99% of the time we access i915_address_space->dev we want the i915
device and not the drm device, so let's store the drm_i915_private
backpointer instead. The only real complication here are the inlines
in i915_vma.h where drm_i915_private is not yet defined and so we have
to choose an alternate path for our asserts.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161129095008.32622-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
In order to prevent a race between the old callback submitting an
incomplete request and i915_gem_set_wedged() installing its nop handler,
we must ensure that the swap occurs when the machine is idle
(stop_machine).
v2: move context lost from out of BKL.
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/20161122144121.7379-4-chris@chris-wilson.co.uk
Since the submit/execute split in commit d55ac5bf97 ("drm/i915: Defer
transfer onto execution timeline to actual hw submission") the
global seqno advance was deferred until the submit_request callback.
After wedging the GPU, we were installing a nop_submit_request handler
(to avoid waking up the dead hw) but I had missed converting this over
to the new scheme. Under the new scheme, we have to explicitly call
i915_gem_submit_request() from the submit_request handler to mark the
request as on the hardware. If we don't the request is always pending,
and any waiter will continue to wait indefinitely and hangcheck will not
be able to resolve the lockup.
References: https://bugs.freedesktop.org/show_bug.cgi?id=98748
Testcase: igt/gem_eio/in-flight
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>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161122144121.7379-3-chris@chris-wilson.co.uk
When a user context is closed, it's file_priv backpointer is replaced by
ERR_PTR(-EBADF); be careful not to chase this invalid pointer after a
hang and a GPU reset.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Fixes: b083a0870c ("drm/i915: Add per client max context ban limit")
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161122144121.7379-1-chris@chris-wilson.co.uk
If we have a bad client submitting unfavourably across different
contexts, creating new ones, the per context scoring of badness
doesn't remove the root cause, the offending client.
To counter, keep track of per client context bans. Deny access if
client is responsible for more than 3 context bans in
it's lifetime.
v2: move ban check to context create ioctl (Chris)
v3: add commentary about hangs needed to reach client ban (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Now when driver has per context scoring of 'hanging badness'
and also subsequent hangs during short windows are allowed,
if there is progress made in between, it does not make sense
to expose a ban timing window as a context parameter anymore.
Let the scoring be the sole indicator for ban policy and substitute
ban period context parameter as a boolean to get/set context
bannable property.
v2: allow non root to opt into being banned (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
As hangcheck score was removed, the active decay of score
was removed also. This removed feature for hangcheck to detect
if the gpu client was accidentally or maliciously causing intermittent
hangs. Reinstate the scoring as a per context property, so that if
one context starts to act unfavourably, ban it.
v2: ban_period_secs as a gate to score check (Chris)
v3: decay in proper spot. scores as tunables (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Hangcheck state accumulation has gained more steps
along the years, like head movement and more recently the
subunit inactivity check. As the subunit sampling is only
done if the previous state check showed inactivity, we
have added more stages (and time) to reach a hang verdict.
Asymmetric engine states led to different actual weight of
'one hangcheck unit' and it was demonstrated in some
hangs that due to difference in stages, simpler engines
were accused falsely of a hang as their scoring was much
more quicker to accumulate above the hang treshold.
To completely decouple the hangcheck guilty score
from the hangcheck period, convert hangcheck score to a
rough period of inactivity measurement. As these are
tracked as jiffies, they are meaningful also across
reset boundaries. This makes finding a guilty engine
more accurate across multi engine activity scenarios,
especially across asymmetric engines.
We lose the ability to detect cross batch malicious attempts
to hinder the progress. Plan is to move this functionality
to be part of context banning which is more natural fit,
later in the series.
v2: use time_before macros (Chris)
reinstate the pardoning of moving engine after hc (Chris)
v3: avoid global state for per engine stall detection (Chris)
v4: take timeline last retirement into account (Chris)
v5: do debug print on pardoning, split out retirement timestamp (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Currently we only clflush the scanout if it is in the CPU domain. Also
flush if we have a pending CPU clflush. We also want to treat the
dirtyfb path similar, and flush any pending writes there as well.
v2: Only send the fb flush message if flushing the dirt on flip
v3: Make flush-for-flip and dirtyfb look more alike since they serve
similar roles as end-of-frame marker.
