This function just check our software flag, while 'is_alive'
may suggest that we are checking runtime firmware status.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@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/20190522193203.23932-5-michal.wajdeczko@intel.com
We were testing full GPU reset in atomic context without correctly
wrapping it by prepare/finish steps. This could confuse our GuC
reset handling code.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190522193203.23932-4-michal.wajdeczko@intel.com
Split igt_atomic_reset selftests into separate full & engines parts,
so we can move former to the dedicated reset selftests file.
While here change engines test to loop first over atomic phases and
then loop over available engines.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190522193203.23932-3-michal.wajdeczko@intel.com
igt_global_reset and igt_wedged_reset testcases are first candidates.
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190522193203.23932-2-michal.wajdeczko@intel.com
Bspec describes that GEN10 only supports capability of YUV 4:2:0 output to
HDMI port and GEN11 supports capability of YUV 4:2:0 output to both DP and
HDMI ports.
v2: Minor style fix.
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-7-gwan-gyeong.mun@intel.com
Data M/N calculations were assumed a bpp as RGB format. But when we are
using YCbCr 4:2:0 output format on DP, we should change bpp calculations
as YCbCr 4:2:0 format. The pipe_bpp value was assumed RGB format,
therefore, it was multiplied with 3. But YCbCr 4:2:0 requires a multiplier
value to 1.5.
Therefore we need to divide pipe_bpp to 2 while DP output uses YCbCr4:2:0
format.
- RGB format bpp = bpc x 3
- YCbCr 4:2:0 format bpp = bpc x 1.5
But Link M/N values are calculated and applied based on the Full Clock for
YCbCr 4:2:0. And DP YCbCr 4:2:0 does not need to pixel clock double for
a dotclock caluation. Only for HDMI YCbCr 4:2:0 needs to pixel clock double
for a dot clock calculation.
It only affects dp and edp port which use YCbCr 4:2:0 output format.
And for now, it does not consider a use case of DSC + YCbCr 4:2:0.
v2:
Addressed review comments from Ville.
Remove a changing of pipe_bpp on intel_ddi_set_pipe_settings().
Because the pipe is running at the full bpp, keep pipe_bpp as RGB
even though YCbCr 4:2:0 output format is used.
Add a link bandwidth computation for YCbCr4:2:0 output format.
v3:
Addressed reivew comments from Ville.
In order to make codes simple, it adds and uses intel_dp_output_bpp()
function.
v6:
Link M/N values are calculated and applied based on the Full Clock for
YCbCr420. The Bit per Pixel needs to be adjusted for YUV420 mode as it
requires only half of the RGB case.
- Link M/N values are calculated and applied based on the Full Clock
- Data M/N values needs to be calculated considering the data is half
due to subsampling
Remove a doubling of pixel clock on a dot clock calculator for
DP YCbCr 4:2:0.
Rebase and remove a duplicate setting of vsc_sdp.DB17.
Add a setting of dynamic range bit to vsc_sdp.DB17.
Change Content Type bit to "Graphics" from "Not defined".
Change a dividing of pipe_bpp to muliplying to constant values on a
switch-case statement.
v7:
Addressed review comments from Ville.
Move a setting of dynamic range bit and a setting of bpc which is based
on pipe_bpp to a "drm/i915/dp: Program VSC Header and DB for Pixel
Encoding/Colorimetry Format" commit.
Change Content Type bit to "Not defined" from "Graphics".
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-6-gwan-gyeong.mun@intel.com
When YCBCR 4:2:0 outputs is used for DP, we should program YCBCR 4:2:0 to
MSA and VSC SDP.
As per DP 1.4a spec section 2.2.4.3 [MSA Field for Indication of Color
Encoding Format and Content Color Gamut] while sending YCBCR 420 signals
we should program MSA MISC1 fields which indicate VSC SDP for the Pixel
Encoding/Colorimetry Format.
v2: Block comment style fix.
v6:
Fix an wrong setting of MSA MISC1 fields for Pixel Encoding/Colorimetry
Format indication. As per DP 1.4a spec Table 2-96 [MSA MISC1 and MISC0
Fields for Pixel Encoding/Colorimetry Format Indication]
When MISC1, bit 6, is Set to 1, a Source device uses a VSC SDP to
indicate the Pixel Encoding/Colorimetry Format. On the wrong version
it set a bit 5 of MISC1, now it set a bit 6 of MISC1.
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-5-gwan-gyeong.mun@intel.com
Function intel_pixel_encoding_setup_vsc handles vsc header and data block
setup for pixel encoding / colorimetry format.
Setup VSC header and data block in function intel_pixel_encoding_setup_vsc
for pixel encoding / colorimetry format as per dp 1.4a spec,
section 2.2.5.7.1, table 2-119: VSC SDP Header Bytes, section 2.2.5.7.5,
table 2-120:VSC SDP Payload for DB16 through DB18.
v2:
Minor style fix. [Maarten]
Refer to commit ids instead of patchwork. [Maarten]
v6: Rebase
v7:
Rebase and addressed review comments from Ville.
