We can use drm_printer to hide the differences between printk and
seq_printf, and so make the i915_engine_info pretty printer able to be
called from different contexts and not just debugfs. For instance, I
want to use the pretty printer to debug kselftests.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171009110301.21705-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
When we write to ELSP, it triggers a context preemption at the earliest
arbitration point (3DPRIMITIVE, some PIPECONTROLs, a few other
operations and the explicit MI_ARB_CHECK). If this is to the same
context, it triggers a LITE_RESTORE where the RING_TAIL is merely
updated (used currently to chain requests from the same context
together, avoiding bubbles). However, if it is to a different context, a
full context-switch is performed and it will start to execute the new
context saving the image of the old for later execution.
Previously we avoided preemption by only submitting a new context when
the old was idle. But now we wish embrace it, and if the new request has
a higher priority than the currently executing request, we write to the
ELSP regardless, thus triggering preemption, but we tell the GPU to
switch to our special preemption context (not the target). In the
context-switch interrupt handler, we know that the previous contexts
have finished execution and so can unwind all the incomplete requests
and compute the new highest priority request to execute.
It would be feasible to avoid the switch-to-idle intermediate by
programming the ELSP with the target context. The difficulty is in
tracking which request that should be whilst maintaining the dependency
change, the error comes in with coalesced requests. As we only track the
most recent request and its priority, we may run into the issue of being
tricked in preempting a high priority request that was followed by a
low priority request from the same context (e.g. for PI); worse still
that earlier request may be our own dependency and the order then broken
by preemption. By injecting the switch-to-idle and then recomputing the
priority queue, we avoid the issue with tracking in-flight coalesced
requests. Having tried the preempt-to-busy approach, and failed to find
a way around the coalesced priority issue, Michal's original proposal to
inject an idle context (based on handling GuC preemption) succeeds.
The current heuristic for deciding when to preempt are only if the new
request is of higher priority, and has the privileged priority of
greater than 0. Note that the scheduler remains unfair!
v2: Disable for gen8 (bdw/bsw) as we need additional w/a for GPGPU.
Since, the feature is now conditional and not always available when we
have a scheduler, make it known via the HAS_SCHEDULER GETPARAM (now a
capability mask).
v3: Stylistic tweaks.
v4: Appease Joonas with a snippet of kerneldoc, only to fuel to fire of
the preempt vs preempting debate.
Suggested-by: Michal Winiarski <michal.winiarski@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171003203453.15692-8-chris@chris-wilson.co.uk
As we emulate execlists on top of the GuC workqueue, it is not
restricted to just 2 ports and we can increase that number arbitrarily
to trade-off queue depth (i.e. scheduling latency) against pipeline
bubbles.
v2: rebase. better commit msg (Chris)
v3: rebase
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-5-mika.kuoppala@intel.com
When first execlist entry is processed, we move the port (contents).
Introduce function for this as execlist and guc use this common
operation.
v2: rebase. s/GEM_DEBUG_BUG/GEM_BUG (Chris)
v3: rebase
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-4-mika.kuoppala@intel.com
Engine's execlist related items have been increasing to
a point where a separate struct is warranted. Carve execlist
specific items to a dedicated struct to add clarity.
v2: add kerneldoc and fix whitespace (Joonas, Chris)
v3: csb_mmio changes, rebase
v4: s/\b(el|execlist)\b/execlists/ (Joonas)
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> (v3)
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v3)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-1-mika.kuoppala@intel.com
When wedging the hw, we want to mark all in-flight requests as -EIO.
This is made slightly more complex by execlists who store the ready but
not yet submitted-to-hw requests on a private queue (an rbtree
priolist). Call into execlists to cancel not only the ELSP tracking for
the submitted requests, but also the queue of unsubmitted requests.
v2: Move the majority of engine_set_wedged to the backends (both legacy
ringbuffer and execlists handling their own lists).