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/20161118211747.25197-1-chris@chris-wilson.co.uk
On the DMA mapping error path, sg may be NULL (it has already been
marked as the last scatterlist entry), and we should avoid dereferencing
it again.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: e227330223 ("drm/i915: avoid leaking DMA mappings")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Imre Deak <imre.deak@intel.com>
Cc: stable@vger.kernel.org
Link: http://patchwork.freedesktop.org/patch/msgid/20161114112930.2033-1-chris@chris-wilson.co.uk
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
I tried to avoid having to track the write for every VMA by only
tracking writes to the ggtt. However, for the purposes of frontbuffer
tracking this is insufficient as we need to invalidate around writes not
just to the the ggtt but all aliased ppgtt views of the framebuffer. By
moving the critical section to the object and only doing so for
framebuffer writes we can reduce the tracking even further by only
watching framebuffers and not vma.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161116190704.5293-1-chris@chris-wilson.co.uk
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
We use DRM_DEBUG() when reporting on user actions, to try and keep
intentional errors out of the CI dmesg. Demote the debug from
i915_gem_open() similarly so that it is only apparent with drm.debug & 1
like its brethren.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20161109104507.21228-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Started with removing INTEL_INFO(dev) and cascaded into a quite
big trickle of function prototype changes. Still, I think it is
for the better.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Plus a small cascade of function prototype changes.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Boost the priority of any rendering required to show the next pageflip
as we want to avoid missing the vblank by being delayed by invisible
workload. We prioritise avoiding jank and jitter in the GUI over
starving background tasks.
v2: Descend dma_fence_array when boosting priorities.
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/20161114204105.29171-10-chris@chris-wilson.co.uk
Track the priority of each request and use it to determine the order in
which we submit requests to the hardware via execlists.
The priority of the request is determined by the user (eventually via
the context) but may be overridden at any time by the driver. When we set
the priority of the request, we bump the priority of all of its
dependencies to match - so that a high priority drawing operation is not
stuck behind a background task.
When the request is ready to execute (i.e. we have signaled the submit
fence following completion of all its dependencies, including third
party fences), we put the request into a priority sorted rbtree to be
submitted to the hardware. If the request is higher priority than all
pending requests, it will be submitted on the next context-switch
interrupt as soon as the hardware has completed the current request. We
do not currently preempt any current execution to immediately run a very
high priority request, at least not yet.
One more limitation, is that this is first implementation is for
execlists only so currently limited to gen8/gen9.
v2: Replace recursive priority inheritance bumping with an iterative
depth-first search list.
v3: list_next_entry() for walking lists
v4: Explain how the dfs solves the recursion problem with PI.
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/20161114204105.29171-8-chris@chris-wilson.co.uk
The scheduler needs to know the dependencies of each request for the
lifetime of the request, as it may choose to reschedule the requests at
any time and must ensure the dependency tree is not broken. This is in
additional to using the fence to only allow execution after all
dependencies have been completed.
One option was to extend the fence to support the bidirectional
dependency tracking required by the scheduler. However the mismatch in
lifetimes between the submit fence and the request essentially meant
that we had to build a completely separate struct (and we could not
simply reuse the existing waitqueue in the fence for one half of the
dependency tracking). The extra dependency tracking simply did not mesh
well with the fence, and keeping it separate both keeps the fence
implementation simpler and allows us to extend the dependency tracking
into a priority tree (whilst maintaining support for reordering the
tree).
To avoid the additional allocations and list manipulations, the use of
the priotree is disabled when there are no schedulers to use it.
v2: Create a dedicated slab for i915_dependency.
Rename the lists.
