If some of the contexts submitting workloads to the GPU have been
configured to shutdown slices/subslices, we might loose the NOA
configurations written in the NOA muxes.
One possible solution to this problem is to reprogram the NOA muxes
when we switch to a new context. We initially tried this in the
workaround batchbuffer but some concerns where raised about the cost
of reprogramming at every context switch. This solution is also not
without consequences from the userspace point of view. Reprogramming
of the muxes can only happen once the powergating configuration has
changed (which happens after context switch). This means for a window
of time during the recording, counters recorded by the OA unit might
be invalid. This requires userspace dealing with OA reports to discard
the invalid values.
Minimizing the reprogramming could be implemented by tracking of the
last programmed configuration somewhere in GGTT and use MI_PREDICATE
to discard some of the programming commands, but the command streamer
would still have to parse all the MI_LRI instructions in the
workaround batchbuffer.
Another solution, which this change implements, is to simply disregard
the user requested configuration for the period of time when i915/perf
is active.
On most platforms there are no issues with this apart from a performance
penality for some media workloads that benefit from running on a partially
powergated GPU. We already prevent RC6 from affecting the programming so
it doesn't sound completely unreasonable to hold on powergating for the
same reason.
On Icelake however there would a functional problem if the slices not-
containing the VME block were left enabled with a running media workload
which explicitly disabled them. To avoid a GPU hang in this case, on
Icelake we lock the enablement to only slices which contain VME blocks.
Downside is that it means degraded GPU performance when OA is active but
there is no known alternative solution for this.
v2: Leave RPCS programming in intel_lrc.c (Lionel)
v3: Update for s/union intel_sseu/struct intel_sseu/ (Lionel)
More to_intel_context() (Tvrtko)
s/dev_priv/i915/ (Tvrtko)
Tvrtko Ursulin:
v4:
* Rebase for make_rpcs changes.
v5:
* Apply OA restriction from make_rpcs directly.
v6:
* Rebase for context image setup changes.
v7:
* Move stream assignment before metric enable.
v8-9:
* Rebase.
v10:
* Squashed with ICL support patch.
Bspec: 21140
Co-developed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v9
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-2-tvrtko.ursulin@linux.intel.com
As GEN8_LR_CONTEXT_ALIGN is I915_GTT_MIN_ALIGNMENT is it functionally
equivalent to 0, and we will not be able to reduce the min-alignment for
the GTT, so passing 0 is and will remain equivalent.
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/20180727092947.1953-1-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: 63572937ce ("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
To ease the frequent and ugly pointer dance of
&request->gem_context->engine[request->engine->id] during request
submission, store that pointer as request->hw_context. One major
advantage that we will exploit later is that this decouples the logical
context state from the engine itself.
v2: Set mock_context->ops so we don't crash and burn in selftests.
Cleanups from Tvrtko.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180517212633.24934-3-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: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180430131503.5375-2-chris@chris-wilson.co.uk
Enhanced Execlists is an upgraded version of execlists which supports
up to 8 ports. The lrcs to be submitted are written to a submit queue
(the ExecLists Submission Queue - ELSQ), which is then loaded on the
HW. When writing to the ELSP register, the lrcs are written cyclically
in the queue from position 0 to position 7. Alternatively, it is
possible to write directly in the individual positions of the queue
using the ELSQC registers. To be able to re-use all the existing code
we're using the latter method and we're currently limiting ourself to
only using 2 elements.
v2: Rebase.
v3: Switch from !IS_GEN11 to GEN < 11 (Daniele Ceraolo Spurio).
v4: Use the elsq registers instead of elsp. (Daniele Ceraolo Spurio)
v5: Reword commit, rename regs to be closer to specs, turn off
preemption (Daniele), reuse engine->execlists.elsp (Chris)
v6: use has_logical_ring_elsq to differentiate the new paths
v7: add preemption support, rename els to submit_reg (Chris)
v8: save the ctrl register inside the execlists struct, drop CSB
handling updates (superseded by preempt_complete_status) (Chris)
v9: s/drm_i915_gem_request/i915_request (Mika)
v10: resolved conflict in inject_preempt_context (Mika)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Thomas Daniel <thomas.daniel@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180302161501.28594-4-mika.kuoppala@linux.intel.com
We only use the preempt context to inject an idle point into execlists.
