Chris Wilson [Thu, 2 Apr 2020 12:42:18 +0000 (13:42 +0100)]
drm/i915/gem: Utilize rcu iteration of context engines
Now that we can peek at GEM->engines[] and obtain a reference to them
using RCU, do so for instances where we can safely iterate the
potentially old copy of the engines. For setting, we can do this when we
know the engine properties are copied over before swapping, so we know
the new engines already have the global property and we update the old
before they are discarded. For reading, we only need to be safe; as we
do so on behalf of the user, their races are their own problem.
Chris Wilson [Wed, 1 Apr 2020 11:04:34 +0000 (12:04 +0100)]
drm/i915/execlists: Peek at the next submission for error interrupts
If we receive the error interrupt before the CS interrupt, we may find
ourselves without an active request to reset, skipping the GPU reset.
All because the attempt to reset was too early.
Chris Wilson [Mon, 30 Mar 2020 21:22:54 +0000 (22:22 +0100)]
drm/i915/uc: Cleanup kerneldoc warnings
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c:205: warning: Excess function parameter 'supported' description in 'intel_uc_fw_init_early'
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c:205: warning: Excess function parameter 'platform' description in 'intel_uc_fw_init_early'
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c:205: warning: Excess function parameter 'rev' description in 'intel_uc_fw_init_early'
drivers/gpu/drm/i915/gt/uc/intel_guc_log.c:696: warning: Function parameter or member 'log' not described in 'intel_guc_log_info'
drivers/gpu/drm/i915/gt/uc/intel_guc_log.c:696: warning: Excess function parameter 'guc' description in 'intel_guc_log_info'
Chris Wilson [Wed, 1 Apr 2020 22:39:24 +0000 (23:39 +0100)]
drm/i915/gem: Drop cached obj->bind_count
We cached the number of vma bound to the object in order to speed up
shrinker decisions. This has been superseded by being more proactive in
removing objects we cannot shrink from the shrinker lists, and so we can
drop the clumsy attempt at atomically counting the bind count and
comparing it to the number of pinned mappings of the object. This will
only get more clumsier with asynchronous binding and unbinding.
Chris Wilson [Wed, 1 Apr 2020 21:01:04 +0000 (22:01 +0100)]
drm/i915/gt: Make fence revocation unequivocal
If we must revoke the fence because the VMA is no longer present, or
because the fence no longer applies, ensure that we do and convert it
into an error if we try but cannot.
Chris Wilson [Wed, 1 Apr 2020 21:01:02 +0000 (22:01 +0100)]
drm/i915/gt: Only wait for GPU activity before unbinding a GGTT fence
Only GPU activity via the GGTT fence is asynchronous, we know that we
control the CPU access directly, so we only need to wait for the GPU to
stop using the fence before we relinquish it.
Chris Wilson [Wed, 1 Apr 2020 19:41:35 +0000 (20:41 +0100)]
drm/i915/gem: Try allocating va from free space
If the current node/entry location is occupied, and the object is not
pinned, try assigning it some free space. We cannot wait here, so if in
doubt, we unreserve and try to grab all at once.
v2: Use the final pin_flags so that we won't have to move the object if
we find the wrong free space.
Chris Wilson [Tue, 31 Mar 2020 16:21:50 +0000 (17:21 +0100)]
drm/i915/gem: Ignore readonly failures when updating relocs
If the user passes in a readonly reloc[], by the time we notice we have
already committed to modifying the execobjects, or have indeed done so
already. Reporting the failure just compounds the issue as we have no
second pass to fall back to anymore.
"Be damned if you do, and damned if you don't."
Testcase: igt/gem_exec_reloc/readonly Fixes: 7dc8f1143778 ("drm/i915/gem: Drop relocation slowpath")
References: fddcd00a49e9 ("drm/i915: Force the slow path after a user-write error") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matthew Auld <matthew.auld@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200331162150.3635-1-chris@chris-wilson.co.uk
Chris Wilson [Tue, 31 Mar 2020 15:23:48 +0000 (16:23 +0100)]
drm/i915/gt: Fill all the unused space in the GGTT
When we allocate space in the GGTT we may have to allocate a larger
region than will be populated by the object to accommodate fencing. Make
sure that this space beyond the end of the buffer points safely into
scratch space, in case the HW tries to access it anyway (e.g. fenced
access to the last tile row).
v2: Preemptively / conservatively guard gen6 ggtt as well.