Reported-by: Michał Winiarski <michal.winiarski@intel.com>
Testcase: igt/gem_eio/in-flight-contexts
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170915173100.26470-1-chris@chris-wilson.co.uk
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
The engine also provides a mirror of the CSB write pointer in the HWSP,
but not of our read pointer. To take advantage of this we need to
remember where we read up to on the last interrupt and continue off from
there. This poses a problem following a reset, as we don't know where
the hw will start writing from, and due to the use of power contexts we
cannot perform that query during the reset itself. So we continue the
current modus operandi of delaying the first read of the context-status
read/write pointers until after the first interrupt. With this we should
now have eliminated all uncached mmio reads in handling the
context-status interrupt, though we still have the uncached mmio writes
for submitting new work, and many uncached mmio reads in the global
interrupt handler itself. Still a step in the right direction towards
reducing our resubmit latency, although it appears lost in the noise!
v2: Cannonlake moved the CSB write index
v3: Include the sw/hwsp state in debugfs/i915_engine_info
v4: Also revert to using CSB mmio for GVT-g
v5: Prevent the compiler reloading tail (Mika)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Acked-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170913085605.18299-6-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
The engine provides a mirror of the CSB in the HWSP. If we use the
cacheable reads from the HWSP, we can shave off a few mmio reads per
context-switch interrupt (which are quite frequent!). Just removing a
couple of mmio is not enough to actually reduce any latency, but a small
reduction in overall cpu usage.
Much appreciation for Ben dropping the bombshell that the CSB was in the
HWSP and for Michel in digging out the details.
v2: Don't be lazy, add the defines for the indices.
v3: Include the HWSP in debugfs/i915_engine_info
v4: Check for GVT-g, it currently depends on intercepting CSB mmio
v5: Fixup GVT-g mmio path
v6: Disable HWSP if VT-d is active as the iommu adds unpredictable
memory latency. (Mika)
v7: Also markup the CSB read with READ_ONCE() as it may still be an mmio
read and we want to stop the compiler from issuing a later (v.slow) reload.
Suggested-by: Ben Widawsky <benjamin.widawsky@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Acked-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170913133534.26927-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
The early gen3 machines (i915g/Grantsdale and i915gm/Alviso) share a lot
of characteristics in their MI/GTT blocks with gen2, and in particular
can only use physical addresses in MI_STORE_DATA_IMM. This makes it
incompatible with our usage, so include those two machines in the
blacklist to prevent usage.
v2: Make it easy for gcc and rewrite it as a switch to save some space.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> #v1
Link: https://patchwork.freedesktop.org/patch/msgid/20170906152859.5304-1-chris@chris-wilson.co.uk
MI_STORE_DWORD_IMM just doesn't work on the video decode engine under
Sandybridge, so refrain from using it. Then switch the selftests over to
using the now common test prior to using MI_STORE_DWORD_IMM.
Fixes: 7dd4f6729f ("drm/i915: Async GPU relocation processing")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.13-rc1+
Link: https://patchwork.freedesktop.org/patch/msgid/20170816085210.4199-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
And store the active request so that we only search for it once.
v2: Check for request completion inside _prepare_engine, don't use
ECANCELED, remove unnecessary null checks (Chris).
v3: Capture active requests during reset_prepare and store it the
engine hangcheck obj.
v4: Rename commit, change i915_gem_reset_request to just confirm the
active_request is still incomplete, instead of engine_stalled (Chris).
v5: With style; pass the active request to gem_reset_engine, keep single
return in reset_prepare_engine (Chris).
v6: Moved before reset-engine code appears (Chris)
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v5)
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170615201828.23144-2-michel.thierry@intel.com
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/20170620095751.13127-3-chris@chris-wilson.co.uk
All the requests at the same priority are executed in FIFO order. They
do not need to be stored in the rbtree themselves, as they are a simple
list within a level. If we move the requests at one priority into a list,
we can then reduce the rbtree to the set of priorities. This should keep
the height of the rbtree small, as the number of active priorities can not
exceed the number of active requests and should be typically only a few.
Currently, we have ~2k possible different priority levels, that may
increase to allow even more fine grained selection. Allocating those in
advance seems a waste (and may be impossible), so we opt for allocating
upon first use, and freeing after its requests are depleted. To avoid
the possibility of an allocation failure causing us to lose a request,
we preallocate the default priority (0) and bump any request to that
priority if we fail to allocate it the appropriate plist. Having a
request (that is ready to run, so not leading to corruption) execute
out-of-order is better than leaking the request (and its dependency
tree) entirely.