We never need to reference its logical state, so tell the GPU never to
load it or save it.
v2: BIT(2) for save-inhibit.
N.B. Daniele mentioned this bit mbz for ICL, and has been moved into the
submission process rather than the context image.
Suggested-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180123210412.17653-1-chris@chris-wilson.co.uk
Execlists and legacy ringbuffer submission are no longer feature
comparable (execlists now offer greater functionality that should
overcome their performance hit) and obsoletes the unsafe module
parameter, i.e. comparing the two modes of execution is no longer
useful, so remove the debug tool.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> #i915_perf.c
Link: https://patchwork.freedesktop.org/patch/msgid/20171120205504.21892-1-chris@chris-wilson.co.uk
Pretty similar to what we have on execlists.
We're reusing most of the GEM code, however, due to GuC quirks we need a
couple of extra bits.
Preemption is implemented as GuC action, and actions can be pretty slow.
Because of that, we're using a mutex to serialize them. Since we're
requesting preemption from the tasklet, the task of creating a workitem
and wrapping it in GuC action is delegated to a worker.
To distinguish that preemption has finished, we're using additional
piece of HWSP, and since we're not getting context switch interrupts,
we're also adding a user interrupt.
The fact that our special preempt context has completed unfortunately
doesn't mean that we're ready to submit new work. We also need to wait
for GuC to finish its own processing.
v2: Don't compile out the wait for GuC, handle workqueue flush on reset,
no need for ordered workqueue, put on a reviewer hat when looking at my own
patches (Chris)
Move struct work around in intel_guc, move user interruput outside of
conditional (Michał)
Keep ring around rather than chase though intel_context
v3: Extract WA for flushing ggtt writes to a helper (Chris)
Keep work_struct in intel_guc rather than engine (Michał)
Use ordered workqueue for inject_preempt worker to avoid GuC quirks.
v4: Drop now unused INTEL_GUC_PREEMPT_OPTION_IMMEDIATE (Daniele)
Drop stray newlines, use container_of for intel_guc in worker,
check for presence of workqueue when flushing it, rather than
enable_guc_submission modparam, reorder preempt postprocessing (Chris)
v5: Make wq NULL after destroying it
v6: Swap struct guc_preempt_work members (Michał)
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@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/20171026133558.19580-1-michal.winiarski@intel.com
Let the listener know that the context we just scheduled out was not
complete, and will be scheduled back in at a later point.
v2: Handle CONTEXT_STATUS_PREEMPTED in gvt by aliasing it to
CONTEXT_STATUS_OUT for the moment, gvt can expand upon the difference
later.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Zhenyu Wang" <zhenyuw@linux.intel.com>
Cc: "Wang, Zhi A" <zhi.a.wang@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171003203453.15692-3-chris@chris-wilson.co.uk
Not only the context image consist of two parts (the PPHWSP, and the
logical context state), but we also allocate a header at the start of
for sharing data with GuC. Thus every lrc looks like this:
| [guc] | [hwsp] [logical state] |
|<- our header ->|<- context image ->|
So far, we have oversimplified whenever we use each of these parts of the
context, just because the GuC header happens to be in page 0, and the
(PP)HWSP is in page 1. But this had led to using the same define for more
than one meaning (as a page index in the lrc and as 1 page).
This patch adds defines for the GuC shared page, the PPHWSP page and the
start of the logical state. It also updated the places where the old
define was being used. Since we are not changing the size (or format) of
the context, there are no functional changes.
v2: Use PPHWSP index for hws again.
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: intel-gvt-dev@lists.freedesktop.org
Link: http://patchwork.freedesktop.org/patch/msgid/20170712193032.27080-1-michel.thierry@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/20170913085605.18299-1-chris@chris-wilson.co.uk
The context descriptor is stored inside the per-engine context state, as
we only need to compute it once and access it frequently. However,
currently only intel_lrc.c has easy access, but i915_guc_submission.c
would like to frequently read it as well, and more so only ever needs
the lower 32bits. Make it an inline as the compiler should be able to
retrieve the value in less instructions than it takes to do the function
call:
add/remove: 0/1 grow/shrink: 1/0 up/down: 8/-45 (-37)
function old new delta
i915_guc_submit 621 629 +8
intel_lr_context_descriptor 45 - -45
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20170912214905.21987-1-chris@chris-wilson.co.uk
Reviewed-by: Oscar Mateo <oscar.mateo@intel.com>
Pre-calculate engine context size based on engine class and device
generation and store it in the engine instance.
