GitHub/moto-9609/android_kernel_motorola_exynos9610.git
8 years agodrm/msm: wire up vmap shrinker
Rob Clark [Fri, 27 May 2016 15:16:28 +0000 (11:16 -0400)]
drm/msm: wire up vmap shrinker

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: change gem->vmap() to get/put
Rob Clark [Thu, 26 May 2016 20:24:35 +0000 (16:24 -0400)]
drm/msm: change gem->vmap() to get/put

Before we can add vmap shrinking, we really need to know which vmap'ings
are currently being used.  So switch to get/put interface.  Stubbed put
fxns for now.

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: shrinker support
Rob Clark [Tue, 17 May 2016 20:19:32 +0000 (16:19 -0400)]
drm/msm: shrinker support

For a first step, only purge obj->madv==DONTNEED objects.  We could be
more agressive and next try unpinning inactive objects..  but that is
only useful if you have swap.

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: add put_iova() helper
Rob Clark [Wed, 1 Jun 2016 15:38:28 +0000 (11:38 -0400)]
drm/msm: add put_iova() helper

We'll need this too for shrinker/purging.

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: add madvise ioctl
Rob Clark [Tue, 17 May 2016 19:44:49 +0000 (15:44 -0400)]
drm/msm: add madvise ioctl

Doesn't do anything too interesting until we wire up shrinker.  Pretty
much lifted from i915.

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: use mutex_lock_interruptible for submit ioctl
Rob Clark [Tue, 17 May 2016 19:43:35 +0000 (15:43 -0400)]
drm/msm: use mutex_lock_interruptible for submit ioctl

Be kinder to things that do lots of signal handling (ie. Xorg)

Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/mdp: Provide details on MDP interface ports
Archit Taneja [Tue, 14 Jun 2016 12:17:26 +0000 (17:47 +0530)]
dt-bindings: msm/mdp: Provide details on MDP interface ports

The MDP4/5 DT node now contains a list of ports that describe how it
connects to external encoder interfaces like DSI and HDMI. These follow
the standard of_graph bindings, and allow us to get rid of the 'connectors'
phandle that contained a list of all the external encoders connected to
MDP.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/mdp5: Add MDP5 display bindings
Archit Taneja [Tue, 14 Jun 2016 10:43:51 +0000 (16:13 +0530)]
dt-bindings: msm/mdp5: Add MDP5 display bindings

Add a new doc for DT bindings for platforms that contain MDP5 display
controller hardware. The doc describes bindings for the top level
MDSS wrapper hardware and MDP5 itself.

Add an example for the bindings as found in MSM8916.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/mdp4: Create a separate binding doc for MDP4
Archit Taneja [Mon, 13 Jun 2016 14:07:37 +0000 (19:37 +0530)]
dt-bindings: msm/mdp4: Create a separate binding doc for MDP4

MDP4 and MDP5 vary a bit in terms of device hierarchy and the properties
they require. Rename the binding doc to mdp4.txt and remove MDP5 specific
pieces. A separate document will be created for MDP5

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/dsi: Don't get DSI index from DT
Archit Taneja [Thu, 23 Jun 2016 09:56:04 +0000 (15:26 +0530)]
drm/msm/dsi: Don't get DSI index from DT

The DSI host and PHY driver currently expects the DT bindings to provide
custom properties "qcom,dsi-host-index" and "qcom,dsi-phy-index" so that
the driver can identify which DSI instance it is.

The binding isn't acceptable, but the driver still needs to figure out
what its instance id. This is now done by storing the mmio starting
addresses for each DSI instance in every SoC version in the driver. The
driver then identifies the index number by trying to match the stored
address with comparing the resource start address we get from DT.

We don't have compatible strings for DSI PHY on each SoC, but only the
DSI PHY type. We only support one SoC version for each PHY type, so we
get away doing the same thing above for the PHY driver. We can revisit
this when we support two SoCs with the same DSI PHY.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Update compatible strings for MDSS/MDP5
Archit Taneja [Mon, 30 May 2016 11:32:00 +0000 (17:02 +0530)]
drm/msm/mdp5: Update compatible strings for MDSS/MDP5

Introduce new compatible strings for the top level MDSS wrapper device,
and the MDP5 device.

Previously, the "qcom,mdp5" and "qcom,mdss_mdp" compatible strings
were used to match the top level platform_device (which was also tied
to the top level drm_device struct). Now, these strings are used
to match the MDP5 platform device.

Use "qcom,mdss" as the compatible string for top level MDSS device.
This is now used to match the top level platform_device (which is
tied to the drm_device struct).

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Drop the gpu binding
Archit Taneja [Thu, 19 May 2016 08:03:52 +0000 (13:33 +0530)]
drm/msm: Drop the gpu binding

The driver currently identifies the GPU components it needs by parsing
a phandle list from the 'gpus' DT property.

This isn't the right binding to go with. So, for now, just search all
device nodes and find the gpu node we need by parsing a list of
compatible strings.

Once we know how to link the kms and gpu drivers, we'll drop this method
and use the correct binding.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Add components for MDP5
Archit Taneja [Mon, 6 Jun 2016 08:15:34 +0000 (13:45 +0530)]
drm/msm: Add components for MDP5

For MDP5 based platforms, the master device isn't the MDP5 platform
device, but the top level MDSS device, which is a parent to MDP5 and
interface (DSI, HDMI, eDP etc) devices.

In order to add components on MDP5 platforms, we first need to populate
the MDSS children, locate the MDP5 child, and then parse its ports to
get the display interfaces.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Add display components by parsing MDP ports
Archit Taneja [Thu, 19 May 2016 05:08:39 +0000 (10:38 +0530)]
drm/msm: Add display components by parsing MDP ports

The kms driver currently identifies all the mdss components it needs by
parsing a phandle list from the 'connectors' DT property.

Instead of this, describe a list of ports that the MDP hardware provides
to the external world. These ports are linked to external encoder
interfaces such as DSI, HDMI. These are also the subcomponent devices
that we need add. This description of ports complies with the generic
graph bindings.

