Kill the blt/render tracking we currently have and use the frontbuffer
tracking infrastructure.

Don't enable things by default yet.

v2: (Rodrigo) Fix small conflict on rebase and typo at subject.
v3: (Paulo) Rebase on RENDER_CS change.
v4: (Paulo) Rebase.
v5: (Paulo) Simplify: flushes don't have origin (Daniel).
            Also rebase due to patch order changes.
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

In invalidate and flush functions of eDP DRRS, if deferred downclock
work starts execution at a time window between acquiring the drrs
mutex and cancellation of the deferred work
(intel_edp_drrs_downclock_work), then deferred work will find
drrs mutex locked and wait for the same.

Meanwhile the function that acquired mutex drrs invalidate/flush will
wait for the completion of the deferred work before releasing the mutex.
Thats a deadlock.

To avoid such deadlock scenario, this change cancels the deferred work
before acquiring the mutex at invalidate and flush functions.
Signed-off-by: NRamalingam C <ramalingam.c@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Adding a debugfs entry to determine if DRRS is supported or not

V2: [By Ram]: Following details about the active crtc will be filled
	in seq-file of the debugfs
	1. Encoder output type
	2. DRRS Support on this CRTC
	3. DRRS current state
	4. Current Vrefresh
Format is as follows:
CRTC 1:  Output: eDP, DRRS Supported: Yes (Seamless), DRRS_State: DRRS_HIGH_RR, Vrefresh: 60
CRTC 2:  Output: HDMI, DRRS Supported : No, VBT DRRS_type: Seamless
CRTC 1:  Output: eDP, DRRS Supported: Yes (Seamless), DRRS_State: DRRS_LOW_RR, Vrefresh: 40
CRTC 2:  Output: HDMI, DRRS Supported : No, VBT DRRS_type: Seamless

V3: [By Ram]: Readability is improved.
	Another error case is covered [Daniel]

V4: [By Ram]: Current status of the Idleness DRRS along with
	the Front buffer bits are added to the debugfs. [Rodrigo]

V5: [By Ram]: Rephrased to make it easy to understand.
	And format is modified. [Rodrigo]

V6: [By Ram]: Modeset mutex are acquired for each crtc along with
	renaming the Idleness detection states  [Daniel]
Signed-off-by: NVandana Kannan <vandana.kannan@intel.com>
Signed-off-by: NRamalingam C <ramalingam.c@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
[danvet: dump full busy_frontbuffer_bits and remove the dubios
computed logical state of DRRS - debugfs is about what is fact,
developers should reach their own conclusion when debugging issues.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Adds a parameter which can be used with DRM_I915_GETPARAM to query the
GPU revision. The intention is to use this in Mesa to implement the
WaDisableSIMD16On3SrcInstr workaround on Skylake but only for
revision 2.
Signed-off-by: NNeil Roberts <neil@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

When logging that full mode switch is necessary, log which connector,
encoder or crtc has caused it, so it is easier to figure out what is
goind on by just looking at the log.
Signed-off-by: NAnder Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We have similar macros for crtcs and encoders, and the pattern happens
often enough to justify the macro.
Signed-off-by: NAnder Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

In the path were there is no state to duplicate, the allocated crtc
state wouldn't have the crtc backpointer initialized.
Signed-off-by: NAnder Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The current minimum vco frequency leaves us with a gap in our supported
frequencies at 233-243 MHz. Your typical 2560x1440@60 display wants a
pixel clock of 241.5 MHz, which is just withing that gap. Reduce the
allowed vco min frequency to 4.8GHz to reduce the gap to 233-240 MHz,
and thus allow such displays to work.

4.8 GHz is actually the documented (at least in some docs) limit of the
PLL, and we just picked 4.86 GHz originally because that was the lowest
value produced by the PLL spreadsheet, which obviously didn't consider
2560x1440 displays.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NVijay Purushothaman <vijay.a.purushothaman@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We need this for FBC, and possibly for PSR too.

v2: Don't only flush: invalidate too (Daniel).
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We want to port FBC to the frontbuffer tracking infrastructure, but
for that we need to know what caused the object invalidation so
we can react accordingly: CPU mmaps need manual, GTT mmaps and
flips don't need handling and ring rendering needs nukes.

v2: - s/ORIGIN_RENDER/ORIGIN_CS/ (Daniel, Rodrigo)
    - Fix copy/pasted wrong documentation
    - Rebase
v3: - Rebase
v4: - Don't pass the operation to flushes (Daniel).
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This reverts commit 3f678c96.

