During request submission, we call the engine->schedule() function so
that we may reorder the active requests as required for inheriting the
new request's priority. This may schedule several tasklets to run on the
local CPU, but we will need to schedule the tasklets again for the new
request. Delay all the local tasklets until the end, so that we only
have to process the queue just once.
v2: Beware PREEMPT_RCU, as then local_bh_disable() is then not a
superset of rcu_read_lock().
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180507135731.10587-2-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
igt_ctx_exec() expects that we retire all active requests/objects before
completing, so that when we clean up the files afterwards they are ready
to be freed. Before we do so, it is then prudent to ensure that we have
indeed retired the GPU activity, raising an error if it fails. If we do
not, we run the risk of triggering an assertion when freeing the object:
__i915_gem_free_objects:4793 GEM_BUG_ON(i915_gem_object_is_active(obj))
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/20180505091014.26126-2-chris@chris-wilson.co.uk
When userspace is passing around swapbuffers using DRI, we frequently
have to open and close the same object in the foreign address space.
This shows itself as the same object being rebound at roughly 30fps
(with a second object also being rebound at 30fps), which involves us
having to rewrite the page tables and maintain the drm_mm range manager
every time.
However, since the object still exists and it is only the local handle
that disappears, if we are lazy and do not unbind the VMA immediately
when the local user closes the object but defer it until the GPU is
idle, then we can reuse the same VMA binding. We still have to be
careful to mark the handle and lookup tables as closed to maintain the
uABI, just allowing the underlying VMA to be resurrected if the user is
able to access the same object from the same context again.
If the object itself is destroyed (neither userspace keeping a handle to
it), the VMA will be reaped immediately as usual.
In the future, this will be even more useful as instantiating a new VMA
for use on the GPU will become heavier. A nuisance indeed, so nip it in
the bud.
v2: s/__i915_vma_final_close/i915_vma_destroy/ etc.
v3: Leave a hint as to why we deferred the unbind on close.
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/20180503195115.22309-1-chris@chris-wilson.co.uk
We need to move to a more flexible timeline that doesn't assume one
fence context per engine, and so allow for a single timeline to be used
across a combination of engines. This means that preallocating a fence
context per engine is now a hindrance, and so we want to introduce the
singular timeline. From the code perspective, this has the notable
advantage of clearing up a lot of mirky semantics and some clumsy
pointer chasing.
By splitting the timeline up into a single entity rather than an array
of per-engine timelines, we can realise the goal of the previous patch
of tracking the timeline alongside the ring.
v2: Tweak wait_for_idle to stop the compiling thinking that ret may be
uninitialised.
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/20180502163839.3248-2-chris@chris-wilson.co.uk
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: 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/20180502163839.3248-1-chris@chris-wilson.co.uk
Due to the latency of the tasklet running from ksoftirqd, by the time we
process the execlist dequeue may be a long time behind the GPU. If the
request was completed when we ran reschedule, we will not have tweaked
its priority, but if it is still listed as being in-flight for dequeue
we will use it as a reference for the rest of the queue, including
requests from its own context which will now be at higher priority. This
can cause us to issue a preempt-to-idle request, even though the request
we want to preempt is already complete.
Reported-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180501122131.19435-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
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: 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/20180430131503.5375-3-chris@chris-wilson.co.uk
In commit 9b6586ae9f ("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: 9b6586ae9f ("drm/i915: Keep a global seqno per-engine")
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/20180430131503.5375-1-chris@chris-wilson.co.uk
Previously, we just reset the ring register in the context image such
that we could skip over the broken batch and emit the closing
breadcrumb. However, on resume the context image and GPU state would be
reloaded, which may have been left in an inconsistent state by the
reset. The presumption was that at worst it would just cause another
reset and skip again until it recovered, however it seems just as likely
to cause an unrecoverable hang. Instead of risking loading an incomplete
context image, restore it back to the default state.
v2: Fix up off-by-one from including the ppHSWP in with the register
state.
v3: Use a ring local to compact a few lines.
v4: Beware setting the ring local before checking for a NULL request.