v2:
- Squash and get rid of hw_context_size (Chris)
v3:
- Move after MMIO init for probing on Gen7 and 8 (Chris)
- Retained rounding (Tvrtko)
v4:
- Rebase for deferred legacy context allocation
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: intel-gvt-dev@lists.freedesktop.org
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
It turns out that we may want to restore the original
engine->submit_request (and engine->schedule) callbacks from more than
just the guc <-> execlists transition. Move this to a vfunc so we can
have a common interface.
v2: Move initial selection to intel_engines_init_common(), repaint vfunc
with engine->set_default_submission (and a similar colour for the
helper).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170316171305.12972-2-chris@chris-wilson.co.uk
In order to reset the GPU early on in the module load sequence, we need
to allocate the basic engine structs (to populate the mmio offsets etc).
Currently, the engine initialisation allocates both the base struct and
also allocate auxiliary objects, which depend upon state setup quite
late in the load sequence. We split off the allocation callback for
later and allow ourselves to allocate the engine structs themselves
early.
v2: Different paint for the unwind following error.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170124110135.6418-1-chris@chris-wilson.co.uk
Start converting over from the byte count to its semantic macro, either
we want to allocate the size of a physical page in main memory or we
want the size of a virtual page in the GTT. 4096 could mean either, but
PAGE_SIZE and I915_GTT_PAGE_SIZE are explicit and should help improve
code comprehension and future changes. In the future, we may want to use
variable GTT page sizes and so have the challenge of knowing which
hardcoded values were used to represent a physical page vs the virtual
page.
v2: Look for a few more 4096s to convert, discover IS_ALIGNED().
v3: 4096ul paranoia, make fence alignment a distinct value of 4096, keep
bdw stolen w/a as 4096 until we know better.
v4: Add asserts that i915_vma_insert() start/end are aligned to GTT page
sizes.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170110144734.26052-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
A fairly trivial move of a matching pair of routines (for preparing a
request for construction) onto an engine vfunc. The ulterior motive is
to be able to create a mock request implementation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161218153724.8439-7-chris@chris-wilson.co.uk
The requests conversion introduced a nasty bug where we could generate a
new request in the middle of constructing a request if we needed to idle
the system in order to evict space for a context. The request to idle
would be executed (and waited upon) before the current one, creating a
minor havoc in the seqno accounting, as we will consider the current
request to already be completed (prior to deferred seqno assignment) but
ring->last_retired_head would have been updated and still could allow
us to overwrite the current request before execution.
We also employed two different mechanisms to track the active context
until it was switched out. The legacy method allowed for waiting upon an
active context (it could forcibly evict any vma, including context's),
but the execlists method took a step backwards by pinning the vma for
the entire active lifespan of the context (the only way to evict was to
idle the entire GPU, not individual contexts). However, to circumvent
the tricky issue of locking (i.e. we cannot take struct_mutex at the
time of i915_gem_request_submit(), where we would want to move the
previous context onto the active tracker and unpin it), we take the
execlists approach and keep the contexts pinned until retirement.
The benefit of the execlists approach, more important for execlists than
legacy, was the reduction in work in pinning the context for each
request - as the context was kept pinned until idle, it could short
circuit the pinning for all active contexts.
We introduce new engine vfuncs to pin and unpin the context
respectively. The context is pinned at the start of the request, and
only unpinned when the following request is retired (this ensures that
the context is idle and coherent in main memory before we unpin it). We
move the engine->last_context tracking into the retirement itself
(rather than during request submission) in order to allow the submission
to be reordered or unwound without undue difficultly.
And finally an ulterior motive for unifying context handling was to
prepare for mock requests.
v2: Rename to last_retired_context, split out legacy_context tracking
for MI_SET_CONTEXT.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161218153724.8439-3-chris@chris-wilson.co.uk
Like GEM init, GUC init, MOCS init and context creation.