The LVDS port is a special case since it is a part of MDP4 itself, and
its output connects directly to the LVDS panel. In this case, we don't
try to add it as a component.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Create separate funcs for adding display/gpu components
Archit Taneja [Thu, 19 May 2016 05:03:57 +0000 (10:33 +0530)]
drm/msm: Create separate funcs for adding display/gpu components

Simplifies some of the code that we'll add later.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Add missing mdp5_enable/disable calls
Archit Taneja [Wed, 15 Jun 2016 10:18:15 +0000 (15:48 +0530)]
drm/msm/mdp5: Add missing mdp5_enable/disable calls

Since runtime PM isn't implemented yet, we need to call
mdp5_enable/disable in a few more places. These would later be
replaced by runtime PM get/put calls.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Call pm_runtime_enable/disable for newly created devices
Archit Taneja [Wed, 15 Jun 2016 12:34:31 +0000 (18:04 +0530)]
drm/msm: Call pm_runtime_enable/disable for newly created devices

With the new device hierarchy for MDP5, we need to enable runtime PM
for both the toplevel MDSS device and the MDP5 device itself. Enable
runtime PM for the new devices.

Since MDP4 and MDP5 now have different places where runtime PM is
enabled, remove the previous pm_runtime_enable/disable calls, and
squash them in the respective kms drivers.

The new device hierarchy (as expressed in the DT bindings) has the GDSC
tied only to the MDSS wrapper device. This GDSC needs to be enabled for
accessing any register in the MDSS sub-blocks. Once every driver is
runtime adapted, the GDSC will be enabled when any sub-block device
calls runtime_get because of the parent-child relationship with MDSS.

Until then, we call pm_runtime_get_sync() once for the MDSS device to
ensure the GDSC is never disabled. This will be removed once all the
drivers are runtime PM adapted.

The error handling paths become a bit tricky when we call these runtime
PM funcs. There doesn't seem to be any helper that checks if runtime PM
is enabled already. Add bool variables in mdp4_kms/mdp5_kms structs to
check if the driver had managed to call pm_runtime_enable before bailing
out.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Update the register offsets of MDP5 sub-blocks
Archit Taneja [Mon, 9 May 2016 06:44:49 +0000 (12:14 +0530)]
drm/msm/mdp5: Update the register offsets of MDP5 sub-blocks

The MDP5 sub-block register offsets are relative to the top level
MDSS register address.

Now that we have the start of MDP5 register address space, provide
the offsets relative to that. This involves subtracting the offsets
with 0x1000 or 0x100 depending on the MDP5 version.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Use updated MDP5 register names
Archit Taneja [Tue, 10 May 2016 05:35:58 +0000 (11:05 +0530)]
drm/msm/mdp5: Use updated MDP5 register names

Since MDSS registers were stuffed within the the MDP5 register
space, we had an __offset_MDP() macro to identify the offset
between the start of MDSS and MDP5 address spaces. This offset
macro expected a MDP index argument, which didn't make much
sense since we don't have multiple MDPs.

The offset is no longer needed now that we have devices for the 2
different register address spaces. Also, remove the "REG_MDP5_MDP_"
prefix to "REG_MDP5_".

Update the generated headers in mdp5.xml.h

We generally update headers as a separate patch, but we need to
do these together to prevent breaking build.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Remove old kms init/destroy funcs
Archit Taneja [Tue, 14 Jun 2016 12:54:54 +0000 (18:24 +0530)]
drm/msm/mdp5: Remove old kms init/destroy funcs

With the new kms_init/destroy funcs in place for MDP5, we can get rid of
the old kms funcs. Some members of the mdp5_kms struct also become
redundant, so we remove those too.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Use the new hierarchy and drop old irq management
Archit Taneja [Sun, 8 May 2016 16:06:28 +0000 (21:36 +0530)]
drm/msm/mdp5: Use the new hierarchy and drop old irq management

Call msm_mdss_init in msm_drv to set up top level registers/irq line.
Start using the new kms_init2/destroy2 funcs to inititalize MDP5 KMS.

With the MDSS interrupt and irqdomain set up, the old MDP5 irq code
can be dropped.

The mdp5_hw_init kms func now uses the platform device tied to MDP5
instead of the one tied to the drm_device/MDSS.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Prepare new kms_init funcs
Archit Taneja [Tue, 14 Jun 2016 12:53:52 +0000 (18:23 +0530)]
drm/msm/mdp5: Prepare new kms_init funcs

With MDP5 as a new device, we need to do less for MDP when initializing
modeset after all the components are bound.

Create mdp5_kms_init2/destroy2 funcs that inits modeset. These will
eventually replace the older kms_init/destroy funcs.

In the new kms_init2, the platform_device used is the one corresponding
to the new MDP5 platform_device. The new change here is that the irq is
now retrieved using irq_of_parse_and_map(), since MDP5 is a child interrupt
of the MDSS interrupt controller.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Create a separate MDP5 device
Archit Taneja [Mon, 30 May 2016 11:06:50 +0000 (16:36 +0530)]
drm/msm/mdp5: Create a separate MDP5 device

In order to have a tree-like device hierarchy between MDSS and its
sub-blocks (MDP5, DSI, HDMI, eDP etc), we need to create a separate
device/driver for MDP5. Currently, MDP5 and MDSS are squashed
together are are tied to the top level platform_device, which is
also the one used to create drm_device.

The mdp5_kms_init code is split into two parts. The part where device
resources are allocated are associated with the MDP5 driver's probe,
the rest is executed later when we initialize modeset.

With this change, unlike MDP4, the MDP5 platform_device isn't tied to
the top level drm_device anymore. The top level drm_device is now
associated with a platform device that corresponds to MDSS wrapper
hardware.

Create mdp5_init/destroy funcs that will be used by the MDP5 driver
probe/remove. Use the HW_VERSION register in the MDP5 register address
space. Both the MDSS and MDP VERSION registers give out identical
version info.

The older mdp5_kms_init code is left as is for now, this would be removed
later when we have all the pieces to support the new device hierarchy.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Add MDSS top level driver
Archit Taneja [Sat, 7 May 2016 17:41:25 +0000 (23:11 +0530)]
drm/msm/mdp5: Add MDSS top level driver

SoCs that contain MDP5 have a top level wrapper called MDSS that manages
clocks, power and irq for the sub-blocks within it.