References: https://bugs.freedesktop.org/show_bug.cgi?id=105304
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> #v2
Link: https://patchwork.freedesktop.org/patch/msgid/20180428111532.15819-1-chris@chris-wilson.co.uk
Even though we weren't injecting guilty requests to be reset, we could
still fall over the issue of resetting the same request too fast -- where
the GPU refuses to start again. (Although it is interesting to note that
reloading the driver is sufficient, suggesting that we could recover if
we delayed the setup after reset?) Continue to paper over the problem by
adding a small delay by waiting for the engine to idle between tests,
and ensure that the engines are idle before starting the idle tests.
v2: Replace single instance of 50 with a magic macro.
References: 028666793a ("drm/i915/selftests: Avoid repeatedly harming the same innocent context")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180411120346.27618-1-chris@chris-wilson.co.uk
Instead of synchronously cancelling the timer and re-enabling it inside
the reset callbacks, keep the timer enabled and let it die on its next
wakeup if no longer required. This allows
intel_engine_reset_breadcrumbs() to be used from an atomic
(timer/softirq) context such as required for resetting an engine.
It also allows us to react better to the user poking around debugfs for
testing missed irqs.
v2: Tighten the order of del_timer_sync as the fake_irq timer
may trigger the hangcheck timer, and so we should cancel it first and
then cancel the hangcheck (Mika)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180424142945.6787-1-chris@chris-wilson.co.uk
If we have more than a few, possibly several thousand request in the
queue, don't show the central portion, just the first few and the last
being executed and/or queued. The first few should be enough to help
identify a problem in execution, and most often comparing the first/last
in the queue is enough to identify problems in the scheduling.
We may need some fine tuning to set MAX_REQUESTS_TO_SHOW for common
debug scenarios, but for the moment if we can avoiding spending more
than a few seconds dumping the GPU state that will avoid a nasty
livelock (where hangcheck spends so long dumping the state, it fires
again and starts to dump the state again in parallel, ad infinitum).
v2: Remember to print last not the stale rq iter after the loop.
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/20180424081600.27544-1-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: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180418184052.7129-3-chris@chris-wilson.co.uk
Add a selftest to ensure that we restore the whitelisted registers after
rewrite the registers everytime they might be scrubbed, e.g. module
load, reset and resume. For the other volatile workaround registers, we
export their presence via debugfs and check in igt/gem_workarounds.
However, we don't export the whitelist and rather than do so, let's test
them directly in the kernel.
The test we use is to read the registers back from the CS (this helps us
be sure that the registers will be valid for MI_LRI etc). In order to
generate the expected list, we split intel_whitelist_workarounds_emit
into two phases, the first to build the list and the second to apply.
Inside the test, we only build the list and then check that list against
the hw.
v2: Filter out pre-gen8 as they do not have RING_NONPRIV.
v3: Drop unused engine parameter, no plans to use it now or future.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Oscar Mateo <oscar.mateo@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180414122754.569-1-chris@chris-wilson.co.uk
We can refine our current execlists->queue_priority if we inspect
ELSP[1] rather than the head of the unsubmitted queue. Currently, we use
the unsubmitted queue and say that if a subsequent request is more
important than the current queue, we will rerun the submission tasklet
to evaluate the need for preemption. However, we only want to preempt if
we need to jump ahead of a currently executing request in ELSP. The
second reason for running the submission tasklet is amalgamate requests
into the active context on ELSP[0] to avoid a stall when ELSP[0] drains.