Use a structure initializer instead of memset().
Fix non-standard comment format.
Remove a referring to specific commit.
Add a setting of dynamic range bit to vsc_sdp.DB17.
Add a setting of bpc which is based on pipe_bpp.
Remove duplicated checking of connector's ycbcr_420_allowed from
intel_pixel_encoding_setup_vsc(). It is already checked from
intel_dp_ycbcr420_config().
Remove comments for VSC_SDP_EXTENSION_FOR_COLORIMETRY_SUPPORTED. It is
already implemented on intel_dp_get_colorimetry_status().
v8:
A missing of setting bpc to VSC setup is the pretty fatal case, it
replaces DRM_DEBUG_KMS() to MISSING_CASE(). [Maarten]
v9: Use a changed member name of struct dp_sdp. it renamed to db from DB.
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-4-gwan-gyeong.mun@intel.com
VSC SDP Payload for PSR is one of data block type of SDP (Secondaray Data
Packet). In order to generalize SDP packet structure name, it renames
struct edp_vsc_psr to struct dp_sdp. And each SDP data blocks have
different usages, each SDP type has different reserved data blocks and
Video_Stream_Configuration Extension VESA SDP might use all of Data Blocks
as Extended INFORFRAME Data Byte. so it makes Data Block variables as
array type. And it adds comments of details of DB of VSC SDP Payload
for Pixel Encoding/Colorimetry Format. This comments follows DP 1.4a spec,
section 2.2.5.7.5, chapter "VSC SDP Payload for Pixel Encoding/Colorimetry
Format".
v7: Addressed review comments from Ville.
v9: Rename a member value name DB to db on struct dp_sdp [Laurent]
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-3-gwan-gyeong.mun@intel.com
This patch checks a support of YCBCR420 outputs on an encoder level.
If the input mode is YCBCR420-only mode then it prepares DP as an YCBCR420
output, else it continues with RGB output mode.
It set output_format to INTEL_OUTPUT_FORMAT_YCBCR420 in order to using
a pipe scaler as RGB to YCbCr 4:4:4.
v2:
Addressed review comments from Ville.
Style fixed with few naming.
%s/config/crtc_state/
%s/intel_crtc/crtc/
If lscon is active, it makes not to call intel_dp_ycbcr420_config()
to avoid to clobber of lspcon_ycbcr420_config() routine.
And it move the 420_only check into the intel_dp_ycbcr420_config().
v3: Fix uninitialized return value and it is reported by Dan Carpenter.
v4:
Addressed review comments from Ville.
In order to avoid the extra indentation, it inverts if-clause on
intel_dp_ycbcr420_config().
Remove the error print where no errors print are allowed.
v6: Rebase
v7:
Move intel_dp_get_colorimetry_status() to intel_dp from intel_psr.
intel_dp_get_colorimetry_status() checks
VSC_SDP_EXTENSION_FOR_COLORIMETRY_SUPPORTED bit in the
DPRX_FEATURE_ENUMERATION_LIST register.
And intel_dp_ycbcr420_config() uses intel_dp_get_colorimetry_status().
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521121721.32010-2-gwan-gyeong.mun@intel.com
Engine discovery query allows userspace to enumerate engines, probe their
configuration features, all without needing to maintain the internal PCI
ID based database.
A new query for the generic i915 query ioctl is added named
DRM_I915_QUERY_ENGINE_INFO, together with accompanying structure
drm_i915_query_engine_info. The address of latter should be passed to the
kernel in the query.data_ptr field, and should be large enough for the
kernel to fill out all known engines as struct drm_i915_engine_info
elements trailing the query.
As with other queries, setting the item query length to zero allows
userspace to query minimum required buffer size.
Enumerated engines have common type mask which can be used to query all
hardware engines, versus engines userspace can submit to using the execbuf
uAPI.
Engines also have capabilities which are per engine class namespace of
bits describing features not present on all engine instances.
v2:
* Fixed HEVC assignment.
* Reorder some fields, rename type to flags, increase width. (Lionel)
* No need to allocate temporary storage if we do it engine by engine.
(Lionel)
v3:
* Describe engine flags and mark mbz fields. (Lionel)
* HEVC only applies to VCS.
v4:
* Squash SFC flag into main patch.
* Tidy some comments.
v5:
* Add uabi_ prefix to engine capabilities. (Chris Wilson)
* Report exact size of engine info array. (Chris Wilson)
* Drop the engine flags. (Joonas Lahtinen)
* Added some more reserved fields.
* Move flags after class/instance.
v6:
* Do not check engine info array was zeroed by userspace but zero the
unused fields for them instead.
v7:
* Simplify length calculation loop. (Lionel Landwerlin)
v8:
* Remove MBZ comments where not applicable.
* Rename ABI flags to match engine class define naming.