Enables them to lose dev_priv locals.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
We assume that the GPU is idle once receiving the seqno via the last
request's user interrupt. In execlist mode the corresponding context
completed interrupt can be delayed though and until this latter
interrupt arrives we consider the request to be pending on the ELSP
submit port. This can cause a problem during system suspend where this
last request will be seen by the resume code as still pending. Such
pending requests are normally replayed after a GPU reset, but during
resume we reset both SW and HW tracking of the ring head/tail pointers,
so replaying the pending request with its stale tail pointer will leave
the ring in an inconsistent state. A subsequent request submission can
lead then to the GPU executing from uninitialized area in the ring
behind the above stale tail pointer.
Fix this by making sure any pending request on the ELSP port is
completed before suspending. I used a polling wait since the completion
time I measured was <1ms and since normally we only need to wait during
system suspend. GPU idling during runtime suspend is scheduled with a
delay (currently 50-100ms) after the retirement of the last request at
which point the context completed interrupt must have arrived already.
The chance of this bug was increased by
commit 1c777c5d1d
Author: Imre Deak <imre.deak@intel.com>
Date: Wed Oct 12 17:46:37 2016 +0300
drm/i915/hsw: Fix GPU hang during resume from S3-devices state
but it could happen even without the explicit GPU reset, since we
disable interrupts afterwards during the suspend sequence.
v2:
- Do an unlocked poll-wait first. (Chris)
v3-4:
- s/intel_lr_engines_idle/intel_execlists_idle/ and move
i915.enable_execlists check to the new helper. (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98470
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1478510405-11799-3-git-send-email-imre.deak@intel.com
Update reset path in preparation for engine reset which requires
identification of incomplete requests and associated context and fixing
their state so that engine can resume correctly after reset.
The request that caused the hang will be skipped and head is reset to the
start of breadcrumb. This allows us to resume from where we left-off.
Since this request didn't complete normally we also need to cleanup elsp
queue manually. This is vital if we employ nonblocking request
submission where we may have a web of dependencies upon the hung request
and so advancing the seqno manually is no longer trivial.
ABI: gem_reset_stats / DRM_IOCTL_I915_GET_RESET_STATS
We change the way we count pending batches. Only the active context
involved in the reset is marked as either innocent or guilty, and not
mark the entire world as pending. By inspection this only affects
igt/gem_reset_stats (which assumes implementation details) and not
piglit.
ARB_robustness gives this guide on how we expect the user of this
interface to behave:
* Provide a mechanism for an OpenGL application to learn about
graphics resets that affect the context. When a graphics reset
occurs, the OpenGL context becomes unusable and the application
must create a new context to continue operation. Detecting a
graphics reset happens through an inexpensive query.
And with regards to the actual meaning of the reset values:
Certain events can result in a reset of the GL context. Such a reset
causes all context state to be lost. Recovery from such events
requires recreation of all objects in the affected context. The
current status of the graphics reset state is returned by
enum GetGraphicsResetStatusARB();
The symbolic constant returned indicates if the GL context has been
in a reset state at any point since the last call to
GetGraphicsResetStatusARB. NO_ERROR indicates that the GL context
has not been in a reset state since the last call.
GUILTY_CONTEXT_RESET_ARB indicates that a reset has been detected
that is attributable to the current GL context.
INNOCENT_CONTEXT_RESET_ARB indicates a reset has been detected that
is not attributable to the current GL context.
UNKNOWN_CONTEXT_RESET_ARB indicates a detected graphics reset whose
cause is unknown.
The language here is explicit in that we must mark up the guilty batch,
but is loose enough for us to relax the innocent (i.e. pending)
accounting as only the active batches are involved with the reset.
In the future, we are looking towards single engine resetting (with
minimal locking), where it seems inappropriate to mark the entire world
as innocent since the reset occurred on a different engine. Reducing the
information available means we only have to encounter the pain once, and
also reduces the information leaking from one context to another.
v2: Legacy ringbuffer submission required a reset following hibernation,
or else we restore stale values to the RING_HEAD and walked over
stolen garbage.
v3: GuC requires replaying the requests after a reset.
v4: Restore engine IRQ after reset (so waiters will be woken!)
Rearm hangcheck if resetting with a waiter.