Currently, the MDSS portions are stuffed into the MDP5 driver. This makes
it hard to represent the DT bindings in the correct way. We create a top
level MDSS helper that handles these parts. This is essentially moving out
some of the mdp5_kms irq code and MDSS register space and keeping it as a
separate entity. We haven't given any clocks to the top level MDSS yet,
but a AHB clock would be added in the future to access registers.

One thing to note is that the resources allocated by this helper are
tied to the top level platform_device (the one that allocates the
drm_device struct too). This device would be the parent to MDSS
sub-blocks like MDP5, DSI, eDP etc.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Get irq number within kms driver itself
Archit Taneja [Wed, 18 May 2016 09:36:03 +0000 (15:06 +0530)]
drm/msm: Get irq number within kms driver itself

The driver gets the irq number using platform_get_irq on the main kms
platform device. This works fine since both MDP4 and MDP5 currently
have a flat device hierarchy. The platform device tied with the
drm_device points to the MDP DT node in both cases.

This won't work when MDP5 supports a tree-like hierarchy. In this
case, the platform device tied to the top level drm_device is the
MDSS DT node, and the irq we need for KMS is the one generated by
MDP5, not MDSS.

Get the irq number from the MDP4/5 kms driver itself. Each driver
can later provide the irq number based on what device hierarchy it
uses.

While we're at it, call drm_irq_install only when we have a valid KMS
driver.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Remove unused fields
Archit Taneja [Fri, 6 May 2016 12:47:15 +0000 (18:17 +0530)]
drm/msm: Remove unused fields

These aren't used. Probably left overs when driver was refactored to
support both MDP4 and MDP5.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Drop the id_table in platform_driver
Archit Taneja [Thu, 5 May 2016 15:56:34 +0000 (21:26 +0530)]
drm/msm: Drop the id_table in platform_driver

This isn't needed as we only support OF.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/dsi: Some binding doc cleanups
Archit Taneja [Thu, 9 Jun 2016 11:37:25 +0000 (17:07 +0530)]
dt-bindings: msm/dsi: Some binding doc cleanups

Some cleanups:

- Use simpler names for DT nodes in the example
- Use references instead of dumping Document links everywhere

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/dsi: Add assigned clocks bindings
Archit Taneja [Thu, 9 Jun 2016 10:21:44 +0000 (15:51 +0530)]
dt-bindings: msm/dsi: Add assigned clocks bindings

The PLL in the DSI PHY block generates 2 clock outputs (Byte and Pixel
clocks) that are fed into the Multimedia Clock Controller (MMCC). The MMCC
uses these as source clocks for some of its RCGs to generate clocks that
finally feed to the DSI host controller.

Use the assigned clocks DT bindings to set up the MMCC RCGs that feed to
the DSI host. Use the DSI PHY provided clocks to set up the parents
of these assigned clocks.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/dsi: Modify port and PHY bindings
Archit Taneja [Thu, 9 Jun 2016 10:16:24 +0000 (15:46 +0530)]
dt-bindings: msm/dsi: Modify port and PHY bindings

The DSI node now has two ports that describe the connection between the
MDP interface output and the DSI input, and the connection between the DSI
output and the connected panel/bridge. Update the properties and the
example.

Also, use generic PHY bindings instead of the custom one.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/dsi: Use standard data lanes binding
Archit Taneja [Thu, 9 Jun 2016 06:21:33 +0000 (11:51 +0530)]
dt-bindings: msm/dsi: Use standard data lanes binding

The "qcom,data-lane-map" binding mentioned in the document is changed to
the more generic "data-lanes" property specified in:

Documentation/devicetree/bindings/media/video-interfaces.txt

The previous binding expressed physical to logical data lane mappings,
the standard "data-lanes" binding uses logical to physical data lane
mappings. Update the docs to reflect this change. The example had the
property incorrectly named as "lanes", update this too.

The MSM DSI DT bindings aren't used anywhere at the moment, so
it's okay to update this property.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/dsi: Use a standard DT binding for data lanes
Archit Taneja [Wed, 8 Jun 2016 10:44:19 +0000 (16:14 +0530)]
drm/msm/dsi: Use a standard DT binding for data lanes

A more standard DT binding describing data lanes already exists here:
Documentation/devicetree/bindings/media/video-interfaces.txt

Use this binding instead of "qcom,data-lane-map". One difference
in the standard binding w.r.t to the existing binding is that it
provides a logical to physical mapping instead of the other way
round. Tweak the code to translate the data the way we want it.

The MSM DSI DT bindings aren't used anywhere at the moment, so
it's okay to update this property.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/dsi: Use generic PHY bindings
Archit Taneja [Fri, 29 Apr 2016 12:34:47 +0000 (18:04 +0530)]
drm/msm/dsi: Use generic PHY bindings

The DSI host links to the DSI PHY device using a custom binding. Switch to
the generic PHY bindings. The DSI PHY driver itself doesn't use the common
PHY framework for now.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/dsi: Modify port parsing
Archit Taneja [Wed, 27 Apr 2016 10:06:53 +0000 (15:36 +0530)]
drm/msm/dsi: Modify port parsing

The DSI interface is going to have two ports defined in its device node.
The first port is always going to be the link between the MDP output
and the input to DSI, the second port is going to be the link between
the DSI output and the connected panel/bridge:

 -----           -----           -------
| MDP | ------> | DSI | ------> | Panel |
 -----           -----           -------
        (Port 0)       (Port 1)

Until now, there was only one Port representing the output. Update the
DSI host driver such that it parses Port #1 for a connected device.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodt-bindings: msm/mdp: Fix up clock related bindings
Archit Taneja [Fri, 10 Jun 2016 06:31:58 +0000 (12:01 +0530)]
dt-bindings: msm/mdp: Fix up clock related bindings

Address some issues wiht clock related bindings. It's okay to change these
since these bindings aren't used in any dtsi files until now.

