops, including the
Swift 3 (SF313-52, SF313-53), Chromebook Spin 513 (CP513-2H) and
others.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405100452.44225-1-angelogioacchino.delregno@collabora.com

drm/tegra: Remove struct tegra_fbdev

Remove struct tegra_fbdev, which is an empty wrapper around struct
drm_fb_helper. Use the latter directly. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Removed fb from struct tegra_fbdev

Fbdev's struct fb_helper stores a pointer to the framebuffer. Remove
struct tegra_fbdev.fb, which contains the same value. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Include <linux/i2c.h>

Include <linux/i2c.h> to get the contained declarations. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Include <linux/of.h>

Include <linux/of.h> to get the contained declarations. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Fix another missing include

Since Tegra now compile tests on other platforms the kernel test robot
started to complain that this here is not pulled in under all
conditions.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304050946.yGGTKkcr-lkp@intel.com/
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/i915/mtl: Add Wa_14017856879

Wa_14017856879 implementation for mtl.

Bspec: 46046

Signed-off-by: Haridhar Kalvala <haridhar.kalvala@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404173220.3175577-1-haridhar.kalvala@intel.com

drm/bridge: ps8640: Use constant sleep time for polling hpd

The default hpd_wait_us in panel_edp.c is 2 seconds. This makes the
sleep time in the polling of _ps8640_wait_hpd_asserted become 200ms.
Change it to a constant 20ms to speed up the function.

Signed-off-by: Pin-yen Lin <treapking@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230331030204.1179524-1-treapking@chromium.org

drm/bridge: ti-sn65dsi83: Do not generate HFP/HBP/HSA and EOT packet

Do not generate the HS front and back porch gaps, the HSA gap and
EOT packet, as per "SN65DSI83 datasheet SLLSEC1I - SEPTEMBER 2012
- REVISED OCTOBER 2020", page 22, these packets are not required.
This makes the TI SN65DSI83 bridge work with Samsung DSIM on i.MX8MN.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403190242.224490-1-marex@denx.de

drm/i915/gt: Hold a wakeref for the active VM

There may be a disconnect between the GT used by the engine and the GT
used for the VM, requiring us to hold a wakeref on both while the GPU is
active with this request.

v2: added explanation to __queue_and_release_pm

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[ahajda: removed not-yet-upstremed wakeref tracking bits]
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330-hold_wakeref_for_active_vm-v2-1-724d201499c2@intel.com

drm/i915: run kernel-doc on headers as part of HDRTEST

Enabling kernel-doc warnings in commit aaee4bbe8a1a ("drm/i915: enable
kernel-doc warnings for CONFIG_DRM_I915_WERROR=y") actually only covers
the .c files. And it's good for avoiding warnings in W= builds. However,
we need something more to check for kernel-doc issues in headers. Add it
as part of the existing HDRTEST.

We have tons of issues, and this unleashes warnings galore on
CONFIG_DRM_I915_WERROR=y. It doesn't fail the build because (at least
for now) we don't pass -Werror to kernel-doc.

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404090528.173075-1-jani.nikula@intel.com

drm: bridge: ldb: add support for using channel 1 only

The LDB driver currently checks whether dual mode is used, otherwise it
assumes only channel 0 is in use. Add support for using only channel 1. In
device tree terms, this means linking port 2 only.

Doing this cleanly requires changing the logic of the probe functions from
this:

 1. use of_graph_get_remote_node() on port 1 to find the panel
 2. use drm_of_lvds_get_dual_link_pixel_order() to detect dual mode

to this:

 1. use of_graph_get_remote_node() twice to find remote ports
 2. reuse the result of the above to know whether each channel is enabled
    and to find the panel
 3. if (both channels as enabled)
        use drm_of_lvds_get_dual_link_pixel_order() to detect dual mode

Also add a dev_dbg() to log the detected mode and log an error in case no
panel was found (no channel enabled).

Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405081058.2347130-2-luca.ceresoli@bootlin.com

drm: bridge: ldb: add missing \n in dev_warn() string

dev_warn() and similar require a training \n.

Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405081058.2347130-1-luca.ceresoli@bootlin.com

drm/bridge: ti-sn65dsi83: use dev_err_probe if host attach failed

There might be cases where the host attach is deferred, use dev_err_probe
to add more detailed information to /sys/kernel/debug/devices_deferred.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230405075223.579461-1-alexander.stein@ew.tq-group.com

drm/i915: Implement UHBR bandwidth check

According to spec, we should check if output_bpp * pixel_rate is less
than DDI clock * 72, if UHBR is used.

v2: - s/pipe_config/crtc_state/ (Jani Nikula)
    - Merged previous patch into that one, to remove empty function(Jani Nikula)

v3: - Make that constraint check to be DSC-related only
    - Limit this to only DISPLAY_VER <= 13

v4: - Move constraint check to the top(Vinod Govindapillai)

HSDES: 1406899791
BSPEC: 49259

Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230324135125.6720-1-stanislav.lisovskiy@intel.com

accel/ivpu: Remove D3hot delay for Meteorlake

VPU on MTL has hardware optimizations and does not require 10ms
D0 - D3hot transition delay imposed by PCI specification (PCIe
r6.0, sec 5.9.) .

The delay removal is traditionally done by adding PCI ID to
quirk_remove_d3hot_delay() in drivers/pci/quirks.c . But since
we do not need that optimization before driver probe and we
can better specify in the ivpu driver on what (future) hardware
use the optimization, we do not use quirk_remove_d3hot_delay()
for that.

Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403121545.2995279-1-stanislaw.gruszka@linux.intel.com

Revert "drm/lima: add usage counting method to ctx_mgr"

This reverts commit bccafec957a5c4b22ac29e53a39e82d0a0008348.

This is due to the depend commit has been reverted on upstream:
commit baad10973fdb ("Revert "drm/scheduler: track GPU active time per entity"")

Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404002601.24136-4-yq882255@163.com

Revert "drm/lima: allocate unique id per drm_file"

This reverts commit 87767de835edf527b879a363d518c33da68adb81.

This is due to the depend commit has been reverted on upstream:
commit baad10973fdb ("Revert "drm/scheduler: track GPU active time per entity"")

Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404002601.24136-3-yq882255@163.com

Revert "drm/lima: add show_fdinfo for drm usage stats"

This reverts commit 4a66f3da99dcb4dcbd28544110636b50adfb0f0d.

This is due to the depend commit has been reverted on upstream:
commit baad10973fdb ("Revert "drm/scheduler: track GPU active time per entity"")

Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404002601.24136-2-yq882255@163.com

drm/vblank: Fix for drivers that do not drm_vblank_init()

This should fix a crash that was reported on ast (and possibly other
drivers which do not initialize vblank).