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-13-chris@chris-wilson.co.uk
Emulate HW to track and manage ELSP queue. A set of SW ports are defined
and requests are assigned to these ports before submitting them to HW. This
helps in cleaning up incomplete requests during reset recovery easier
especially after engine reset by decoupling elsp queue management. This
will become more clear in the next patch.
In the engine reset case we want to resume where we left-off after skipping
the incomplete batch which requires checking the elsp queue, removing
element and fixing elsp_submitted counts in some cases. Instead of directly
manipulating the elsp queue from reset path we can examine these ports, fix
up ringbuffer pointers using the incomplete request and restart submissions
again after reset.
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1470414607-32453-3-git-send-email-arun.siluvery@linux.intel.com
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-6-chris@chris-wilson.co.uk
Space for flushing the GPU cache prior to completing the request is
preallocated and so cannot fail - the GPU caches will always be flushed
along with the completed request. This means we no longer have to track
whether the GPU cache is dirty between batches like we had to with the
outstanding_lazy_seqno.
With the removal of the duplication in the per-backend entry points for
emitting the obsolete lazy flush, we can then further unify the
engine->emit_flush.
v2: Expand a bit on the legacy of gpu_caches_dirty
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1469432687-22756-18-git-send-email-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470174640-18242-7-git-send-email-chris@chris-wilson.co.uk
'ring' is an old deprecated term for a GPU engine. Here we make the
terminology more consistent by renaming the 'ring' parameter of lots of
macros that calculate addresses within the MMIO space of an engine.
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1469034967-15840-2-git-send-email-david.s.gordon@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Ringbuffers are now being written to either through LLC or WC paths, so
treating them as simply iomem is no longer adequate. However, for the
older !llc hardware, the hardware is documentated as treating the TAIL
register update as serialising, so we can relax the barriers when filling
the rings (but even if it were not, it is still an uncached register write
and so serialising anyway.).
For simplicity, let's ignore the iomem annotation.
v2: Remove iomem from ringbuffer->virtual_address
v3: And for good measure add iomem elsewhere to keep sparse happy
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> #v2
Link: http://patchwork.freedesktop.org/patch/msgid/1469005202-9659-8-git-send-email-chris@chris-wilson.co.uk
Link: http://patchwork.freedesktop.org/patch/msgid/1469017917-15134-7-git-send-email-chris@chris-wilson.co.uk
Common code deserves to be put in a separate file from legacy and
execlists implementation for clarity and ease of maintenance.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris-wilson.co.uk>
With the unified common engine setup done, and the execlist engine
initialization loop clearly split into two phases, we can eliminate
the separate legacy engine initialization code.
v2: Fix cleanup path for legacy.
v3: Rename constructors. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Chris Wilson <chris-wilson.co.uk>
This patch introduces an approach to track the execlist context status
change.
GVT-g uses GVT context as the "shadow context". The content inside GVT
context will be copied back to guest after the context is idle. And GVT-g
has to know the status of the execlist context.
This function is configurable when creating a new GEM context. Currently,
Only GVT-g will create the "status-change-notification" enabled GEM
context.
v10:
- Fix the identation. (Joonas)
v8:
- Remove the boolean flag in struct i915_gem_context. (Joonas)
v7:
- Remove per-engine ctx status notifiers. Use one status notifier for all
engines. (Joonas)
- Add prefix "INTEL_" for related definitions. (Joonas)
- Refine the comments in execlists_context_status_change(). (Joonas)
v6:
- When !CONFIG_DRM_I915_GVT, make GVT code as dead code then compiler
could automatically eliminate them for us. (Chris)
- Always initialize the notifier header, so it could be switched on/off
at runtime. (Chris)
v5:
- Only compile this feature when CONFIG_DRM_I915_GVT is enabled.(Tvrtko)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> (v8)
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1466078825-6662-8-git-send-email-zhi.a.wang@intel.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
struct intel_context contains two substructs, one for the legacy RCS and
one for every execlists engine. Since legacy RCS is a subset of the
execlists engine support, just combine the two substructs.
