They now also work on vlv, which has the regs somewhere else. And
daring a glance into the looking glass it seems like this
functionality will continue to work the same for the next few hardware
platforms.

So it's better to just remove that misleading prefix and have a bit
shorter code for better readability.

The only exceptions are the panel/backlight functions shared with
intel_ddi.c, those get an intel_ prefix.

While at it make the vdd_on/off functions static.

And one straggler was missing the edp_ in the name, so make everything
neatly OCD.

Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The eDP spec defines some points where after you do action A, you have
to wait some time before action B. The thing is that in our driver
action B does not happen exactly after action A, but we still use
msleep() calls directly. What this patch does is that we record the
timestamp of when action A happened, then, just before action B, we
look at how much time has passed and only sleep the remaining amount
needed.

With this change, I am able to save about 5-20ms (out of the total
200ms) of the backlight_off delay and completely skip the 1ms
backlight_on delay. The 600ms vdd_off delay doesn't happen during
normal usage anymore due to a previous patch.

v2: - Rename ironlake_wait_jiffies_delay to intel_wait_until_after and
      move it to intel_display.c
    - Fix the msleep call: diff is in jiffies
v3: - Use "tmp_jiffies" so we don't need to worry about the value of
      "jiffies" advancing while we're doing the math.
v4: - Rename function again.
    - Move function to i915_drv.h.
    - Store last_power_cycle at edp_panel_off too.
    - Use msecs_to_jiffies_timeout, then replace the msleep with an
      open-coded version that avoids the extra +1 jiffy.
    - Try to add units to every variable name so we don't confuse
      jiffies with milliseconds.
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Add a new_config pointer to intel_crtc which will point to the new pipe
config for said crtc while intel_crtc.config will still contain the old
config during first parts of the modeset operation. This is a step
towards having the entire new state available during the compute phase,
so that we can make accurate decisions about global resource usage.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Add 'new_enabled' to intel_crtc and precompute it alongside new_encoder
and new_crtc. This will allow making decisions about shared resources
that are affected by the set of active pipes, before we've clobbered
anything for real.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

On older generations (gen2, gen3) the GPU requires fences for many
operations, such as blits. The display hardware also requires fences for
scanouts and this leads to a situation where an arbitrary number of
fences may be pinned by old scanouts following a pageflip but before we
have executed the unpin workqueue. This is unpredictable by userspace
and leads to random EDEADLK when submitting an otherwise benign
execbuffer. However, we can detect when we have an outstanding flip and
so cause userspace to wait upon their completion before finally
declaring that the system is starved of fences. This is really no worse
than forcing the GPU to stall waiting for older execbuffer to retire and
release their fences before we can reallocate them for the next
execbuffer.

v2: move the test for a pending fb unpin to a common routine for
later reuse during eviction

Reported-and-tested-by: dimon@gmx.net
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=73696Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It's an accident waiting to happen.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

That we can use for debugging purposes.

v2: Use designated initializers for the 'names' array (Paulo Zanoni,
    Jani Nikula).
    Add a check in case the array has a hole (which can now remain
    unnoticed with designated initializers) (Jani Nikula)

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com> (for v1)
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The first piece, intel_ddi_pll_select, finds a PLL and assigns it to
the CRTC, but doesn't write any register. It can also fail in case it
doesn't find a PLL.

The second piece, intel_ddi_pll_enable, uses the information stored by
intel_ddi_pll_select to actually enable the PLL by writing to its
register. This function can't fail. We also have some refcount sanity
checks here.

The idea is that one day we'll remove all the functions that touch
registers from haswell_crtc_mode_set to haswell_crtc_enable, so we'll
call intel_ddi_pll_select at haswell_crtc_mode_set and then call
intel_ddi_pll_enable at haswell_crtc_enable. Since I'm already
touching this code, let's take care of this particular split today.

v2: - Clock on the debug message is in KHz
    - Add missing POSTING_READ
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
[danvet: Bikeshed comments.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It was supposed to have been killed on the same commit that killed the
function, e1264ebe, but I guess the
intel_drv.h reorganization accidentally brought it back.
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This patch adds the initial infrastructure to allow a Runtime PM
implementation that sets the device to its D3 state. The patch just
adds the necessary callbacks and the initial infrastructure.