Reported-by: Imre Deak <imre.deak@intel.com>
References: https://gitlab.freedesktop.org/drm/intel/-/issues/1554 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matthew Auld <matthew.auld@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: stable@vger.kernel.org Reviewed-by: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200331152348.26946-1-chris@chris-wilson.co.uk
Chris Wilson [Tue, 31 Mar 2020 09:14:57 +0000 (10:14 +0100)]
drm/i915/execlists: Pause CS flow before reset
Since we may be attempting to reset an active engine, we try to freeze
it in place before resetting -- to be on the safe side. We can go one
step further if we are using the CS flow semaphore to prevent the
context switching into the next.
Chris Wilson [Tue, 31 Mar 2020 11:48:52 +0000 (12:48 +0100)]
drm/i915: Defer kicking the tasklet until all rescheduling is complete
Since we may kick more than engine, and may kick each one a couple of
times, coalesce the tasklet execution to the end. This also ensures that
we have the chance to run the tasklet immediately after priority
bumping.
DDI ports have its encoders initialized with INTEL_OUTPUT_DDI type and
later eDP ports that have the type changed to INTEL_OUTPUT_EDP.
But for all other DDI ports it can drive HDMI or DP depending on what
user connects to the ports.
ehl_get_combo_buf_trans() and tgl_get_combo_buf_trans() was checking
for INTEL_OUTPUT_DP that was never true, causing wrong vswing tables
being used.
So here replacing the INTEL_OUTPUT_DP checks by the valid output types
that this functions receives as parameters. HDMI cases will be
correctly handled as it do not use encoder->type, instead it calls the
functions with INTEL_OUTPUT_HDMI as type parameter and HDMI don't have
retraining.
v2:
changed INTEL_OUTPUT_DDI to INTEL_OUTPUT_EDP and INTEL_OUTPUT_HDMI
Fixes: bd3cf6f7ce20 ("drm/i915/dp/tgl+: Update combo phy vswing tables") Cc: Clinton A Taylor <clinton.a.taylor@intel.com> Cc: Matt Roper <matthew.d.roper@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200330210044.130510-1-jose.souza@intel.com
In this sequence, we can see that although we have submitted the next
request [ff84:45156] to HW (via ELSP[]) it has not yet reported the
lite-restore. Instead, we see the completion event of the currently
active request [ff84:45155] but at the time of processing that event,
the breadcrumb has not yet been written. Though by the time we do print
out the debug info, the seqno write of ff84:45156 has landed!
Therefore there is a serialisation problem between the seqno writes and
CS events, not just between the CS buffer and its head/tail pointers as
previously observed on Icelake.
This is not a huge problem, as we don't strictly rely on the breadcrumb
to determine HW activity, but it may indicate that interrupt delivery is
before the seqno write, aka bringing back the plague of missed
interrupts from yesteryear. However, there is no indication of this
wider problem, so let's just flush the seqno read before reporting an
error. If it persists after the fresh read we can worry again.
Chris Wilson [Mon, 30 Mar 2020 12:58:27 +0000 (13:58 +0100)]
drm/i915/execlists: Explicitly reset both reg and context runtime
Upon a GPU reset, we copy the default context image over top of the
guilty image. This will rollback the CTX_TIMESTAMP register to before
our value of ce->runtime.last. Reset both back to 0 so that we do not
encounter an underflow on the next schedule out after resume.
This should not be a huge issue in practice, as hangs should be rare in
correct code.
Chris Wilson [Mon, 30 Mar 2020 13:37:10 +0000 (14:37 +0100)]
drm/i915/gem: Split eb_vma into its own allocation
Use a separate array allocation for the execbuf vma, so that we can
track their lifetime independently from the copy of the user arguments.
With luck, this has a secondary benefit of splitting the malloc size to
within reason and avoid vmalloc. The downside is that we might require
two separate vmallocs -- but much less likely.
In the process, this prevents a memory leak on the ww_mutex error
unwind.
Chris Wilson [Fri, 27 Mar 2020 11:22:12 +0000 (11:22 +0000)]
drm/i915/perf: Schedule oa_config after modifying the contexts
We wish that the scheduler emit the context modification commands prior
to enabling the oa_config, for which we must explicitly inform it of the
ordering constraints. This is especially important as we now wait for
the final oa_config setup to be completed and as this wait may be on a
distinct context to the state modifications, we need that command packet
to be always last in the queue.
We borrow the i915_active for its ability to track multiple timelines
and the last dma_fence on each; a flexible dma_resv. Keeping track of
each dma_fence is important for us so that we can efficiently schedule
the requests and reprioritise as required.
Chris Wilson [Fri, 27 Mar 2020 11:22:11 +0000 (11:22 +0000)]
drm/i915: Wrap i915_active in a simple kreffed struct
For conveniences of callers that just want to use an i915_active to
track a wide array of concurrent timelines, wrap the base i915_active
struct inside a kref. This i915_active will self-destruct after use.
