drm/nouveau/acpi: allow non-optimus setups to load vbios from acpi

There appear to be a crop of new hardware where the vbios is not
available from PROM/PRAMIN, but there is a valid _ROM method in ACPI.
The data read from PCIROM almost invariably contains invalid
instructions (still has the x86 opcodes), which makes this a low-risk
way to try to obtain a valid vbios image.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76475
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: <stable@vger.kernel.org> # v2.6.35+
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: fix missing newline

Add a missing newline at the end of a DRM_INFO message.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/bios: fetch the vbios from PROM using only aligned 32-bit accesses

Other kind of accesses are unreliable on Kepler cards. As advised by NVIDIA,
let's only use 32-bit accesses to fetch the vbios from PROM.

This fixes vbios fetching on my nve7 which failed in certain specific
conditions.

I suggest we Cc stable, for all kernels they still maintain after the big
rewrite.

Suggested-by: Christian Zander <czander@nvidia.com>
Signed-off-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/therm: let the vbios decide on the automatic fan management mode

This should fix automatic fan management on fermi cards who do not have
0x46 entries in the thermal table.

On my nve6, the blob sets the default linear range from 40°C to 100°C
but my nvcf's default values are 40°C to 85°C. Let's keep 85 as a default
for everyone.

Signed-off-by: Martin Peres <martin.peres@labri.fr>
Tested-by: Timothée Ravier <tim@siosm.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nvd7/therm: handle another kind of PWM fans

This should fix fan management on many nvd7+ chipsets.

Signed-off-by: Martin Peres <martin.peres@labri.fr>
Tested-by: Timothée Ravier <tim@siosm.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/pm/fan: drop the fan lock in fan_update() before rescheduling

This should fix a deadlock that has been reported to us where fan_update()
would hold the fan lock and try to grab the alarm_program_lock to reschedule
an update. On an other CPU, the alarm_program_lock would have been taken
before calling fan_update(), leading to a deadlock.

We should Cc: <stable@vger.kernel.org> # 3.9+

Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Tested-by: Timothée Ravier <tim@siosm.fr>
Tested-by: Boris Fersing (IRC nick fersingb, no public email address)
Signed-off-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: fix small thinko in vblank timestamping.

*hpos horizontal scanout position doesn't need to be corrected
to count the pixels between hactive end and htotal negative.
That is only needed for *vpos to count lines until end of
vblank for the vblank timestamping.

Use hpos as is without correction.

Removes occassional spikes in timestamps of up to 1 scanline
duration, thereby improves accuracy to about +/- 2 usecs instead
of +/- 12 usecs, wrt. true onset time as measured with high
precision equipment on NV-A5.

Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/therm: check for sensor presence with requested mode, not current

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

drm/nouveau/disp/dp: allow 540MHz data rate

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76319
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: recognise higher link rate for available dp bw calculations

I should resurrect/merge that cleanup branch to remove the weird
duplication.. One day.

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

drm/nouveau/disp: limit dp capabilities as per dcb

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

drm/nva3/fbram: restrict training pattern setup to GT218

It doesn't look like the others have the registers.

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

drm/nva3/devinit: restrict script access to some PFB regs

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

drm/nouveau/devinit: add interface to check if a mmio access by scripts is ok

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

drm/nouveau/bios: have strap reads show on devinit spam debug level

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

drm/nv50/gpio: fixup reset for gpios >= 16

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

drm/nv50/gpio: exclude sense value from mask when changing registers

Shouldn't effect anything, was just momentarily confusing while looking
at traces.

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

drm/gk104/gr: therm magic needed on some kepler boards

Not needed everywhere, and potentially not safe to do depending on how
the rest of PTHERM is configured...

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

drm/gm107/gr: initial support

Our ucode only partially works at this point, so requiring binary fw
image for now.

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

drm/gf100-/gf: fix a stupid typo, waiting on wrong signal for mmctx

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

drm/nouveau/bios: parsing of some random table needed to bring up gr

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

drm/nv50/bar: fix plymouth issues on certain efi macbooks

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

drm/nouveau: fix fbcon not being accelerated after suspend

This does *not* (and is not intended to) fix the issue reported by
Christoph Rudorff on the nouveau mailinglist.