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/20161114204105.29171-7-chris@chris-wilson.co.uk
In order to simplify the lockdep annotation, as they become more complex
in the future with deferred execution and multiple paths through the
same functions, create a separate lockclass for the user timeline and
the hardware execution timeline.
We should only ever be locking the user timeline and the execution
timeline in parallel so we only need to create two lock classes, rather
than a separate class for every timeline.
v2: Rename the lock classes to be more consistent with other lockdep.
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/20161114204105.29171-2-chris@chris-wilson.co.uk
When we release the shmem backing storage, we make sure that the pages
are coherent with the cpu cache. However, our clflush routine was
skipping the flush as the object had no pages at release time. Fix this by
explicitly flushing the sg_table we are decoupling.
Fixes: 03ac84f183 ("drm/i915: Pass around sg_table to get_pages/put_pages backend")
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/20161111145809.9701-2-chris@chris-wilson.co.uk
A small selection of macros which can only accept dev_priv from
now on and a resulting trickle of fixups.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: David Weinehall <david.weinehall@linux.intel.com>
As a side product, had to split two other files;
- i915_gem_fence_reg.h
- i915_gem_object.h (only parts that needed immediate untanglement)
I tried to move code in as big chunks as possible, to make review
easier. i915_vma_compare was moved to a header temporarily.
v2:
- Use i915_gem_fence_reg.{c,h}
v3:
- Rebased
v4:
- Fix building when DEBUG_GEM is enabled by reordering a bit.
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1478861034-30643-1-git-send-email-joonas.lahtinen@linux.intel.com
At the moment we allocate enough sg table entries assuming we
will not be able to do any coalescing. But since in practice
we most often can, and more so very effectively, this ends up
wasting a lot of memory.
A simple and effective way of trimming the over-allocated
entries is to copy the table over to a new one allocated to the
exact size.
Experiments on my freshly logged and idle desktop (KDE) showed
that by doing this we can save approximately 1 MiB of RAM, or
when running a typical benchmark like gl_manhattan I have
even seen a 6 MiB saving.
More complicated techniques such as only copying the last used
page and freeing the rest are left to the reader.
v2:
* Update commit message.
* Use temporary sg_table on stack. (Chris Wilson)
v3:
* Commit message update.
* Comment added.
* Replace memcpy with copy assignment.
(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/1478704423-7447-1-git-send-email-tvrtko.ursulin@linux.intel.com
We always flush the chipset prior to executing with the GPU, so we can
skip the flush during ordinary domain management.
This should help mitigate some of the potential performance regressions,
but likely trivial, from doing the flush unconditionally before execbuf
introduced in commit dcd79934b0 ("drm/i915: Unconditionally flush any
chipset buffers before execbuf")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20161106130001.9509-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
During resume we will reset the SW/HW tracking for each ring head/tail
pointers and so are not prepared to replay any pending requests (as
opposed to GPU reset time). Add an assert for this both to the suspend
and the resume code.
v2:
- Check for ELSP port idle already during suspend and check !gt.awake
during resume. (Chris)
v3:
- Move the !gt.awake check to i915_gem_resume().
v4:
- s/intel_lr_engines_idle/intel_execlists_idle/ (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
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/1478510405-11799-4-git-send-email-imre.deak@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
There is a small race where a new request can be submitted and retired
after the idle worker started to run which leads to idling the GPU too
early. Fix this by deferring the idling to the pending instance of the
worker.
This scenario was pointed out by Chris.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
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/1478510405-11799-2-git-send-email-imre.deak@intel.com
Valleyview appears to be limited to only scanning out from the first 512MiB
of the Global GTT. Lets presume that this behaviour was inherited from the
display block copied from g4x (not Ironlake) and all earlier generations
are similarly affected, though testing suggests different symptoms. For
simplicity, impose that these platforms must scanout from the mappable
region. (For extra simplicity, use HAS_GMCH_DISPLAY even though this
catches Cherryview which does not appear to be limited to the low
aperture for its scanout.)
v2: Use HAS_GMCH_DISPLAY() to more clearly convey my intent about
limiting this workaround to the old style of display engine.