Chris Wilson [Mon, 30 Mar 2020 12:16:44 +0000 (13:16 +0100)]
drm/i915/selftests: Check timeout before flush and cond checks
Allow a bit of leniency for the CPU scheduler to be distracted while we
flush the tasklet and so ensure that we always check the status of the
request once more before timing out.
v2: Wait until the HW acked the submit, and we do any secondary actions
for the submit (e.g. timeslices)
Michal Wajdeczko [Mon, 30 Mar 2020 11:33:02 +0000 (13:33 +0200)]
drm/i915/huc: Add more errors for I915_PARAM_HUC_STATUS
There might be many reasons why we failed to successfully
load and authenticate HuC firmware, but today we only use
single error in case of no HuC hardware. Add some more
error codes for most common cases (disabled, not installed,
corrupted or mismatched firmware).
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Tony Ye <tony.ye@intel.com> Cc: Robert M. Fosha <robert.m.fosha@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/20200330113302.1670-1-michal.wajdeczko@intel.com
Chris Wilson [Fri, 27 Mar 2020 20:14:33 +0000 (20:14 +0000)]
drm/i915/execlists: Workaround switching back to a completed context
In what seems remarkably similar to the w/a required to not reload an
idle context with HEAD==TAIL, it appears we must prevent the HW from
switching to an idle context in ELSP[1], while simultaneously trying to
preempt the HW to run another context and a continuation of the idle
context (which is no longer idle).
We can achieve this by preventing the context from completing while we
reload a new ELSP (by applying ring_set_paused(1) across the whole of
dequeue), except this eventually fails due to a lite-restore into a
waiting semaphore does not generate an ACK. Instead, we try to avoid
making the GPU do anything too challenging and not submit a new ELSP
while the interrupts + CSB events appear to have fallen behind the
completed contexts. We expect it to catch up shortly so we queue another
tasklet execution and hope for the best.
Ville Syrjälä [Fri, 13 Mar 2020 16:48:19 +0000 (18:48 +0200)]
drm/i915/mst: Use .compute_config_late() to compute master transcoder
Use the recently introduced encoder .compute_config_late() hook to
do the MST master transcoder assignment. Avoids having to do it
in a funny way before we know the CPU transcoder of each pipe.
And now we can also properly use hw.active instead of uapi.active
since it too has been calculated earlier for everyone.
Chris Wilson [Thu, 26 Mar 2020 14:27:27 +0000 (14:27 +0000)]
drm/i915: Differentiate between aliasing-ppgtt and ggtt pinning
Userptr causes lockdep to complain when we are using the aliasing-ppgtt
(and ggtt, but for that it is rightfully so to complain about) in that
when we revoke the userptr we take a mutex which we also use to revoke
the mmaps. However, we only revoke mmaps for GGTT bindings and we never
allow userptr to create a GGTT binding so the warning should be false
and is simply caused by our conflation of the aliasing-ppgtt with the
ggtt. So lets try treating the binding into the aliasing-ppgtt as a
separate lockclass from the ggtt. The downside is that we are
deliberately suppressing lockdep;s ability to warn us of cycles.
Ville Syrjälä [Tue, 10 Mar 2020 20:27:52 +0000 (22:27 +0200)]
drm/i915: Get rid of silly void* from MST code
Not sure why this thing is trying to avoid declaring the proper
type for these pointers. But since these are used only once let's
just get rid of the local variable entirely.
Kai Vehmanen [Tue, 24 Mar 2020 15:32:12 +0000 (17:32 +0200)]
drm/i915: use forced codec wake on all gen9+ platforms
Commit 632f3ab95fe2 ("drm/i915/audio: add codec wakeup override
enabled/disable callback"), added logic to toggle Codec Wake on gen9.
This is used by audio driver when it resets the HDA controller.
It seems explicit toggling of the wakeline can help to fix problems
with probe failing on some gen12 platforms. And based on specs, there
is no reason why this programming sequence should not be applied to all
gen9+ platforms. No side-effects are seen on gen10/11. So apply
the wake-logic to all gen9+ platforms.
drm/i915/perf: add new open param to configure polling of OA buffer
This new parameter let's the application choose how often the OA
buffer should be checked on the CPU side for data availability. Longer
polling period tend to reduce CPU overhead if the application does not
care about somewhat real time data collection.
v2: Allow disabling polling completely with 0 value (Lionel)
v3: Version the new parameter (Joonas)
v4: Rebase (Umesh)
v5: Make poll delay value of 0 invalid (Umesh)
v6:
- Describe poll_oa_period (Ashutosh)
- Fix comment for new poll parameter (Lionel)
- Drop open_flags in read_properties_unlocked (Lionel)
- Rename uapi parameter (Ashutosh)
v7: Reword the comment in uapi (Ashutosh)
We're about to introduce an options to open the perf stream, giving
the user ability to configure how often it wants the kernel to poll
the OA registers for available data.
