Commentary from Chris Wilson's original version:

> I was looking at some wait_for() timeouts on a slow system, with lots of
> debug enabled (KASAN, lockdep, mmio_debug). Thinking that we were
> mishandling the timeout, I tried to ensure that we loop at least once
> after first testing COND. However, the double test of COND either side
> of the timeout check makes that unlikely. But we can do an equivalent
> loop, that keeps the COND check after testing for timeout (required so
> that we are not preempted between testing COND and then testing for a
> timeout) without expanding COND twice.
>
> The advantage of only expanding COND once is a dramatic reduction in
> code size:
>
>    text	   data	    bss	    dec	    hex
>    1308733	   5184	   1152	1315069	 1410fd	before
>    1305341	   5184	   1152	1311677	 1403bd	after

but it turned out that due to a missing iniitialiser, gcc had "gone
wild trimming undefined code" :( This version acheives a rather more
modest (but still worthwhile) gain of ~550 bytes.
Signed-off-by: NDave Gordon <david.s.gordon@intel.com>
Original-idea-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Zanoni, Paulo R <paulo.r.zanoni@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1473855033-26980-1-git-send-email-david.s.gordon@intel.comReviewed-by: NPaulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Turns out
commit a0562819 ("drm/i915: Get panel_type from OpRegion panel
details") has regressed quite a few machines. So it looks like we
can't use the panel type from OpRegion on all systems, and yet we
absolutely must use it on some specific systems.

Despite trying, I was unable to find any automagic way to determine
if the OpRegion panel type is respectable or not. The only glimmer
of hope I had was bit 8 in the SCIC response, but that turned out to
not work either (it was always 0 on both types of systems).

So, to fix the regressions without breaking the machine we know to need
the OpRegion panel type, let's just add a quirk for this. Only specific
machines known to require the OpRegion panel type will therefore use
it. Everyone else will fall bck to the VBT panel type.

The only known machine so far is a "Conrac GmbH IX45GM2". The PCI
subsystem ID on this machine is just a generic 8086:2a42, so of no use.
Instead we'll go with a DMI match.

I suspect we can now also revert
commit aeddda06 ("drm/i915: Ignore panel type from OpRegion on SKL")
but let's leave that to a separate patch.

v2: Do the DMI match in the opregion code directly, as dev_priv->quirks
    gets populated too late

Cc: Rob Kramer <rob@solution-space.com>
Cc: Martin van Es <martin@mrvanes.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Marco Krüger <krgsch@gmail.com>
Cc: Sean Greenslade <sean@seangreenslade.com>
Cc: Trudy Tective <bertslany@gmail.com>
Cc: Robin Müller <rm1990@gmx.de>
Cc: Alexander Kobel <a-kobel@a-kobel.de>
Cc: Alexey Shumitsky <alexey.shumitsky@gmail.com>
Cc: Emil Andersen Lauridsen <mine809@gmail.com>
Cc: oceans112@gmail.com
Cc: James Hogan <james@albanarts.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: stable@vger.kernel.org
References: https://lists.freedesktop.org/archives/intel-gfx/2016-August/105545.html
References: https://lists.freedesktop.org/archives/dri-devel/2016-August/116888.html
References: https://lists.freedesktop.org/archives/intel-gfx/2016-June/098826.html
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94825
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97060
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97443
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97363
Fixes: a0562819 ("drm/i915: Get panel_type from OpRegion panel details")
Tested-by: NMarco Krüger <krgsch@gmail.com>
Tested-by: NAlexey Shumitsky <alexey.shumitsky@gmail.com>
Tested-by: NSean Greenslade <sean@seangreenslade.com>
Tested-by: NEmil Andersen Lauridsen <mine809@gmail.com>
Tested-by: NRobin Müller <rm1990@gmx.de>
Tested-by: oceans112@gmail.com
Tested-by: NRob Kramer <rob@solution-space.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1473758539-21565-1-git-send-email-ville.syrjala@linux.intel.com
References: http://patchwork.freedesktop.org/patch/msgid/1473602239-15855-1-git-send-email-adrienverge@gmail.comAcked-by: NJani Nikula <jani.nikula@intel.com>

This reverts

commit 1c80c25f
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed May 18 18:47:12 2016 +0200

    drm/i915/psr: Make idle_frames sensible again

There are panels that needs 4 idle frames before entering PSR,
but VBT is unproperly set.