The patch proposed (which is similar to this one, but also reorders
whether we disable accel or call fb_set_suspend first), papers over
another problem entirely by avoiding touching the framebuffer.

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

drm/gf100-/gr: split ppc state into its subunits

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

drm/gf100-/gf: split tpc state into its subunits

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

drm/gf100-/gr: split gpc state into its subunits

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

drm/gf100-/gr: tidy reg/ctx initval lists, mostly by giving them names

Unit names come from the Android GK20A driver.

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

drm/gk110/gr: minor adjustment to some random initval

Change from traces of a newer binary driver version.

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

drm/gk104/gr: minor adjustment to some random initval

Change from traces of a newer binary driver version.

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

drm/gf119/gr: fix bug in some random initval

Bug noticed vs traces.

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

drm/gf110/gr: fixup gpc/tpc initvals lists

Differences noted vs traces.

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

drm/gk208/gr: minor adjustment to some random initval

Change from traces of a newer binary driver version.

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

drm/gf108/gr: minor adjustment to some random initval

Bug noticed vs traces.

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

drm/gf104/gr: rename gf104 (nvc4), it came before gf106 (nvc3)

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

drm/nvc0-/graph: fix gpccs fuc stack setup

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

drm/nouveau/bios: add more checks to PRAMIN image fetching

Prevents an attempt to access VRAM on an un-posted board, which, on a
particular system with a GRID K1 installed, causes a MCE and chokes
the entire system.

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

support for platform devices

Upcoming mobile Kepler GPUs (such as GK20A) use the platform bus instead
of PCI to which Nouveau is tightly dependent. This patch allows Nouveau
to handle platform devices by:

- abstracting PCI-dependent functions that were typically used for
  resource querying and page mapping,
- introducing a nv_device_is_pci() function that allows to make
  PCI-dependent code conditional,
- providing a nouveau_drm_platform_probe() function that takes a GPU
  platform device to be probed.

Core code as well as engine/subdev drivers are updated wherever possible
to make use of these functions. Some older drivers are too dependent on
PCI to be properly updated, but all newer code on which future chips may
depend should at least be runnable with platform devices.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: support modesetting on GM107

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

drm/gm100/device: recognise GM107

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

drm/gm107/disp: initial implementation

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

drm/gm107/ltcg: initial implementation

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

drm/nouveau/bios: add HDMI-C (mini) connector type

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

drm/gm107/fb: initial implementation

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

drm/gk20a/timer: initial implementation

A bit different from NVIDIA's RFC patch, but I want this now for GM107.

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

drm/gm100/bios: hw disable register has moved

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

drm/gm107/devinit: initial implementation

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

drm/nve0/fifo: bind intr

Error code names from Android GK20A driver.

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

drm/nve0/fifo: attempt to recover from engine ctxsw timeouts

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

drm/nve0/fifo: attempt to recover engines from mmu faults

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

drm/nve0/fifo: allow copy engine channel to be looked up by instance

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

drm/nve0/fifo: use runlist event instead of polling

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

drm/nve0/fifo: allow channels to be marked as unrunnable

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

drm/nve0/fifo: single printk for sched error data

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

drm/nve0/fifo: single printk for mmu fault data

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

drm/nve0/fifo: ack pb intr individually after handling each unit

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

drm/nve0/fifo: runlist intr

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

drm/nve0/fifo: engine intr

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

drm/nve0/fifo: mask unhandled intr bits when seen, rather than all intrs

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

drm/nvc0/fifo: attempt to recover from engine ctxsw timeouts

My test cases don't seem to trigger this on all Fermi boards, not sure
if they're broken tests or it didn't work until later versions.

GF119 definitely works.