Right now the workaround against the OA tail pointer race condition
requires at least twice the internal kernel polling timer to make any
data available.
This changes introduce checks on the OA data written into the circular
buffer to make as much data as possible available on the first
iteration of the polling timer.
v2: Use OA_TAKEN macro without the gtt_offset (Lionel)
v3: (Umesh)
- Rebase
- Change report to report32 from below review
https://patchwork.freedesktop.org/patch/330704/?series=66697&rev=1
v4: (Ashutosh, Lionel)
- Fix checkpatch errors
- Fix aging_timestamp initialization
- Check for only one valid landed report
- Fix check for unlanded report
v5: (Ashutosh)
- Fix bug in accurately determining landed report.
- Optimize the check for landed reports by going as far as the
previously determined aged tail.
drm/i915: Cast remain to unsigned long in eb_relocate_vma
A recent commit in clang added -Wtautological-compare to -Wall, which is
enabled for i915 after -Wtautological-compare is disabled for the rest
of the kernel so we see the following warning on x86_64:
../drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c:1433:22: warning:
result of comparison of constant 576460752303423487 with expression of
type 'unsigned int' is always false
[-Wtautological-constant-out-of-range-compare]
if (unlikely(remain > N_RELOC(ULONG_MAX)))
~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
../include/linux/compiler.h:78:42: note: expanded from macro 'unlikely'
# define unlikely(x) __builtin_expect(!!(x), 0)
^
1 warning generated.
It is not wrong in the case where ULONG_MAX > UINT_MAX but it does not
account for the case where this file is built for 32-bit x86, where
ULONG_MAX == UINT_MAX and this check is still relevant.
Cast remain to unsigned long, which keeps the generated code the same
(verified with clang-11 on x86_64 and GCC 9.2.0 on x86 and x86_64) and
the warning is silenced so we can catch more potential issues in the
future.
We need to keep the GuC error logs around to debug the load failure,
so we can't clean them in the error unwind, which includes uc_fini().
Moving the cleanup to driver remove ensures that the logs stick around
long enough for us to dump them.
drm/i915/uc: Move uC debugfs to its own folder under GT
uC is a component of the GT, so it makes sense for the uC debugfs files
to be in the GT folder. A subfolder has been used to keep the same
structure we have for the code.
v2: use intel_* prefix (Jani), rebase on new gt_debugfs_register_files,
fix permissions for writable debugfs files.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Andi Shyti <andi.shyti@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Tony Ye <tony.ye@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> #v2 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20200326181121.16869-6-daniele.ceraolospurio@intel.com
drm/i915/debugfs: move uC printers and update debugfs file names
Move the printers to the respective files for clarity. The
guc_load_status debugfs has been squashed in the guc_info one, has
having separate ones wasn't very useful. The HuC debugfs has been
renamed huc_info to match.
Suggested-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Tony Ye <tony.ye@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Reviewed-by: John Harrison <John.C.Harrison@Intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20200326181121.16869-5-daniele.ceraolospurio@intel.com
drm/i915/huc: make "support huc" reflect HW capabilities
We currently initialize HuC support based on GuC being enabled in
modparam; this means that huc_is_supported() can return false on HW that
does have a HuC when enable_guc=0. The rationale for this behavior is
that HuC requires GuC for authentication and therefore is not supported
by itself. However, we do not allow defining HuC fw wthout GuC fw and
selecting HuC in modparam implicitly selects GuC as well, so we can't
actually hit a scenario where HuC is selected alone. Therefore, we can
flip the support check to reflect the HW capabilities and fw
availability, which is more intuitive and will make it cleaner to log
HuC the difference between not supported in HW and not selected.
Removing the difference between GuC and HuC also allows us to simplify
the init_early, since we don't need to differentiate the support based
on the type of uC.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Reviewed-by: John Harrison <John.C.Harrison@Intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20200326181121.16869-4-daniele.ceraolospurio@intel.com
The pool will be private to GuC in the new submission scheme, so we
won't be able to print it and we can just drop the current legacy code.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20200326181121.16869-3-daniele.ceraolospurio@intel.com
Andi Shyti [Thu, 26 Mar 2020 18:11:16 +0000 (11:11 -0700)]
drm/i915/gt: allow setting generic data pointer
When registering debugfs files the intel gt debugfs library
forces a 'struct *gt' private data on the caller.