Also lately it was identified that idle frame count calculated at HW
can be off by 1, what makes the minimum of 2, at least.

Without the current vbt+1 we are with the risk of having HW calculating
0 idle frames and entering PSR when it shouldn't. Regardless the lack
of link training.

[Jani: there is some disagreement on the explanation, but the commit
regresses so revert it is.]

References: http://marc.info/?i=20160904191153.GA2328@light.dominikbrodowski.net
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Fixes: 1c80c25f ("drm/i915/psr: Make idle_frames sensible again")
Cc: drm-intel-fixes@lists.freedesktop.org # v4.8-rc1+
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1473295351-8766-1-git-send-email-rodrigo.vivi@intel.com

This adds support for KBL in the new function added in commit ID:
commit <f169660e> that returns a
shared pll in case of DDI platforms.
Signed-off-by: NManasi Navare <manasi.d.navare@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1473728663-14355-1-git-send-email-manasi.d.navare@intel.com

drm already provides fallback versions of readq and writeq.
Signed-off-by: NMatthew Auld <matthew.auld@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1473451373-9852-1-git-send-email-matthew.auld@intel.com

Fix the number of tries in channel euqalization link training sequence
according to DP 1.2 Spec. It returns a boolean depending on channel
equalization pass or failure.
Signed-off-by: NDhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: NManasi Navare <manasi.d.navare@intel.com>
Reviewed-by: NMika Kahola <mika.kahola@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

This function cleans up clock recovery loop in link training compliant
tp Dp Spec 1.2. It tries the clock recovery 5 times for the same voltage
or until max voltage swing is reached and removes the additional non
compliant retries. This function now returns a boolean values based on
if clock recovery passed or failed.

v3:
* Better Debug prints in case of failures (Mika Kahola)
v2:
* Rebased on top of new revision of vswing patch (Manasi Navare)
Signed-off-by: NDhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: NManasi Navare <manasi.d.navare@intel.com>
Reviewed-by: NMika Kahola <mika.kahola@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Wrap the max. vswing check in a separate function.
This makes the clock recovery phase of DP link training cleaner

v3:
Fixed the paranthesis warning (Mika Kahola)
v2:
Fixed the Compiler warning (Mika Kahola)
Signed-off-by: NDhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: NManasi Navare <manasi.d.navare@intel.com>
Reviewed-by: NMika Kahola <mika.kahola@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Add the PLL selection code for HSW/BDW/BXT/SKL into a stand-alone function
in order to allow for the implementation of a platform neutral upfront
link training function.

v4:
* Removed dereferencing NULL pointer in  case of failure (Dhinakaran Pandiyan)
v3:
* Add Hooks for all DDI platforms into this standalone function

v2:
* Change the macro to use dev_priv instead of dev (David Weinehall)
Reviewed-by: NDurgadoss R <durgadoss.r@intel.com>
Signed-off-by: NManasi Navare <manasi.d.navare@intel.com>
Signed-off-by: NJim Bride <jim.bride@linux.intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Recently I have been applying an optimisation to avoid stalling and
clflushing GGTT objects based on their current binding. That is we only
set-to-gtt-domain upon first bind. However, on hibernation the objects
remain bound, but they are in the CPU domain. Currently (since commit
975f7ff4 ("drm/i915: Lazily migrate the objects after hibernation"))
we only flush scanout objects as all other objects are expected to be
flushed prior to use. That breaks down in the face of the runtime
optimisation above - and we need to flush all GGTT pinned objects
(essentially ringbuffers).

To reduce the burden of extra clflushes, we only flush those objects we
cannot discard from the GGTT. Everything pinned to the scanout, or
current contexts or ringbuffers will be flushed and rebound. Other
objects, such as inactive contexts, will be left unbound and in the CPU
domain until first use after resuming.