   fbcon: Taking over console
   Unable to handle kernel NULL pointer dereference at virtual address 0000000000000074
   Mem abort info:
     ESR = 0x0000000096000004
     EC = 0x25: DABT (current EL), IL = 32 bits
     SET = 0, FnV = 0
     EA = 0, S1PTW = 0
     FSC = 0x04: level 0 translation fault
   Data abort info:
     ISV = 0, ISS = 0x00000004
     CM = 0, WnR = 0
   user pgtable: 4k pages, 48-bit VAs, pgdp=0000080009d16000
   [0000000000000074] pgd=0000000000000000, p4d=0000000000000000
   Internal error: Oops: 0000000096000004 [#1] SMP
   Modules linked in: ip6table_nat tun nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 rfkill ip_set nf_tables nfnetlink qrtr sunrpc binfmt_misc vfat fat xfs snd_usb_audio snd_hwdep snd_usbmidi_lib snd_seq snd_pcm snd_rawmidi snd_timer snd_seq_device snd soundcore joydev mc ipmi_ssif ipmi_devintf ipmi_msghandler arm_spe_pmu arm_cmn arm_dsu_pmu arm_dmc620_pmu cppc_cpufreq loop zram crct10dif_ce polyval_ce nvme polyval_generic ghash_ce sbsa_gwdt igb nvme_core ast nvme_common i2c_algo_bit xgene_hwmon gpio_dwapb scsi_dh_rdac scsi_dh_emc scsi_dh_alua ip6_tables ip_tables dm_multipath fuse
   CPU: 12 PID: 469 Comm: kworker/12:1 Not tainted 6.3.0-rc2-00008-gd39e48ca80c0 #1
   Hardware name: ADLINK AVA Developer Platform/AVA Developer Platform, BIOS TianoCore 2.04.100.07 (SYS: 2.06.20220308) 09/08/2022
   Workqueue: events fbcon_register_existing_fbs
   pstate: 20400009 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
   pc : drm_crtc_next_vblank_start+0x2c/0x98
   lr : drm_atomic_helper_wait_for_fences+0x90/0x240
   sp : ffff80000d583960
   x29: ffff80000d583960 x28: ffff07ff8fc187b0 x27: 0000000000000000
   x26: ffff07ff99c08c00 x25: 0000000000000038 x24: ffff07ff99c0c000
   x23: 0000000000000001 x22: 0000000000000038 x21: 0000000000000000
   x20: ffff07ff9640a280 x19: 0000000000000000 x18: ffffffffffffffff
   x17: 0000000000000000 x16: ffffb24d2eece1c0 x15: 0000003038303178
   x14: 3032393100000048 x13: 0000000000000000 x12: 0000000000000000
   x11: 0000000000000000 x10: 0000000000000000 x9 : ffffb24d2eeeaca0
   x8 : ffff80000d583628 x7 : 0000080077783000 x6 : 0000000000000000
   x5 : ffff80000d584000 x4 : ffff07ff99c0c000 x3 : 0000000000000130
   x2 : 0000000000000000 x1 : ffff80000d5839c0 x0 : ffff07ff99c0cc08
   Call trace:
    drm_crtc_next_vblank_start+0x2c/0x98
    drm_atomic_helper_wait_for_fences+0x90/0x240
    drm_atomic_helper_commit+0xb0/0x188
    drm_atomic_commit+0xb0/0xf0
    drm_client_modeset_commit_atomic+0x218/0x280
    drm_client_modeset_commit_locked+0x64/0x1a0
    drm_client_modeset_commit+0x38/0x68
    __drm_fb_helper_restore_fbdev_mode_unlocked+0xb0/0xf8
    drm_fb_helper_set_par+0x44/0x88
    fbcon_init+0x1e0/0x4a8
    visual_init+0xbc/0x118
    do_bind_con_driver.isra.0+0x194/0x3a0
    do_take_over_console+0x50/0x70
    do_fbcon_takeover+0x74/0xf8
    do_fb_registered+0x13c/0x158
    fbcon_register_existing_fbs+0x78/0xc0
    process_one_work+0x1ec/0x478
    worker_thread+0x74/0x418
    kthread+0xec/0x100
    ret_from_fork+0x10/0x20
   Code: f9400004 b9409013 f940a082 9ba30a73 (b9407662)
   ---[ end trace 0000000000000000 ]---

v2: Use drm_dev_has_vblank()

Reported-by: Nathan Chancellor <nathan@kernel.org>
Fixes: d39e48ca80c0 ("drm/atomic-helper: Set fence deadline for vblank")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403160314.1210533-1-robdclark@gmail.com

drm/tegra: dsi: Clear enable register if powered by bootloader

In cases where the DSI module is left on by the bootloader
some panels may fail to initialize if the enable register is not cleared
before the panel's initialization sequence is sent, so clear it if that
is the case.

Signed-off-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: plane: Improve use of dev_err_probe()

Return dev_err_probe() directly, because the return value of
dev_err_probe() is the appropriate error code, and it can
reduce code size, simplify the code.

Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: dsi: Make use of the helper function dev_err_probe()

When possible use dev_err_probe help to properly deal with the
PROBE_DEFER error, the benefit is that DEFER issue will be logged
in the devices_deferred debugfs file.
And using dev_err_probe() can reduce code size, the error value
gets printed.

Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: sor: Make use of the helper function dev_err_probe()

When possible use dev_err_probe help to properly deal with the
PROBE_DEFER error, the benefit is that DEFER issue will be logged
in the devices_deferred debugfs file.
And using dev_err_probe() can reduce code size, the error value
gets printed.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: sor: Remove redundant error logging

A call to platform_get_irq() already prints an error on failure within
its own implementation. So printing another error based on its return
value in the caller is redundant and should be removed. The clean up
also makes if condition block braces unnecessary. Remove that as well.

Issue identified using platform_get_irq.cocci coccicheck script.