v3: Update changelog to reflect testing by Ville Syrjälä
v4: Include the changes to the comments as well
Reported-by: Luis Botello <luis.botello.ortega@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98036
Fixes: 2efb813d53 ("drm/i915: Fallback to using unmappable memory for scanout")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Akash Goel <akash.goel@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.9-rc1+
Link: http://patchwork.freedesktop.org/patch/msgid/20161107110128.28762-1-chris@chris-wilson.co.uk
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com
When we split a large object up into chunks for GTT faulting (because we
can't fit the whole object into the aperture) we have to align our cuts
with the fence registers. Each partial VMA must cover a complete set of
tile rows or the offset into each partial VMA is not aligned with the
whole image. Currently we enforce a minimum size on each partial VMA,
but this minimum size itself was not aligned to the tile row causing
distortion.
Reported-by: Andreas Reis <andreas.reis@gmail.com>
Reported-by: Chris Clayton <chris2553@googlemail.com>
Reported-by: Norbert Preining <preining@logic.at>
Tested-by: Norbert Preining <preining@logic.at>
Tested-by: Chris Clayton <chris2553@googlemail.com>
Fixes: 03af84fe7f ("drm/i915: Choose partial chunksize based on tile row size")
Fixes: a61007a83a ("drm/i915: Fix partial GGTT faulting") # enabling patch
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98402
Testcase: igt/gem_mmap_gtt/medium-copy-odd
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.9-rc1+
Link: http://patchwork.freedesktop.org/patch/msgid/20161107105443.27855-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
commit bc0629a767 ("drm/i915: Track pages pinned due to swizzling
quirk") fixed one problem, but revealed a whole lot more. The root cause
of the pin count mismatch for the swizzle quirk (for L-shaped memory on
gen3/4) was that we were incrementing the pages_pin_count upon getting
the backing pages but then overwriting the pages_pin_count to set it to
1 afterwards. With a little bit of adjustment to satisfy the GEM_BUG_ON
sanitychecks, the fix is to replace the explicit atomic_set with an
atomic_inc.
v2: Consistently use atomics (not mix atomics and helpers) within the
lowlevel get_pages routines. This makes the atomic operations much
clearer.
Fixes: 1233e2db19 ("drm/i915: Move object backing storage manipulation")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161104103001.27643-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Commit 1bec9b0bda ("drm/i915/shrinker: Only shmemfs objects
are backed by swap") stopped considering the userptr objects
in shrinker callbacks.
Restore that so idle userptr objects can be discarded in order
to free up memory.
One change further to what was introduced in 1bec9b0bda is
to start considering userptr objects in oom but that should
also be a correct thing to do.
v2: Introduce I915_GEM_OBJECT_IS_SHRINKABLE. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 1bec9b0bda ("drm/i915/shrinker: Only shmemfs objects are backed by swap")
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: <stable@vger.kernel.org>
Link: http://patchwork.freedesktop.org/patch/msgid/1478011450-6634-1-git-send-email-tvrtko.ursulin@linux.intel.com
The shrinker may appear to recurse into obj->mm.lock as the shrinker may
be called from a direct reclaim path whilst handling get_pages. We
filter out recursing on the same obj->mm.lock by inspecting
obj->mm.pages, but we do want to take the lock on a second object in
order to reap their pages. lockdep spots the recursion on the same
lockclass and needs annotation to avoid a false positive. To keep the
two paths distinct, create an enum to indicate which subclass of
obj->mm.lock we are using. This removes the false positive and avoids
masking real bugs.
Suggested-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161101121134.27504-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
With full-ppgtt one of the main bottlenecks is the lookup of the VMA
underneath the object. For execbuf there is merit in having a very fast
direct lookup of ctx:handle to the vma using a hashtree, but that still
leaves a large number of other lookups. One way to speed up the lookup
would be to use a rhashtable, but that requires extra allocations and
may exhibit poor worse case behaviour. An alternative is to use an
embedded rbtree, i.e. no extra allocations and deterministic behaviour,
but at the slight cost of O(lgN) lookups (instead of O(1) for
rhashtable). The major of such tree will be very shallow and so not much
slower, and still scales much, much better than the current unsorted
list.
v2: Bump vma_compare() to return a long, as we return the result of
comparing two pointers.