Fixes: 7abc98fa ("drm/i915: Only change the context object's domain...")
Fixes: 57e88531 ("drm/i915: Use VMA for ringbuffer tracking")
References: https://bugs.freedesktop.org/show_bug.cgi?id=94722Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: David Weinehall <david.weinehall@intel.com>
Reviewed-by: NMatthew Auld <matthew.auld@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909201957.2499-1-chris@chris-wilson.co.uk

In an attempt to keep the hibernation image as same as possible, let's
try and discard any unwanted pages and our own page arrays.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909190218.16831-1-chris@chris-wilson.co.uk

Now that we can wait upon fences before emitting the request, it becomes
trivial to wait upon any implicit fence provided by the dma-buf
reservation object.

To protect against failure, we force any asynchronous waits on a foreign
fence to timeout after 10s - so that a stall in another driver does not
permanently cripple ourselves. Still unpleasant though!

Testcase: igt/prime_vgem/fence-wait
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJohn Harrison <john.c.harrison@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-21-chris@chris-wilson.co.uk

Now that we have fences in place to drive request submission, we can
employ those to queue requests after their dependencies as opposed to
stalling in the middle of an execbuf ioctl. (However, we still choose to
spin before enabling the IRQ as that is faster - though contentious.)

v2: Do the fence ordering first, where we can still fail.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-20-chris@chris-wilson.co.uk

If we are waiting upon an external fence, from the pov of hangcheck the
engine is stuck on the last submitted seqno. Currently we give a small
increment to the hangcheck score in order to catch a stuck waiter /
driver. Now that we both have an independent wait hangcheck and may be
stuck waiting on an external fence, resetting the GPU has little effect
on that external fence. As we cannot advance by resetting, skip
incrementing the hangcheck score.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-19-chris@chris-wilson.co.uk

If we find a ring waiting on a semaphore for another assigned but not yet
emitted request, treat it as valid and waiting.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-18-chris@chris-wilson.co.uk

Currently the presumption is that the request construction and its
submission to the GuC are all under the same holding of struct_mutex. We
wish to relax this to separate the request construction and the later
submission to the GuC. This requires us to reserve some space in the
GuC command queue for the future submission. For flexibility to handle
out-of-order request submission we do not preallocate the next slot in
the GuC command queue during request construction, just ensuring that
there is enough space later.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-17-chris@chris-wilson.co.uk

We are about to specialize object synchronisation to enable nonblocking
execbuf submission. First we make a copy of the current object
synchronisation for execbuffer. The general i915_gem_object_sync() will
be removed following the removal of CS flips in the near future.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJohn Harrison <john.c.harrison@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-16-chris@chris-wilson.co.uk

Let's avoid mixing sealing the hardware commands for the request and
adding the request to the software tracking.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-15-chris@chris-wilson.co.uk

Drive final request submission from a callback from the fence. This way
the request is queued until all dependencies are resolved, at which
point it is handed to the backend for queueing to hardware. At this
point, no dependencies are set on the request, so the callback is
immediate.

A side-effect of imposing a heavier-irqsafe spinlock for execlist
submission is that we lose the softirq enabling after scheduling the
execlists tasklet. To compensate, we manually kickstart the softirq by
disabling and enabling the bh around the fence signaling.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJohn Harrison <john.c.harrison@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-14-chris@chris-wilson.co.uk

Update reset path in preparation for engine reset which requires
identification of incomplete requests and associated context and fixing
their state so that engine can resume correctly after reset.

The request that caused the hang will be skipped and head is reset to the
start of breadcrumb. This allows us to resume from where we left-off.
Since this request didn't complete normally we also need to cleanup elsp
queue manually. This is vital if we employ nonblocking request
submission where we may have a web of dependencies upon the hung request
and so advancing the seqno manually is no longer trivial.

ABI: gem_reset_stats / DRM_IOCTL_I915_GET_RESET_STATS

We change the way we count pending batches. Only the active context
involved in the reset is marked as either innocent or guilty, and not
mark the entire world as pending. By inspection this only affects
igt/gem_reset_stats (which assumes implementation details) and not
piglit.

ARB_robustness gives this guide on how we expect the user of this
interface to behave:

 * Provide a mechanism for an OpenGL application to learn about
   graphics resets that affect the context.  When a graphics reset
   occurs, the OpenGL context becomes unusable and the application
   must create a new context to continue operation. Detecting a
   graphics reset happens through an inexpensive query.