To be open to different usages make the new
"intel_gt_debugfs_register_files()"[*] function more generic by
converting the 'struct *gt' pointer to a 'void *' type.
I take the chance to rename the functions by using "intel_gt_" as
prefix instead of "debugfs_", so that "debugfs_gt_register_files()"
becomes "intel_gt_debugfs_register_files()".
Uma Shankar [Thu, 26 Mar 2020 12:51:11 +0000 (18:21 +0530)]
drm/i915/display: Fix mode private_flags comparison at atomic_check
This patch fixes the private_flags of mode to be checked and
compared against uapi.mode and not from hw.mode. This helps
properly trigger modeset at boot if desired by driver.
It helps resolve audio_codec initialization issues if display
is connected at boot. Initial discussion on this issue has happened
on below thread:
https://patchwork.freedesktop.org/series/74828/
v2: No functional change. Fixed the Closes tag and added
Maarten's RB.
v3: Added Fixes tag.
Cc: Ville Syrjä <ville.syrjala@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Kai Vehmanen <kai.vehmanen@linux.intel.com> Cc: Souza, Jose <jose.souza@intel.com> Fixes: 58d124ea2739 ("drm/i915: Complete crtc hw/uapi split, v6.") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1363 Suggested-by: Ville Syrjä <ville.syrjala@linux.intel.com> Signed-off-by: Uma Shankar <uma.shankar@intel.com> Signed-off-by: SweeAun Khor <swee.aun.khor@intel.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200326125111.11081-1-uma.shankar@intel.com
Fixes: 419190429cd1 ("drm/i915/hdmi: use struct drm_device based logging") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Wambui Karuga <wambui.karugax@gmail.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: "Ville Syrjälä" <ville.syrjala@linux.intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200326082838.16357-1-chris@chris-wilson.co.uk
Chris Wilson [Wed, 25 Mar 2020 13:00:59 +0000 (13:00 +0000)]
drm/i915/gt: Stage the transfer of the virtual breadcrumb
We move the virtual breadcrumb from one physical engine to the next, if
the next virtual request is scheduled on a new physical engine. Since
the virtual context can only be in one signal queue, we need it to track
the current physical engine for the new breadcrumbs. However, to move
the list we need both breadcrumb locks -- and since we cannot take both
at the same time (unless we are careful and always ensure consistent
ordering) stage the movement of the signaler via the current virtual
request.
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1510 Fixes: 6d06779e8672 ("drm/i915: Load balancing across a virtual engine") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200325130059.30600-1-chris@chris-wilson.co.uk
Chris Wilson [Wed, 25 Mar 2020 12:02:27 +0000 (12:02 +0000)]
drm/i915: Immediately execute the fenced work
If the caller allows and we do not have to wait for any signals,
immediately execute the work within the caller's process. By doing so we
avoid the overhead of scheduling a new task, and the latency in
executing it, at the cost of pulling that work back into the immediate
context. (Sometimes we still prefer to offload the task to another cpu,
especially if we plan on executing many such tasks in parallel for this
client.)
Chris Wilson [Wed, 25 Mar 2020 12:02:26 +0000 (12:02 +0000)]
drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission
We dropped calling process_csb prior to handling direct submission in
order to avoid the nesting of spinlocks and lift process_csb() and the
majority of the tasklet out of irq-off. However, we do want to avoid
ksoftirqd latency in the fast path, so try and pull the interrupt-bh
local to direct submission if we can acquire the tasklet's lock.
v2: Document the read of pending[0] from outside the tasklet with
READ_ONCE.
Chris Wilson [Wed, 25 Mar 2020 10:15:02 +0000 (10:15 +0000)]
drm/i915/selftests: Measure the energy consumed while in RC6
Measure and compare the energy consumed, as reported by the rapl MSR,
by the GPU while in RC0 and RC6 states. Throw an error if RC6 does not
at least halve the energy consumption of RC0, as this more than likely
means we failed to enter RC0 correctly.
If we can't measure the energy draw with the MSR, then it will report 0
for both measurements. Since the measurement works on all gen6+, this seems
worth flagging as an error.
Chris Wilson [Wed, 25 Mar 2020 10:13:58 +0000 (10:13 +0000)]
drm/i915/execlists: Drop setting sibling priority hint on virtual engines
We set the priority hint on execlists to avoid executing the tasklet for
when we know that there will be no change in execution order. However,
as we set it from the virtual engine for all siblings, but only one
physical engine may respond, we leave the hint set on the others
stopping direct submission that could take place.