We've been a bit too optimistic with this one here :(

The trouble is that internally we're still using these plane
update/disable hooks. Which was totally ok pre-atomic since the drm
core did all the book-keeping updating and these just mostly updated
hw state. But with atomic there's lots more going on, and it causes
heaps of trouble with the load detect code.

This one specifically cause a deadlock since both the load detect code
and the nested plane atomic helper functions tried to grab the same
locks. It only blows up because of the evil tricks though we play with
the implicit ww acquire context.

Applying this revert unearths the NULL deref on already freed
framebuffer objects reported as a regression in 4.0 by various people.

Fixing this will be fairly invasive, hence revert even for the
4.1-next queue.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Bolle <pebolle@tiscali.nl>
Acked-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

This translation entry was updated after electrical validation by the hw
team. The other entries are removed from existence as they aren't
validated and because the sole use of a certain type of level shifter
for SKL products is anticipated.

v2: Remove all the other entries and force the use of the 800mv+2dB
    config (Sonika)
Suggested-by: NSonika Jindal <sonika.jindal@intel.com>
Cc: Sonika Jindal <sonika.jindal@intel.com>
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Reviewed-by: NSonika Jindal <sonika.jindal@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Implicit usage of local variables in macros isn't exactly the greatest
thing in the world, especially when that variable is the drm device and
we want to move towards a broader use of the i915 device structure.

Let's make for_each_sprite() take dev_priv as its first argument then.
Suggested-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Chris Wilson <chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Implicit usage of local variables in macros isn't exactly the greatest
thing in the world, especially when that variable is the drm device and
we want to move towards a broader use of the i915 device structure.

Let's make for_each_plane() take dev_priv as its first argument then.
Suggested-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Chris Wilson <chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

With the two-step reset counter increments which braket the actual
reset code and the subsequent wake-up we're guaranteeing that all the
lockless waiters _will_ be woken up. And since we unconditionally bail
out of waits with -EAGAIN (or -EIO) in that case there is not risk of
lost interrupt enabling bits when the lockless wait code races against
a gpu reset.

Let's remove this FIXME as resolved then.
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Change 'mutliple' to 'multiple'
Change 'mutlipler' to 'multiplier'
Change 'Haswel' to 'Haswell'
Signed-off-by: NYannick Guerrini <yguerrini@tomshardware.fr>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

plane->fb is a legacy pointer that not always be up-to-date (or updated
early enough).  Make sure the watermark code uses plane->state->fb so
that we're always doing our calculations based on the correct
framebuffers.

This patch was generated by Coccinelle with the following semantic
patch:

        @@
        struct drm_plane *P;
        @@
        - P->fb
        + P->state->fb

v2: Rebase
Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The cursor size fields in intel_crtc just duplicate the data from
cursor->state.crtc_{w,h} so we don't need them any more.  Worse, their
use in the watermark code actually introduces a subtle bug since they
don't get updated to mirror the state values until the plane commit
stage, which is *after* we've already used them to calculate new
watermark values.  This happens because we had to move watermark updates
slightly earlier (outside vblank evasion) in commit

        commit 32b7eeec
        Author: Matt Roper <matthew.d.roper@intel.com>
        Date:   Wed Dec 24 07:59:06 2014 -0800

            drm/i915: Refactor work that can sleep out of commit (v7)

Dropping the intel_crtc fields and just using the state values (which
are properly updated by the time watermark updates happen) should solve
the problem.

Aside from the actual removal of the struct fields (which are formatted
in a way that I couldn't figure out how to match in Coccinelle), the
rest of this patch was generated via the following semantic patch:

        // Drop assignment
        @@
        struct intel_crtc *C;
        struct drm_plane_state S;
        @@
        (
        - C->cursor_width = S.crtc_w;
        |
        - C->cursor_height = S.crtc_h;
        )

        // Replace usage
        @@
        struct intel_crtc *C;
        expression E;
        @@
        (
        - C->cursor_width
        + C->base.cursor->state->crtc_w
        |
        - C->cursor_height
        + C->base.cursor->state->crtc_h
        |
        - to_intel_crtc(E)->cursor_width
        + E->cursor->state->crtc_w
        |
        - to_intel_crtc(E)->cursor_height
        + E->cursor->state->crtc_h
        )

v2: Rebase

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joe Konno <joe.konno@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89346Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Drivers implementing the universal planes API report the list of
supported pixel formats for the primary plane. Make sure the fb passed
to the setcrtc ioctl is compatible.

