In the future, we want to move a request between engines. To achieve
this, we first realise that we have two timelines in effect here. The
first runs through the GTT is required for ordering vma access, which is
tracked currently by engine. The second is implied by sequential
execution of commands inside the ringbuffer. This timeline is one that
maps to userspace's expectations when submitting requests (i.e. given the
same context, batch A is executed before batch B). As the rings's
timelines map to userspace and the GTT timeline an implementation
detail, move the timeline from the GTT into the ring itself (per-context
in logical-ring-contexts/execlists, or a global per-engine timeline for
the shared ringbuffers in legacy submission.

The two timelines are still assumed to be equivalent at the moment (no
migrating requests between engines yet) and so we can simply move from
one to the other without adding extra ordering.

v2: Reinforce that one isn't allowed to mix the engine execution
timeline with the client timeline from userspace (on the ring).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180502163839.3248-1-chris@chris-wilson.co.uk

We don't need to track every ring for its lifetime as they are managed
by the contexts/engines. What we do want to track are the live rings so
that we can sporadically clean up requests if userspace falls behind. We
can simply restrict the gt->rings list to being only gt->live_rings.

v2: s/live/active/ for consistency with gt.active_requests
Suggested-by: NTvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180430131503.5375-4-chris@chris-wilson.co.uk

In the next patch, rings are the central timeline as requests may jump
between engines. Therefore in the future as we retire in order along the
engine timeline, we may retire out-of-order within a ring (as the ring now
occurs along multiple engines), leading to much hilarity in miscomputing
the position of ring->head.

As an added bonus, retiring along the ring reduces the penalty of having
one execlists client do cleanup for another (old legacy submission
shares a ring between all clients). The downside is that slow and
irregular (off the critical path) process of cleaning up stale requests
after userspace becomes a modicum less efficient.

In the long run, it will become apparent that the ordered
ring->request_list matches the ring->timeline, a fun challenge for the
future will be unifying the two lists to avoid duplication!

v2: We need both engine-order and ring-order processing to maintain our
knowledge of where individual rings have completed upto as well as
knowing what was last executing on any engine. And finally by decoupling
retiring the contexts on the engine and the timelines along the rings,
we do have to keep a reference to the context on each request
(previously it was guaranteed by the context being pinned).

v3: Not just a reference to the context, but we need to keep it pinned
as we manipulate the rings; i.e. we need a pin for both the manipulation
of the engine state during its retirements, and a separate pin for the
manipulation of the ring state.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180430131503.5375-3-chris@chris-wilson.co.uk

Make life easier in upcoming patches by moving the context_pin and
context_unpin vfuncs into inline helpers.

v2: Fixup mock_engine to mark the context as pinned on use.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180430131503.5375-2-chris@chris-wilson.co.uk

In commit 9b6586ae ("drm/i915: Keep a global seqno per-engine"), we
moved from a global inflight counter to per-engine counters in the
hope that will be easy to run concurrently in future. However, with the
advent of the desire to move requests between engines, we do need a
global counter to preserve the semantics that no engine wraps in the
middle of a submit. (Although this semantic is now only required for gen7
semaphore support, which only supports greater-then comparisons!)

v2: Keep a global counter of all requests ever submitted and force the
reset when it wraps.

References: 9b6586ae ("drm/i915: Keep a global seqno per-engine")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180430131503.5375-1-chris@chris-wilson.co.uk

Today we only want to pass along the priority to engine->schedule(), but
in the future we want to have much more control over the various aspects
of the GPU during a context's execution, for example controlling the
frequency allowed. As we need an ever growing number of parameters for
scheduling, move those into a struct for convenience.

v2: Move the anonymous struct into its own function for legibility and
ye olde gcc.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180418184052.7129-3-chris@chris-wilson.co.uk

Having moved the priotree struct into i915_scheduler.h, identify it as
the scheduling element and rebrand into i915_sched. This becomes more
useful as we start attaching more information we require to propagate
through the scheduler.

v2: Use i915_sched_node for future distinctiveness
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180418184052.7129-2-chris@chris-wilson.co.uk

Include fence context and seqno in low level tracing so it is easier to
follow flows of individual requests when things go bad.