* Rename SFC ABI flag to reflect it applies to VCS and VECS.
* SFC is wired to even _logical_ engine instances.
* SFC applies to VCS and VECS.
* HEVC is present on all instances on Gen11. (Tony)
* Simplify length calculation even more. (Chris Wilson)
* Move info_ptr assigment closer to loop for clarity. (Chris Wilson)
* Use vdbox_sfc_access from runtime info.
* Rebase for RUNTIME_INFO.
* Refactor for lower indentation.
* Rename uAPI class/instance to engine_class/instance to avoid C++
keyword.
v9:
* Rebase for s/num_rings/num_engines/ in RUNTIME_INFO.
v10:
* Use new copy_query_item.
v11:
* Consolidate with struct i915_engine_class_instnace.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jon Bloomfield <jon.bloomfield@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> # v7
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190522090054.6007-1-tvrtko.ursulin@linux.intel.com
Test context workarounds have been correctly applied in newly created
contexts.
To accomplish this the existing engine_wa_list_verify helper is extended
to take in a context from which reading of the workaround list will be
done.
Context workaround verification is done from the existing subtests, which
have been renamed to reflect they are no longer only about GT and engine
workarounds.
v2:
* Test after resets and refactor to use intel_context more. (Chris)
v3:
* Use ce->engine->i915 instead of ce->gem_context->i915. (Chris)
* gem_engine_iter.idx is engine->id + 1. (Chris)
v4:
* Make local function static.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190520142546.12493-1-tvrtko.ursulin@linux.intel.com
There is a desire to split a task onto two engines and have them run at
the same time, e.g. scanline interleaving to spread the workload evenly.
Through the use of the out-fence from the first execbuf, we can
coordinate secondary execbuf to only become ready simultaneously with
the first, so that with all things idle the second execbufs are executed
in parallel with the first. The key difference here between the new
EXEC_FENCE_SUBMIT and the existing EXEC_FENCE_IN is that the in-fence
waits for the completion of the first request (so that all of its
rendering results are visible to the second execbuf, the more common
userspace fence requirement).
Since we only have a single input fence slot, userspace cannot mix an
in-fence and a submit-fence. It has to use one or the other! This is not
such a harsh requirement, since by virtue of the submit-fence, the
secondary execbuf inherit all of the dependencies from the first
request, and for the application the dependencies should be common
between the primary and secondary execbuf.
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Testcase: igt/gem_exec_fence/parallel
Link: https://github.com/intel/media-driver/pull/546
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/20190521211134.16117-10-chris@chris-wilson.co.uk
Some users require that when a master batch is executed on one particular
engine, a companion batch is run simultaneously on a specific slave
engine. For this purpose, we introduce virtual engine bonding, allowing
maps of master:slaves to be constructed to constrain which physical
engines a virtual engine may select given a fence on a master engine.
For the moment, we continue to ignore the issue of preemption deferring
the master request for later. Ideally, we would like to then also remove
the slave and run something else rather than have it stall the pipeline.
With load balancing, we should be able to move workload around it, but
there is a similar stall on the master pipeline while it may wait for
the slave to be executed. At the cost of more latency for the bonded
request, it may be interesting to launch both on their engines in
lockstep. (Bubbles abound.)
Opens: Also what about bonding an engine as its own master? It doesn't
break anything internally, so allow the silliness.
v2: Emancipate the bonds
v3: Couple in delayed scheduling for the selftests
v4: Handle invalid mutually exclusive bonding
v5: Mention what the uapi does
v6: s/nbond/num_bonds/
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/20190521211134.16117-9-chris@chris-wilson.co.uk
In the next patch, we will want to configure the slave request
depending on which physical engine the master request is executed on.
For this, we introduce a callback from the execute fence to convey this
information.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521211134.16117-8-chris@chris-wilson.co.uk
Allow the user to direct which physical engines of the virtual engine
they wish to execute one, as sometimes it is necessary to override the
load balancing algorithm.
v2: Only kick the virtual engines on context-out if required
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/20190521211134.16117-7-chris@chris-wilson.co.uk
Having allowed the user to define a set of engines that they will want
to only use, we go one step further and allow them to bind those engines
into a single virtual instance. Submitting a batch to the virtual engine
will then forward it to any one of the set in a manner as best to
distribute load. The virtual engine has a single timeline across all
engines (it operates as a single queue), so it is not able to concurrently
run batches across multiple engines by itself; that is left up to the user
to submit multiple concurrent batches to multiple queues. Multiple users
will be load balanced across the system.
The mechanism used for load balancing in this patch is a late greedy
balancer. When a request is ready for execution, it is added to each
engine's queue, and when an engine is ready for its next request it
claims it from the virtual engine. The first engine to do so, wins, i.e.
the request is executed at the earliest opportunity (idle moment) in the
system.