And with regards to the actual meaning of the reset values:

   Certain events can result in a reset of the GL context. Such a reset
   causes all context state to be lost. Recovery from such events
   requires recreation of all objects in the affected context. The
   current status of the graphics reset state is returned by

	enum GetGraphicsResetStatusARB();

   The symbolic constant returned indicates if the GL context has been
   in a reset state at any point since the last call to
   GetGraphicsResetStatusARB. NO_ERROR indicates that the GL context
   has not been in a reset state since the last call.
   GUILTY_CONTEXT_RESET_ARB indicates that a reset has been detected
   that is attributable to the current GL context.
   INNOCENT_CONTEXT_RESET_ARB indicates a reset has been detected that
   is not attributable to the current GL context.
   UNKNOWN_CONTEXT_RESET_ARB indicates a detected graphics reset whose
   cause is unknown.

The language here is explicit in that we must mark up the guilty batch,
but is loose enough for us to relax the innocent (i.e. pending)
accounting as only the active batches are involved with the reset.

In the future, we are looking towards single engine resetting (with
minimal locking), where it seems inappropriate to mark the entire world
as innocent since the reset occurred on a different engine. Reducing the
information available means we only have to encounter the pain once, and
also reduces the information leaking from one context to another.

v2: Legacy ringbuffer submission required a reset following hibernation,
or else we restore stale values to the RING_HEAD and walked over
stolen garbage.

v3: GuC requires replaying the requests after a reset.

v4: Restore engine IRQ after reset (so waiters will be woken!)
    Rearm hangcheck if resetting with a waiter.

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-13-chris@chris-wilson.co.uk

Since we have a cooperative mode now with a direct reset, we can avoid
the contention on struct_mutex and instead try then sleep on the
I915_RESET_IN_PROGRESS bit. If the mutex is held and that bit is
cleared, all is fine. Otherwise, we sleep for a bit and try again. In
the worst case we sleep for an extra second waiting for the mutex to be
released (no one touching the GPU is allowed the struct_mutex whilst the
I915_RESET_IN_PROGRESS bit is set). But when we have a direct reset,
this allows us to clean up the reset worker faster.

v2: Remember to call wake_up_bit() after changing (for the faster wakeup
as promised)
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-12-chris@chris-wilson.co.uk

If a waiter is holding the struct_mutex, then the reset worker cannot
reset the GPU until the waiter returns. We do not want to return -EAGAIN
form i915_wait_request as that breaks delicate operations like
i915_vma_unbind() which often cannot be restarted easily, and returning
-EIO is just as useless (and has in the past proven dangerous). The
remaining WARN_ON(i915_wait_request) serve as a valuable reminder that
handling errors from an indefinite wait are tricky.

We can keep the current semantic that knowing after a reset is complete,
so is the request, by performing the reset ourselves if we hold the
mutex.

uevent emission is still handled by the reset worker, so it may appear
slightly out of order with respect to the actual reset (and concurrent
use of the device).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-11-chris@chris-wilson.co.uk

In the next patch we want to handle reset directly by a locked waiter in
order to avoid issues with returning before the reset is handled. To
handle the reset, we must first know whether we hold the struct_mutex.
If we do not hold the struct_mtuex we can not perform the reset, but we do
not block the reset worker either (and so we can just continue to wait for
request completion) - otherwise we must relinquish the mutex.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-10-chris@chris-wilson.co.uk

We need finer control over wakeup behaviour during i915_wait_request(),
so expand the current bool interruptible to a bitmask.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-9-chris@chris-wilson.co.uk

Access to intel_init_emon() is strictly ordered by gt_powersave, using
struct_mutex around it is overkill (and will conflict with the caller
holding struct_mutex themselves).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-8-chris@chris-wilson.co.uk

In preparation for introducing a per-engine reset, we can first separate
the mixing of the reset state from the global reset counter.

The loss of atomicity in updating the reset state poses a small problem
for handling the waiters. For requests, this is solved by advancing the
seqno so that a waiter waking up after the reset knows the request is
complete. For pending flips, we still rely on the increment of the
global reset epoch (as well as the reset-in-progress flag) to signify
when the hardware was reset.