Also added tracing on the reset side of things.

v2:
 Chris Wilson:
 * Standardize global_seqno and seqno as global.
 * Include current hws seqno in execlists_cancel_port_requests.

v3:
 * Fix port printk format for all builds.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v2
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180406123514.5809-1-tvrtko.ursulin@linux.intel.com

Currently, we rely on inspecting the hangcheck state from within the
i915_reset() routines to determine which engines were guilty of the
hang. This is problematic for cases where we want to run
i915_handle_error() and call i915_reset() independently of hangcheck.
Instead of relying on the indirect parameter passing, turn it into an
explicit parameter providing the set of stalled engines which then are
treated as guilty until proven innocent.

While we are removing the implicit stalled parameter, also make the
reason into an explicit parameter to i915_reset(). We still need a
back-channel for i915_handle_error() to hand over the task to the locked
waiter, but let's keep that its own channel rather than incriminate
another.

This leaves stalled/seqno as being private to hangcheck, with no more
nefarious snooping by reset, be it whole-device or per-engine. \o/

The only real issue now is that this makes it crystal clear that we
don't actually do any testing of hangcheck per se in
drv_selftest/live_hangcheck, merely of resets!
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NMichel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180406220354.18911-2-chris@chris-wilson.co.uk

We will want to park GEM before disengaging the drive^W^W^W unwedging.
Since we already do the work for idling, expose the guts as a new
function that we can then reuse.

v2: Just skip if already parked; makes it more forgiving to use by
future callers.
v3: Extract mark_busy, rename it to i915_gem_unpark and place it next to
i915_gem_park so that we can evaluate it for symmetry more easily.
Calling GEM from inside i915_request looks to be a bit of a layering
violation, for the moment I am imaging them as being notify_cb.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> #v1
Reviewed-by: NMichal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180406155144.27791-1-chris@chris-wilson.co.uk

When we include a request's global_seqno in a GEM_TRACE it often helps
to know how that relates to the current breadcrumb as seen by the
hardware.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180327210157.16896-3-chris@chris-wilson.co.ukReviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>

Both request_submit and request_unsubmit deal with transferring the
request from the client's timeline onto the execution timeline and back
again. As both functions deal with a pair of timeline's, using a
shorthand for just one of them is slightly confusing, especially as the
different functions use the shorthand for the alternate timeline.
Instead, use the full version of each timeline so it should be easier to
keep track of the transfer between the request/client and the engine.

v2: Refactor the common lock+list_move
v3: Be clear we require the other timeline list to be locked as well.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180322131034.6036-1-chris@chris-wilson.co.uk

We pre-increment the timeline->seqno when handing it to the request,
make sure the GEM_TRACE takes this into account. Otherwise, it appears
that we go backwards over a preemption point:

1d..1 157681077us : __i915_request_unsubmit: vcs0 fence 75e:3 <- global_seqno 17
0d.s1 157681113us : __i915_request_submit: vcs0 fence 75e:3 -> global_seqno 16

Fixes: d9b13c4d ("drm/i915: Trace GEM steps between submit and wedging")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180322110059.4467-1-chris@chris-wilson.co.uk

Not all callers want the GPU error to handled in the same way, so expose
a control parameter. In the first instance, some callers do not want the
heavyweight error capture so add a bit to request the state to be
captured and saved.

v2: Pass msg down to i915_reset/i915_reset_engine so that we include the
reason for the reset in the dev_notice(), superseding the earlier option
to not print that notice.
v3: Stash the reason inside the i915->gpu_error to handover to the direct
reset from the blocking waiter.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: NMichel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180320100449.1360-2-chris@chris-wilson.co.uk

Similar to the staging around handling of engine->submit_request, we
need to stop adding to the execlists->queue prior to calling
engine->cancel_requests. cancel_requests will move requests from the
queue onto the timeline, so if we add a request onto the queue after that
point, it will be lost.

