| Age | Commit message (Collapse) | Author |
|---|
|
Both microcontrollers are part of the GT HW and are closely related to
GT operations. To keep all the files cleanly together, they've been
placed in their own subdir inside the gt/ folder
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-6-chris@chris-wilson.co.uk
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Having taken the first step in encapsulating the functionality by moving
the related files under gt/, the next step is to start encapsulating by
passing around the relevant structs rather than the global
drm_i915_private. In this step, we pass intel_gt to intel_reset.c
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190712192953.9187-1-chris@chris-wilson.co.uk
|
|
drivers/gpu/drm/i915/intel_guc_submission.c:799: warning: Excess function parameter 'ctx' description in 'guc_client_alloc'
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190711162415.2938-1-chris@chris-wilson.co.uk
|
|
We originally added support, in some cases partial, for different modes
of operations via guc clients:
- proxy vs direct submission;
- variable engine mask per-client.
We only ever used one flow (all submissions via a single proxy), so the
other code paths haven't been exercised and are most likely
non-functional. The guc firmware interface is also in the process of
being updated to better fit the i915 flow and our client abstraction
will need to change accordingly (or possibly go away entirely), so these
old unused paths can be considered dead and removed.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Acked-by: Matthew Brost <Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190710005437.3496-3-daniele.ceraolospurio@intel.com
|
|
Preemption via GuC submission is not being supported with its current
legacy incarnation. The current FW does support a similar pre-emption
flow via H2G, but it is class-based instead of being instance-based,
which doesn't fit well with the i915 tracking. To fix this, the
firmware is being updated to better support our needs with a new flow,
so we can safely remove the old code.
v2 (Daniele): resurrect & rebase, reword commit message, remove
preempt_context as well
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Acked-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190710005437.3496-2-daniele.ceraolospurio@intel.com
|
|
We want to set this flag in the next commit on requests containing
perf queries so that the result of the perf query can just be a delta
of global counters, rather than doing post processing of the OA
buffer.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
[ickle: add basic selftest for nopreempt]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190709164227.25859-1-chris@chris-wilson.co.uk
|
|
When using a global seqno, we required a precise stop-the-workd event to
handle preemption and unwind the global seqno counter. To accomplish
this, we would preempt to a special out-of-band context and wait for the
machine to report that it was idle. Given an idle machine, we could very
precisely see which requests had completed and which we needed to feed
back into the run queue.
However, now that we have scrapped the global seqno, we no longer need
to precisely unwind the global counter and only track requests by their
per-context seqno. This allows us to loosely unwind inflight requests
while scheduling a preemption, with the enormous caveat that the
requests we put back on the run queue are still _inflight_ (until the
preemption request is complete). This makes request tracking much more
messy, as at any point then we can see a completed request that we
believe is not currently scheduled for execution. We also have to be
careful not to rewind RING_TAIL past RING_HEAD on preempting to the
running context, and for this we use a semaphore to prevent completion
of the request before continuing.
To accomplish this feat, we change how we track requests scheduled to
the HW. Instead of appending our requests onto a single list as we
submit, we track each submission to ELSP as its own block. Then upon
receiving the CS preemption event, we promote the pending block to the
inflight block (discarding what was previously being tracked). As normal
CS completion events arrive, we then remove stale entries from the
inflight tracker.
v2: Be a tinge paranoid and ensure we flush the write into the HWS page
for the GPU semaphore to pick in a timely fashion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190620142052.19311-1-chris@chris-wilson.co.uk
|
|
To continue the onslaught of removing the assumption of a global
execution ordering, another casualty is the engine->timeline. Without an
actual timeline to track, it is overkill and we can replace it with a
much less grand plain list. We still need a list of requests inflight,
for the simple purpose of finding inflight requests (for retiring,
resetting, preemption etc).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190614164606.15633-3-chris@chris-wilson.co.uk
|
|
Only a few call sites remain which have been converted to uncore mmio
accessors and so the macro can be removed.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190611104548.30545-2-tvrtko.ursulin@linux.intel.com
|
|
Remove a couple dev_priv locals as a consequence.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190607084521.16845-1-tvrtko.ursulin@linux.intel.com
|
|
Continuing the theme of separating out the GEM clutter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190528092956.14910-8-chris@chris-wilson.co.uk
|
|
With newer GuC firmware it is always ok to ask GuC to update power
domain states. Make it an unconditional initialization step.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Spotswood <john.a.spotswood@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: John Spotswood <john.a.spotswood@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190527183613.17076-6-michal.wajdeczko@intel.com
|
|
This function just check our software flag, while 'is_alive'
may suggest that we are checking runtime firmware status.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190522193203.23932-5-michal.wajdeczko@intel.com
|
|
Currently there is an underlying assumption that i915_request_unsubmit()
is synchronous wrt the GPU -- that is the request is no longer in flight
as we remove it. In the near future that may change, and this may upset
our signaling as we can process an interrupt for that request while it
is no longer in flight.