We still don't have any platform that actually uses this
infrastructure, we still don't call get/put in all the places we need
to, and we don't have any function to save/restore the state of the
registers. This is not a problem since no platform uses the code added
by this patch. We have a few people simultaneously working on runtime
PM, so this initial code could help everybody make their plans.

V2: - Move some functions to intel_pm.c
    - Remove useless pm_runtime_allow() call at init
    - Remove useless pm_runtime_mark_last_busy() call at get
    - Use pm_runtime_get_sync() instead of 2 calls
    - Add a WARN to check if we're really awake

V3: - Rebase.

V4: - Don't need to call pci_{save,restore}_state and
      pci_set_power_sate, since they're already called by the PCI
      layer
    - Remove wrong pm_runtime_enable() call at init_runtime_pm
Signed-off-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@gmail.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We don't modify the packed infoframe data, so we should keep the
const qualifier in place. Just pass the buffer as 'const void *'
instead of 'const uint8_t *' and we can drop the cast entirely.

v2: Do intel_sdvo_write_infoframe() as well
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

If one mode of a internal panel has more than one refresh rate, then a reduced
clock is found for the LFP (LVDS/eDP). This enables switching between low
and high frequency dynamically. Moving downclock calculation to intel_panel
so that it is common for LVDS and eDP.
Signed-off-by: NVandana Kannan <vandana.kannan@intel.com>
Signed-off-by: NPradeep Bhat <pradeep.bhat@intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Shovel a bit more of the the code into the setup function, and call
it earlier. Otherwise lockdep is unhappy since we cancel the delayed
resume work before it's initialized.

While at it also shovel the pc8 setup code into the same functions.
I wanted to also ditch the header declaration of the hws pc8 functions,
but for unfathomable reasons that stuff is in intel_display.c instead
of intel_pm.c.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71980Tested-by: NGuo Jinxian <jinxianx.guo@intel.com>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Atm we call intel_display_power_enabled() from
i915_capture_error_state() in IRQ context and then take a mutex. To fix
this add a new intel_display_power_enabled_sw() which returns the domain
state based on software tracking as opposed to reading the actual HW
state.

Since we use domain_use_count for this without locking on the reader
side make sure we increase the counter only after enabling all required
power wells and decrease it before disabling any of these power wells.

Regression introduced in
commit 1b02383464b4a915627ef3b8fd0ad7f07168c54c
Author: Imre Deak <imre.deak@intel.com>
Date:   Tue Sep 24 16:17:09 2013 +0300

    drm/i915: support for multiple power wells

Note that atm we depend on the value returned by
intel_display_power_enabled_sw() in i915_capture_error_state() to avoid
unclaimed register access reports. This was never guaranteed though,
since another thread can disable the power concurrently. If this is a
problem we need another explicit way to disable the reporting during
error captures.

v2:
- remove barriers as the caller can't depend on the value
  returned from i915_capture_error_state_sw() anyway (Ville)
- dump the state of pipe/transcoder power domain state (Daniel)
Reported-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

VLV can have eDP on either port B or C, or even both. Based on the
VBT spec, intel_dpd_is_edp() should work on VLV too, assuming we
check the correct ports.

So instead of hardcoding port D, rename the function to
intel_dp_is_edp() and pass the port as a parameter, and use it
on VLV ports B and C.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71051Tested-by: NRobert Hooker <robert.hooker@canonical.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NJesse Barnes <jbarnes@virtuousgeek.org>
[danvet: Wrestle the patch to apply and compile properly.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Prepare for being able to use the information at enable.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The backlight code has grown rather hairy, not least because the
hardware registers and bits have repeatedly been shuffled around. And
this isn't expected to get any easier with new hardware. Make things
easier for our (read: my) poor brains, and split the code up into chip
specific functions.