Fixes: af7a8ffa ("drm/i915: Use rcu instead of stop_machine in set_wedged")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180307134226.25492-5-chris@chris-wilson.co.uk
(cherry picked from commit 47650db0)
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

We still have an odd race with wedging/unwedging as shown by igt/gem_eio
that defies expectations. Add some more trace_printks to try and
visualize the flow over the precipice.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180315131451.4060-1-chris@chris-wilson.co.uk

There is some redundancy between dma_fence->ops->enable_signaling (via
i915_fence_enable_signaling) and our backend,
intel_engine_enable_signaling() in that both levels recheck the fence
status multiple times. If we convert intel_engine_enable_signaling() to
return the information desired by dma_fence->ops->enable_signaling, we
can reduce i915_fence_enable_signaling to a simple stub and avoid
trying to reinterpret the same information.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180308140732.25090-1-chris@chris-wilson.co.uk

Similar to the staging around handling of engine->submit_request, we
need to stop adding to the execlists->queue prior to calling
engine->cancel_requests. cancel_requests will move requests from the
queue onto the timeline, so if we add a request onto the queue after that
point, it will be lost.

Fixes: af7a8ffa ("drm/i915: Use rcu instead of stop_machine in set_wedged")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180307134226.25492-5-chris@chris-wilson.co.uk

When wedged, we do not update the ring->tail as we submit the requests
causing us to leak the ring->space upon cleaning up the wedged driver.
We can just use the value stored in rq->tail, and keep the submission
backend details away from set-wedge.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180307134226.25492-3-chris@chris-wilson.co.uk

Previously, we would spin waiting for all waiters to wake up and notice
their request had completed before we would reset the seqno upon
wraparound.  However, we can mark their waits as complete and wake them
up directly using the existing machinery for handling the flushing of
missed wakeups when idling.
Suggested-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180306130143.13312-2-chris@chris-wilson.co.uk

Since commit fd10e2ce ("drm/i915/breadcrumbs: Ignore unsubmitted
signalers"), we cancel the signaler when retiring the request and so
upon wraparound, where we wait for all requests to be retired, we no
longer need to spin waiting for the signaling thread to release its
references to the in-flight requests, and so we can assert that the
signaler is idle.

References: fd10e2ce ("drm/i915/breadcrumbs: Ignore unsubmitted signalers")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180306130143.13312-1-chris@chris-wilson.co.uk

Mostly doc/print messages that were not updated after commit e61e0f51
("drm/i915: Rename drm_i915_gem_request to i915_request").
Signed-off-by: NMichel Thierry <michel.thierry@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180222172405.11386-1-michel.thierry@intel.com

We want to de-emphasize the link between the request (dependency,
execution and fence tracking) from GEM and so rename the struct from
drm_i915_gem_request to i915_request. That is we may implement the GEM
user interface on top of requests, but they are an abstraction for
tracking execution rather than an implementation detail of GEM. (Since
they are not tied to HW, we keep the i915 prefix as opposed to intel.)

In short, the spatch:
@@

@@
- struct drm_i915_gem_request
+ struct i915_request

A corollary to contracting the type name, we also harmonise on using
'rq' shorthand for local variables where space if of the essence and
repetition makes 'request' unwieldy. For globals and struct members,
'request' is still much preferred for its clarity.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180221095636.6649-1-chris@chris-wilson.co.ukReviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NMichał Winiarski <michal.winiarski@intel.com>
Acked-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Commit fe49789f ("drm/i915: Deconstruct execute fence") re-arranged
the code and moved the i915_gem_request_execute tracepoint to before the
global seqno is assigned to the request.

We need to move the tracepoint a bit later so this information is once
again available.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: fe49789f ("drm/i915: Deconstruct execute fence")
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: intel-gfx@lists.freedesktop.org
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180220104742.565-1-tvrtko.ursulin@linux.intel.com

drivers/gpu/drm/i915/i915_gem_request.c:941: warning: No description found for parameter 'write'
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NLionel Landwerlin <lionel.g.landwerlin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180208111453.32567-1-chris@chris-wilson.co.uk

We only need to wake up the RPS worker once when initially enabling the
client boost, it remains in effect then until the last client no longer
requires the boost.