As not all HW is created equal, the user is still able to skip the
virtual engine and execute the batch on a specific engine, all within the
same queue. It will then be executed in order on the correct engine,
with execution on other virtual engines being moved away due to the load
detection.
A couple of areas for potential improvement left!
- The virtual engine always take priority over equal-priority tasks.
Mostly broken up by applying FQ_CODEL rules for prioritising new clients,
and hopefully the virtual and real engines are not then congested (i.e.
all work is via virtual engines, or all work is to the real engine).
- We require the breadcrumb irq around every virtual engine request. For
normal engines, we eliminate the need for the slow round trip via
interrupt by using the submit fence and queueing in order. For virtual
engines, we have to allow any job to transfer to a new ring, and cannot
coalesce the submissions, so require the completion fence instead,
forcing the persistent use of interrupts.
- We only drip feed single requests through each virtual engine and onto
the physical engines, even if there was enough work to fill all ELSP,
leaving small stalls with an idle CS event at the end of every request.
Could we be greedy and fill both slots? Being lazy is virtuous for load
distribution on less-than-full workloads though.
Other areas of improvement are more general, such as reducing lock
contention, reducing dispatch overhead, looking at direct submission
rather than bouncing around tasklets etc.
sseu: Lift the restriction to allow sseu to be reconfigured on virtual
engines composed of RENDER_CLASS (rcs).
v2: macroize check_user_mbz()
v3: Cancel virtual engines on wedging
v4: Commence commenting
v5: Replace 64b sibling_mask with a list of class:instance
v6: Drop the one-element array in the uabi
v7: Assert it is an virtual engine in to_virtual_engine()
v8: Skip over holes in [class][inst] so we can selftest with (vcs0, vcs2)
Link: https://github.com/intel/media-driver/pull/283
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/20190521211134.16117-6-chris@chris-wilson.co.uk
A usecase arose out of handling context recovery in mesa, whereby they
wish to recreate a context with fresh logical state but preserving all
other details of the original. Currently, they create a new context and
iterate over which bits they want to copy across, but it would much more
convenient if they were able to just pass in a target context to clone
during creation. This essentially extends the setparam during creation
to pull the details from a target context instead of the user supplied
parameters.
The ideal here is that we don't expose control over anything more than
can be obtained via CONTEXT_PARAM. That is userspace retains explicit
control over all features, and this api is just convenience.
For example, you could replace
struct context_param p = { .param = CONTEXT_PARAM_VM };
param.ctx_id = old_id;
gem_context_get_param(&p.param);
new_id = gem_context_create();
param.ctx_id = new_id;
gem_context_set_param(&p.param);
gem_vm_destroy(param.value); /* drop the ref to VM_ID handle */
with
struct create_ext_param p = {
{ .name = CONTEXT_CREATE_CLONE },
.clone_id = old_id,
.flags = CLONE_FLAGS_VM
}
new_id = gem_context_create_ext(&p);
and not have to worry about stray namespace pollution etc.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521211134.16117-5-chris@chris-wilson.co.uk
The SINGLE_TIMELINE flag can be used to create a context such that all
engine instances within that context share a common timeline. This can
be useful for mixing operations between real and virtual engines, or
when using a composite context for a single client API context.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521211134.16117-4-chris@chris-wilson.co.uk
Allow the user to specify a local engine index (as opposed to
class:index) that they can use to refer to a preset engine inside the
ctx->engine[] array defined by an earlier I915_CONTEXT_PARAM_ENGINES.
This will be useful for setting SSEU parameters on virtual engines that
are local to the context and do not have a valid global class:instance
lookup.
Note that due to the ambiguity in using class:instance with
ctx->engines[], if a user supplied engine map is active the user must
specify the engine to alter by its index into the ctx->engines[].
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/20190521211134.16117-3-chris@chris-wilson.co.uk
Over the last few years, we have debated how to extend the user API to
support an increase in the number of engines, that may be sparse and
even be heterogeneous within a class (not all video decoders created
equal). We settled on using (class, instance) tuples to identify a
specific engine, with an API for the user to construct a map of engines
to capabilities. Into this picture, we then add a challenge of virtual
engines; one user engine that maps behind the scenes to any number of
physical engines. To keep it general, we want the user to have full
control over that mapping. To that end, we allow the user to constrain a
context to define the set of engines that it can access, order fully
controlled by the user via (class, instance). With such precise control
in context setup, we can continue to use the existing execbuf uABI of
specifying a single index; only now it doesn't automagically map onto
the engines, it uses the user defined engine map from the context.
v2: Fixup freeing of local on success of get_engines()
v3: Allow empty engines[]
v4: s/nengine/num_engines/
v5: Replace 64 limit on num_engines with a note that execbuf is
currently limited to only using the first 64 engines.
v6: Actually use the engines_mutex to guard the ctx->engines.