If we do not set the hint, we may attempt direct submission even if we
don't expect to submit. If we set the hint, we may not do any submission
until the tasklet is run (and sometimes we may park the engine before
that has had a chance). Ergo there's only a minor ill-effect on mixed
virtual/physical engine workloads where we may try and fail to do direct
submission more often than required. (Pure virtual / engine workloads
will have redundant tasklet execution suppressed as normal.)
drm/i915/display/fbc: Make fences a nice-to-have for GEN9+
dGFX has local memory so it does not have aperture or support
CPU fences but even for iGFX it have a small number of fences.
As replacement for fences to track frontbuffer modifications by CPU
we have a software tracking that is already in used by FBC and PSR.
PSR don't support fences so it shows that this tracking is reliable.
So lets make fences a nice-to-have to activate FBC for GEN9+, this
will allow us to enable FBC for dGFXs and iGFXs even when there is no
available fence.
We do not set fences to rotated planes but FBC only have restrictions
against 16bpp, so adding it here.
Also adding a new check for the tiling format, fences are only set
to X and Y tiled planes but again FBC don't have any restrictions
against tiling so adding linear as supported as well, other formats
should be added after tested but IGT only supports drawing in thse
3 formats.
intel_fbc_hw_tracking_covers_screen() maybe can also have the same
treatment as fences but BSpec is not clear if the size limitation is
for hardware tracking or general use of FBC and I don't have a 5K
display to test it, so keeping as is for safety.
v2:
- Added tiling and pixel format rotation checks
- Changed the GEN version not requiring fences to 11 from 9, DDX
needs some changes but it don't have support for GEN11+
v3:
- Changed back to GEN9+
- Moved GEN test to inside of tiling_is_valid()
v4:
- moved rotation check to its own functions
v5:
- renamed rotations_is_valid to rotation_is_valid
- moved pre-g4x rotation check to rotation_is_valid()
Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200319211535.114625-1-jose.souza@intel.com
Chris Wilson [Tue, 24 Mar 2020 13:42:32 +0000 (13:42 +0000)]
drm/i915/gt: Select the deepest available parking mode for rc6
On Ivybridge, we can go lower than rc6 to rc6p. And this is required for
Ivybridge to hit the same minimum power consumption as rc6 on other
platforms, so make it so.
v2: Update selftest to include all rc6 residency counters
Note that Andi did mention that we should be converting the magic
numbers into opaque magic macros, so if they ever get reused (unlikely
given only Ivybridge used the extra modes) we'll need to pay back the
technical debt.
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1518 Fixes: 730eaeb52426 ("drm/i915/gt: Manual rc6 entry upon parking")
Testcase: igt/i915_pm_rc6_residency/rc6-idle Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Andi Shyti <andi.shyti@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Imre Deak <imre.deak@intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200324134232.8773-1-chris@chris-wilson.co.uk
Vandita Kulkarni [Thu, 12 Mar 2020 05:38:35 +0000 (11:08 +0530)]
drm/i915/dsi: Add cmd mode flags in display mode private flags
Adding TE flags and periodic command mode flags
as part of private flags to indicate what TE interrupts
we would be getting instead of vblanks in case of mipi dsi
command mode.
Vandita Kulkarni [Thu, 12 Mar 2020 05:38:34 +0000 (11:08 +0530)]
drm/i915/dsi: Add vblank calculation for command mode
Transcoder timing calculation differ for command mode.
v2: Use is_vid_mode, and use same I915_WRITE (Jani)
v3: Adjust the calculations to reflect dsc compression ratio
v4: Rearrange the vertical and horizontal timing calc, optimize
local variables usage. (Jani)
v5: Fix the values used for calculation, use afe_clk for
byte clock calculation, use intel_de_write/read (Jani)
Vandita Kulkarni [Thu, 12 Mar 2020 05:38:33 +0000 (11:08 +0530)]
drm/i915/dsi: Configure transcoder operation for command mode.
Configure the transcoder to operate in TE GATE command mode
and take TE events from GPIO.
Also disable the periodic command mode, that GOP would have
programmed.
v2: Disable util pin (Jani)
v3: Use intel_de_write (Jani)
Chris Wilson [Mon, 23 Mar 2020 19:20:29 +0000 (19:20 +0000)]
drm/i915/gt: Only delay the context barrier pm
It is strictly sufficient to only delay the intel_engine_pm_put from the
context barrier (and not from the context exit) in order to prevent the
gem_exec_nop contention. Adding the delay to the context exit incurs
noticably extra penalty for soft-rc6.