MDP5:
- Don't ask for source clock

MDP4:
- Give a better name for MDP_TV_CLK
- Remove TV_SRC
- Add MDP_AXI_CLK

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp4: Clean up some MDP4 clocks
Archit Taneja [Fri, 10 Jun 2016 06:25:43 +0000 (11:55 +0530)]
drm/msm/mdp4: Clean up some MDP4 clocks

Fix some issues with MDP4 clocks:

- mdp4_dtv_encoder tries to get "src_clk", which is a RCG(TV_SRC) in
  MSM8960 and APQ8064. This isn't something the driver should access or
  configure. Instead of this, configure the "mdp_clk" (MDP_TV_CLK), a
  branch clock in MMCC that has the TV_SRC as its parent. Setting
  rate/enabling the "mdp_clk" will eventually configure "src_clk", which
  is what we want.
- Rename "mdp_clk" to "tv_clk" because that's slightly less confusing.
- Rename "mdp_axi_clk" to "bus_clk" because that's what we do elsewhere
  too.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm/mdp5: Don't get source of MDP core clock
Archit Taneja [Thu, 9 Jun 2016 11:41:08 +0000 (17:11 +0530)]
drm/msm/mdp5: Don't get source of MDP core clock

The driver expects DT to provide the parent to MDP core clock. The only
operation done to the parent clock is to set a rate. This can be
achieved by setting the rate on the core clock itsef. Don't try to
get the parent clock anymore.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Print the correct virtual addresses in map/unmap funcs
Archit Taneja [Mon, 18 Apr 2016 06:56:21 +0000 (12:26 +0530)]
drm/msm: Print the correct virtual addresses in map/unmap funcs

The msm_iommu_map/unmap funcs have debug prints to show the list of
VA:PA mappings. Use the correct variable to print the VAs.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agodrm/msm: Use correct type for physical addresses
Archit Taneja [Sun, 17 Apr 2016 14:58:43 +0000 (20:28 +0530)]
drm/msm: Use correct type for physical addresses

The u32 type used to pass the physical addresses to iommu_map can't
accommodate 64 bit addresses. Move to dma_addr_t to ensure wrong
addresses aren't provided to the IOMMU driver.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
8 years agoMerge tag 'drm-vc4-next-2016-07-15' of https://github.com/anholt/linux into drm-next
Dave Airlie [Sat, 16 Jul 2016 01:25:11 +0000 (11:25 +1000)]
Merge tag 'drm-vc4-next-2016-07-15' of https://github.com/anholt/linux into drm-next

This pull request brings in vc4 shader validation for branching,
allowing GLSL shaders with non-unrolled loops.

* tag 'drm-vc4-next-2016-07-15' of https://github.com/anholt/linux:
  drm/vc4: Fix a "the the" typo in a comment.
  drm/vc4: Fix definition of QPU_R_MS_REV_FLAGS
  drm/vc4: Add a getparam to signal support for branches.
  drm/vc4: Add support for branching in shader validation.
  drm/vc4: Add a bitmap of branch targets during shader validation.
  drm/vc4: Move validation's current/max ip into the validation struct.
  drm/vc4: Add a getparam ioctl for getting the V3D identity regs.

8 years agoMerge tag 'drm/panel/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux into...
Dave Airlie [Sat, 16 Jul 2016 01:24:29 +0000 (11:24 +1000)]
Merge tag 'drm/panel/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux into drm-next

drm/panel: Changes for v4.8-rc1

This set of changes contains a few cleanups for existing panels as well
as improved handling of certain backlights. In addition there's support
for a few new simple panels.

* tag 'drm/panel/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux:
  drm/panel: simple: Add support for Starry KR122EA0SRA panel
  dt-bindings: Add Starry KR122EA0SRA panel binding
  dt-bindings: Add vendor prefix for Starry
  dt-bindings: display: Add Sharp LQ101K1LY04 panel binding
  drm/panel: simple: Add support for Sharp LQ101K1LY04
  drm/panel: simple: Add support for LG LP079QX1-SP0V panel
  dt-bindings: Add support for LG LP079QX1-SP0V panel
  drm/panel: simple: Add support for Sharp LQ123P1JX31 panel
  dt-bindings: Add Sharp LQ123P1JX31 panel binding
  drm/panel: simple: Add support for Samsung LSN122DL01-C01 panel
  dt-bindings: Add Samsung LSN122DL01-C01 panel binding
  drm/panel: simple: Add support for LG LP097QX1-SPA1 panel
  dt-bindings: Add LG LP097QX1-SPA1 panel binding
  drm/panel: simple: Update backlight state property
  drm/panel: simple: Remove gratuitous blank line
  drm/panel: simple: Fix a couple of physical sizes

8 years agoMerge tag 'drm/tegra/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux into...
Dave Airlie [Sat, 16 Jul 2016 01:23:50 +0000 (11:23 +1000)]
Merge tag 'drm/tegra/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux into drm-next

drm/tegra: Changes for v4.8-rc1

This set of changes contains a bunch of cleanups to the host1x driver as
well as the addition of a pin controller for DPAUX, which is required by
boards to configure the DPAUX pads in AUX mode (for DisplayPort) or I2C
mode (for HDMI and DDC).

Included is also a bit of rework of the SOR driver in preparation to add
DisplayPort support as well as some refactoring and cleanup.

Finally, all output drivers are converted to runtime PM, which greatly
simplifies the handling of clocks and resets.

* tag 'drm/tegra/for-4.8-rc1' of git://anongit.freedesktop.org/tegra/linux: (35 commits)
  drm/tegra: sor: Reject HDMI 2.0 modes
  drm/tegra: sor: Prepare for generic PM domain support
  drm/tegra: dsi: Prepare for generic PM domain support
  drm/tegra: sor: Make XBAR configurable per SoC
  drm/tegra: sor: Use sor1_src clock to set parent for HDMI
  dt-bindings: display: tegra: Add source clock for SOR
  drm/tegra: sor: Implement sor1_brick clock
  drm/tegra: sor: Implement runtime PM
  drm/tegra: hdmi: Implement runtime PM
  drm/tegra: dsi: Implement runtime PM
  drm/tegra: dc: Implement runtime PM
  drm/tegra: hdmi: Enable audio over HDMI
  drm/tegra: sor: Do not support deep color modes
  drm/tegra: sor: Extract tegra_sor_mode_set()
  drm/tegra: sor: Split out tegra_sor_apply_config()
  drm/tegra: sor: Rename tegra_sor_calc_config()
  drm/tegra: sor: Factor out tegra_sor_set_parent_clock()
  drm/tegra: dpaux: Add pinctrl support
  dt-bindings: Add bindings for Tegra DPAUX pinctrl driver
  drm/tegra: Prepare DPAUX for supporting generic PM domains
  ...