Testcase: igt/gem_ctx_engines
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/20190521211134.16117-2-chris@chris-wilson.co.uk
Having hid the partially exposed new ABI from the PR, put it back again
for completion of context recovery. A significant part of context
recovery is the ability to reuse as much of the old context as is
feasible (to avoid expensive reconstruction). The biggest chunk kept
hidden at the moment is fine-control over the ctx->ppgtt (the GPU page
tables and associated translation tables and kernel maps), so make
control over the ctx->ppgtt explicit.
This allows userspace to create and share virtual memory address spaces
(within the limits of a single fd) between contexts they own, along with
the ability to query the contexts for the vm state.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190521211134.16117-1-chris@chris-wilson.co.uk
With gtt remapping in place we can use arbitrarily large
framebuffers. Let's bump the limits to 16kx16k on gen7+.
The limit was chosen to match the maximum 2D surface size
of the 3D engine.
With the remapping we could easily go higher than that for the
display engine. However the modesetting ddx will blindly assume
it can handle whatever is reported via kms. The oversized
buffer dimensions are not caught by glamor nor Mesa until
finally an assert will trip when genxml attempts to pack the
SURFACE_STATE. So we pick a safe limit to avoid the X server
from crashing (or potentially misbehaving if the genxml asserts
are compiled out).
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110187
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-9-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
With gtt remapping plugged in we can simply raise the stride
limit on gen4+. Let's just pick the limit to match the render
engine max stride (256KiB on gen7+, 128KiB on gen4+).
No remapping CCS because the virtual address of each page actually
matters due to the new hash mode
(WaCompressedResourceDisplayNewHashMode:skl,kbl etc.), and no
remapping on gen2/3 due extra complications from fence alignment
and gen2 2KiB GTT tile size. Also no real benefit since the
display engine limits already match the other limits.
v2: Rebase due to is_ccs_modifier()
v3: Tweak the comment and commit msg
v4: Fix gen4+ stride limit to be 128KiB
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> #v3
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-8-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Align dumb buffer stride to 4k if the fb will be big enough to
require gtt remapping.
v2: Leave the stride alone for buffers that look to be for the cursor
v3: Make it not a hack (Daniel)
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-7-ville.syrjala@linux.intel.com
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
The display engine stride limits are getting in our way. On SKL+
we are limited to 8k pixels, which is easily exceeded with three
4k displays. To overcome this limitation we can remap the pages
in the GTT to provide the display engine with a view of memory
with a smaller stride.
The code is mostly already there as We already play tricks with
the plane surface address and x/y offsets.
A few caveats apply:
* linear buffers need the fb stride to be page aligned, as
otherwise the remapped lines wouldn't start at the same
spot
* compressed buffers can't be remapped due to the new
ccs hash mode causing the virtual address of the pages
to affect the interpretation of the compressed data. IIRC
the old hash was limited to the low 12 bits so if we were
using that mode we could remap. As it stands we just refuse
to remapp with compressed fbs.
* no remapping gen2/3 as we'd need a fence for the remapped
vma, which we currently don't have. Need to deal with the
fence POT requirements, and do something about the gen2
gtt page size vs tile size difference
v2: Rebase due to is_ccs_modifier()
Fix up the skl+ stride_mult mess
memset() the gtt_view because otherwise we could leave
junk in plane[1] when going from 2 plane to 1 plane format
v3: intel_check_plane_stride() was split out
v4: Drop the aligned viewport stuff, it was meant for ccs which
can't be remapped anyway
v5: Introduce intel_plane_can_remap()
Reorder the code so that plane_state->view gets filled
even for invisible planes, otherwise we'd keep using
stale values and could explode during remapping. The new
logic never remaps invisible planes since we don't have
a viewport, and instead pins the full fb instead
v6: Fix plane src coord checks after remapping by moving
plane_state->base.src to the final plane x/y offsets.
Allow intel_plane_check_stride() to fail even with
remapping (can happen at least with a linear 64bpp
fb with a 4k plane and a suitably inconvenient src
coordinates).
Improve aux plane FIXME (Daniel)
Move some code shuffling into a separate patch (Daniel)
Testcase: igt/kms_big_fb
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-6-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reorganize some fb stride checking code a bit to prepare for
gtt remapping. And do a bit of s/pitch/stride/ renaming in the
process for a bit more uniformity (apart from the whole
fb->pitches[] thing).
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-5-ville.syrjala@linux.intel.com
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Add a live selftest to excercise rotated/remapped vmas. We simply
write through the rotated/remapped vma, and confirm that the data
appears in the right page when read through the normal vma.
Not sure what the fallout of making all rotated/remapped vmas
mappable/fenceable would be, hence I just hacked it in the test.
v2: Grab rpm reference (Chris)
GEM_BUG_ON(view.type not as expected) (Chris)
Allow CAN_FENCE for rotated/remapped vmas (Chris)
Update intel_plane_uses_fence() to ask for a fence
only for normal vmas on gen4+
v3: Deal with intel_wakeref_t
v4: Rebase
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-4-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Extend the rotated vma mock selftest to cover remapped vmas as
well.