References: https://bugs.freedesktop.org/show_bug.cgi?id=102250
References: 7b92c1bd ("drm/i915: Avoid keeping waitboost active for signaling threads")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: NMichał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206143137.15509-1-chris@chris-wilson.co.uk

If we submit a request and see that the previous request on this
timeline was already signaled, we first do not need to add the
dependency tracker for that completed request and secondly we know that
we there is then a large backlog in retiring requests affecting this
timeline. Given that we just submitted more work to the HW, now would be
a good time to catch up on those retirements.

v2: Try to sum up the compromises involved in flushing the retirement
queue after submission.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180207084350.3929-1-chris@chris-wilson.co.uk

If the last request on the timeline is already complete, we do not need
to emit the serialisation barriers.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180207084350.3929-2-chris@chris-wilson.co.uk

If we remember to cancel the signaler on a request when retiring it
(after we know that the request has been signaled), we do not need to
carry an additional request in the signaler itself. This prevents an
issue whereby the signaler threads may be delayed and hold on to
thousands of request references, causing severe memory fragmentation and
premature oom (most noticeable on 32b snb due to the limited GFP_KERNEL
and frequent use of inter-engine fences).

v2: Rename first_signal(), document reads outside of locks.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180203101914.24880-1-chris@chris-wilson.co.ukReviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>

Previously, we relied on only running the hangcheck while somebody was
waiting on the GPU, in order to minimise the amount of time hangcheck
had to run. (If nobody was watching the GPU, nobody would notice if the
GPU wasn't responding -- eventually somebody would care and so kick
hangcheck into action.) However, this falls apart from around commit
4680816b ("drm/i915: Wait first for submission, before waiting for
request completion"), as not all waiters declare themselves to hangcheck
and so we could switch off hangcheck and miss GPU hangs even when
waiting under the struct_mutex.

If we enable hangcheck from the first request submission, and let it run
until the GPU is idle again, we forgo all the complexity involved with
only enabling around waiters. We just have to remember to be careful that
we do not declare a GPU hang when idly waiting for the next request to
be come ready, as we will run hangcheck continuously even when the
engines are stalled waiting for external events. This should be true
already as we should only be tracking requests submitted to hardware for
execution as an indicator that the engine is busy.

Fixes: 4680816b ("drm/i915: Wait first for submission, before waiting for request completion"
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104840Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180129144104.3921-1-chris@chris-wilson.co.ukReviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
(cherry picked from commit 88923048)
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Previously, we relied on only running the hangcheck while somebody was
waiting on the GPU, in order to minimise the amount of time hangcheck
had to run. (If nobody was watching the GPU, nobody would notice if the
GPU wasn't responding -- eventually somebody would care and so kick
hangcheck into action.) However, this falls apart from around commit
4680816b ("drm/i915: Wait first for submission, before waiting for
request completion"), as not all waiters declare themselves to hangcheck
and so we could switch off hangcheck and miss GPU hangs even when
waiting under the struct_mutex.

If we enable hangcheck from the first request submission, and let it run
until the GPU is idle again, we forgo all the complexity involved with
only enabling around waiters. We just have to remember to be careful that
we do not declare a GPU hang when idly waiting for the next request to
be come ready, as we will run hangcheck continuously even when the
engines are stalled waiting for external events. This should be true
already as we should only be tracking requests submitted to hardware for
execution as an indicator that the engine is busy.

Fixes: 4680816b ("drm/i915: Wait first for submission, before waiting for request completion"
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104840Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180129144104.3921-1-chris@chris-wilson.co.ukReviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>

Assert that we do not try to unsubmit a completed request, as should we
try to resubmit it later, the ring is already past the request's
breadcrumb and the breadcrumb will not be updated.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180129094912.14428-1-chris@chris-wilson.co.ukReviewed-by: NMichał Winiarski <michal.winiarski@intel.com>

By counting the number of times we have woken up, we have a very simple
means of defining an epoch, which will come in handy if we want to
perform deferred tasks at the end of an epoch (i.e. while we are going
to sleep) without imposing on the next activity cycle.

v2: No reason to specify precise number of bits here.
v3: Take Tvrtko's advice and reserve 0 as an invalid epoch.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180124113608.14909-1-chris@chris-wilson.co.uk