Drivers not implementing the universal planes API will have no format
reported for the primary plane, skip the check in that case.
Signed-off-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Since the beginning, sysfs/connector/status has done a heavyweight
detection of the current connector status. But no user, such as upowerd
or logind, has ever desired to initiate a probe. Move the probing into a
new attribute so that existing readers get the behaviour they desire.

v2: David Herrmann suggested using "echo detect > /sys/.../status" to
trigger the probing, which is a fine idea. This extends that to also
allow the user to apply the force detection overrides at runtime.

v3: Now with airlied's email address fixed! Requires sysfs_streq()
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: David Herrmann <dh.herrmann@gmail.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Alex Deucher <alexdeucher@gmail.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We shouldn't tempt driver writers into using this since it uses a
default format list which is likely wrong. And when that's done we can
simplify the code a bit, too.

Noticed while reviewing a patch from Laurent.

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

Both the legacy and atomic helpers need to check whether a plane
supports a given pixel format. The code is currently duplicated, share
it.
Signed-off-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
[danvet: Slightly extend the docbook.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Change 'pixes' to 'pixels'
Change 'enabel' to 'enable'
Change 'enabeling' to 'enabling'
Signed-off-by: NYannick Guerrini <yguerrini@tomshardware.fr>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Originally it was impossible to be dropping the last refcount in this
function since there was always one around still from the idr. But in

commit 83f45fc3
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed Aug 6 09:10:18 2014 +0200

    drm: Don't grab an fb reference for the idr

we've switched to weak references, broke that assumption but forgot to
fix it up.

Since we still force-disable planes it's only possible to hit this
when racing multiple rmfb with fbdev restoring or similar evil things.
As long as userspace is nice it's impossible to hit the BUG_ON.

But the BUG_ON would most likely be hit from fbdev code, which usually
invovles the console_lock besides all modeset locks. So very likely
we'd never get the bug reports if this was hit in the wild, hence
better be safe than sorry and backport.

Spotted by Matt Roper while reviewing other patches.

Cc: stable@vger.kernel.org
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

So no need to have code which never gets called in the driver.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Commit 1da30627 "drm: Add rotation value to
plane state" moved the rotation property to DRM core but only did the set
property part. This does the get property part as well.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: NMatt Roper <matthew.d.roper@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Copypaste-fail from cleanup_planes. Reported by Tvrtko.

Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

When enabling atomic state object for this ioctl in

commit abd69c55
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Tue Nov 25 23:50:05 2014 +0100

    drm: Handle atomic state properly in kms getfoo ioctl

I've forgotten to remove this hunk in one of the later revisions.
drm_encoder_get_crtc already does this.

Cc: Rob Clark <robdclark@gmail.com>
Cc: Sean Paul <seanpaul@chromium.org>
Reviewed-by: NRob Clark <robdclark@gmail.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

commit a8c6ecb3
Merge: 8dd0eb35 9eccca08
Author: Dave Airlie <airlied@redhat.com>
Date:   Mon Mar 9 19:58:30 2015 +1000

    Merge tag 'v4.0-rc3' into drm-next

managed to pick the wrong code to resolve the conflict and left us with
a mutex_lock(struct_mutex) without the mutex_unlock(struct_mutex) leading
to a deadlock. Fix the problem by recovering the correct code which doesn't
need the lock.

Cc: Dave Airlie <airlied@redhat.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Use cases like rotation require these hooks to have some context so they
know how to prepare and cleanup the frame buffer correctly.

For i915 specifically, object backing pages need to be mapped differently
for different rotation modes and the driver needs to know which mapping to
instantiate and which to tear down when transitioning between them.

v2: Made passed in states const. (Daniel Vetter)

[airlied: add mdp5 and atmel fixups]
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: NRob Clark <robdclark@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

In kexec environment, we are more likely to encounter irq's already
enabled from previous environment.  At which point we find that writes
to disable/clear pending irq's are slightly less than useless without
first enabling clocks.