(Though repeatedly amalgamating requests into the active context and
triggering many lite-restore is off question gain, the goal really is to
put a context into ELSP[1] to cover the interrupt.) So if instead of
looking at the head of the queue, we look at the context in ELSP[1] we
can answer both of the questions more accurately -- we don't need to
rerun the submission tasklet unless our new request is important enough
to feed into, at least, ELSP[1].
v2: Add some comments from the discussion with Tvrtko.
v3: More commentary to cross-reference queue_request()
References: f6322eddaf ("drm/i915/preemption: Allow preemption between submission ports")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180411103929.27374-1-chris@chris-wilson.co.uk
%phn is not a valid specifier, and the trailing 'n' is being eaten by
the format-specifier and defaulting to the ' ' separator. Avoid angering
smatch by using the unknown specifier, and use the default we expect.
drivers/gpu/drm/i915/intel_guc_ct.c:616 ctb_read() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:616 ctb_read() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:616 ctb_read() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:669 ct_handle_response() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:679 ct_handle_response() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:693 ct_handle_response() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:707 ct_handle_response() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:727 ct_process_request() warn: '%ph' cannot be followed by 'n'
drivers/gpu/drm/i915/intel_guc_ct.c:803 ct_handle_request() warn: '%ph' cannot be followed by 'n'
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180410111417.27563-1-chris@chris-wilson.co.uk
Resetting the GPU doesn't affect the RPS/RC6 state, so we can stop
forcibly reloading the registers.
Ville suggested this many moons ago, I said at that time that sanitizing
was no harm and meant that our bookkeeping was kept consistent with the
HW. However, in a forthcoming series, we want to split rps/rc6 GT
powermanagement and one of the key simplifications is the control of
when we enable it. Performing a crude sanitize in the middle of
i915_gem_reset() is then a huge wart.
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180410133354.13425-1-chris@chris-wilson.co.uk
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: Chris 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: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180406220354.18911-2-chris@chris-wilson.co.uk
In commit 932066a15335 ("tracing: Default to using trace_global_clock if
sched_clock is unstable"), the logic for deciding to override the
default clock if unstable was reversed from the earlier posting. I was
trying to reduce the width of the message by using an early return
rather than a if-block, but reverted back to using the if-block and
accidentally left the predicate inverted.
Link: http://lkml.kernel.org/r/20180404212450.26646-1-chris@chris-wilson.co.uk
Fixes: 932066a15335 ("tracing: Default to using trace_global_clock if sched_clock is unstable")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Across suspend, we may see a very large drift in timestamps if the sched
clock is unstable, prompting the global trace's ringbuffer code to warn
and suggest switching to the global clock. Preempt this request by
detecting when the sched clock is unstable (determined during
late_initcall) and automatically switching the default clock over to
trace_global_clock.
This should prevent requiring user interaction to resolve warnings such
as:
Delta way too big! 18446743856563626466 ts=18446744054496180323 write stamp = 197932553857
If you just came from a suspend/resume,
please switch to the trace global clock:
echo global > /sys/kernel/debug/tracing/trace_clock
Link: http://lkml.kernel.org/r/20180330150132.16903-1-chris@chris-wilson.co.uk
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
We don't handle resetting the kernel context very well, or presumably any
context executing its breadcrumb commands in the ring as opposed to the
batchbuffer and flush. If we trigger a device reset twice in quick
succession while the kernel context is executing, we may end up skipping
the breadcrumb. This is really only a problem for the selftest as
normally there is a large interlude between resets (hangcheck), or we
focus on resetting just one engine and so avoid repeatedly resetting
innocents.
Something to try would be a preempt-to-idle to quiesce the engine before
reset, so that innocent contexts would be spared the reset.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
CC: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180330131801.18327-1-chris@chris-wilson.co.uk
If we skip the intel_prepare_reset(), we should also skip the
intel_display_reset(). If we we use a flag set by intel_prepare_reset()
then we do not have to second guess based on external user controlled
state whether or not the prepare was called before deciding to finish
it after the reset. igt/gem_eio is one such example that may tweak
i915.reset faster than the code is expecting, leading to
[ 190.233528] =====================================
[ 190.233534] WARNING: bad unlock balance detected!