References: https://bugs.freedesktop.org/show_bug.cgi?id=87726
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/20161101115400.15647-1-chris@chris-wilson.co.uk
If we have a tiled object and an unknown CPU swizzle pattern, we pin the
pages to prevent the object from being swapped out (and us corrupting
the contents as we do not know the access pattern and so cannot convert
it to linear and back to tiled on reuse). This requires us to remember
to drop the extra pinning when freeing the object, or else we trigger
warnings about the pin leak. In commit fbbd37b36f ("drm/i915: Move
object release to a freelist + worker"), the object free path was
deferred to a worker, but the unpinning of the quirk, along with marking
the object as reclaimable, was left on the immediate path (so that if
required we could reclaim the pages under memory pressure as early as
possible). However, this split introduced a bug where the pages were no
longer being unpinned if they were marked as unneeded.
[ 231.800401] WARNING: CPU: 1 PID: 90 at drivers/gpu/drm/i915/i915_gem.c:4275 __i915_gem_free_objects+0x326/0x3c0 [i915]
[ 231.800403] WARN_ON(i915_gem_object_has_pinned_pages(obj))
[ 231.800405] Modules linked in:
[ 231.800406] snd_hda_intel i915 snd_hda_codec_generic mei_me snd_hda_codec coretemp snd_hwdep mei lpc_ich snd_hda_core snd_pcm e1000e ptp pps_core [last unloaded: i915]
[ 231.800426] CPU: 1 PID: 90 Comm: kworker/1:4 Tainted: G U 4.9.0-rc2-CI-CI_DRM_1780+ #1
[ 231.800428] Hardware name: LENOVO 7465CTO/7465CTO, BIOS 6DET44WW (2.08 ) 04/22/2009
[ 231.800456] Workqueue: events __i915_gem_free_work [i915]
[ 231.800459] ffffc9000034fc80 ffffffff8142dd65 ffffc9000034fcd0 0000000000000000
[ 231.800465] ffffc9000034fcc0 ffffffff8107e4e6 000010b300000001 0000000000001000
[ 231.800469] ffff88011d3db740 ffff880130ef0000 0000000000000000 ffff880130ef5ea0
[ 231.800474] Call Trace:
[ 231.800479] [<ffffffff8142dd65>] dump_stack+0x67/0x92
[ 231.800484] [<ffffffff8107e4e6>] __warn+0xc6/0xe0
[ 231.800487] [<ffffffff8107e54a>] warn_slowpath_fmt+0x4a/0x50
[ 231.800491] [<ffffffff811d12ac>] ? kmem_cache_free+0x2dc/0x340
[ 231.800520] [<ffffffffa009ef36>] __i915_gem_free_objects+0x326/0x3c0 [i915]
[ 231.800548] [<ffffffffa009effe>] __i915_gem_free_work+0x2e/0x50 [i915]
[ 231.800552] [<ffffffff8109c27c>] process_one_work+0x1ec/0x6b0
[ 231.800555] [<ffffffff8109c1f6>] ? process_one_work+0x166/0x6b0
[ 231.800558] [<ffffffff8109c789>] worker_thread+0x49/0x490
[ 231.800561] [<ffffffff8109c740>] ? process_one_work+0x6b0/0x6b0
[ 231.800563] [<ffffffff8109c740>] ? process_one_work+0x6b0/0x6b0
[ 231.800566] [<ffffffff810a2aab>] kthread+0xeb/0x110
[ 231.800569] [<ffffffff810a29c0>] ? kthread_park+0x60/0x60
[ 231.800573] [<ffffffff818164a7>] ret_from_fork+0x27/0x40
Moving to a separate flag for tracking the quirked pin is overkill for
the bug (since we only have to interchange the two tests in
i915_gem_free_object) but it does reduce a complicated test on all
objects and provide a sanitycheck for uncommon code paths.
