Now that we can use atomic configurations, all the legacy callbacks
of CRTCs, encoders and connectors can be switched to the atomic version.
For the imx-ldb driver, there is a clock parent setting mismatch bewteen
->enable and ->disable after the switch, so a fixup is added.  For the
imx-tve driver, since the encoder's callback ->dpms is replaced by
->disable, we need to move the setting for the IPU_CLK_EN bit(in register
TVE_COM_CONF_REG) from ->enable/->disable to ->mode_set, otherwise, the
relevant CRTC cannot be disabled correctly with a warning on DC stop timeout.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Replacing drm_crtc_helper_set_config() by drm_atomic_helper_set_config()
and converting the suspend/resume operations to atomic make us be able
to use atomic configurations.  All of these allow us to remove the
crtc_funcs->mode_set callback as it is no longer used.  Also, change
the plane_funcs->update/disable_plane callbacks from the transitional
version to the atomic version.  Furthermore, switching to the pure atomic
version of set_config callback means that we may implement CRTC/plane
atomic checks by using the new CRTC/plane states instead of the legacy
ones and we may remove the private ipu_crtc->enabled state which was left
there for the transitional atomic helpers in phase 1.  Page flip is also
switched to the atomic version.  Last, the legacy function
drm_helper_disable_unused_functions() is removed from ->load in order
not to confuse the atomic driver.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The main task of imx encoders' ->prepare callbacks is to set bus_format,
bus_flags, di_vsync_pin and di_hsync_pin.  We may create a structure named
imx_encoder to cache them.  The atomic encoder callback ->disable may
replace ->prepare later, so let's remove ->prepare.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Use drm_atomic_set_fb_for_plane() in the legacy ->page_flip path to track
the pointer plane_state->fb correctly.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Wire up CRTCs', planes' and connectors' ->reset, ->duplicate and ->destroy state
hooks to use the default implementations from the atomic helper library.
The helpers track each DRM object state.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Use the drm_plane_helper_update/disable() and drm_helper_crtc_mode_set()
transitional atomic helpers.  The crtc->mode_set_nofb callback is added
so that the primary plane is no longer tied to the CRTC.  Check/update
logics are separated to make sure crtc->mode_set_nofb and plane->atomic_update
are always successful.  Also, some necessary logics are tweaked for a smooth
transition.
Signed-off-by: NLiu Ying <gnuiyl@gmail.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

This patch allows panels to set pixel clock and data enable pin polarity
other than the default of driving data at the falling pixel clock edge
and active high display enable.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The component master driver imx-drm-core matches component devices using
their of_node. Since commit 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc
module autoloading"), the imx-ipuv3-crtc dev->of_node is not set during
probing. Before that, of_node was set and caused an of: modalias to be
used instead of the platform: modalias, which broke module autoloading.

On the other hand, if dev->of_node is not set yet when the imx-ipuv3-crtc
probe function calls component_add, component matching in imx-drm-core
fails. While dev->of_node will be set once the next component tries to
bring up the component master, imx-drm-core component binding will never
succeed if one of the crtc devices is probed last.

Add of_node to the component platform data and match against the
pdata->of_node instead of dev->of_node in imx-drm-core to work around
this problem.

Cc: <stable@vger.kernel.org> # 4.4.x
Fixes: 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc module autoloading")
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Tested-by: NFabio Estevam <fabio.estevam@nxp.com>
Tested-by: NLothar Waßmann <LW@KARO-electronics.de>
Tested-by: NHeiko Schocher <hs@denx.de>
Tested-by: NChris Ruehl <chris.ruehl@gtsys.com.hk>

The vblank IRQ is only needed to trigger page flip work, so we
might as well disable it when there is no work to do.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

If the FB is backed by a GEM object with an dma-buf attached
we need to wait for any pending fences to signal before executing
the page flip.

The implementation is straight forward by deferring the flip to
a workqueue in that case.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The DRM core only references the currently queued/active framebuffer.
So there is a period of time where the flip is not completed, but
the GEM object backing the FB is already unreferenced and could be
destroyed if userspace closes its handle.

Make sure to keep a reference to the GEM object until the flip is
actually executed clean things up in a worker running behind the
flip execution.

Also move the page flip event into the context of this worker, so
it gets cleaned up automatically.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Start tracking the flip state explicitly, as opposed to inferring
it from the presence if a new FB. This is a preparatory step to
introduce an new immediate state, where we can wait for a fence to
signal.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The core already does the correct replacemet if the driver
page flip function returns without an error, so there is no
need to do it here.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Make sure the DRM core is aware that there will be no vblank interrupts
incoming if the CRTC is disabled. That way the core will reject any
attempts from userspace to wait on a vblank event on a disabled CRTC.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

So this one is special, since it tries to prevent races when userspace
crashes simply by disabling the vblank machinery. Well except that imx
always has vblanks enabled, and the disable_vblank hook actually just
tries to cancel a pending pageflip. Without any locking whatsoever. Of
course this is wrong, since it'll result in the hw not actually
displaying what drm thinks is the current frontbuffer.