There should be no functional changes.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

ALthough usually there's only one connector that supports backlight,
this also finds the correct connector. Before, we only updated the
connector on pipe A, which might not be the one with backlight. (This
only made a difference on BYT.)
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Move from dev_priv to connector->panel. We still don't allow multiple
sysfs interfaces, though.

There should be no functional changes, except for a slight reordering of
connector backlight and sysfs destroy calls. (This change happens now
that the backlight device is actually per-connector, even though the
destroy calls became per-connector earlier.)
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

I've always felt the backlight device conditional build has been all
backwards. Make it feel right.

Gently move things towards connector based stuff while at it.

There should be no functional changes.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NImre Deak <imre.deak@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

vlv_dpio_read/write should be describe more in PHY centric instead of
display controller centric.
Create a enum dpio_channel for channel index and enum dpio_phy for PHY
index.  This should better to gather for upcoming platform.

v2: Rebase the code based on
drm/i915/vlv: Fix typo in the DPIO register define.

v3: Rename vlv_phy to dpio_phy_iosf_port and define additional macro
DPIO_PHY, and remove unrelated change. (Ville)
Suggested-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NChon Ming Lee <chon.ming.lee@intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

On VLV/BYT, backlight controls a per-pipe, so when adjusting the
backlight we need to pass the correct info.  So make the externally
visible backlight functions take a connector argument, which can be used
internally to figure out the pipe backlight to adjust.

v2: make connector pipe lookup check for NULL crtc (Jani)
    fixup connector check in ASLE code (Jani)
v3: make sure we take the mode config lock around lookups (Daniel)
v4: fix double unlock in panel_get_brightness (Daniel)
v5: push ASLE work into a work queue (Daniel)
v6: separate ASLE work to a prep patch, rebase (Jani)
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It's possible that the CCK clock could run at a different rate than the
DDR clock, so use the same method to get CCK as the GMBUS code does when
calculating the new CDclk divider in the VLV display code.
Reported-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Similarly rename the other related functions in the power domain
interface.

Higher level driver code calling these functions knows only about power
domains, not the underlying power wells which may be different on
different platforms. Also these functions really init/cleanup/resume
power domains and only through that all related power wells, so rename
them accordingly.

Note that I left i915_{request,release}_power_well as is, since that
really changes the state only of a single power well (and is HSW
specific). It should also get a better name once we make it more
generic by controlling things through a new audio power domain.

v4:
- use intel prefix instead of i915 everywhere (Paulo)
- use a $prefix_$block_$action format (Daniel)
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Call intel_ddi_get_config() to get the pipe_bpp settings from
DDI.

The sync polarity settings from DDI are irrelevant for CRT
output, so override them with data from the ADPA register.

Note: This is already merged in drm-intel-next-queued as

commit 6801c18c
Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date:   Tue Sep 24 14:24:05 2013 +0300

    drm/i915: Add HSW CRT output readout support

but is required for the following edp bpp bugfix.

v2: Extract intel_crt_get_flags()

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69691Tested-by: NQingshuai Tian <qingshuai.tian@intel.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Currently we make sure that all power domains are enabled during driver
init and turn off unneded ones only after the first modeset. Similarly
during suspend we enable all power domains, which will remain on through
the following resume until the first modeset.

This logic is supported by intel_set_power_well() in the power domain
framework. It would be nice to simplify the API, so that we only have
get/put functions and make it more explicit on the higher level how this
"power well on during init" logic works. This will make it also easier
if in the future we want to shorten the time the power wells are on.

For this add a new device private flag tracking whether we have the
power wells on because of init/suspend and use only
intel_display_power_get()/put(). As nothing else uses
intel_set_power_well() we can remove it.