CPU0 CPU1
intel_engine_breadcrumbs_irq
(queue request completion)
i915_request_cancel_signaling
... ...
i915_request_enable_signaling
dma_fence_signal
Hence in the time it took us to drop the lock to signal the request, a
preemption event may have occurred and re-queued the request. In the
process, that request would have seen I915_FENCE_FLAG_SIGNAL clear and
so reused the rq->signal_link that was in use on CPU0, leading to bad
pointer chasing in intel_engine_breadcrumbs_irq.
A related issue was that if someone started listening for a signal on a
completed but no longer in-flight request, we missed the opportunity to
immediately signal that request.
Furthermore, as intel_contexts may be immediately released during
request retirement, in order to be entirely sure that
intel_engine_breadcrumbs_irq may no longer dereference the intel_context
(ce->signals and ce->signal_link), we must wait for irq spinlock.
In order to prevent the race, we use a bit in the fence.flags to signal
the transfer onto the signal list inside intel_engine_breadcrumbs_irq.
For simplicity, we use the DMA_FENCE_FLAG_SIGNALED_BIT as it then
quickly signals to any outside observer that the fence is indeed signaled.
v2: Sketch out potential dma-fence API for manual signaling
v3: And the test_and_set_bit()
Fixes: 52c0fdb25c7c ("drm/i915: Replace global breadcrumbs with per-context interrupt tracking")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190508112452.18942-1-chris@chris-wilson.co.uk
|
|
If we couple the scheduler more tightly with the execlists policy, we
can apply the preemption policy to the question of whether we need to
kick the tasklet at all for this priority bump.
v2: Rephrase it as a core i915 policy and not an execlists foible.
v3: Pull the kick together.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190507122544.12698-1-chris@chris-wilson.co.uk
|
|
Tidy up the cleanup sequence by always ensure that the tasklet is
flushed on parking (before we cleanup). The parking provides a
convenient point to ensure that the backend is truly idle.
v2: Do the full check for idleness before parking, to be sure we flush
any residual interrupt.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190503080942.30151-1-chris@chris-wilson.co.uk
|
|
We switched to a tree of per-engine HW context to accommodate the
introduction of virtual engines. However, we plan to also support
multiple instances of the same engine within the GEM context, defeating
our use of the engine as a key to looking up the HW context. Just
allocate a logical per-engine instance and always use an index into the
ctx->engines[]. Later on, this ctx->engines[] may be replaced by a user
specified map.
v2: Add for_each_gem_engine() helper to iterator within the engines lock
v3: intel_context_create_request() helper
v4: s/unsigned long/unsigned int/ 4 billion engines is quite enough.
v5: Push iterator locking to caller
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190426163336.15906-7-chris@chris-wilson.co.uk
|
|
Start partitioning off the code that talks to the hardware (GT) from the
uapi layers and move the device facing code under gt/
One casualty is s/intel_ringbuffer.h/intel_engine.h/ with the plan to
subdivide that header and body further (and split out the submission
code from the ringbuffer and logical context handling). This patch aims
to be simple motion so git can fixup inflight patches with little mess.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190424174839.7141-1-chris@chris-wilson.co.uk
|
|
Before causing guc and execlists to diverge further (breaking guc in the
process), take a copy of the current reset procedure and make it local to
the guc submission backend
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190411130515.20716-1-chris@chris-wilson.co.uk
|
|
Circumvent the dance we currently perform to find the preempt_client and
lookup its HW context for this engine, as we know we have already pinned
the preempt_context on the engine.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190408091728.20207-15-chris@chris-wilson.co.uk
|
|
Replace the WARN with a simple if() + error message to squech the sparse
warning that entire wait_for() macro was being stringified:
drivers/gpu/drm/i915/intel_guc_submission.c:658:9: error: too long token expansion
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190408091728.20207-2-chris@chris-wilson.co.uk
|
|
Use the engine->flags to store whether we want to kick the submission
tasklet on receipt of a breadcrumb interrupt, so that this decision can
be made by the submission backend and not dependent on a limited feature
test within the interrupt handler. This should make it easier to adapt to
different submission backends.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Xiaolin Zhang <xiaolin.zhang@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190329154912.13781-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
If we use the STORE_DATA_INDEX function we can use a fixed offset and
avoid having to lookup up the engine HWS address. A step closer to being
able to emit the final breadcrumb during request_add rather than later
in the submission interrupt handler.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190318095204.9913-9-chris@chris-wilson.co.uk
|
|
In preparation for an ever growing number of engines and so ever
increasing static array of HW contexts within the GEM context, move the
array over to an rbtree, allocated upon first use.