Fixes: edee52c927ef ("drm/i915/gt: Delay release of engine-pm after last retirement")
Testcase: igt/i915_pm_rc6_residency/rc6-idle Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200323192029.20723-1-chris@chris-wilson.co.uk
Matthew Auld [Mon, 23 Mar 2020 13:08:21 +0000 (13:08 +0000)]
drm/i915/selftests: mark huge_gem_object as not shrinkable
It looks like some callers expect a non-volatile object, that they do not
want the contents of the pages lost if they happen to not be looking at it.
The shrinker however sees that we mark the pages as DONTNEED and
believes that it can freely reap them. However, since the huge object
use plain pages, they cannot be swapped out as they have no backing
storge, and the only way we can shrink them is by discarding the
contents. In light of the callers wanting to keep the contents around,
both IS_SHRINKABLE and marking the pages as volatile are incorrect.
If we drop the IS_SHRINKABLE flag we avoid the immediate issue of the
shrinker accidentally removing valuable content. We will have to
remember that a huge object is not suitable for exercising the shrinker
interaction -- although we can introduce a shrinkable one if we require.
Chris Wilson [Mon, 23 Mar 2020 09:28:37 +0000 (09:28 +0000)]
drm/i915/gt: Delay release of engine-pm after last retirement
Keep the engine-pm awake until the next jiffie, to avoid immediate
ping-pong under moderate load. (Forcing the idle barrier excerbates the
moderate load, dramatically increasing the driver overhead.) On the
other hand, delaying the idle-barrier slightly incurs longer rc6-off
and so more power consumption.
Chris Wilson [Mon, 23 Mar 2020 10:32:21 +0000 (10:32 +0000)]
drm/i915: Extend intel_wakeref to support delayed puts
In some cases we want to hold onto the wakeref for a little after the
last user so that we can avoid having to drop and then immediately
reacquire it. Allow the last user to specify if they would like to keep
the wakeref alive for a short hysteresis.
Chris Wilson [Mon, 23 Mar 2020 09:28:38 +0000 (09:28 +0000)]
drm/i915: Rely on direct submission to the queue
Drop the pretense of kicking the tasklet (used only for the defunct guc
submission backend, it should just take ownership of the submit!) and so
remove the bh-kicking from around submission.
Chris Wilson [Mon, 23 Mar 2020 09:28:41 +0000 (09:28 +0000)]
drm/i915/gem: Avoid gem_context->mutex for simple vma lookup
As we store the handle lookup inside a radix tree, we do not need the
gem_context->mutex except until we need to insert our lookup into the
common radix tree. This takes a small bit of rearranging to ensure that
the lut we insert into the tree is ready prior to actually inserting it
(as soon as it is exposed via the radixtree, it is visible to any other
submission).
v2: For brownie points, remove the goto spaghetti.
v3: Tighten up the closed-handle checks.
Chris Wilson [Mon, 23 Mar 2020 09:28:35 +0000 (09:28 +0000)]
drm/i915: Avoid live-lock with i915_vma_parked()
Abuse^W Take advantage that we know we are inside the GT wakeref and
that prevents any client execbuf from reopening the i915_vma in order to
claim all the vma to close without having to drop the spinlock to free
each one individually. By keeping the spinlock, we do not have to
restart if we run concurrently with i915_gem_free_objects -- which
causes them both to restart continually and make very very slow
progress.
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1361 Fixes: 77853186e547 ("drm/i915: Claim vma while under closed_lock in i915_vma_parked()") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200323092841.22240-2-chris@chris-wilson.co.uk
Chris Wilson [Mon, 23 Mar 2020 09:28:34 +0000 (09:28 +0000)]
drm/i915/gt: Mark timeline->cacheline as destroyed after rcu grace period
Since we take advantage of RCU for some i915_active objects, like the
intel_timeline_cacheline, we need to delay the i915_active_fini until
after the RCU grace period and we perform the kfree -- that is until
after all RCU protected readers.
Anshuman Gupta [Wed, 18 Mar 2020 08:18:37 +0000 (13:48 +0530)]
drm/i915/edp: Ignore short pulse when panel powered off
Few edp panels like Sharp is triggering short and long
hpd pulse after panel is getting powered off.
Currently driver is already ignoring long pulse for eDP
panel but in order to process the short pulse, it turns on
the VDD which requires panel power_cycle_delay + panel_power_on_delay
these delay on Sharp panel introduced the responsiveness overhead
of 800ms in the modeset sequence and as well is in suspend
sequence.
Ignoring any short pulse if panel is powered off.
FIXME: It requires to wait for panel_power_off delay in order
to check the panel power status due to pps_lock because panel triggers
short pulse immediately after writing PP_OFF to PP_CTRL register and
wait_panel_off waits for panel_power_off delay with pps_lock held.