There should be a benefit to reducing execlists_dequeue() to principally
using a simple list (and reducing the frequency of both rbtree iteration
and balancing on erase) but for typical workloads, request coalescing
should be small enough that we don't notice any change. The main gain is
from improving PI calls to schedule, and the explicit list within a
level should make request unwinding simpler (we just need to insert at
the head of the list rather than the tail and not have to make the
rbtree search more complicated).
v2: Avoid use-after-free when deleting a depleted priolist
v3: Michał found the solution to handling the allocation failure
gracefully. If we disable all priority scheduling following the
allocation failure, those requests will be executed in fifo and we will
ensure that this request and its dependencies are in strict fifo (even
when it doesn't realise it is only a single list). Normal scheduling is
restored once we know the device is idle, until the next failure!
Suggested-by: Michał Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-8-chris@chris-wilson.co.uk
add/remove: 1/1 grow/shrink: 5/4 up/down: 391/-578 (-187)
function old new delta
execlists_submit_ports 262 471 +209
port_assign.isra - 136 +136
capture 6344 6359 +15
reset_common_ring 438 452 +14
execlists_submit_request 228 238 +10
gen8_init_common_ring 334 341 +7
intel_engine_is_idle 106 105 -1
i915_engine_info 2314 2290 -24
__i915_gem_set_wedged_BKL 485 411 -74
intel_lrc_irq_handler 1789 1604 -185
execlists_update_context 294 - -294
The most important change there is the improve to the
intel_lrc_irq_handler and excclist_submit_ports (net improvement since
execlists_update_context is now inlined).
v2: Use the port_api() for guc as well (even though currently we do not
pack any counters in there, yet) and hide all port->request_count inside
the helpers.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-5-chris@chris-wilson.co.uk
Typically, there is space available within the ring and if not we have
to wait (by definition a slow path). Rearrange the code to reduce the
number of branches and stack size for the hotpath, accomodating a slight
growth for the wait.
v2: Fix the new assert that packets are not larger than the actual ring.
v3: Make the parameters unsigned as well to make usage.
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/20170504130846.4807-3-chris@chris-wilson.co.uk
Some callers immediately want to know the current ring->space after
calling intel_ring_update_space(), which we can freely provide via the
return parameter.
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/20170504130846.4807-2-chris@chris-wilson.co.uk
Since unifying ringbuffer/execlist submission to use
engine->pin_context, we ensure that the intel_ring is available before
we start constructing the request. We can therefore move the assignment
of the request->ring to the central i915_gem_request_alloc() and not
require it in every engine->request_alloc() callback. Another small step
towards simplification (of the core, but at a cost of handling error
pointers in less important callers of engine->pin_context).
v2: Rearrange a few branches to reduce impact of PTR_ERR() on gcc's code
generation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Oscar Mateo <oscar.mateo@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170504093308.4137-1-chris@chris-wilson.co.uk
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>
If we are enabling the breadcrumbs signaling prior to submitting the
request, we know that we cannot have missed the interrupt and can
therefore skip immediately waking the signaler to check.
This reduces a significant chunk of the __i915_gem_request_submit()
overhead for inter-engine synchronisation, for example in gem_exec_whisper.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170426080659.28771-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
We need to keep track of the last location we ask the hw to read up to
(RING_TAIL) separately from our last write location into the ring, so
that in the event of a GPU reset we do not tell the HW to proceed into
a partially written request (which can happen if that request is waiting
for an external signal before being executed).
v2: Refactor intel_ring_reset() (Mika)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100144
Testcase: igt/gem_exec_fence/await-hang
Fixes: 821ed7df6e ("drm/i915: Update reset path to fix incomplete requests")
Fixes: d55ac5bf97 ("drm/i915: Defer transfer onto execution timeline to actual hw submission")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170425130049.26147-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
We want to refer to the index of the engine consistently throughout the
userspace ABI. We already have such an index through the execbuffer
engine specifier, that needs to be able to refer to each engine
specifically, so rename it the index to uabi_id to reflect its
generality beyond execbuf.