TODO: reindent the loops I guess? Left like this for now to
ease review
v2: Include the vma type in the error message (Chris)
v3: Deal with trimmed sg
v4: Drop leftover debugs
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-3-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To overcome display engine stride limits we'll want to remap the
pages in the GTT. To that end we need a new gtt_view type which
is just like the "rotated" type except not rotated.
v2: Use intel_remapped_plane_info base type
s/unused/unused_mbz/ (Chris)
Separate BUILD_BUG_ON()s (Chris)
Use I915_GTT_PAGE_SIZE (Chris)
v3: Use i915_gem_object_get_dma_address() (Chris)
Trim the sg (Tvrtko)
v4: Actually trim this time. Limit the max length
to one row of pages to keep things simple
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509122159.24376-2-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
With the disappearance of NEWCLIENT, we no longer need to provide the
priority boost on preemption in order to prevent repeated gazumping,
and we can remove the dead code.
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/20190515130052.4475-5-chris@chris-wilson.co.uk
Commit 1413b2bc07 ("drm/i915: Trim NEWCLIENT boosting") had the
intended consequence of not allowing a sequence of work that merely
crossed into a new engine the privilege to be promoted to NEWCLIENT
status. It also had the unintended consequence of actually making
NEWCLIENT effective on heavily oversubscribed transcode machines and
impacting upon their throughput.
If we consider a client packet composed of (rcsA, rcsB, vcs) and 30 of
those clients, using the NEWCLIENT boost that will be scheduled as
rcsA x 30, (rcsB, vcs) x 30
where as before it would have been
(rcsA, rcsB, vcs) x 30
That is with NEWCLIENT only boosting the first request of each client,
we would execute all rcsA requests prior to running on the vcs engines;
acruing a lot of dead time as compared to the previous case where the
vcs engine would be started in parallel to processing the second client.
The previous patch has the effect of delaying submission until it is
required by a third party (either the user with an explicit wait, or by
another client/engine). We reduce the NEWCLIENT bump to a mere WAIT,
which has the effect of removing its preemptive grant and reducing it to
the same level as any other user interaction -- that it will not be
promoted above the interengine dependencies, and so preventing NEWCLIENTS
from starving other engines. This a large nerf to the rrul properties of
the current NEWCLIENT, but it still does give prioritised submission to
new requests from light workloads.
References: b16c765122 ("drm/i915: Priority boost for new clients")
Fixes: 1413b2bc07 ("drm/i915: Trim NEWCLIENT boosting") # customer impact
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Dmitry Ermilov <dmitry.ermilov@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190515130052.4475-4-chris@chris-wilson.co.uk
The handling of the no-preemption priority level imposes the restriction
that we need to maintain the implied ordering even though preemption is
disabled. Otherwise we may end up with an AB-BA deadlock across multiple
engine due to a real preemption event reordering the no-preemption
WAITs. To resolve this issue we currently promote all requests to WAIT
on unsubmission, however this interferes with the timeslicing
requirement that we do not apply any implicit promotion that will defeat
the round-robin timeslice list. (If we automatically promote the active
request it will go back to the head of the queue and not the tail!)
So we need implicit promotion to prevent reordering around semaphores
where we are not allowed to preempt, and we must avoid implicit
promotion on unsubmission. So instead of at unsubmit, if we apply that
implicit promotion on adding the dependency, we avoid the semaphore
deadlock and we also reduce the gains made by the promotion for user
space waiting. Furthermore, by keeping the earlier dependencies at a
higher level, we reduce the search space for timeslicing without
altering runtime scheduling too badly (no dependencies at all will be
assigned a higher priority for rrul).
v2: Limit the bump to external edges (as originally intended) i.e.
between contexts and out to the user.
Testcase: igt/gem_concurrent_blit
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190515130052.4475-3-chris@chris-wilson.co.uk
Smatch spotted:
drivers/gpu/drm/i915//intel_hdcp.c:1406 hdcp2_authenticate_repeater_topology() warn: should this be a bitwise op?
and indeed looks to be suspect that we do need to use a bitwise or to
combine the two register fields into one counter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ramalingam C <ramalingam.c@intel.com>
Reviewed-by: Ramalingam C <ramalingam.c@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190517102225.3069-3-chris@chris-wilson.co.uk
Just to squelch an smatch warning that doesn't see the with_() being
taken unconditionally:
drivers/gpu/drm/i915//intel_dp.c:230 intel_dp_get_fia_supported_lane_count() error: uninitialized symbol 'lane_info'.
drivers/gpu/drm/i915//intel_dp.c:5338 intel_digital_port_connected() error: uninitialized symbol 'is_connected'.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190517102225.3069-2-chris@chris-wilson.co.uk
In commit b7404c7ecb ("drm/i915: Bump ready tasks ahead of
busywaits"), I tried cutting a corner in order to not install a signal
for each of our dependencies, and only listened to requests on which we
were intending to busywait. The compromise that was made was that
instead of then being able to promote the request with a full
NOSEMAPHORE like its non-busywaiting brethren, as we had not ensured we
had cleared the semaphore chain, we settled for only using the NEWCLIENT
boost. With an over saturated system with multiple NEWCLIENTS in flight
at any time, this was found to be an inadequate promotion and left us
with a much poorer scheduling order than prior to using semaphores.