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

drm/nvc0/fifo: attempt to recover engines from mmu faults

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

drm/nvc0/fifo: use subdev identifiers for bar/ifb fault recovery cases

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

drm/nvc0/fifo: use runlist event instead of polling

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

drm/nvc0/fifo: allow channels to be marked as unrunnable

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

drm/nvc0/fifo: sched intr

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

drm/nvc0/fifo: single printk for mmu fault data

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

drm/nvc0/fifo: ack pb intr individually after handling each unit

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

drm/nvc0/fifo: runlist intr

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

drm/nvc0/fifo: engine intr

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

drm/nvc0/fifo: mask unhandled intr bits when seen, rather than all intrs

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

drm/nvc0/fifo: rename a couple of units

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

drm/nvd0/disp: add debugging to show which supervisor actions are taken

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

drm/nv50-/disp: dump channel state when update method fails

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

drm/nv50-/disp: allow dumping core channel state at first supervisor intr

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

drm/nv50-/disp: add method descriptions for debugging

Lists of known methods for the DMA channel classes, and mappings to
their priv register addresses (where known).

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

drm/nv50/disp: decode the known error codes to human readable form

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

drm/nv50/disp: preparation for storing static class data

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

drm/nouveau/core: extend width of engine mask for namedb

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

drm/nouveau/bios: fix INDEX_ADDRESS_LATCHED trace printout

Having a \n in the middle of a format string means that the next line
doesn't get the prefixes unlike every other line printed by the trace.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nv50/gr: decode texture trap status code

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: use nv_debug for NV_DEBUG, make DRM a separate subflag

It's really confusing for NV_DEBUG's printing to be controlled via
drm.debug while everything else is controlled via nouveau.debug. These
messages can be turned on with nouveau.debug=DRM=debug.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: make hdmi device finding failure prints debug level

The hdmi device is required for runtime pm. However it is not available
on many esp older devices, which were all seeing these error messages.
Take this opportunity to also convert to nv_debug instead of the DRM_*
messages, like the rest of nouveau does.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nve0/fifo: allocate usermem as needed

Memory was always allocated for 4096 channels. Change this to allocate
what we actually need according to the number of channels we use.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: handle -EACCES runtime PM return code

pm_runtime_get*() may return -EACCES to indicate a device does not have
runtime PM enabled. This is currently the case with platform devices
on Nouveau, and is not an error in that context. Handle this case
without failure.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/abi16: fix handles past the 32nd one

abi16->handles is a u64, so make sure to use 1ULL << val when modifying.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau: replace ffsll with __ffs64

The ffsll function is a lot slower than the __ffs64 built-in which
compiles to a single instruction on 64-bit. It's also nice to avoid
custom versions of standard functions. Note that __ffs == ffs - 1.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/hwmon: replace strict_strtol() with kstrtol()

The usage of strict_strtol() is not preferred, because
strict_strtol() is obsolete. Thus, kstrtol() should be
used.

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nouveau/hwmon: remove some redundant checks

No need to check "ret" twice in a row.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/nv50/graph: update status enum names

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

drm/bridge/ptn3460: fix modular build

This failed to build =m, quick fix.

Signed-off-by: Dave Airlie <airlied@redhat.com>

Merge branch 'exynos-drm-next' of git://git./linux/kernel/git/daeinki/drm-exynos into drm-next

Highlights
----------

Re-factoring works over the exynos drm framework.
  - drm_crtc, drm_encoder/drm_connector are implemented by sub drivers
    directly.
  - Removing pm interfaces from each sub driver, and implementing them
    at top level of exynos drm.
Add DisplayPort Transmitter driver.
  - Just moving existing driver from drivers/vides/exynos into
    drivers/gpu/drm/exynos.
Add new LVDS bridge driver, PTN3460.
  - Placed in drivers/gpu/drm/bridge, and this device is used to transfer
    image signal from DP(DisplayPort) to LVDS Panel.
    So this driver will be used with DP driver moved into exynos drm.
Add parallel panel support
  - With the re-factoring patch series, existing parallel panel support was
    broken by moving exynos_drm_display ops into each real connector driver,
    DP. So this patch series adds a new parallel panel module,
    exynos_drm_dpi, for supporting parallel panel, and also adds relevant
    bindings.
Some fixups and cleanups.