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/20170411124306.15448-1-chris@chris-wilson.co.uk
Not really needed, but makes the next change a little bit more compact.
v2:
- Use zero-based numbering for engine names: xcs0, xcs1.. xcsN (Tvrtko, Chris)
- Make sure the mock engine name is null-terminated (Tvrtko, Chris)
v3: Because I'm stupid (Chris)
v4: Verify engine name wasn't truncated (Michal)
v5:
- Kill the warning in mock engine (Chris)
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>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491834873-9345-4-git-send-email-oscar.mateo@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
If we needed to do something different for the init functions, we could
always look at the engine instance to make the distinction. But, in any
case, the two functions are virtually identical already (please notice
that BSD2_RING is only used from gen8 onwards).
With this, the init functions depends excusively on the engine class
(a fact that we will use soon).
v2: Commit message
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>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491834873-9345-3-git-send-email-oscar.mateo@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
In such a way that vcs and vcs2 are just two different instances (0 and 1)
of the same engine class (VIDEO_DECODE_CLASS).
v2: Align the instance types (Tvrtko)
v3: Don't use enums for bspec-defined stuff (Michal)
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491834873-9345-2-git-send-email-oscar.mateo@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Or rather it is used only by intel_ring_pin() to extract the
drm_i915_private which we can easily pass in. As this is a relatively
rare operation, save the space in the struct, and as such it is even
break even in the extra code for passing around the parameter:
add/remove: 0/0 grow/shrink: 2/3 up/down: 15/-15 (0)
function old new delta
intel_init_ring_buffer 906 918 +12
execlists_context_pin 1308 1311 +3
mock_engine 407 403 -4
intel_engine_create_ring 367 363 -4
intel_ring_pin 326 319 -7
Total: Before=1261794, After=1261794, chg +0.00%
v2: Reorder intel_init_ring_buffer to keep the ring setup together:
add/remove: 0/0 grow/shrink: 2/3 up/down: 9/-15 (-6)
function old new delta
intel_init_ring_buffer 906 912 +6
execlists_context_pin 1308 1311 +3
mock_engine 407 403 -4
intel_engine_create_ring 367 363 -4
intel_ring_pin 326 319 -7
Total: Before=1261794, After=1261788, chg -0.00%
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/20170403113426.25707-1-chris@chris-wilson.co.uk
Whilst I like having the assertions clearly visible in the code, they
are quite repetitious! As we find new limits we want to incorporate into
the set of assertions, it make sense to refactor them to a common
routine.
v2: Add a guc holdout.
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/20170327131412.20293-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
If the request->wa_tail is 0 (because it landed exactly on the end of
the ringbuffer), when we reconstruct request->tail following a reset we
fill in an illegal value (-8 or 0x001ffff8). As a result, RING_HEAD is
never able to catch up with RING_TAIL and the GPU spins endlessly. If
the ring contains a couple of breadcrumbs, even our hangcheck is unable
to catch the busy-looping as the ACTHD and seqno continually advance.
v2: Move the wrap into a common intel_ring_wrap().
Fixes: a3aabe86a3 ("drm/i915/execlists: Reinitialise context image after GPU hang")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: <stable@vger.kernel.org> # v4.10+
Link: http://patchwork.freedesktop.org/patch/msgid/20170327130009.4678-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
intel_flush_status_page() is defunct since commit f8dd2934c4
("drm/i915: Remove BXT incoherent seqno write workaround"), time to
remove it.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170324163540.31981-2-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Not all of our target platforms have clflush. For those without, just
assume the status page is sufficiently coherent that we do not need our
paranoia.
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Fixes: 14a6bbf9e5 ("drm/i915: Replace irq_seqno_barrier on hws write with a clflush")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170324163540.31981-1-chris@chris-wilson.co.uk
Tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Storing the position of the breadcrumb of the last retired request as
a separate last_retired_head is superfluous as we always copy that into
head prior to recalculation of the intel_ring.space.
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/20170321102552.24357-1-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
GVTg has introduced the context status notifier to schedule the GVTg
workload. At that time, the notifier is bound to GVTg context only,
so GVTg is not aware of host workloads.
Now we are going to improve GVTg's guest workload scheduler policy,
and add Guc emulation support for new Gen graphics. Both these two
features require acknowledgment for all contexts running on hardware.
(But will not alter host workload.) So here try to make some change.
The change is simple:
1. Move the context status notifier head from i915_gem_context to
intel_engine_cs. Which means there is a notifier head per engine
instead of per context. Execlist driver still call notifier for
each context sched-in/out events of current engine.