v2: Only pin the default context for legacy mode (the object only exists
for legacy, but adding i915.enable_execlists provides symmetry with the
cleanup functions).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1464098023-3294-8-git-send-email-chris@chris-wilson.co.uk
Our goal is to rename the anonymous per-engine struct beneath the
current intel_context. However, after a lively debate resolving around
the confusion between intel_context_engine and intel_engine_context, the
realisation is that the two structs target different users. The outer
struct is API / user facing, and so carries the higher level GEM
information. The inner struct is hw facing. Thus we want to name the
inner struct intel_context and the outer one i915_gem_context. As the
first step, we need to rename the current struct:
s/struct intel_context/struct i915_gem_context/
which fits much better with its constructors already conveying the
i915_gem_context prefix!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1464098023-3294-1-git-send-email-chris@chris-wilson.co.uk
text data bss dec hex filename
6309351 3578714 696320 10584385 a18141 vmlinux
6308391 3578714 696320 10583425 a17d81 vmlinux
Almost 1KiB of code reduction.
v2: More s/INTEL_INFO()->gen/INTEL_GEN()/ and IS_GENx() conversions
text data bss dec hex filename
6304579 3578778 696320 10579677 a16edd vmlinux
6303427 3578778 696320 10578525 a16a5d vmlinux
Now over 1KiB!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1462545621-30125-3-git-send-email-chris@chris-wilson.co.uk
With the previous patch having extended the pinned lifetime of
contexts by referencing the previous context from the current
request until the latter is retired (completed by the GPU),
we can now remove usage of execlist retired queue entirely.
This is because the above now guarantees that all execlist
object access requirements are satisfied by this new tracking,
and we can stop taking additional references and stop keeping
request on the execlists retired queue.
The latter was a source of significant scalability issues in
the driver causing performance hits on some tests. Most
dramatical of which was igt/gem_close_race which had run time
in tens of minutes which is now reduced to tens of seconds.
Signed-off-by: Tvrtko Ursulin <tvrtko@ursulin.net>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1461833819-3991-24-git-send-email-chris@chris-wilson.co.uk
Combine the near identical implementations of intel_logical_ring_begin()
and intel_ring_begin() - the only difference is that the logical wait
has to check for a matching ring (which is assumed by legacy).
In the process some debug messages are culled as there were following a
WARN if we hit an actual error.
v2: Updated commentary
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461833819-3991-12-git-send-email-chris@chris-wilson.co.uk
Separate out the layers of includes (linux, drm, intel, i915) so that it
is a little easier to order our definitions between our multiple
reentrant headers. A couple of headers needed fixes to make them more
standalone (forgotten includes, forward declarations etc).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1460565315-7748-2-git-send-email-chris@chris-wilson.co.uk
We can use the new pin/lazy unpin API for simplicity
and more performance in the execlist submission paths.
v2:
* Fix error handling and convert more users.
* Compact some names for readability.
v3:
* intel_lr_context_free was not unpinning.
* Special case for GPU reset which otherwise unbalances
the HWS object pages pin count by running the engine
initialization only (not destructors).
v4:
* Rebased on top of hws setup/init split.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1460472042-1998-1-git-send-email-tvrtko.ursulin@linux.intel.com
[tursulin: renames: s/hwd/hws/, s/obj_addr/vaddr/]
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Rather than blindly waking up all forcewake domains on command
submission, we can teach each engine what is (or are) the correct
one to take.
On platforms with multiple forcewake domains like VLV, CHV, SKL
and BXT, this has the potential of lowering the GPU and CPU
power use and submission latency.
To implement it we add a function named
intel_uncore_forcewake_for_reg whose purpose is to query which
forcewake domains need to be taken to read or write a specific
register with raw mmio accessors.
These enables the execlists engine setup to query which
forcewake domains are relevant per engine on the currently
running platform.
v2:
* Kerneldoc.
* Split from intel_uncore.c macro extraction, WARN_ON,
no warns on old platforms. (Chris Wilson)
v3:
* Single domain per engine, mention all registers,
bi-directional function and a new name, fix handling
of gen6 and gen7 writes. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1460468251-14069-1-git-send-email-tvrtko.ursulin@linux.intel.com
Doing a lot of work in the interrupt handler introduces huge
latencies to the system as a whole.