Unfortunately, this imposes an rbtree lookup at a few frequent callsites,
but we should be able to mitigate those by moving over to using the HW
context as our primary type and so only incur the lookup on the boundary
with the user GEM context and engines.
v2: Check for no HW context in guc_stage_desc_init
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190308132522.21573-4-chris@chris-wilson.co.uk
|
|
In the next patch, we are introducing a broad virtual engine to encompass
multiple physical engines, losing the 1:1 nature of BIT(engine->id). To
reflect the broader set of engines implied by the virtual instance, lets
store the full bitmask.
v2: Use intel_engine_mask_t (s/ring_mask/engine_mask/)
v3: Tvrtko voted for moah churn so teach everyone to not mention ring
and use $class$instance throughout.
v4: Comment upon the disparity in bspec for using VCS1,VCS2 in gen8 and
VCS[0-4] in later gen. We opt to keep the code consistent and use
0-index naming throughout.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190305180332.30900-1-chris@chris-wilson.co.uk
|
|
WAIT is occasionally suppressed by virtue of preempted requests being
promoted to NEWCLIENT if they have not all ready received that boost.
Make this consistent for all WAIT boosts that they are not allowed to
preempt executing contexts and are merely granted the right to be at the
front of the queue for the next execution slot. This is in keeping with
the desire that the WAIT boost be a minor tweak that does not give
excessive promotion to its user and open ourselves to trivial abuse.
The problem with the inconsistent WAIT preemption becomes more apparent
as the preemption is propagated across the engines, where one engine may
preempt and the other not, and we be relying on the exact execution
order being consistent across engines (e.g. using HW semaphores to
coordinate parallel execution).
v2: Also protect GuC submission from false preemption loops.
v3: Build bug safeguards and better debug messages for st.
v4: Do the priority bumping in unsubmit (i.e. on preemption/reset
unwind), applying it earlier during submit causes out-of-order execution
combined with execute fences.
v5: Call sw_fence_fini for our dummy request (Matthew)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190228220639.3173-1-chris@chris-wilson.co.uk
|
|
As kmem_caches share the same properties (size, allocation/free behaviour)
for all potential devices, we can use global caches. While this
potential has worse fragmentation behaviour (one can argue that
different devices would have different activity lifetimes, but you can
also argue that activity is temporal across the system) it is the
default behaviour of the system at large to amalgamate matching caches.
The benefit for us is much reduced pointer dancing along the frequent
allocation paths.
v2: Defer shrinking until after a global grace period for futureproofing
multiple consumers of the slab caches, similar to the current strategy
for avoiding shrinking too early.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190228102035.5857-1-chris@chris-wilson.co.uk
|
|
Having weaned the interrupt handling off using a single global execution
queue, we no longer need to emit a global_seqno. Note that we still have
a few assumptions about execution order along engine timelines, but this
removes the most obvious artefact!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190226094922.31617-3-chris@chris-wilson.co.uk
|
|
In order to avoid preempting ourselves, we currently refuse to schedule
the tasklet if we reschedule an inflight context. However, this glosses
over a few issues such as what happens after a CS completion event and
we then preempt the newly executing context with itself, or if something
else causes a tasklet_schedule triggering the same evaluation to
preempt the active context with itself.
However, when we avoid preempting ELSP[0], we still retain the preemption
value as it may match a second preemption request within the same time period
that we need to resolve after the next CS event. However, since we only
store the maximum preemption priority seen, it may not match the
subsequent event and so we should double check whether or not we
actually do need to trigger a preempt-to-idle by comparing the top
priorities from each queue. Later, this gives us a hook for finer
control over deciding whether the preempt-to-idle is justified.