8 years agoMerge branch 'upstream/analogix-dp-20160705' of git://github.com/yakir-Yang/linux...
Dave Airlie [Sat, 16 Jul 2016 01:21:50 +0000 (11:21 +1000)]
Merge branch 'upstream/analogix-dp-20160705' of git://github.com/yakir-Yang/linux into drm-next

Please consider merging this tag, which contains the v4 misc fixes and add RK3399 eDP support patches[0] I sent on 2016-06-29, rebased onto v4.7-rc5.

* 'upstream/analogix-dp-20160705' of git://github.com/yakir-Yang/linux:
  dt-bindings: analogix_dp: rockchip: correct the wrong compatible name
  drm/rockchip: analogix_dp: introduce the pclk for grf
  drm/bridge: analogix_dp: fix no drm hpd event when panel plug in
  drm/rockchip: analogix_dp: update the comments about why need to hardcode VOP output mode
  drm/rockchip: analogix_dp: correct the connector display color format and bpc
  drm/bridge: analogix_dp: passing the connector as an argument in .get_modes()
  drm/rockchip: analogix_dp: make panel detect to an optional action
  drm/rockchip: analogix_dp: add rk3399 eDP support
  drm/bridge: analogix_dp: some rockchip chips need to flip REF_CLK bit setting
  drm/bridge: analogix_dp: correct the register bit define error in ANALOGIX_DP_PLL_REG_1
  drm/rockchip: analogix_dp: split the lcdc select setting into device data

8 years agoMerge tag 'imx-drm-next-2016-07-14' of git://git.pengutronix.de/git/pza/linux into...
Dave Airlie [Sat, 16 Jul 2016 01:19:29 +0000 (11:19 +1000)]
Merge tag 'imx-drm-next-2016-07-14' of git://git.pengutronix.de/git/pza/linux into drm-next

imx-drm updates

- atomic mode setting conversion
- replace DMFC FIFO allocation mechanism with a fixed allocation
  that is good enough for all cases
- support for external bridges connected to parallel-display
- improved error handling in imx-ldb, imx-tve, and parallel-display
- some code cleanup in imx-tve

* tag 'imx-drm-next-2016-07-14' of git://git.pengutronix.de/git/pza/linux:
  drm/imx: parallel-display: add bridge support
  drm/imx: parallel-display: check return code from of_get_drm_display_mode()
  gpu: ipu-v3: ipu-dc: don't bug out on invalid bus_format
  drm/imx: imx-tve: fix the error message
  drm/imx: imx-tve: remove unneeded 'or' operation
  drm/imx: imx-tve: check the value returned by regulator_set_voltage()
  drm/imx: imx-ldb: check return code on panel attach
  drm/imx: turn remaining container_of macros into inline functions
  drm/imx: store internal bus configuration in crtc state
  drm/imx: remove empty mode_set encoder callbacks
  drm/imx: atomic phase 3 step 3: Advertise DRIVER_ATOMIC
  drm/imx: atomic phase 3 step 2: Legacy callback fixups
  drm/bridge: dw-hdmi: Remove the legacy drm_connector_funcs structure
  drm/imx: atomic phase 3 step 1: Use atomic configuration
  drm/imx: Remove encoders' ->prepare callbacks
  drm/imx: atomic phase 2 step 2: Track plane_state->fb correctly in ->page_flip
  drm/imx: atomic phase 2 step 1: Wire up state ->reset, ->duplicate and ->destroy
  drm/imx: atomic phase 1: Use transitional atomic CRTC and plane helpers
  gpu: ipu-v3: ipu-dmfc: Use static DMFC FIFO allocation mechanism
  drm/imx: ipuv3 plane: Check different types of plane separately

8 years agodrm/vc4: Fix a "the the" typo in a comment.
Eric Anholt [Sat, 2 Jul 2016 16:58:41 +0000 (09:58 -0700)]
drm/vc4: Fix a "the the" typo in a comment.

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agodrm/vc4: Fix definition of QPU_R_MS_REV_FLAGS
Eric Anholt [Mon, 4 Jul 2016 19:59:18 +0000 (12:59 -0700)]
drm/vc4: Fix definition of QPU_R_MS_REV_FLAGS

We don't use it in shader validation currently, so it had no effect,
but best to fix it anyway in case we do some day.

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agodrm/vc4: Add a getparam to signal support for branches.
Eric Anholt [Sat, 2 Jul 2016 21:14:27 +0000 (14:14 -0700)]
drm/vc4: Add a getparam to signal support for branches.

Userspace needs to know if it can create shaders that do branching.
Otherwise, for backwards compatibility with old kernels it needs to
lower if statements to conditional assignments.

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agodrm/vc4: Add support for branching in shader validation.
Eric Anholt [Sat, 2 Jul 2016 19:17:10 +0000 (12:17 -0700)]
drm/vc4: Add support for branching in shader validation.

We're already checking that branch instructions are between the start
of the shader and the proper PROG_END sequence.  The other thing we
need to make branching safe is to verify that the shader doesn't read
past the end of the uniforms stream.

To do that, we require that at any basic block reading uniforms have
the following instructions:

load_imm temp, <next offset within uniform stream>
add unif_addr, temp, unif

The instructions are generated by userspace, and the kernel verifies
that the load_imm is of the expected offset, and that the add adds it
to a uniform.  We track which uniform in the stream that is, and at
draw call time fix up the uniform stream to have the address of the
start of the shader's uniforms at that location.

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agodrm/vc4: Add a bitmap of branch targets during shader validation.
Eric Anholt [Sat, 2 Jul 2016 17:10:24 +0000 (10:10 -0700)]
drm/vc4: Add a bitmap of branch targets during shader validation.