2. At GVTg side, it binds a notifier_block for each physical engine
at GVTg initialization period. Then GVTg can hear all context
status events.
In this patch, GVTg do nothing for host context event, but later
will add a function there. But in any case, the notifier callback is
a noop if this is no active vGPU.
Since intel_gvt_init() is called at early initialization stage and
require the status notifier head has been initiated, I initiate it in
intel_engine_setup().
v2: remove a redundant newline. (chris)
Fixes: 3c7ba6359d ("drm/i915: Introduce execlist context status change notification")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100232
Signed-off-by: Changbin Du <changbin.du@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170313024711.28591-1-changbin.du@intel.com
Acked-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Generally we are using macros for any hardware identifiers as these
may change between Gens. Do the same with hardware engine ids.
v2: move hw engine defs to i915_reg.h (Chris)
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170301202615.118632-1-michal.wajdeczko@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we now take the breadcrumbs spinlock within the interrupt handler, we
wish to minimise its hold time. During the interrupt we do not care
about the state of the full rbtree, only that of the first element, so
we can guard that with a separate lock.
v2: Rename first_wait to irq_wait to make it clearer that it is guarded
by irq_lock.
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/20170303190824.1330-1-chris@chris-wilson.co.uk
During reset_all_global_seqno() on seqno rollover, we have to update the
HWS. This causes all in flight requests to be completed, so first we
wait. However, we were only waiting for the requests themselves to be
completed and clearing out the waiter rbtrees - what I had missed was
the extra reference in execlists->port[]. Since commit fe9ae7a3bf
("drm/i915/execlists: Detect an out-of-order context switch") we can
detect when the request is retired before the context switch interrupt
is completed. The impact should be neglible outside of debugging.
Testcase: igt/gem_exec_whisper
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170303121947.20482-1-chris@chris-wilson.co.uk
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
A significant cost in setting up a wait is the overhead of enabling the
interrupt. As we disable the interrupt whenever the queue of waiters is
empty, if we are frequently waiting on alternating batches, we end up
re-enabling the interrupt on a frequent basis. We do want to disable the
interrupt during normal operations as under high load it may add several
thousand interrupts/s - we have been known in the past to occupy whole
cores with our interrupt handler after accidentally leaving user
interrupts enabled. As a compromise, leave the interrupt enabled until
the next IRQ, or the system is idle. This gives a small window for a
waiter to keep the interrupt active and not be delayed by having to
re-enable the interrupt.
v2: Restore hangcheck/missed-irq detection for continuations
v3: Be more careful restoring the hangcheck timer after reset
v4: Be more careful restoring the fake irq after reset (if required!)
v5: Redo changes to intel_engine_wakeup()
v6: Factor out __intel_engine_wakeup()
v7: Improve commentary for declaring a missed wakeup
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: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170227205850.2828-4-chris@chris-wilson.co.uk
As execlists and other non-semaphore multi-engine devices coordinate
between engines using interrupts, we can shave off a few 10s of
microsecond of scheduling latency by doing the fence signaling from the
interrupt as opposed to a RT kthread. (Realistically the delay adds
about 1% to an individual cross-engine workload.) We only signal the
first fence in order to limit the amount of work we move into the
interrupt handler. We also have to remember that our breadcrumbs may be
unordered with respect to the interrupt and so we still require the
waiter process to perform some heavyweight coherency fixups, as well as
traversing the tree of waiters.
v2: No need for early exit in irq handler - it breaks the flow between
patches and prevents the tracepoint
v3: Restore rcu hold across irq signaling of request
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: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170227205850.2828-2-chris@chris-wilson.co.uk
The two users of the return value from intel_engine_wakeup() are
expecting different results. In the breadcrumbs hangcheck, we are using
it to determine whether wake_up_process() detected the waiter was
currently running (and if so we presume that it hasn't yet missed the
interrupt). However, in the fake_irq path, we are using the return value
as a check as to whether there are any waiters, and so we may
incorrectly stop the fake-irq if that waiter was currently running.