Well since the core takes care of the disappearing DRM fd now. So we
can nuke all this confused code without ill side-effects.

Someone else needs to audit the locking for ->newfb and
->page_flip_event and fix it up. Common approach is to reuse
dev->event_lock for this.

Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: NDaniel Stone <daniels@collabora.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1453756616-28942-7-git-send-email-daniel.vetter@ffwll.ch

Moves a bunch of junk to .rodata from .data.

 drivers/gpu/drm/imx/imxdrm.ko:
-.rodata                      624
+.rodata                      652
-.data                        372
+.data                        344

 drivers/gpu/drm/imx/imx-ipuv3-crtc.ko:
-.rodata                      224
+.rodata                      280
-.data                        184
+.data                        128

 drivers/gpu/drm/imx/imx-ldb.ko:
-.rodata                      660
+.rodata                      784
-.data                        240
+.data                        116

 drivers/gpu/drm/imx/imx-tve.ko:
-.rodata                      400
+.rodata                      524
-.data                        416
+.data                        292

 drivers/gpu/drm/imx/parallel-display.ko:
-.rodata                      400
+.rodata                      524
-.data                        216
+.data                         92
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1450178476-26284-22-git-send-email-boris.brezillon@free-electrons.comSigned-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

ipu_crtc_handle_pageflip() was calling drm_send_vblank_event() with
a pipe argument of -1.  Commit cc1ef118 ("drm/irq: Make pipe
unsigned and name consistent") now makes this error obvious, as we
now may get a warning from:

	if (WARN_ON(pipe >= dev->num_crtcs))

in drm_vblank_count_and_time().  Prior to this change, we would end
up making out-of-bounds array accesses via:

	struct drm_vblank_crtc *vblank = &dev->vblank[crtc];
and
	*vblanktime = vblanktimestamp(dev, pipe, cur_vblank);

So, this has been broken for a very long time, and is not a result
of the above commit.  Since we don't care about the staging versions,
I've tagged this with the earliest mainline commit where we do care,
even though this commit did not introduce the bug.

Fixes: 6556f7f8 ("drm: imx: Move imx-drm driver out of staging")
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The crtc child device driver shouldn't modify the of_node of its platform
device in the probe function. Instead, since the previous patch, the IPU
core driver sets the of_node when the platform device is created.

Drop the now unused custom imx_drm_get_port_by_id function.
Suggested-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NRussell King <rmk+kernel@arm.linux.org.uk>

For primary plane initialization failure cases, ipu_plane_init() may return
a pointer encoded by ERR_PTR(). So, we should bailout instead of using that
pointer blindly.
Signed-off-by: NLiu Ying <Ying.Liu@freescale.com>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Use drm_universal_plane_init to create the planes, create the primary
plane first and use drm_crtc_init_with_planes to associate it with
the crtc.
This gets rid of the unused fallback primary plane previously created
by drm_crtc_init and fixes a NULL pointer dereference issue that can
be triggered by a modeset from userspace when fbdev helpers are
enabled [1].

[1] https://lkml.org/lkml/2015/11/4/107Reported-by: NLiu Ying <Ying.Liu@freescale.com>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NLiu Ying <Ying.Liu@freescale.com>

This patch consolidates the different interface_pix_fmt, pixel_fmt, pix_fmt,
and pixfmt variables to a common name "bus_format" wherever they describe the
pixel format on the bus between display controller and encoder hardware.
At the same time, it renames imx_drm_panel_format to imx_drm_set_bus_format.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Tested-by: NEmil Renner Berthing <kernel@esmil.dk>

This patch allows the IPU to divide the 27 MHz input clock from
the TVE by two to obtain the 13.5 MHz pixel clock needed for
NTSC/PAL SD modes.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

This patch allows interlaced frame buffer scanout for interlaced output
via HDMI or TV-Encoder.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

We do not need to track the state of the IPU DI's clock flags by having
each display bridge calling back into imx-drm-core, and then back out
into ipuv3-crtc.c.

ipuv3-crtc can instead just scan the list of encoders to retrieve their
type, and build up a picture of which types of encoders are attached.
We can then use this information to configure the IPU DI clocking mode
without any uncertainty - if we have multiple bridges connected to the
same DI, if one of them requires a synchronous DI clock, that's what we
must use.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

This patch changes struct ipu_di_signal_cfg to use struct videomode
to define video timings and flags.
Signed-off-by: NSteve Longerbeam <steve_longerbeam@mentor.com>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

Ask the IPU display interface, via ipu_di_adjust_videomode(), to
adjust a video mode to meet any DI restrictions. The function takes
a subsystem independent videomode, so the drm_display_mode must be
converted to videomode first, and then the adjusted mode converted
back to a drm_display_mode.
Signed-off-by: NSteve Longerbeam <steve_longerbeam@mentor.com>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>

The imx-drm driver was put into staging mostly for the following reasons,
all of which have been addressed or superseded:
 - convert the irq driver to use linear irq domains
 - work out the device tree bindings, this lead to the common of_graph
   bindings being used
 - factor out common helper functions, this mostly resulted in the
   component framework and drm of_graph helpers.