This isn't used yet, it's just a first step toward loop validation.
During the main parsing of instructions, we need to know when we hit a
new basic block so that we can reset validated state.

v2: Fix a stray semicolon after an if block.  (caught by kbuild test).

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agoMerge branch 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daein...
Dave Airlie [Fri, 15 Jul 2016 04:05:41 +0000 (14:05 +1000)]
Merge branch 'exynos-drm-next' of git://git./linux/kernel/git/daeinki/drm-exynos into drm-next

   This pull request adds to the rework patch series for IOMMU
   integration to support ARM64bit architecture with DMA-IOMMU
   glue code.

   With this patch series, Exynos DRM works well on Exynos5433 SoC
   with IOMMU enabled.

* 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos:
  drm/exynos: iommu: add support for ARM64 specific code for IOMMU glue
  drm/exynos: iommu: move ARM specific code to exynos_drm_iommu.h
  drm/exynos: iommu: remove unused entries from exynos_drm_private strcuture
  drm/exynos: iommu: add a check if all sub-devices have iommu controller
  drm/exynos: iommu: move dma_params configuration code to separate functions

8 years agoMerge tag 'drm-vc4-next-2016-07-12' of https://github.com/anholt/linux into drm-next
Dave Airlie [Fri, 15 Jul 2016 03:56:11 +0000 (13:56 +1000)]
Merge tag 'drm-vc4-next-2016-07-12' of https://github.com/anholt/linux into drm-next

This pull request brings in new vc4 plane formats for Android, precise
vblank timestamping, and a couple of small cleanups.

* tag 'drm-vc4-next-2016-07-12' of https://github.com/anholt/linux:
  drm/vc4: remove redundant ret status check
  drm/vc4: Implement precise vblank timestamping.
  drm/vc4: Bind the HVS before we bind the individual CRTCs.
  gpu: drm: vc4_hdmi: add missing of_node_put after calling of_parse_phandle
  drm: vc4: enable XBGR8888 and ABGR8888 pixel formats
  drm/vc4: clean up error exit path on failed dpi_connector allocation

8 years agoMerge tag 'drm-intel-next-2016-07-11' of git://anongit.freedesktop.org/drm-intel...
Dave Airlie [Fri, 15 Jul 2016 03:50:58 +0000 (13:50 +1000)]
Merge tag 'drm-intel-next-2016-07-11' of git://anongit.freedesktop.org/drm-intel into drm-next

- select igt testing depencies for CONFIG_DRM_I915_DEBUG (Chris)
- track outputs in crtc state and clean up all our ad-hoc connector/encoder
  walking in modest code (Ville)
- demidlayer drm_device/drm_i915_private (Chris Wilson)
- thundering herd fix from Chris Wilson, with lots of help from Tvrtko Ursulin
- piles of assorted clean and fallout from the thundering herd fix
- documentation and more tuning for waitboosting (Chris)
- pooled EU support on bxt (Arun Siluvery)
- bxt support is no longer considered prelimary!
- ring/engine vfunc cleanup from Tvrtko
- introduce intel_wait_for_register helper (Chris)
- opregion updates (Jani Nukla)
- tuning and fixes for wait_for macros (Tvrkto&Imre)
- more kabylake pci ids (Rodrigo)
- pps cleanup and fixes for bxt (Imre)
- move sink crc support over to atomic state (Maarten)
- fix up async fbdev init ordering (Chris)
- fbc fixes from Paulo and Chris

* tag 'drm-intel-next-2016-07-11' of git://anongit.freedesktop.org/drm-intel: (223 commits)
  drm/i915: Update DRIVER_DATE to 20160711
  drm/i915: Select DRM_VGEM for igt
  drm/i915: Select X86_MSR for igt
  drm/i915: Fill unused GGTT with scratch pages for VT-d
  drm/i915: Introduce Kabypoint PCH for Kabylake H/DT.
  drm/i915:gen9: implement WaMediaPoolStateCmdInWABB
  drm/i915: Check for invalid cloning earlier during modeset
  drm/i915: Simplify hdmi_12bpc_possible()
  drm/i915: Kill has_dsi_encoder
  drm/i915: s/INTEL_OUTPUT_DISPLAYPORT/INTEL_OUTPUT_DP/
  drm/i915: Replace some open coded intel_crtc_has_dp_encoder()s
  drm/i915: Kill has_dp_encoder from pipe_config
  drm/i915: Replace manual lvds and sdvo/hdmi counting with intel_crtc_has_type()
  drm/i915: Unify intel_pipe_has_type() and intel_pipe_will_have_type()
  drm/i915: Add output_types bitmask into the crtc state
  drm/i915: Remove encoder type checks from MST suspend/resume
  drm/i915: Don't mark eDP encoders as MST capable
  drm/i915: avoid wait_for_atomic() in non-atomic host2guc_action()
  drm/i915: Group the irq breadcrumb variables into the same cacheline
  drm/i915: Wake up the bottom-half if we steal their interrupt
  ...

8 years agoMerge tag 'topic/drm-misc-2016-07-14' of git://anongit.freedesktop.org/drm-intel...
Dave Airlie [Fri, 15 Jul 2016 01:01:37 +0000 (11:01 +1000)]
Merge tag 'topic/drm-misc-2016-07-14' of git://anongit.freedesktop.org/drm-intel into drm-next

I recovered dri-devel backlog from my vacation, more misc stuff:
- of_put_node fixes from Peter Chen (not all yet)
- more patches from Gustavo to use kms-native drm_crtc_vblank_* funcs
- docs sphinxification from Lukas Wunner
- bunch of fixes all over from Dan Carpenter
- more follow up work from Chris register/unregister rework in various
  places
- vgem dma-buf export (for writing testcases)
- small things all over from tons of different people