The sequence of events where we end up preempting for no avail is:
1. Queue requests/contexts A, B
2. Priority boost A; no preemption as it is executing, but keep hint
3. After CS switch, B is less than hint, force preempt-to-idle
4. Resubmit B after idling
v2: We can simplify a bunch of tests based on the knowledge that PI will
ensure that earlier requests along the same context will have the highest
priority.
v3: Demonstrate the stale preemption hint with a selftest
References: a2bf92e8cc16 ("drm/i915/execlists: Avoid kicking priority on the current context")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190129185452.20989-4-chris@chris-wilson.co.uk
|
|
After noticing that we trigger preemption events for currently executing
requests, as well as requests that complete before the preemption and
attempting to suppress those preemption events, it is wise to not
consider the queue_priority to be authoritative. As we only track the
maximum priority seen between dequeue passes, if the maximum priority
request is no longer available for dequeuing (it completed or is even
executing on another engine), we have no knowledge of the previous
queue_priority as it would require us to keep a full history of enqueued
requests -- but we already have that history in the priolists!
Rename the queue_priority to queue_priority_hint so that we do not
confuse it as being exactly the maximum priority in the queue, but merely
an indication that we have seen a new maximum priority value and as such
we should check whether it should preempt the currently running request.
v2: s/preempt_priority_hint/queue_priority_hint/ as preempt implies it
being only used for the singular task of preemption and not the wider
question of waking up due to a change in the queue.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190129185452.20989-3-chris@chris-wilson.co.uk
|
|
Currently we only allocate an object and vma if we are using a GGTT
virtual HWSP, and a plain struct page for a physical HWSP. For
convenience later on with global timelines, it will be useful to always
have the status page being tracked by a struct i915_vma. Make it so.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-4-chris@chris-wilson.co.uk
|
|
Now that the submission backends are controlled via their own spinlocks,
with a wave of a magic wand we can lift the struct_mutex requirement
around GPU reset. That is we allow the submission frontend (userspace)
to keep on submitting while we process the GPU reset as we can suspend
the backend independently.
The major change is around the backoff/handoff strategy for performing
the reset. With no mutex deadlock, we no longer have to coordinate with
any waiter, and just perform the reset immediately.
Testcase: igt/gem_mmap_gtt/hang # regresses
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190125132230.22221-3-chris@chris-wilson.co.uk
|
|
Simplify by using sizeof(u32) to convert from the index inside the HWSP
to the byte offset. This has the advantage of not only being shorter
(and so not upsetting checkpatch!) but that it matches use where we are
writing to byte addresses using other commands than MI_STORE_DWORD_IMM.
v2: Drop the now superfluous MI_STORE_DWORD_INDEX_SHIFT, it appears to
be a local invention so keeping it after the final use does not help to
clarify the GPU instruction.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190125120005.25191-2-chris@chris-wilson.co.uk
|
|
In preparation for removing the manual EMIT_FLUSH prior to emitting the
breadcrumb implement the flush inline with writing the breadcrumb for
execlists. Using one command to both flush and write the breadcrumb is
naturally a tiny bit faster than splitting it into two.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20181228153114.4948-1-chris@chris-wilson.co.uk
|
|
A collection of very small cleanups/improvements around doorbell checking
that do not deserve their own patch:
- Move doorbell-related HW defs to intel_guc_reg.h
- use GUC_NUM_DOORBELLS instead of GUC_DOORBELL_INVALID where
appropriate
- do not stop on error in guc_verify_doorbells
- do not print drbreg on error: the only content of the register
apart from the valid bit is the lower part of the physical memory
address, which we can't use even if valid because we don't know
which descriptor it came from (since the doorbell is in an unexpected
state)
- Move the checking of doorbell valid bit to a common helper.
v2: add more cleanups (move defs, use GUC_NUM_DOORBELLS, don't stop in
guc_verify_doorbells) (Michal)
v3: move more things to intel_guc_reg, redefine
GUC_DOORBELL_INVALID (Michal), drop guc_doorbell_qw since it just
duplicates guc_doorbell_info
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20181022230427.5616-3-daniele.ceraolospurio@intel.com
|
|
Cacheline selection is only needed if we actually manage to reserve a
doorbell.
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20181022230427.5616-2-daniele.ceraolospurio@intel.com
|
|
The 2 functions don't create or destroy anything, they just update the
doorbell state in memory. Use init and fini instead for clarity.