If we fail to allocate a new request, make sure we recover the pages
that are in the process of being freed by inserting an RCU barrier.

v2: Comment before the shrink and barrier in the error path.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180119144657.22606-1-chris@chris-wilson.co.uk

Back up our comment that all signalers should have been signaled before
we ourselves were retired with an assert to that effect.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NMichał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180102151235.3949-9-chris@chris-wilson.co.uk

We should only attempt to remove requests from the execution queue that
are on the execution queue. These are the requests that have been
assigned a global_seqno, so we can assert that we only attempt to remove
requests with a nonzero global_seqno. Afterwards we assert that we
remove them in order, i.e. the global_seqno matches the engine's seqno,
but that leaves a small loophole for an unattached request on an unused
engine.

We can then make the same assertion on queuing the request to the
execution engine, it must have a zero global_seqno or else we are queuing
the same request twice.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171222141959.3006-1-chris@chris-wilson.co.ukReviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>

If we fail to allocate a request, we can reap the outstanding requests
and push them to the request's slab's freelist before trying again. This
forces us to ratelimit malicious clients that tie up all of the system
resources in requests, instead of causing a system-wide oom.

Testcase: igt/gem_shrink/execbuf1
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171212180652.22061-3-chris@chris-wilson.co.uk

If a fence allocation fails in a blocking context, we will sleep on the
fence as a last resort. We can therefore allow ourselves to fail and
sleep on the fence instead of triggering a system-wide oom. This allows
us to throttle malicious clients that are consuming lots of system
resources by capping the amount of memory used by fences.

Testcase: igt/gem_shrink/execbufX
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171212180652.22061-2-chris@chris-wilson.co.uk

It seems that the DMC likes to transition between the DC states a lot when
there are no connected displays (no active power domains) during command
submission.

This activity on DC states has a negative impact on the performance of the
chip with huge latencies observed in the interrupt handlers and elsewhere.
Simple tests like igt/gem_latency -n 0 are slowed down by a factor of
eight.

Work around it by introducing a new power domain named,
POWER_DOMAIN_GT_IRQ, associtated with the "DC off" power well, which is
held for the duration of command submission activity.

CNL has the same problem which will be addressed as a follow-up. Doing
that requires a fix for a DC6 context corruption problem in the CNL DMC
firmware which is yet to be released.

v2:
 * Add commit text as comment in i915_gem_mark_busy. (Chris Wilson)
 * Protect macro body with braces. (Jani Nikula)

v3:
 * Add dedicated power domain for clarity. (Chris, Imre)
 * Commit message and comment text updates.
 * Apply to all big-core GEN9 parts apart for Skylake which is pending DMC
   firmware release.

v4:
 * Power domain should be inner to device runtime pm. (Chris)
 * Simplify NEEDS_CSR_GT_PERF_WA macro. (Chris)
 * Handle async DMC loading by moving the GT_IRQ power domain logic into
   intel_runtime_pm. (Daniel, Chris)
 * Include small core GEN9 as well. (Imre)

v5
 * Special handling for async DMC load is not needed since on failure the
   power domain reference is kept permanently taken. (Imre)

v6:
 * Drop the NEEDS_CSR_GT_PERF_WA macro since all firmwares have now been
   deployed. (Imre, Chris)
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100572
Testcase: igt/gem_exec_nop/headless
Cc: Imre Deak <imre.deak@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk> (v2)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> (v5)
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
[Imre: Add note about applying the WA on CNL as a follow-up]
Signed-off-by: NImre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171205132854.26380-1-tvrtko.ursulin@linux.intel.com

The legacy context switch for ringbuffer submission is multistaged,
where each of those stages may fail. However, we were updating global
state after some stages, and so we had to force the incomplete request
to be submitted because we could not unwind. Save the global state
before performing the switches, and so enable us to unwind back to the
previous global state should any phase fail. We then must cancel the
request instead of submitting it should the construction fail.

v2: s/saved_ctx/from_ctx/; s/ctx/to_ctx/ etc.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171123152631.31385-1-chris@chris-wilson.co.uk