The outcome of this patch, is that all requests have NOSEMAPHORE
priority when they have no dependencies and are ready to run and not
busywait, restoring the pre-semaphore ordering on saturated systems.
We can demonstrate the effect of poor scheduling order by oversaturating
the system using gem_wsim on a system with multiple vcs engines
(i.e running the same workloads across more clients than required for
peak throughput, e.g. media_load_balance_17i7.wsim -c4 -b context):
x v5.1 (normalized)
+ tip
* fix
+------------------------------------------------------------------------+
| x |
| x |
| x |
| x |
| %x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %%x |
| %#x |
| %#x |
| %#x |
| %#x |
| %#x |
| + %#xx |
| + %#xx |
| + %%#xx |
| + %%#xx |
| + %%#xx |
| + %%#xx |
| + %%##x |
| +++ %%##x |
| +++ %%##x |
| +++ %%##x |
| ++++ %%##x |
| ++++ %%##x |
| ++++ %%##xx |
| ++++ %###xx |
| ++++ %###xx |
| ++++ %###xx |
| ++++ %###xx |
| ++++ + %#O#xx |
| ++++ + %#O#xx |
| ++++++ + %#O#xx |
| ++++++++++ %OOOxxx|
| ++++++++++ + %#OOO#xx|
| + ++++++++++++ ++ +++++ + ++ @@OOOO#xx|
| |A_| |
||__________M_______A____________________| |
| |A_| |
+------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 120 0.99456 1.00628 0.999985 1.0001545 0.0024387139
+ 120 0.873021 1.00037 0.884134 0.90148752 0.039190862
Difference at 99.5% confidence
-0.098667 +/- 0.0110762
-9.86517% +/- 1.10745%
(Student's t, pooled s = 0.0277657)
% 120 0.990207 1.00165 0.9970265 0.99699748 0.0021024
Difference at 99.5% confidence
-0.003157 +/- 0.000908245
-0.315651% +/- 0.0908105%
(Student's t, pooled s = 0.00227678)
Fixes: b7404c7ecb ("drm/i915: Bump ready tasks ahead of busywaits")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Dmitry Ermilov <dmitry.ermilov@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190515130052.4475-2-chris@chris-wilson.co.uk
Avoid charging us for the presumed busywait if the request was preempted
after successfully using semaphores to reduce inter-engine latency.
v2: Bump the priority to reflect the lack of semaphores now required.
References: ca6e56f654 ("drm/i915: Disable semaphore busywaits on saturated systems")
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/20190515130052.4475-1-chris@chris-wilson.co.uk
Add an assert that we don't use TypeC ports for eDP. That may in theory
be possible on TypeC legacy ports, but I'm not sure if that's a
practical scenario, so let's deal with that only if there's a use case.
Adding support for that wouldn't be too difficult, since TypeC mode
switching is not possible on TypeC legacy ports.
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-12-imre.deak@intel.com
On ICL we have to make sure that we enable the AUX power domain in a
controlled way (corresponding to the port's actual TypeC mode). Since
the PPS lock - which takes an AUX power ref - is only needed on
eDP on all platforms and eDP/DP on VLV/CHV avoid taking it in all
other cases.
v2:
- Clarify commit log about the condition for taking the PPS lock.
(Ville)
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-11-imre.deak@intel.com
There isn't a separate power domain specific to PLLs. When programming
them we require the same power domain to be enabled which is needed when
accessing other display core parts (not specific to any
pipe/port/transcoder). This corresponds to the DISPLAY_CORE domain added
previously in this patchset, so use that instead to save bits in the
power domain mask.
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-10-imre.deak@intel.com
The power get/put was added in
commit 1c767b339b ("drm/i915: take display port power domain in DP HPD handler")
Author: Imre Deak <imre.deak@intel.com>
Date: Mon Aug 18 14:42:42 2014 +0300
to account for the HW access in ibx_digital_port_connected(). This
latter call was in turn removed in
commit 7d23e3c37b ("drm/i915: Cleaning up intel_dp_hpd_pulse")
Author: Shubhangi Shrivastava <shubhangi.shrivastava@intel.com>
Date: Wed Mar 30 18:05:23 2016 +0530
after which we didn't actually need the power reference.
One way we are accessing the HW during HPD pulse handling is via DP AUX
transfers, but the transfer function takes its own reference, so doesn't
need the reference in intel_dp_hpd_pulse().