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20181022230427.5616-1-daniele.ceraolospurio@intel.com
|
|
If the HW has not processed the db invalidation request yet, clearing
the cookie can generate a db ring. We clear the cookie when we
(re-)allocate the doorbell so no need to do it on destroy as well as no
one is going to look at it while the doorbell is inactive
v2: fix typo in patch title (Michal)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20181002215430.15049-2-daniele.ceraolospurio@intel.com
|
|
GuC stores some data in there, which might be stale after a reset.
We already reset the WQ head and tail, but more things are being moved
to the descriptor with the interface updates. Instead of trying to track
them one by one, always memset and init the descriptors from scratch
after GuC is loaded.
The code is also reorganized so that the above operations and the
doorbell creation are grouped as "client enabling"
v2: add proc_desc_fini for symmetry (Daniele), remove unneeded var init,
add guc_is_alive() (Michal)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20181002215430.15049-1-daniele.ceraolospurio@intel.com
|
|
As we are about to allow ourselves to slightly bump the user priority
into a few different sublevels, packthose internal priority lists
into the same i915_priolist to keep the rbtree compact and avoid having
to allocate the default user priority even after the internal bumping.
The downside to having an requests[] rather than a node per active list,
is that we then have to walk over the empty higher priority lists. To
compensate, we track the active buckets and use a small bitmap to skip
over any inactive ones.
v2: Use MASK of internal levels to simplify our usage.
v3: Prevent overflow when SHIFT is zero.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20181001123204.23982-4-chris@chris-wilson.co.uk
|
|
If the request is currently on the HW (in port 0), then we do not need
to kick the submission tasklet to evaluate whether we should be
preempting itself in order to execute it again.
In the case that was annoying me:
execlists_schedule: rq(18:211173).prio=0 -> 2
need_preempt: last(18:211174).prio=0, queue.prio=2
We are bumping the priority of the first of a pair of requests running
in the current context. Then when evaluating preempt, we would see that
that our priority request is higher than the last executing request in
ELSP0 and so trigger preemption, not realising that our intended request
was already executing.
v2: As we assume state of the execlists->port[] that is only valid while
we hold the timeline lock we have to repeat some earlier tests that on
the validity of the node.
v3: Wrap guc submission under the timeline.lock as is now the way of all
things.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180925083205.2229-2-chris@chris-wilson.co.uk
|
|
Stolen memory is lost across S4 (hibernate) or S3-RST as it is a portion
of ordinary volatile RAM. As we allocate our rings from stolen, this may
include the rings used for our preempt context and their breadcrumb
instructions. In order to allow preemption following hibernation and
loss of stolen memory, we therefore need to repopulate the instructions
inside the lost ring upon resume. To handle both module load and resume,
we simply defer constructing the ring to first use.
Testcase: igt/drv_selftest/live_gem
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180919205432.18394-1-chris@chris-wilson.co.uk
|
|
We currently verify that all doorbells can be registered with GuC and
HW but don't check that all works as expected after a db ring.
Do a nop ring of all doorbells to make sure we haven't misprogrammed
any WQ or stage descriptor data. This will also help validating
upcoming changes in the db programming flow.
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Acked-by: Katarzyna Dec <katarzyna.dec@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180827223614.22789-1-daniele.ceraolospurio@intel.com
|
|
A reasonably common operation is to pin the map of the vma alongside the
vma itself for the lifetime of the vma, and so release both pins at the
same time as destroying the vma. It is common enough to pull into the
release function, making that central function more attractive to a
couple of other callsites.
The continual ulterior motive is to sweep over errors on module load
aborting...
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180721125037.20127-1-chris@chris-wilson.co.uk
|
|
We make a decision at module load whether to use the GuC backend or not,
but lose that setup across set-wedge. Currently, the guc doesn't
override the engine->set_default_submission hook letting execlists sneak
back in temporarily on unwedging leading to an unbalanced park/unpark.
v2: Remove comment about switching back temporarily to execlists on
guc_submission_disable(). We currently only call disable on shutdown,
and plan to also call disable before suspend and reset, in which case we
will either restore guc submission or mark the driver as wedged, making
the reset back to execlists pointless.