Before adding new fixes, and certainly before adding new features,
move it into its proper place below drivers/gpu/drm.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Linux already includes a copy of the GPL, checkpatch compains about the address.
Remove it from the license text.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This is a patch to the ipuv3-crtc.c file that fixes up two "lines
over 80 characters" warnings found by the checkpatch.pl tool,
keeping the alignment of the x/y/width/height parameters for
readability.
Signed-off-by: NYannis Damigos <giannis.damigos@gmail.com>
Acked-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The i.MX Image Processing Unit (IPU) contains a number of image processing
blocks that sit right in the middle between DRM and V4L2. Some of the modules,
such as Display Controller, Processor, and Interface (DC, DP, DI) or CMOS
Sensor Interface (CSI) and their FIFOs could be assigned to either framework,
but others, such as the dma controller (IDMAC) and image converter (IC) can
be used by both.
The IPUv3 core driver provides an internal API to access the modules, to be
used by both DRM and V4L2 IPUv3 drivers.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Now that ipu_dc_disable_channel correctly waits for the channel to finish,
we can reorder the enable/disable order to first stop the DC and DI and
only then disable the IDMAC. Enabling is done the other way around: IDMAC
first, then DC, then DI.

This avoids an issue where sometimes the channel would not correctly start,
leading to non-working LVDS displays.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

According to the datasheet, setting the di0_polarity_disp_clk
field in the GENERAL di register sets the output clock polarity
to active high.
Signed-off-by: NDenis Carikli <denis@eukrea.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Now that CRTC's have a primary plane, there's no need to track the
framebuffer in the CRTC.  Replace all references to the CRTC fb with the
primary plane's fb.

This patch was generated by the Coccinelle semantic patching tool using
the following rules:

        @@ struct drm_crtc C; @@
        -   (C).fb
        +   C.primary->fb

        @@ struct drm_crtc *C; @@
        -   (C)->fb
        +   C->primary->fb

v3: Generate patch via coccinelle.  Actual removal of crtc->fb has been
    moved to a subsequent patch.

v2: Fixup several lingering crtc->fb instances that were missed in the
    first patch iteration.  [Rob Clark]
Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Reviewed-by: NRob Clark <robdclark@gmail.com>

This patch adds support to find the involved components connected to
the IPU display interface ports using the OF graph bindings documented
in Documentation/devicetree/bindings/media/video-interfaces.txt.
It makes use of the of_graph (formerly v4l2_of) parsing helpers and
thus depends on the patch that moves those out to drivers/of.

Each display interface needs to have an associated port node in the
device tree. We can associate this node with the crtc platform device
and use it to find the crtc corresponding to a given port node instead
of using a combination of parent device node and id number, as before.

Explicitly converting the void* cookie to the port device tree node
allows to get rid of the ipu_id and di_id fields. The multiplexer
setting on i.MX6 now can be obtained from the port id (reg property)
in the device tree.

The imx-drm node now needs a ports property that contains phandles
to each of the IPU display interface port nodes. From there, all
attached encoders are scanned and enabled encoders are added to a
waiting list.
The bind order makes sure that once all components are probed, crtcs
are bound before encoders, so that imx_drm_encoder_parse_of can be
called from the encoder bind callbacks.

For parsing the OF graph, temporary copies of the V4L2 OF graph
helpers are used, that can be removed again once those are available
at a generic place.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

The component helper provides us the drm_device which is being
registered.  Rather than having to reference a global in imx-drm-core,
use this to get the imxdrm device, and also use it to register the CRTC
against.

This means we never have CRTCs/encoders/connectors without the drivers
private data being accessible.

Remove the module owner field as well; this provides no protection
against the device being unbound.
Acked-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NShawn Guo <shawn.guo@linaro.org>
Reviewed-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Use the componentised device support for imx-drm.  This requires all
the sub-components and the master device to register with the component
device support.
Acked-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NShawn Guo <shawn.guo@linaro.org>
Reviewed-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

The encoder format passed into imx_drm_crtc_panel_format*() is the
encoder format used for DRM in most cases; the HDMI encoder sets
this to none, but this is incorrect, it should be TMDS.

Since this is the case, we can pass the drm_encoder structure
directly into this function and use the supplied fields there to
configure the CRTC.
Acked-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Acked-by: NShawn Guo <shawn.guo@linaro.org>
Reviewed-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

CRTC offsets are only set with the initial modeset, any
subseqent pageflips assume them to be kept the same, so
we need to remember the current state until another modeset
changes it.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The correct way for a driver to specify the coherent DMA mask is
not to directly access the field in the struct device, but to use
dma_set_coherent_mask().  Only arch and bus code should access this
member directly.

Convert all direct write accesses to using the correct API.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>