[ 190.233540] 4.16.0-rc7-g335ef9849310-drmtip_10+ #1 Tainted: G U
[ 190.233547] -------------------------------------
[ 190.233553] gem_eio/1348 is trying to release lock (crtc_ww_class_acquire) at:
[ 190.233569] [<ffffffff895c7810>] drm_modeset_acquire_fini+0x0/0x60
[ 190.233575] but there are no more locks to release!
[ 190.233580]
other info that might help us debug this:
[ 190.233588] 3 locks held by gem_eio/1348:
[ 190.233592] #0: (&f->f_pos_lock){+.+.}, at: [<00000000ab90c784>] __fdget_pos+0x3a/0x50
[ 190.233607] #1: (sb_writers#11){.+.+}, at: [<00000000e1529265>] vfs_write+0x188/0x1a0
[ 190.233622] #2: (&attr->mutex){+.+.}, at: [<0000000011f40afe>] simple_attr_write+0x36/0xd0
[ 190.233635]
stack backtrace:
[ 190.233644] CPU: 0 PID: 1348 Comm: gem_eio Tainted: G U 4.16.0-rc7-g335ef9849310-drmtip_10+ #1
[ 190.233655] Hardware name: Dell Inc. OptiPlex GX280 /0G8310, BIOS A04 02/09/2005
[ 190.233664] Call Trace:
[ 190.233674] dump_stack+0x67/0x95
[ 190.233682] ? drm_modeset_backoff+0x1b0/0x1b0
[ 190.233690] print_unlock_imbalance_bug+0xd2/0xe0
[ 190.233698] ? drm_modeset_backoff+0x1b0/0x1b0
[ 190.233704] lock_release+0x23e/0x300
[ 190.233712] drm_modeset_acquire_fini+0x16/0x60
[ 190.233835] intel_finish_reset+0x72/0x160 [i915]
[ 190.233894] i915_reset_device+0x1e9/0x240 [i915]
[ 190.233953] ? __intel_get_crtc_scanline+0x1c0/0x1c0 [i915]
[ 190.233962] ? work_on_cpu_safe+0x50/0x50
[ 190.234020] i915_handle_error+0x1f2/0x470 [i915]
[ 190.234031] ? __might_fault+0x39/0x90
[ 190.234037] ? __might_fault+0x39/0x90
[ 190.234099] i915_wedged_set+0x7f/0xc0 [i915]
[ 190.234107] simple_attr_write+0xb0/0xd0
[ 190.234117] full_proxy_write+0x51/0x80
[ 190.234125] __vfs_write+0x21/0x140
[ 190.234133] ? rcu_read_lock_sched_held+0x6f/0x80
[ 190.234140] ? rcu_sync_lockdep_assert+0x29/0x50
[ 190.234147] ? __sb_start_write+0x152/0x1f0
[ 190.234152] ? __sb_start_write+0x168/0x1f0
[ 190.234159] vfs_write+0xbd/0x1a0
[ 190.234166] SyS_write+0x40/0xa0
[ 190.234173] ? do_syscall_64+0x19/0x1b0
[ 190.234180] do_syscall_64+0x6b/0x1b0
[ 190.234188] entry_SYSCALL_64_after_hwframe+0x42/0xb7
[ 190.234196] RIP: 0033:0x7f84c1b392b7
[ 190.234201] RSP: 002b:00007f84b6755b00 EFLAGS: 00000293 ORIG_RAX: 0000000000000001
[ 190.234211] RAX: ffffffffffffffda RBX: 0000000000000046 RCX: 00007f84c1b392b7
[ 190.234218] RDX: 0000000000000002 RSI: 000055ec20abc8d6 RDI: 0000000000000046
[ 190.234225] RBP: 000055ec20abc8d6 R08: 0000000000000000 R09: 0000000000000000
[ 190.234231] R10: 0000000000000000 R11: 0000000000000293 R12: 0000000000000002
[ 190.234238] R13: 0000000000000000 R14: 00007f84b0000b20 R15: 000055ec20ce4eb8
Testcase: igt/gem_eio
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180405123714.3638-1-chris@chris-wilson.co.uk
Only sleep and repeat when asked for a full device reset (ALL_ENGINES)
and avoid using sleeping waits when asked for a per-engine reset. The
goal is to be able to use a per-engine reset from hardirq/softirq/timer
context. A consequence is that our individual wait timeouts are a
thousand times shorter, on the order of a hundred microseconds rather
than hundreds of millisecond. This may make hitting the timeouts more
common, but hopefully the fallover to the full-device reset will be
sufficient to pick up the pieces.