This also fixes

commit 6efdf354
Author: Imre Deak <imre.deak@intel.com>
Date:   Wed Oct 16 17:25:52 2013 +0300

    drm/i915: enable only the needed power domains during modeset

where removing intel_set_power_well() resulted in not releasing the
reference on the power well that was taken during init and thus leaving
the power well on all the time. Regression reported by Paulo.

v2:
- move the init_power_on flag to the power_domains struct (Daniel)

v3:
- add note about this being a regression fix too (Paulo)
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

So far the modeset code enabled all power domains if it needed any. It
wasn't a problem since HW generations so far only had one always-on
power well and one dynamic power well that can be enabled/disabled. For
domains powered by always-on power wells (panel fitter on pipe A and the
eDP transcoder) we didn't do anything, for all other domains we just
enabled the single dynamic power well.

Future HW generations will change this, as they add multiple dynamic
power wells. Support for these will be added later, this patch prepares
for those by making sure we only enable the required domains.

Note that after this change on HSW we'll enable all power domains even
if it was the domain for the panel fitter on pipe A or the eDP
transcoder. This isn't a problem since the power domain framework
already checks if the domain is on an always-on power well and doesn't
do anything in this case.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Fill out the HSW watermark s/w tracking structures with the current
hardware state in intel_modeset_setup_hw_state(). This allows us to skip
the HW state readback during watermark programming and just use the values
we keep around in dev_priv->wm. Reduces the overhead of the watermark
programming quite a bit.

v2: s/init_wm/wm_get_hw_state
    Remove stale comment about sprites
    Make DDB partitioning readout safer
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
[danvet: Fix whitespace fail.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Introduce a new struct intel_pipe_wm which contains all the
watermarks for a single pipe. Use it to unify the LP0 and LP1+
watermark computations so that we can just iterate through the
watermark levels neatly and call ilk_compute_wm_level() for each.

Also add another tool ilk_wm_merge() that merges the LP1+ watermarks
from all pipes. For that, embed one intel_pipe_wm inside intel_crtc that
contains the currently valid watermarks for each pipe.

This is mainly preparatory work for pre-computing the watermarks for
each pipe and merging them at a later time. For now the merging still
happens immediately.

v2: Add some comments about level 0 DDB split and intel_wm_config
    Add WARN_ON for level 0 being disabled
    s/lp_wm/merged
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This file is all about the legacy fbdev support. If we want to extract
framebuffer functions, we better put those into a separate file.

Also rename functions accordingly, only two have used the intel_fb_
prefix anyway.
Reviewed-by: NChon Ming Lee <chon.ming.lee@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Boots Just Fine (tm)!

The only glitch seems to be that at least on Fedora the boot splash
gets confused and doesn't display much at all.

And since there's no ugly console flickering anymore in between, the
flicker while switching between X servers (VT support is still enabled)
is even more jarring.

Also, I'm unsure whether we don't need to somehow kick out vgacon, now
that nothing else gets in the way. But stuff seems to work, so I
don't care. Also everything still works as well with VGA_CONSOLE=n

Also the #ifdef mess needs a bit of a cleanup, follow-up patches will
do just that.

To keep the Kconfig tidy, extract all the i915 options into its own
file.

v2:
- Rebase on top of the preliminary hw support option and the
  intel_drv.h cleanup.
- Shut up warnings in i915_debugfs.c

v3: Use the right CONFIG variable, spotted by Chon Ming.

Cc: Lee, Chon Ming <chon.ming.lee@intel.com>
Cc: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: NChon Ming Lee <chon.ming.lee@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This reverts commit 6e1b4fda.

This is part of a revert due to a userspace breakage, better explained in the revert of 1a1a4cbf4906a13c0c377f708df5d94168e7b582.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Let's try to avoid these confusing negated booleans.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Something already got misplaced (although it's from a patch from
before Paulo's cleanup). Move it to the right spot.

v2: Remove the line to keep a neat block, requested by Paulo.
Reported-by: NPaulo Zanoni <przanoni@gmail.com>
Cc: Paulo Zanoni <przanoni@gmail.com>
Reviewed-by: NPaulo Zanoni <przanoni@gmail.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The intel_flush_primary_plane name actually tells us which plane
we're talking about.