* 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos: (45 commits)
  drm/exynos: fimd: remove unused variable
  ARM: dts: exynos4210-universal: add exynos/fimd node
  drm/exynos: restore parallel output interface support
  exynos/fimd: add parallel output related bindings
  drm/exynos: correct timing porch conversion
  drm/exynos: init kms poll after creation of connectors
  drm/exynos: delay fbdev initialization until an output is connected
  drm/exynos: fix unnecessary resource cleanup
  drm/exynos: hdmi: use i2c_adapter instead of i2c_client
  drm/exynos: hdmi: consider APB PHY
  drm/exynos: Remove the exynos_drm_connector shim
  drm/exynos: Implement lvds bridge discovery to DP driver
  drm/bridge: Add PTN3460 bridge driver
  drm/exynos: Implement drm_connector directly in vidi driver
  drm/exynos: Implement drm_connector directly in dp driver
  drm/exynos: Implement drm_connector in hdmi directly
  drm/exynos: Add create_connector callback
  drm/exynos: Consolidate suspend/resume in drm_drv
  drm/exynos: Clean up FIMD power on/off routines
  drm/exynos: Implement dpms display callback in DP
  ...

drm/gma500: add locking to fixed panel edid probing

With the recent addition of locking checks in

commit 62ff94a5492175759546f8bc61383189d6b49122
Author:     Daniel Vetter <daniel.vetter@ffwll.ch>
AuthorDate: Thu Jan 23 22:18:47 2014 +0100

    drm/crtc-helper: remove LOCKING from kerneldoc

drm_add_edid_modes started to WARN about the mode_config.mutex not
being held in the lvds and dp initialization code.

Now since this is init code locking is fairly redudant if it wouldn't
be for the drm core registering sysfs files a bit early. And the
locking WARNINGs nicely enforce that indeed all access to the mode
lists are properly protected. And a full audit shows that only i915
and gma500 touch the modes lists at init time.

Hence I've opted to wrap up this entire mode detection sequence for
fixed panels with the mode_config mutex for both lvds and edp outputs.

Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drm/i915: add locking to fixed panel edid probing

With the recent addition of locking checks in

commit 62ff94a5492175759546f8bc61383189d6b49122
Author:     Daniel Vetter <daniel.vetter@ffwll.ch>
AuthorDate: Thu Jan 23 22:18:47 2014 +0100

    drm/crtc-helper: remove LOCKING from kerneldoc

drm_add_edid_modes started to WARN about the mode_config.mutex not
being held in the lvds and dp initialization code.

Now since this is init code locking is fairly redudant if it wouldn't
be for the drm core registering sysfs files a bit early. And the
locking WARNINGs nicely enforce that indeed all access to the mode
lists are properly protected. And a full audit shows that only i915
and gma500 touch the modes lists at init time.

Hence I've opted to wrap up this entire mode detection sequence for
fixed panels with the mode_config mutex for both lvds and edp outputs.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drm/exynos: fimd: remove unused variable

The patch removes unused vidcon0 field from fimd_context structure.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

ARM: dts: exynos4210-universal: add exynos/fimd node

The patch adds fimd node with display timings for exynos4210-universal device.
It also makes LCD regulators always on. This allow to re-use panel initialized
by boot loader.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

drm/exynos: restore parallel output interface support

The patch adds parallel output interface to FIMD device driver.
It also restores support for panels initialized by boot loader,
but without proper kernel driver.
Driver uses video interface bindings to find connected panel.
It uses drm_panel interface to interact with the panel.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

exynos/fimd: add parallel output related bindings

The patch adds bindings required to add support
for parallel output.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

drm/exynos: correct timing porch conversion

The patch corrects porch calculation. It should
be calculated as a difference between adjacent
respective fields of drm_display_mode.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

drm/exynos: init kms poll after creation of connectors

KMS poll init helper should be run when connectors are created,
otherwise it will not schedule connection detector.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

drm/exynos: delay fbdev initialization until an output is connected

In case fbdev is initialized before any output is connected,
fb resolution defaults to 1024x768. After that any output with
bigger resolution is ignored and fbdev is not displayed.
The patch postpones fbdev initialization to avoid such situation.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

drm/exynos: fix unnecessary resource cleanup

This patch removes unnecessary drm_mode_config_cleanup call.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>