Note, that the sleeps inside older gen (pre-gen8) have been left as they
are only used in full device reset mode.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
CC: Michel Thierry <michel.thierry@intel.com>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180329224519.13598-1-chris@chris-wilson.co.uk
Tvrtko uncovered a fun issue with recovering from a wedge device. In his
tests, he wedged the driver by injecting an unrecoverable hang whilst a
batch was spinning. As we reset the gpu in the middle of the spinner,
when resumed it would continue on from the next instruction in the ring
and write it's breadcrumb. However, on wedging we updated our
bookkeeping to indicate that the GPU had completed executing and would
restart from after the breadcrumb; so the emission of the stale
breadcrumb from before the reset came as a bit of a surprise.
A simple fix is to when rebinding the context into the GPU, we update
the ring register state in the context image to match our bookkeeping.
We already have to update the RING_START and RING_TAIL, so updating
RING_HEAD as well is trivial. This works because whenever we unbind the
context, we keep the bookkeeping in check; and on wedging we unbind all
contexts.
Testcase: igt/gem_eio
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180327210136.16750-1-chris@chris-wilson.co.uk
Tested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
When cancelling the requests and clearing out the ports following a
successful preemption completion, also clear the active flag. I had
assumed that all preemptions would be followed by an immediate dequeue
(preserving the active user flag), but under rare circumstances we may
be triggering a preemption for the second port only for it to have
completed before the preemotion kicks in; leaving execlists->active set
even though the system is now idle.
We can clear the flag inside the common execlists_cancel_port_requests()
as the other users also expect the semantics of active being cleared.
Fixes: f6322eddaf ("drm/i915/preemption: Allow preemption between submission ports")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180324125829.27026-1-chris@chris-wilson.co.uk
After resetting the GPU (or subset of engines), call synchronize_irq()
to flush any pending irq before proceeding with the cleanup. For a
device level reset, we disable the interupts around the reset, but when
resetting just one engine, we have to avoid such global disabling. This
leaves us open to an interrupt arriving for the engine as we try to
reset it. We already do try to flush the IIR following the reset, but we
have to ensure that the in-flight interrupt does not land after we start
cleaning up after the reset; enter synchronize_irq().
As it current stands, we very rarely, but fatally, see sequences such as:
2.... 57964564us : execlists_reset_prepare: rcs0
2.... 57964613us : execlists_reset: rcs0 seqno=424
0d.h1 57964615us : gen8_cs_irq_handler: rcs0 CS active=1
2d..1 57964617us : __i915_request_unsubmit: rcs0 fence 29:1056 <- global_seqno 1060
2.... 57964703us : execlists_reset_finish: rcs0
0..s. 57964705us : execlists_submission_tasklet: rcs0 awake?=1, active=0, irq-posted?=1
v2: Move the sync into the execlists reset handler so that we coordinate
the flush with disabling the interrupt handling and canceling the
pending interrupt.
v3: Just use synchronize_hardirq() to avoid the might_sleep(), we do not
yet have threaded-irq to worry about.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180322073533.5313-4-chris@chris-wilson.co.uk
Reviewed-by: Jeff McGee <jeff.mcgee@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: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180322131034.6036-1-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: Chris 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: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180320100449.1360-2-chris@chris-wilson.co.uk
Chris Wilson