* tag 'topic/drm-misc-2016-07-14' of git://anongit.freedesktop.org/drm-intel: (52 commits)
  drm: Don't overwrite user ioctl arg unless requested
  dma-buf/sync_file: improve Kconfig description for Sync Files
  MAINTAINERS: add entry for the Sync File Framework
  drm: Resurrect atomic rmfb code
  drm/vgem: Use PAGE_KERNEL in place of x86-specific PAGE_KERNEL_IO
  qxl: silence uninitialized variable warning
  qxl: check for kmap failures
  vga_switcheroo: Sphinxify docs
  drm: Restore double clflush on the last partial cacheline
  gpu: drm: rockchip_drm_drv: add missing of_node_put after calling of_parse_phandle
  gpu: drm: sti_vtg: add missing of_node_put after calling of_parse_phandle
  gpu: drm: sti_hqvdp: add missing of_node_put after calling of_parse_phandle
  gpu: drm: sti_vdo: add missing of_node_put after calling of_parse_phandle
  gpu: drm: sti_compositor: add missing of_node_put after calling of_parse_phandle
  drm/tilcdc: use drm_crtc_handle_vblank()
  drm/rcar-du: use drm_crtc_handle_vblank()
  drm/nouveau: use drm_crtc_handle_vblank()
  drm/atmel: use drm_crtc_handle_vblank()
  drm/armada: use drm_crtc_handle_vblank()
  drm: make drm_vblank_count_and_time() static
  ...

8 years agodrm/vc4: Move validation's current/max ip into the validation struct.
Eric Anholt [Sat, 2 Jul 2016 16:57:07 +0000 (09:57 -0700)]
drm/vc4: Move validation's current/max ip into the validation struct.

Reduces the argument count for some of the functions, and will be used
more with the upcoming looping support.

Signed-off-by: Eric Anholt <eric@anholt.net>
8 years agodrm/vc4: Add a getparam ioctl for getting the V3D identity regs.
Eric Anholt [Fri, 1 Jul 2016 20:10:38 +0000 (13:10 -0700)]
drm/vc4: Add a getparam ioctl for getting the V3D identity regs.

As I extend the driver to support different V3D revisions, userspace
needs to know what version it's targeting.  This is most easily
detected using the V3D identity registers.

v2: Make sure V3D is runtime PM on when reading the registers.
v3: Switch to a 64-bit param value (suggested by Rob Clark in review)

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> (v2)
Reviewed-by: Rob Clark <robdclark@gmail.com> (v3, over irc)
8 years agodrm/tegra: sor: Reject HDMI 2.0 modes
Thierry Reding [Tue, 12 Jul 2016 14:52:22 +0000 (16:52 +0200)]
drm/tegra: sor: Reject HDMI 2.0 modes

Enabling HDMI 2.0 modes requires extra programming and will not work
with the current driver, so reject all those modes.

Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/tegra: sor: Prepare for generic PM domain support
Jon Hunter [Fri, 1 Jul 2016 13:21:38 +0000 (14:21 +0100)]
drm/tegra: sor: Prepare for generic PM domain support

The SOR driver for Tegra requires the SOR power partition to be enabled.
Now that Tegra supports the generic PM domain framework we manage the
SOR power partition via this framework. However, the sequence for
gating/ungating the SOR power partition requires that the SOR reset is
asserted/de-asserted at the time the SOR power partition is
gated/ungated, respectively. Now that the reset control core assumes
that resets are exclusive, the Tegra generic PM domain code and the SOR
driver cannot request the same reset unless we mark the reset as shared.
Sharing resets will not work in this case because we cannot guarantee
that the reset will be asserted/de-asserted at the appropriate time.
Therefore, given that the Tegra generic PM domain code will handle the
resets, do not request the reset in the SOR driver if the SOR device has
a PM domain associated.

Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/tegra: dsi: Prepare for generic PM domain support
Jon Hunter [Fri, 1 Jul 2016 13:21:37 +0000 (14:21 +0100)]
drm/tegra: dsi: Prepare for generic PM domain support

The DSI driver for Tegra requires the SOR power partition to be enabled.
Now that Tegra supports the generic PM domain framework we manage the
SOR power partition via this framework. However, the sequence for
gating/ungating the SOR power partition requires that the DSI reset is
asserted/de-asserted at the time the SOR power partition is
gated/ungated, respectively. Now that the reset control core assumes
that resets are exclusive, the Tegra generic PM domain code and the DSI
driver cannot request the same reset unless we mark the reset as shared.
Sharing resets will not work in this case because we cannot guarantee
that the reset will be asserted/de-asserted at the appropriate time.
Therefore, given that the Tegra generic PM domain code will handle the
resets, do not request the reset in the DSI driver if the DSI device has
a PM domain associated.

Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/tegra: sor: Make XBAR configurable per SoC
Thierry Reding [Mon, 3 Aug 2015 13:50:32 +0000 (15:50 +0200)]
drm/tegra: sor: Make XBAR configurable per SoC

Provide a per-SoC mapping of lanes which can be used to configure the
XBAR.

Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/tegra: sor: Use sor1_src clock to set parent for HDMI
Thierry Reding [Thu, 9 Jun 2016 15:53:57 +0000 (17:53 +0200)]
drm/tegra: sor: Use sor1_src clock to set parent for HDMI

When running in HDMI mode, the sor1 IP block needs to use the sor1_src
as parent clock, and in turn configure the sor1_src to use pll_d2_out0
as its parent.

Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodt-bindings: display: tegra: Add source clock for SOR
Thierry Reding [Thu, 23 Jun 2016 10:04:24 +0000 (12:04 +0200)]
dt-bindings: display: tegra: Add source clock for SOR

The SOR clock can have various sources, with the most commonly used
being the sor_safe, pll_d2_out0, pll_dp and sor_brick clocks. These
are configured using a three level mux, of which the first 2 levels
can be treated as one. The direct parents of the SOR clock are the
sor_safe, sor_brick and sor_src clocks, whereas the pll_d2_out0 and
pll_dp clocks can be selected as parents of the sor_src clock via a
second mux.

Previous generations of Tegra have only supported eDP and LVDS with
the SOR, where LVDS was never used on publicly available hardware.
Clocking for this only ever required the first level mux (to select
between sor_safe and sor_brick).

Tegra210 has a new revision of the SOR that supports HDMI and hence
needs to support the second level mux to allow selecting pll_d2_out0
as the SOR clock's parent. This second mux is knows as sor_src, and
operating system software needs a reference to it in order to select
the proper parent.

Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/tegra: sor: Implement sor1_brick clock
Thierry Reding [Thu, 1 Oct 2015 12:25:03 +0000 (14:25 +0200)]
drm/tegra: sor: Implement sor1_brick clock

sor1_brick is a clock that can be used as a source for the sor1 clock.
The registers to control the clock output are part of the sor1 IP block
and hence the sor driver is the best place to implement it.

Signed-off-by: Thierry Reding <treding@nvidia.com>
8 years agodrm/imx: parallel-display: add bridge support
Philipp Zabel [Mon, 27 Jun 2016 13:26:31 +0000 (15:26 +0200)]
drm/imx: parallel-display: add bridge support

Add support for bridge chips connected externally to the i.MX
DISP0/DISP1 DPI interfaces.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
8 years agodrm: Don't overwrite user ioctl arg unless requested
Chris Wilson [Tue, 12 Jul 2016 14:59:50 +0000 (15:59 +0100)]
drm: Don't overwrite user ioctl arg unless requested

Currently, we completely ignore the user when it comes to the in/out
direction of the ioctl argument, as we simply cannot trust userspace.
(For example, they might request a copy of the modified ioctl argument
when the driver is not expecting such and so leak kernel stack.)
However, blindly copying over the target address may also lead to a
spurious EFAULT, and a failure after the ioctl was completed
successfully. This is important in order to avoid an ABI break when
extending an ioctl from IOR to IORW. Similar to how we only copy the
intersection of the kernel arg size and the user arg size, we only want
to copy back the kernel arg data iff both the kernel and userspace
request the copy.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1468335590-21023-1-git-send-email-chris@chris-wilson.co.uk
8 years agoMerge branch 'linux-4.8' of git://github.com/skeggsb/linux into drm-next
Dave Airlie [Thu, 14 Jul 2016 02:14:49 +0000 (12:14 +1000)]
Merge branch 'linux-4.8' of git://github.com/skeggsb/linux into drm-next

Here's an initial drm-next pull for nouveau 4.8, highlights:
- GK20A/GM20B volt and clock improvements.
- Initial support for GP100/GP104 GPUs, GP104 will not yet support
acceleration due to NVIDIA having not released firmware for them as of yet.

* 'linux-4.8' of git://github.com/skeggsb/linux: (97 commits)
  drm/nouveau/bus: remove cpu_coherent flag
  drm/nouveau/ttm: remove special handling of coherent objects
  drm/nouveau: check for supported chipset before booting fbdev off the hw
  drm/nouveau/ce/gp104: initial support
  drm/nouveau/fifo/gp104: initial support
  drm/nouveau/disp/gp104: initial support
  drm/nouveau/dma/gp104: initial support
  drm/nouveau/ltc/gp104: initial support
  drm/nouveau/ibus/gp104: initial support
  drm/nouveau/i2c/gp104: initial support
  drm/nouveau/gpio/gp104: initial support
  drm/nouveau/fuse/gp104: initial support
  drm/nouveau/bus/gp104: initial support
  drm/nouveau/bar/gp104: initial support
  drm/nouveau/mmu/gp104: initial support
  drm/nouveau/fb/gp104: initial support
  drm/nouveau/imem/gp104: initial support
  drm/nouveau/devinit/gp104: initial support
  drm/nouveau/bios/gp104: initial support
  drm/nouveau/tmr/gp104: initial support
  ...

8 years agodrm/nouveau/bus: remove cpu_coherent flag
Alexandre Courbot [Wed, 13 Jul 2016 06:29:36 +0000 (15:29 +0900)]
drm/nouveau/bus: remove cpu_coherent flag

This flag's only remaining function is to ignore the uncached flag for
BOs on coherent architectures.

However the reason for allocating an object uncache on a non-coherent
architecture (namely because the cost of doing explicit flushes/
invalidations is higher than the benefit of caching the data because
accesses are few and far between) should also apply on architectures for
which coherency is maintained implicitly. Thus allocate coherent objects
as uncached on all architectures.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ttm: remove special handling of coherent objects
Alexandre Courbot [Wed, 13 Jul 2016 06:29:35 +0000 (15:29 +0900)]
drm/nouveau/ttm: remove special handling of coherent objects

TTM-allocated coherent objects were populated using the DMA API and
accessed using the mapping it returned to workaround coherency
issues. These issues seem to have been solved, thus remove this extra
case to handle and use the regular kernel mapping functions.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau: check for supported chipset before booting fbdev off the hw
Ben Skeggs [Tue, 12 Jul 2016 01:57:07 +0000 (11:57 +1000)]
drm/nouveau: check for supported chipset before booting fbdev off the hw

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Cc: stable@vger.kernel.org
8 years agodrm/nouveau/ce/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ce/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/fifo/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/fifo/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/disp/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/disp/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/dma/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/dma/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ltc/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ltc/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ibus/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ibus/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/i2c/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/i2c/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/gpio/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/gpio/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/fuse/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/fuse/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/bus/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/bus/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/bar/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/bar/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/mmu/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/mmu/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/fb/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/fb/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/imem/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/imem/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/devinit/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/devinit/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/bios/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/bios/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/tmr/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/tmr/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/pci/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/pci/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/mc/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/mc/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/top/gp104: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/top/gp104: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/core: recognise GP104 chipset
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/core: recognise GP104 chipset

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/sw/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/sw/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/gr/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/gr/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ce/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ce/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/fifo/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/fifo/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/disp/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/disp/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/dma/gp100: initial implementation
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/dma/gp100: initial implementation

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/secboot/gm200: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/secboot/gm200: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ltc/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ltc/gp100: initial support

Due to the GPU preventing us from touching NV_PLTCG_LTCS_LTSS_CBC_BASE,
we cannot provide CBC/ZBC support without signed PMU firmware to handle
the task for us...

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/ibus/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/ibus/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/i2c/gm204: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/i2c/gm204: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
8 years agodrm/nouveau/gpio/gp100: initial support
Ben Skeggs [Sat, 9 Jul 2016 00:41:01 +0000 (10:41 +1000)]
drm/nouveau/gpio/gp100: initial support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>