v3: Move reset.prepare across
Fixes: 63572937cebf ("drm/i915/execlists: Flush pending preemption events during reset")
Testcase: igt/drv_module_reload/basic-reload-inject
Testcase: igt/gem_eio
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180717202932.1423-1-chris@chris-wilson.co.uk
|
|
Inject a failure into preemption completion to pretend as if the HW
didn't successfully handle preemption and we are forced to do a reset in
the middle.
v2: Wait for preemption, to force testing with the missed preemption.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180716132154.12539-1-chris@chris-wilson.co.uk
|
|
On reset/wedging, we cancel all pending replies from the HW and we also
want to cancel an outstanding preemption event. Since we use the same
function to cancel the pending replies for reset and for a preemption
event, we can simply clear the active tracking for all.
v2: Keep execlists_user_end() markup for wedging
v3: Move assignment to inline to hide the bare assignment.
Fixes: 60a943245413 ("drm/i915/execlists: Drop clear_gtiir() on GPU reset")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180716125424.5715-1-chris@chris-wilson.co.uk
|
|
Hopefully the final hack to get guc fault-injection happy before we can
clean it up again, starting from a known good baseline...
[ 383.017530] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a0
[ 383.017556] Oops: 0000 [#1] PREEMPT SMP PTI
[ 383.017566] CPU: 7 PID: 4725 Comm: drv_module_relo Tainted: G U 4.18.0-rc4-CI-CI_DRM_4485+ #1
[ 383.017581] Hardware name: Micro-Star International Co., Ltd. MS-7B54/Z370M MORTAR (MS-7B54), BIOS 1.10 12/28/2017
[ 383.017664] RIP: 0010:guc_stage_desc_pool_destroy+0x17/0xe0 [i915]
[ 383.017674] Code: 59 a0 c6 05 02 59 18 00 01 e8 5e 01 c3 e0 eb b1 0f 1f 00 53 48 89 fb 48 81 c7 90 02 00 00 e8 60 64 45 e1 48 8b 83 80 02 00 00 <48> 8b 80 a0 00 00 00 48 8b 90 68 02 00 00 48 83 ea 01 48 81 fa ff
[ 383.017771] RSP: 0018:ffffc900004bbdd0 EFLAGS: 00010282
[ 383.017782] RAX: 0000000000000000 RBX: ffff88012ff41300 RCX: 0000000000000000
[ 383.017794] RDX: 0000000000000000 RSI: ffffc900004bbd80 RDI: 0000000000000000
[ 383.017805] RBP: ffff88012ff40000 R08: 00000000d876ee11 R09: 0000000000000000
[ 383.017817] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88012ff47770
[ 383.017828] R13: ffff88012ff40068 R14: ffff880264392ef8 R15: ffffffffa0639950
[ 383.017840] FS: 00007fb9c18c8980(0000) GS:ffff8802663c0000(0000) knlGS:0000000000000000
[ 383.017853] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 383.017864] CR2: 00000000000000a0 CR3: 00000001df6cc003 CR4: 00000000003606e0
[ 383.017875] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 383.017887] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 383.017898] Call Trace:
[ 383.017962] intel_uc_fini+0x34/0xd0 [i915]
[ 383.018020] i915_gem_fini+0x5c/0x100 [i915]
[ 383.018093] i915_driver_unload+0xd2/0x110 [i915]
[ 383.018150] i915_pci_remove+0x10/0x20 [i915]
[ 383.018165] pci_device_remove+0x36/0xb0
[ 383.018179] device_release_driver_internal+0x185/0x250
[ 383.018193] driver_detach+0x35/0x70
[ 383.018205] bus_remove_driver+0x53/0xd0
[ 383.018217] pci_unregister_driver+0x25/0xa0
[ 383.018232] __se_sys_delete_module+0x162/0x210
[ 383.018245] ? do_syscall_64+0xd/0x190
[ 383.018257] do_syscall_64+0x55/0x190
[ 383.018270] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 383.018282] RIP: 0033:0x7fb9c0f7c1b7
[ 383.018290] Code: 73 01 c3 48 8b 0d d1 8c 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a1 8c 2c 00 f7 d8 64 89 01 48
[ 383.018408] RSP: 002b:00007fffa01c2aa8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0
[ 383.018425] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fb9c0f7c1b7
[ 383.018440] RDX: 0000000000000000 RSI: 0000000000000800 RDI: 0000560b96856d48
[ 383.018454] RBP: 0000560b96856ce0 R08: 0000560b96856d4c R09: 00007fffa01c2ae8
[ 383.018468] R10: 00007fffa01c1aa4 R11: 0000000000000206 R12: 0000560b954f7470
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180713172658.14070-1-chris@chris-wilson.co.uk
|
|
After aborting a module load, we may try and disable guc before we have
finished setting it. Long term plan is to ensure perfect onion unwind,
but in the short term we want to fix the oops to re-enable
drv_module_reload.