This still creates responsiveness overhead of panel_power_off delay.
v2:
- checking vdd along with panel power to ignore the hpd. [Jani,Ville]
v3:
- safer side check to ignore the long hpd when eDP have power,
adding type of hpd to debug log. [Jani]
Chris Wilson [Sun, 22 Mar 2020 16:32:25 +0000 (16:32 +0000)]
drm/i915/gt: Leave rps->cur_freq on unpark
Don't override our previous frequency we used after parking, and avoid
continually spiking back to the efficient frequency for mostly idle
workloads. Trust our ability to autotune across a workload switch.
Chris Wilson [Sun, 22 Mar 2020 16:32:24 +0000 (16:32 +0000)]
drm/i915/gt: Treat idling as a RPS downclock event
If we park/unpark faster than we can respond to RPS events, we never
will process a downclock event after expiring a waitboost, and thus we
will forever restart the GPU at max clocks even if the workload switches
and doesn't justify full power.
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1500 Fixes: 3e7abf814193 ("drm/i915: Extract GT render power state management") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Andi Shyti <andi.shyti@intel.com> Cc: Lyude Paul <lyude@redhat.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200322163225.28791-1-chris@chris-wilson.co.uk Cc: <stable@vger.kernel.org> # v5.5+
Ville Syrjälä [Wed, 18 Mar 2020 17:45:15 +0000 (19:45 +0200)]
drm/i915: Fix crtc nv12 etc. plane bitmasks for DPMS off
We only consider crtc_state->enable when initially calculating plane
visibility. Later on we try to override the plane's state to invisible
if the crtc is in DPMS off state (crtc_state->active==false).
Unfortunately the code doing that only updates the plane_state.visible
flag and the crtc_state.active_planes bimask, but forgets to update
some of the other plane bitmasks stored in the crtc_state. Namely
crtc_state.nv12_planes is left set up based on the original visibility
check which makes icl_check_nv12_planes() pick a slave plane for the
flagged plane in the bitmask. Later on we hit the watermark code
which sees a plane with a slave assigned and it then makes the
logical assumption that the master plane must itself be visible.
Since the master's plane_state.visible flag was already cleared
we get a WARN.
Fix the problem by clearing all the plane bitmasks for DPMS off.
This is more or less the wrong approach and instead we should
calculate all the plane related state purely based crtc_state->enable
(to guarantee that the subsequent DPMS on can't fail). However in
the past we definitely had some roadblocks to making that happen.
Not sure how many are left these days, but let's stick to the current
approach since it's a much simpler fix to the immediate problem
(the WARN).
v2: Keep the visible=false, it's important (Rodrigo)
Gwan-gyeong Mun [Tue, 11 Feb 2020 07:46:43 +0000 (09:46 +0200)]
drm/i915/dp: Add writing of DP SDPs
It adds routines that write DP VSC SDP and DP HDR Metadata Infoframe SDP.
In order to pack DP VSC SDP, it adds intel_dp_vsc_sdp_pack() function.
It follows DP 1.4a spec. [Table 2-116: VSC SDP Header Bytes] and
[Table 2-117: VSC SDP Payload for DB16 through DB18]
In order to pack DP HDR Metadata Infoframe SDP, it adds
intel_dp_hdr_metadata_infoframe_sdp_pack() function.
And it follows DP 1.4a spec.
([Table 2-125: INFOFRAME SDP v1.2 Header Bytes] and
[Table 2-126: INFOFRAME SDP v1.2 Payload Data Bytes - DB0 through DB31])
and CTA-861-G spec. [Table-42 Dynamic Range and Mastering InfoFrame].
A mechanism and a naming rule of intel_dp_set_infoframes() function
references intel_encoder->set_infoframes() of intel_hdmi.c .
VSC SDP is used for PSR and Pixel Encoding and Colorimetry Formats cases.
Because PSR routine has its own routine of writing a VSC SDP, when the PSR
is enabled, intel_dp_set_infoframes() does not write a VSC SDP.
v3:
- Explicitly disable unused DIPs (AVI, GCP, VS, SPD, DRM. They will be
used for HDMI), when intel_dp_set_infoframes() function will be called.
- Replace a structure name to drm_dp_vsc_sdp from intel_dp_vsc_sdp.
v4: Use struct drm_device logging macros
v5:
- use intel_de_*() functions for register access
- Addressed review comments from Uma
Polish commit message and comments
Add 6bpc to packing of VSC SDP
projects / mirror_ubuntu-hirsute-kernel.git / log