Most dramatic effect can be seen by running an all engine
stress test like igt/gem_exec_nop/all where, when the kernel
config is lean enough, the whole system can be brought into
multi-second periods of complete non-interactivty. That can
look for example like this:
NMI watchdog: BUG: soft lockup - CPU#0 stuck for 23s! [kworker/u8:3:143]
Modules linked in: [redacted for brevity]
CPU: 0 PID: 143 Comm: kworker/u8:3 Tainted: G U L 4.5.0-160321+ #183
Hardware name: Intel Corporation Broadwell Client platform/WhiteTip Mountain 1
Workqueue: i915 gen6_pm_rps_work [i915]
task: ffff8800aae88000 ti: ffff8800aae90000 task.ti: ffff8800aae90000
RIP: 0010:[<ffffffff8104a3c2>] [<ffffffff8104a3c2>] __do_softirq+0x72/0x1d0
RSP: 0000:ffff88014f403f38 EFLAGS: 00000206
RAX: ffff8800aae94000 RBX: 0000000000000000 RCX: 00000000000006e0
RDX: 0000000000000020 RSI: 0000000004208060 RDI: 0000000000215d80
RBP: ffff88014f403f80 R08: 0000000b1b42c180 R09: 0000000000000022
R10: 0000000000000004 R11: 00000000ffffffff R12: 000000000000a030
R13: 0000000000000082 R14: ffff8800aa4d0080 R15: 0000000000000082
FS: 0000000000000000(0000) GS:ffff88014f400000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fa53b90c000 CR3: 0000000001a0a000 CR4: 00000000001406f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Stack:
042080601b33869f ffff8800aae94000 00000000fffc2678 ffff88010000000a
0000000000000000 000000000000a030 0000000000005302 ffff8800aa4d0080
0000000000000206 ffff88014f403f90 ffffffff8104a716 ffff88014f403fa8
Call Trace:
<IRQ>
[<ffffffff8104a716>] irq_exit+0x86/0x90
[<ffffffff81031e7d>] smp_apic_timer_interrupt+0x3d/0x50
[<ffffffff814f3eac>] apic_timer_interrupt+0x7c/0x90
<EOI>
[<ffffffffa01c5b40>] ? gen8_write64+0x1a0/0x1a0 [i915]
[<ffffffff814f2b39>] ? _raw_spin_unlock_irqrestore+0x9/0x20
[<ffffffffa01c5c44>] gen8_write32+0x104/0x1a0 [i915]
[<ffffffff8132c6a2>] ? n_tty_receive_buf_common+0x372/0xae0
[<ffffffffa017cc9e>] gen6_set_rps_thresholds+0x1be/0x330 [i915]
[<ffffffffa017eaf0>] gen6_set_rps+0x70/0x200 [i915]
[<ffffffffa0185375>] intel_set_rps+0x25/0x30 [i915]
[<ffffffffa01768fd>] gen6_pm_rps_work+0x10d/0x2e0 [i915]
[<ffffffff81063852>] ? finish_task_switch+0x72/0x1c0
[<ffffffff8105ab29>] process_one_work+0x139/0x350
[<ffffffff8105b186>] worker_thread+0x126/0x490
[<ffffffff8105b060>] ? rescuer_thread+0x320/0x320
[<ffffffff8105fa64>] kthread+0xc4/0xe0
[<ffffffff8105f9a0>] ? kthread_create_on_node+0x170/0x170
[<ffffffff814f351f>] ret_from_fork+0x3f/0x70
[<ffffffff8105f9a0>] ? kthread_create_on_node+0x170/0x170
I could not explain, or find a code path, which would explain
a +20 second lockup, but from some instrumentation it was
apparent the interrupts off proportion of time was between
10-25% under heavy load which is quite bad.
When a interrupt "cliff" is reached, which was >~320k irq/s on
my machine, the whole system goes into a terrible state of the
above described multi-second lockups.
By moving the GT interrupt handling to a tasklet in a most
simple way, the problem above disappears completely.