Signed-off-by: Deepak R Varma <drv@mailo.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: dc: Remove set but unused variable 'state'

Fixes the following W=1 kernel build warning(s):

 drivers/gpu/drm/tegra/dc.c: In function ‘tegra_crtc_calculate_memory_bandwidth’:
 drivers/gpu/drm/tegra/dc.c:2384:38: warning: variable ‘old_state’ set but not used [-Wunused-but-set-variable]

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: David Airlie <airlied@gmail.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: dri-devel@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Signed-off-by: Lee Jones <lee@kernel.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: Allow compile test on !ARM v2

This compile tests on x86 just perfectly fine.

v2: fix missing include complained by kernel test robot

Signed-off-by: Christian König <christian.koenig@amd.com>
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Jonathan Hunter <jonathanh@nvidia.com>
CC: linux-tegra@vger.kernel.org
Signed-off-by: Thierry Reding <treding@nvidia.com>

fbdev/radeon: use pci aperture helpers

It's not exactly the same since the open coded version doesn't set
primary correctly. But that's a bugfix, so shouldn't hurt really.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: linux-fbdev@vger.kernel.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230111154112.90575-7-daniel.vetter@ffwll.ch

staging/lynxfb: Use pci aperture helper

It exists! Note that since this is an exact copy, there shouldn't be
any functional difference here.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Cc: Teddy Wang <teddy.wang@siliconmotion.com>
Cc: linux-fbdev@vger.kernel.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230111154112.90575-6-daniel.vetter@ffwll.ch

drm/ast: Use drm_aperture_remove_conflicting_pci_framebuffers

It's just open coded and matches.

Note that Thomas said that his version apparently failed for some
reason, but hey maybe we should try again.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Tested-by: Thomas Zimmmermann <tzimmermann@suse.de>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230111154112.90575-1-daniel.vetter@ffwll.ch

drm/tegra: vic: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: sor: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: nvdec: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: hub: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: hdmi: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: gr3d: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: gr2d: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: dsi: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: dpaux: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: dc: Convert to platform remove callback returning void

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

drm/tegra: rgb: Make tegra_dc_rgb_remove() return void

This function returned zero unconditionally. Make it return no value and
simplify all callers accordingly.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>

gpu: host1x: Fix memory leak of device names

The device names allocated by dev_set_name() need be freed
before module unloading, but they can not be freed because
the kobject's refcount which was set in device_initialize()
has not be decreased to 0.

As comment of device_add() says, if it fails, use only
put_device() drop the refcount, then the name will be
freed in kobejct_cleanup().

device_del() and put_device() can be replaced with
device_unregister(), so call it to unregister the added
successfully devices, and just call put_device() to the
not added device.

Add a release() function to device to avoid null release()
function WARNING in device_release(), it's empty, because
the context devices are freed together in
host1x_memory_context_list_free().

Fixes: 8aa5bcb61612 ("gpu: host1x: Add context device management code")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

gpu: host1x: Fix potential double free if IOMMU is disabled

If context device has no IOMMU, the 'cdl->devs' is freed in
error path, but host1x_memory_context_list_init() doesn't
return an error code, so the module can be loaded successfully,
when it's unloading, the host1x_memory_context_list_free() is
called in host1x_remove(), it will cause double free. Set the
'cdl->devs' to NULL after freeing it to avoid double free.

Fixes: 8aa5bcb61612 ("gpu: host1x: Add context device management code")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

gpu: host1x: mipi: Use devm_platform_get_and_ioremap_resource()

Convert platform_get_resource(), devm_ioremap_resource() to a single
call to devm_platform_get_and_ioremap_resource(), as this is exactly
what this function does.

Signed-off-by: Ye Xingchen <ye.xingchen@zte.com.cn>
Signed-off-by: Thierry Reding <treding@nvidia.com>

gpu: host1x: Don't rely on dma_fence_wait_timeout return value

dma_fence_wait_timeout (along with a host of other jiffies-based
timeouting functions) returns zero both in case of timeout and when
the wait completes during the last jiffy before timeout. As such,
we can't rely on it to distinguish between success and timeout.

To prevent confusing callers by returning -EAGAIN before the timeout
period has elapsed, check if the fence got signaled again after
the wait.

Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

gpu: host1x: Make host1x_client_unregister() return void

This function returned zero unconditionally. Make it return no value and
simplify all callers accordingly.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Thierry Reding <treding@nvidia.com>

MAINTAINERS: Add Mikko as backup maintainer for Tegra DRM

Mikko has been involved as the primary author of the host1x driver and
has volunteered to help out with maintenance.

Signed-off-by: Thierry Reding <treding@nvidia.com>