[ 317.401239] BUG: unable to handle kernel NULL pointer dereference at 0000000000000030
[ 317.401279] Oops: 0000 [#1] PREEMPT SMP PTI
[ 317.401294] CPU: 5 PID: 4275 Comm: drv_module_relo Tainted: G U 4.18.0-rc4-CI-CI_DRM_4476+ #1
[ 317.401317] Hardware name: System manufacturer System Product Name/Z170M-PLUS, BIOS 3610 03/29/2018
[ 317.401440] RIP: 0010:unreserve_doorbell+0x0/0x80 [i915]
[ 317.401454] Code: bb e0 48 8b 35 21 4d 18 00 49 c7 c0 a8 e5 62 a0 b9 cc 00 00 00 48 c7 c2 d8 41 5f a0 48 c7 c7 c9 f6 53 a0 e8 a2 3d c2 e0 0f 0b <0f> b7 47 30 66 3d 00 01 74 20 48 8b 57 18 48 0f a3 82 40 05 00 00
[ 317.401602] RSP: 0018:ffffc900003d3da0 EFLAGS: 00010246
[ 317.401619] RAX: ffffffff8223b300 RBX: 0000000000000000 RCX: 0000000000000000
[ 317.401636] RDX: 0000001fffffffc0 RSI: ffff880219f115f0 RDI: 0000000000000000
[ 317.401654] RBP: ffff880219f11838 R08: 0000000000000000 R09: 0000000000000000
[ 317.401671] R10: 0000000000000000 R11: 0000000000000000 R12: ffff880219f11300
[ 317.401689] R13: ffff880219f17770 R14: ffff88022c1daef8 R15: ffffffffa06ae950
[ 317.401707] FS: 00007febf77a9980(0000) GS:ffff880236d40000(0000) knlGS:0000000000000000
[ 317.401727] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 317.401743] CR2: 0000000000000030 CR3: 0000000222072003 CR4: 00000000003606e0
[ 317.401761] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 317.401779] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 317.401796] Call Trace:
[ 317.401894] guc_client_free+0x9/0x130 [i915]
[ 317.401993] intel_guc_submission_fini+0x50/0x90 [i915]
[ 317.402092] intel_uc_fini+0x34/0xd0 [i915]
[ 317.402179] i915_gem_fini+0x5c/0x100 [i915]
[ 317.402249] i915_driver_unload+0xd2/0x110 [i915]
[ 317.402321] i915_pci_remove+0x10/0x20 [i915]
[ 317.402341] pci_device_remove+0x36/0xb0
[ 317.402357] device_release_driver_internal+0x185/0x250
[ 317.402374] driver_detach+0x35/0x70
[ 317.402390] bus_remove_driver+0x53/0xd0
[ 317.402404] pci_unregister_driver+0x25/0xa0
[ 317.402423] __se_sys_delete_module+0x162/0x210
[ 317.402439] ? do_syscall_64+0xd/0x190
[ 317.402454] do_syscall_64+0x55/0x190
[ 317.402470] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 317.402485] RIP: 0033:0x7febf6e5d1b7
[ 317.402496] Code: 73 01 c3 48 8b 0d d1 8c 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a1 8c 2c 00 f7 d8 64 89 01 48
[ 317.402646] RSP: 002b:00007fffb5e72798 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0
[ 317.402667] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007febf6e5d1b7
[ 317.402686] RDX: 0000000000000000 RSI: 0000000000000800 RDI: 0000562da1addd98
[ 317.402703] RBP: 0000562da1addd30 R08: 0000562da1addd9c R09: 00007fffb5e727d8
[ 317.402721] R10: 00007fffb5e71794 R11: 0000000000000206 R12: 0000562da0ff6470
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712202027.19801-1-chris@chris-wilson.co.uk
|
|
If we fail the module load, we may try and cleanup before we even
allocate the GuC clients. KISS in order to try and re-enable
drv_module_reload for BAT.
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712105830.20390-1-chris@chris-wilson.co.uk
|