The advantage, now that we do not inspect the reset state during reset
itself i.e. we no longer emit requests during reset, is that we can use
the atomic updates of the state flags to ensure that only one reset
worker is active.

v2: Mika spotted that I transformed the i915_gem_wait_for_error() wakeup
into a waiter wakeup.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470414607-32453-6-git-send-email-arun.siluvery@linux.intel.comReviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-7-chris@chris-wilson.co.uk

Emulate HW to track and manage ELSP queue. A set of SW ports are defined
and requests are assigned to these ports before submitting them to HW. This
helps in cleaning up incomplete requests during reset recovery easier
especially after engine reset by decoupling elsp queue management. This
will become more clear in the next patch.

In the engine reset case we want to resume where we left-off after skipping
the incomplete batch which requires checking the elsp queue, removing
element and fixing elsp_submitted counts in some cases. Instead of directly
manipulating the elsp queue from reset path we can examine these ports, fix
up ringbuffer pointers using the incomplete request and restart submissions
again after reset.

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1470414607-32453-3-git-send-email-arun.siluvery@linux.intel.comReviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-6-chris@chris-wilson.co.uk

Just rearrange the code to reduce churn in the next patch.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-5-chris@chris-wilson.co.uk

Similar to the issue with reading from the context status buffer,
see commit 26720ab9 ("drm/i915: Move CSB MMIO reads out of the
execlists lock"), we frequently write to the ELSP register (4 writes per
interrupt) and know we hold the required spinlock and forcewake throughout.
We can further reduce the cost of writing these registers beyond the
I915_WRITE_FW() by precomputing the address of the ELSP register. We also
note that the subsequent read serves no purpose here, and are happy to
see it go.

v2: Address I915_WRITE mistakes in changelog

   text    data     bss     dec     hex filename
1259784    4581     576 1264941  134d2d drivers/gpu/drm/i915/i915.ko
1259720    4581     576 1264877  134ced drivers/gpu/drm/i915/i915.ko

Saves 64 bytes of address recomputation.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-4-chris@chris-wilson.co.uk

Rather than blindly assuming we need to advance the tail for
resubmitting the request via the ELSP, record the position.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-3-chris@chris-wilson.co.uk

Leave the more complicated request dequeueing to the tasklet and instead
just kick start the tasklet if we detect we are adding the first
request.

v2: Play around with list operators until we agree upon something
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-2-chris@chris-wilson.co.uk

This is really a core kernel struct in disguise until we can finally
place it in kernel/. There is an immediate need for a fence collection
mechanism that is more flexible than fence-array, in particular being
able to easily drive request submission via events (and not just
interrupt driven). The same mechanism would be useful for handling
nonblocking and asynchronous atomic modesets, parallel execution and
more, but for the time being just create a local sw fence for execbuf.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-1-chris@chris-wilson.co.uk

Moving all GPU features to the platform definition allows for
        - standard place when adding new features from new platform
        - possible to see supported features when dumping struct
          definitions
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Make the .hws_needs_physical the exception by switching the flag
on earlier platforms since they are fewer to support. Remove the flag on
later GPUs hardware since they all use GTT hws by default.

Switch the logic as well in the driver to reflect this change
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Moving all GPU features to the platform definition allows for
        - standard place when adding new features from new platforms
        - possible to see supported features when dumping struct
          definitions
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Moving all GPU features to the platform definition allows for
        - standard place when adding new features from new platforms
        - possible to see supported features when dumping struct
          definitions
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Moving all GPU features to the platform definition allows for
        - standard place when adding new features from new platforms
        - possible to see supported features when dumping struct
          definitions
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Moving all GPU features to the platform definition allows for
	- standard place when adding new features from new platforms
	- possible to see supported features when dumping struct
	  definitions
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Introducing a GEN2_FEATURES macro to simplify the struct definitions by
platforms given that most of the features are common. Inspired by the
GEN7_FEATURES macro done by Ben W. and others.

Use it for 830, 845g, i85x, i865g.

CC: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Introducing a GEN3_FEATURES macro to simplify the struct definitions by
platforms given that most of the features are common. Inspired by the
GEN7_FEATURES macro done by Ben W. and others.

Use it for i915g, i915gm, i945g, i945gm, g33 and pnv.

CC: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: NCarlos Santa <carlos.santa@intel.com>
Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>