Testing the effect on sytem-wide latencies using
igt/gem_syslatency shows the following before this patch:
gem_syslatency: cycles=1532739, latency mean=416531.829us max=2499237us
gem_syslatency: cycles=1839434, latency mean=1458099.157us max=4998944us
gem_syslatency: cycles=1432570, latency mean=2688.451us max=1201185us
gem_syslatency: cycles=1533543, latency mean=416520.499us max=2498886us
This shows that the unrelated process is experiencing huge
delays in its wake-up latency. After the patch the results
look like this:
gem_syslatency: cycles=808907, latency mean=53.133us max=1640us
gem_syslatency: cycles=862154, latency mean=62.778us max=2117us
gem_syslatency: cycles=856039, latency mean=58.079us max=2123us
gem_syslatency: cycles=841683, latency mean=56.914us max=1667us
Showing a huge improvement in the unrelated process wake-up
latency. It also shows an approximate halving in the number
of total empty batches submitted during the test. This may
not be worrying since the test puts the driver under
a very unrealistic load with ncpu threads doing empty batch
submission to all GPU engines each.
Another benefit compared to the hard-irq handling is that now
work on all engines can be dispatched in parallel since we can
have up to number of CPUs active tasklets. (While previously
a single hard-irq would serially dispatch on one engine after
another.)
More interesting scenario with regards to throughput is
"gem_latency -n 100" which shows 25% better throughput and
CPU usage, and 14% better dispatch latencies.
I did not find any gains or regressions with Synmark2 or
GLbench under light testing. More benchmarking is certainly
required.
v2:
* execlists_lock should be taken as spin_lock_bh when
queuing work from userspace now. (Chris Wilson)
* uncore.lock must be taken with spin_lock_irq when
submitting requests since that now runs from either
softirq or process context.
v3:
* Expanded commit message with more testing data;
* converted missed locking sites to _bh;
* added execlist_lock comment. (Chris Wilson)
v4:
* Mention dispatch parallelism in commit. (Chris Wilson)
* Do not hold uncore.lock over MMIO reads since the block
is already serialised per-engine via the tasklet itself.
(Chris Wilson)
* intel_lrc_irq_handler should be static. (Chris Wilson)
* Cancel/sync the tasklet on GPU reset. (Chris Wilson)
* Document and WARN that tasklet cannot be active/pending
on engine cleanup. (Chris Wilson/Imre Deak)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Imre Deak <imre.deak@intel.com>
Testcase: igt/gem_exec_nop/all
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94350
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1459768316-6670-1-git-send-email-tvrtko.ursulin@linux.intel.com
Previously intel_lr_context_(un)pin were operating on requests
which is in conflict with their names.
If we make them take a context and an engine, it makes the names
make more sense and it also makes future fixes possible.
v2: Rebase for default_context/kernel_context change.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Nick Hoath <nicholas.hoath@intel.com>
LRC code was calling GEM API like i915_gem_obj_ggtt_offset from
places where the struct_mutex cannot be grabbed (irq handlers).
To avoid that this patch caches some interesting bits and values
in the engine and context structures.
Some usages are also removed where they are not needed like a
few asserts which are either impossible or have been checked
already during engine initialization.
Side benefit is also that interrupt handlers and command
submission stop evaluating invariant conditionals, like what
Gen we are running on, on every interrupt and every command
submitted.
This patch deals with logical ring context id and descriptors
while subsequent patches will deal with the remaining issues.
v2:
* Cache the VMA instead of the address. (Chris Wilson)
* Incorporate Dave Gordon's good comments and function name.
v3:
* Extract ctx descriptor template to a function and group
functions dealing with ctx descriptor & co together near
top of the file. (Dave Gordon)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dave Gordon <david.s.gordon@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1452870629-13830-1-git-send-email-tvrtko.ursulin@linux.intel.com
I think this patch is a worthwhile cleanup even if it might look only marginally
useful. It gets more useful in upcoming patches and for handling of future GEN
platforms.
The only non-mechanical part of this is the removal of the extra & operation on
the ring->next_context_status_buffer. This is safe because right above this, we
already did a modulus operation.
Signed-off-by: Ben Widawsky <benjamin.widawsky@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1452018609-10142-2-git-send-email-benjamin.widawsky@intel.com
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The GuC code needs to know the size of a logical context, so we
expose get_lr_context_size(), renaming it intel_lr_context__size()
to fit the naming conventions for nonstatic functions.
For: VIZ-2021
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1450468812-4882-2-git-send-email-yu.dai@intel.com
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