To handle the two different needs, return both bits of information! We
uninline it from the irq path in preparation for the next patch which
makes the irq hotpath special and relegates intel_engine_wakeup() to the
slow fixup paths.
v2: s/ret/result/
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: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170227205850.2828-1-chris@chris-wilson.co.uk
The plan in the near-future is to allow requests to be removed from the
signaler. We can no longer then rely on holding a reference to the
request for the duration it is in the signaling tree, and instead must
obtain a reference to the request for the current operation using RCU.
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/20170223074422.4125-10-chris@chris-wilson.co.uk
A request is assigned a global seqno only when it is on the hardware
execution queue. The global seqno can be used to maintain a list of
requests on the same engine in retirement order, for example for
constructing a priority queue for waiting. Prior to its execution, or
if it is subsequently removed in the event of preemption, its global
seqno is zero. As both insertion and removal from the execution queue
may operate in IRQ context, it is not guarded by the usual struct_mutex
BKL. Instead those relying on the global seqno must be prepared for its
value to change between reads. Only when the request is complete can
the global seqno be stable (due to the memory barriers on submitting
the commands to the hardware to write the breadcrumb, if the HWS shows
that it has passed the global seqno and the global seqno is unchanged
after the read, it is indeed complete).
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/20170223074422.4125-9-chris@chris-wilson.co.uk
Replace the global device seqno with one for each engine, and account
for in-flight seqno on each separately. This is consistent with
dma-fence as each timeline has separate fence-contexts for each engine
and a seqno is only ordered within a fence-context (i.e. seqno do not
need to be ordered wrt to other engines, just ordered within a single
engine). This is required to enable request rewinding for preemption on
individual engines (we have to rewind the global seqno to avoid
overflow, and we do not have to rewind all engines just to preempt one.)
v2: Rename active_seqno to inflight_seqnos to more clearly indicate that
it is a counter and not equivalent to the existing seqno. Update
functions that operated on active_seqno similarly.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170223074422.4125-3-chris@chris-wilson.co.uk
When the timer expires for checking on interrupt processing, check to
see if any interrupts arrived within the last time period. If real
interrupts are still being delivered, we can be reassured that we
haven't missed the final interrupt as the waiter will still be woken.
Only once all activity ceases, do we have to worry about the waiter
never being woken and so need to install a timer to kick the waiter for
a slow arrival of a seqno.
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: http://patchwork.freedesktop.org/patch/msgid/20170217151304.16665-2-chris@chris-wilson.co.uk
We have a few open coded instances in the execlists code and an
almost suitable helper in intel_ringbuf.c
We can consolidate to a single helper if we change the existing
helper to emit directly to ring buffer memory and move the space
reservation outside it.
v2: Drop memcpy for memset. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170216122325.31391-2-tvrtko.ursulin@linux.intel.com
It is only used within intel_ringbuffer.c
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.oc.uk>
This removes the usage of intel_ring_emit in favour of
directly writing to the ring buffer.
intel_ring_emit was preventing the compiler for optimising
fetch and increment of the current ring buffer pointer and
therefore generating very verbose code for every write.
It had no useful purpose since all ringbuffer operations
are started and ended with intel_ring_begin and
intel_ring_advance respectively, with no bail out in the
middle possible, so it is fine to increment the tail in
intel_ring_begin and let the code manage the pointer
itself.
Useless instruction removal amounts to approximately
two and half kilobytes of saved text on my build.
Not sure if this has any measurable performance
implications but executing a ton of useless instructions
on fast paths cannot be good.
v2:
* Change return from intel_ring_begin to error pointer by
popular demand.
* Move tail increment to intel_ring_advance to enable some
error checking.
v3:
* Move tail advance back into intel_ring_begin.
* Rebase and tidy.
v4:
* Complete rebase after a few months since v3.
v5:
* Remove unecessary cast and fix !debug compile. (Chris Wilson)
v6:
* Make intel_ring_offset take request as well.
* Fix recording of request postfix plus a sprinkle of asserts.
(Chris Wilson)
v7:
* Use intel_ring_offset to get the postfix. (Chris Wilson)
* Convert GVT code as well.
v8:
* Rename *out++ to *cs++.
v9:
* Fix GVT out to cs conversion in GVT.
v10:
* Rebase for new intel_ring_begin in selftests.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170214113242.29241-1-tvrtko.ursulin@linux.intel.com