Also reorganize the internals a bit and add a missing POSTING_READ()
to make sure the hardware has seen the changes by the time we
return from the function.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

IPS should be OK as long as one plane is enabled on the pipe, but
it does seem to cause problems when going between primary only and
sprite only.

This needs more investigations, but for now just disable IPS whenever
the primary plane is disabled.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It's not really needed, rather just adds another place to hold
intermediate values that could go wrong, and it's not clear that the
training pattern set or training lane set should be written at this
point at all.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

If we encounter a situation where the CPU blocks waiting for results
from the GPU, give the GPU a kick to boost its the frequency.

This should work to reduce user interface stalls and to quickly promote
mesa to high frequencies - but the cost is that our requested frequency
stalls high (as we do not idle for long enough before rc6 to start
reducing frequencies, nor are we aggressive at down clocking an
underused GPU). However, this should be mitigated by rc6 itself powering
off the GPU when idle, and that energy use is dependent upon the workload
of the GPU in addition to its frequency (e.g. the math or sampler
functions only consume power when used). Still, this is likely to
adversely affect light workloads.

In particular, this nearly eliminates the highly noticeable wake-up lag
in animations from idle. For example, expose or workspace transitions.
(However, given the situation where we fail to downclock, our requested
frequency is almost always the maximum, except for Baytrail where we
manually downclock upon idling. This often masks the latency of
upclocking after being idle, so animations are typically smooth - at the
cost of increased power consumption.)

Stéphane raised the concern that this will punish good applications and
reward bad applications - but due to the nature of how mesa performs its
client throttling, I believe all mesa applications will be roughly
equally affected. To address this concern, and to prevent applications
like compositors from permanently boosting the RPS state, we ratelimit the
frequency of the wait-boosts each client recieves.

Unfortunately, this techinique is ineffective with Ironlake - which also
has dynamic render power states and suffers just as dramatically. For
Ironlake, the thermal/power headroom is shared with the CPU through
Intelligent Power Sharing and the intel-ips module. This leaves us with
no GPU boost frequencies available when coming out of idle, and due to
hardware limitations we cannot change the arbitration between the CPU and
GPU quickly enough to be effective.

v2: Limit each client to receiving a single boost for each active period.
    Tested by QA to only marginally increase power, and to demonstrably
    increase throughput in games. No latency measurements yet.

v3: Cater for front-buffer rendering with manual throttling.

v4: Tidy up.

v5: Sadly the compositor needs frequent boosts as it may never idle, but
due to its picking mechanism (using ReadPixels) may require frequent
waits. Those waits, along with the waits for the vrefresh swap, conspire
to keep the GPU at low frequencies despite the interactive latency. To
overcome this we ditch the one-boost-per-active-period and just ratelimit
the number of wait-boosts each client can receive.
Reported-and-tested-by: NPaul Neumann <paul104x@yahoo.de>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68716Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: Stéphane Marchesin <stephane.marchesin@gmail.com>
Cc: Owen Taylor <otaylor@redhat.com>
Cc: "Meng, Mengmeng" <mengmeng.meng@intel.com>
Cc: "Zhuang, Lena" <lena.zhuang@intel.com>
Reviewed-by: NJesse Barnes <jbarnes@virtuousgeek.org>
[danvet: No extern for function prototypes in headers.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

struct drm_mode_display now has a separate crtc_ version of the clock to
be used when we're talking about the timings given to the harwadre (was
far as the mode is concerned).

This commit is really the result of a git grep adjusted_mode.*clock and
replacing those by adjusted_mode.crtc_clock. No functional change.

v2: Rebased on drm-intel-queued-next
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDamien Lespiau <damien.lespiau@intel.com>
Acked-by: NDave Airlie <airlied@gmail.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>