TODO: full blown state read-in so kexec'd kernel can inherit the mode
already setup.
Signed-off-by: NRob Clark <robdclark@gmail.com>

Seems like we just want BLEND_EN and not BLEND_TRANSP_EN (setting the
latter results in black pixels in the cursor image treated as
transparent).
Signed-off-by: NRob Clark <robdclark@gmail.com>

If cursor is set near the edge of the screen, it is not valid to use the
new cursor width/height as the ROI dimensions.  Split out the ROI calc
and use it both cursor_set and cursor_move.
Signed-off-by: NRob Clark <robdclark@gmail.com>

If the atomic commit fails due to completion wait interruption the
atomic commit object is not freed and is thus leaked. Free it.
Signed-off-by: NLaurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: NRob Clark <robdclark@gmail.com>

When a CRTC is disabled, no CTL is allocated to it (CRTC->ctl == NULL);
in that case we should not try to FLUSH registers and do nothing instead.

This can happen when we try to move a cursor but the CRTC's CTL
(CONTROL) has not been allocated yet (inactive CRTC).
It can also happens when we .atomic_check()/.atomic_flush() on a
disabled CRTC.

A CTL needs to be kept as long as the CRTC is alive. Releasing it
after the last VBlank is safer than in .atomic_flush().
Signed-off-by: NStephane Viau <sviau@codeaurora.org>
Signed-off-by: NRob Clark <robdclark@gmail.com>

Some target have up to 6 layer mixers (LM).
Let the header file access the last LM's base address.
Signed-off-by: NStephane Viau <sviau@codeaurora.org>
Signed-off-by: NRob Clark <robdclark@gmail.com>

Commit 0b776d45 ("drm/msm: fix fallout of atomic dpms
changes") has a typo in both mdp5_encoder_helper_funcs and
mdp5_crtc_helper_funcs definitions:

	.dpms entry should be replaced by .disable and .enable

Also fixed a typo in mdp5_encoder_enable().

Note that these typos are only present for MDP5. MDP4 is fine.
Signed-off-by: NStephane Viau <sviau@codeaurora.org>
Signed-off-by: NRob Clark <robdclark@gmail.com>

We need to store device offsets in 64 bit as the device
address space may be larger than the CPU's.

Fixes GPU init failures on radeons with 4GB or more of
vram on 32 bit kernels.  We put vram at the start of the
GPU's address space so the gart aperture starts at 4 GB
causing all GPU addresses in the gart aperture to get
truncated.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=89072

[airlied: fix warning on nouveau build]
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: thellstrom@vmware.com
Acked-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

The current implementation is limited by the number of addresses that
fit into an unsigned long. This causes problems on 32-bit Tegra where
unsigned long is 32-bit but drm_mm is used to manage an IOVA space of
4 GiB. Given the 32-bit limitation, the range is limited to 4 GiB - 1
(or 4 GiB - 4 KiB for page granularity).

This commit changes the start and size of the range to be an unsigned
64-bit integer, thus allowing much larger ranges to be supported.

[airlied: fix i915 warnings and coloring callback]
Signed-off-by: NThierry Reding <treding@nvidia.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDave Airlie <airlied@redhat.com>

fixupo

Bjørn reported that his machine hang during hibernation and eventually
bisected the problem to the following commit:

commit da2bc1b9
Author: Imre Deak <imre.deak@intel.com>
Date:   Thu Oct 23 19:23:26 2014 +0300

    drm/i915: add poweroff_late handler

The problem seems to be that after the kernel puts the device into D3
the BIOS still tries to access it, or otherwise assumes that it's in D0.
This is clearly bogus, since ACPI mandates that devices are put into D3
by the OSPM if they are not wake-up sources. In the future we want to
unify more of the driver's runtime and system suspend paths, for example
by skipping all the system suspend/hibernation hooks if the device is
runtime suspended already. Accordingly for all other platforms the goal
is still to properly power down the device during hibernation.

v2:
- Another GEN4 Lenovo laptop had the same issue, while platforms from
  other vendors (including mobile and desktop, GEN4 and non-GEN4) seem
  to work fine. Based on this apply the workaround on all GEN4 Lenovo
  platforms.
- add code comment about failing platforms (Ville)

Reference: http://lists.freedesktop.org/archives/intel-gfx/2015-February/060633.htmlReported-and-bisected-by: NBjørn Mork <bjorn@mork.no>
Cc: stable@vger.kernel.org # v3.19
Signed-off-by: NImre Deak <imre.deak@intel.com>
Acked-by: NDaniel Vetter <daniel.vetter@intel.com>
Signed-off-by: NJani Nikula <jani.nikula@intel.com>