Fixes: fbbd37b36f ("drm/i915: Move object release to a freelist + worker")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161101100317.11129-2-chris@chris-wilson.co.uk
With the infrastructure converted over to tracking multiple timelines in
the GEM API whilst preserving the efficiency of using a single execution
timeline internally, we can now assign a separate timeline to every
context with full-ppgtt.
v2: Add a comment to indicate the xfer between timelines upon submission.
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/20161028125858.23563-35-chris@chris-wilson.co.uk
A restriction on our global seqno is that they cannot wrap, and that we
cannot use the value 0. This allows us to detect when a request has not
yet been submitted, its global seqno is still 0, and ensures that
hardware semaphores are monotonic as required by older hardware. To
meet these restrictions when we defer the assignment of the global
seqno, we must check that we have an available slot in the global seqno
space during request construction. If that test fails, we wait for all
requests to be completed and reset the hardware back to 0.
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/20161028125858.23563-33-chris@chris-wilson.co.uk
Though we will have multiple timelines, we still have a single timeline
of execution. This we can use to provide an execution and retirement order
of requests. This keeps tracking execution of requests simple, and vital
for preserving a single waiter (i.e. so that we can order the waiters so
that only the earliest to wakeup need be woken). To accomplish this we
distinguish the seqno used to order requests per-context (external) and
that used internally for execution.
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/20161028125858.23563-26-chris@chris-wilson.co.uk
Before suspend, we wait for the switch to the kernel context. In order
for all the other context images to be complete upon suspend, that
switch must be the last operation by the GPU (i.e. this idling request
must not overtake any pending requests). To make this request execute last,
we make it depend on every other inflight request.
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/20161028125858.23563-24-chris@chris-wilson.co.uk
Our timelines are more than just a seqno. They also provide an ordered
list of requests to be executed. Due to the restriction of handling
individual address spaces, we are limited to a timeline per address
space but we use a fence context per engine within.
Our first step to introducing independent timelines per context (i.e. to
allow each context to have a queue of requests to execute that have a
defined set of dependencies on other requests) is to provide a timeline
abstraction for the global execution queue.
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/20161028125858.23563-23-chris@chris-wilson.co.uk
In preparation to support many distinct timelines, we need to expand the
activity tracking on the GEM object to handle more than just a request
per engine. We already use the struct reservation_object on the dma-buf
to handle many fence contexts, so integrating that into the GEM object
itself is the preferred solution. (For example, we can now share the same
reservation_object between every consumer/producer using this buffer and
skip the manual import/export via dma-buf.)
v2: Reimplement busy-ioctl (by walking the reservation object), postpone
the ABI change for another day. Similarly use the reservation object to
find the last_write request (if active and from i915) for choosing
display CS flips.
Caveats:
* busy-ioctl: busy-ioctl only reports on the native fences, it will not
warn of stalls (in set-domain-ioctl, pread/pwrite etc) if the object is
being rendered to by external fences. It also will not report the same
busy state as wait-ioctl (or polling on the dma-buf) in the same
circumstances. On the plus side, it does retain reporting of which
*i915* engines are engaged with this object.
* non-blocking atomic modesets take a step backwards as the wait for
render completion blocks the ioctl. This is fixed in a subsequent
patch to use a fence instead for awaiting on the rendering, see
"drm/i915: Restore nonblocking awaits for modesetting"
* dynamic array manipulation for shared-fences in reservation is slower
than the previous lockless static assignment (e.g. gem_exec_lut_handle
runtime on ivb goes from 42s to 66s), mainly due to atomic operations
(maintaining the fence refcounts).
* loss of object-level retirement callbacks, emulated by VMA retirement
tracking.
* minor loss of object-level last activity information from debugfs,
could be replaced with per-vma information if desired
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/20161028125858.23563-21-chris@chris-wilson.co.uk