The other spot is in
intel_psr_short_pulse()->intel_psr_disable_locked()
but that can only happen when the panel is enabled with the
corresponding modeset already holding the required power reference.
v2:
- Remove the unneeded power get/put from intel_psr_disable_locked().
(Ville)
- Checkpatch commit quoting format fix in the commit log.
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-9-imre.deak@intel.com
We don't need the AUX power for the whole duration of the detect, only
when we're doing AUX transfers. The AUX transfer function takes its own
reference on the AUX power domain already. The two places during detect
which access display core registers (not specific to a
pipe/port/transcoder) only need the power domain that is required for
that access. That power domain is equivalent to the device global power
domain on most platforms (enabled whenever we hold a runtime PM
reference) except on CHV/VLV where it's equivalent to the display power
well.
Add a new power domain that reflects the above, and use this at the two
spots accessing registers. With that we can avoid taking the AUX
reference for the whole duration of the detect function.
Put the domains asynchronously to avoid the unneeded on-off-on toggling.
Also adapt the idea from with_intel_runtime_pm et al. for making it easy
to write short sequences where a display power ref is needed.
v2: (Ville)
- Add with_intel_display_power() helper to simplify things.
- s/bool res/bool is_connected/
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-8-imre.deak@intel.com
In a follow-up patch we will restrict holding the reference on the AUX
power domain to the AUX transfer function. To avoid the unnecessary
on-off-on power togglings drop the reference asynchronously.
There is no reason we couldn't do this in general and also put the
reference asynchronously in pps_unlock(); but that's a separate change
that can be done as a follow-up.
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-6-imre.deak@intel.com
By disabling a power domain asynchronously we can restrict holding a
reference on that power domain to the actual code sequence that
requires the power to be on for the HW access it's doing, by also
avoiding unneeded on-off-on togglings of the power domain (since the
disabling happens with a delay).
One benefit is potential power saving due to the following two reasons:
1. The fact that we will now be holding the reference only for the
necessary duration by the end of the patchset. While simply not
delaying the disabling has the same benefit, it has the problem that
frequent on-off-on power switching has its own power cost (see the 2.
point below) and the debug trace for power well on/off events will
cause a lot of dmesg spam (see details about this further below).
2. Avoiding the power cost of freuqent on-off-on power switching. This
requires us to find the optimal disabling delay based on the measured
power cost of on->off and off->on switching of each power well vs.
the power of keeping the given power well on.
In this patchset I'm not providing this optimal delay for two
reasons:
a) I don't have the means yet to perform the measurement (with high
enough signal-to-noise ratio, or with the help of an energy
counter that takes switching into account). I'm currently looking
for a way to measure this.
b) Before reducing the disabling delay we need an alternative way for
debug tracing powerwell on/off events. Simply avoiding/throttling
the debug messages is not a solution, see further below.
Note that even in the case where we can't measure any considerable
power cost of frequent on-off switching of powerwells, it still would
make sense to do the disabling asynchronously (with 0 delay) to avoid
blocking on the disabling. On VLV I measured this disabling time
overhead to be 1ms on average with a worst case of 4ms.
In the case of the AUX power domains on ICL we would also need to keep
the sequence where we hold the power reference short, the way it would
be by the end of this patchset where we hold it only for the actual AUX
transfer. Anything else would make the locking we need for ICL TypeC
ports (whenever we hold a reference on any AUX power domain) rather
problematic, adding for instance unnecessary lockdep dependencies to
the required TypeC port lock.
I chose the disabling delay to be 100msec for now to avoid the unneeded
toggling (and so not to introduce dmesg spamming) in the DP MST sideband
signaling code. We could optimize this delay later, once we have the
means to measure the switching power cost (see above).
Note that simply removing/throttling the debug tracing for power well
on/off events is not a solution. We need to know the exact spots of
these events and cannot rely only on incorrect register accesses caught
(due to not holding a wakeref at the time of access). Incorrect
powerwell enabling/disabling could lead to other problems, for instance
we need to keep certain powerwells enabled for the duration of modesets
and AUX transfers.
v2:
- Clarify the commit log parts about power cost measurement and the
problem of simply removing/throttling debug tracing. (Chris)
- Optimize out local wakeref vars at intel_runtime_pm_put_raw() and
intel_display_power_put_async() call sites if
CONFIG_DRM_I915_DEBUG_RUNTIME_PM=n. (Chris)
- Rebased on v2 of the wakeref w/o power-on guarantee patch.
- Add missing docbook headers.
v3:
- Checkpatch spelling/missing-empty-line fix.
v4:
- Fix unintended local wakeref var optimization when using
call-arguments with side-effects, by using inline funcs instead of
macros. In this patch in particular this will fix the
intel_display_power_grab_async_put_ref()->intel_runtime_pm_put_raw()
call).
No size change in practice (would be the same disregarding the
corresponding change in intel_display_power_grab_async_put_ref()):
$ size i915-macro.ko
text data bss dec hex filename
2455190 105890 10272 2571352 273c58 i915-macro.ko
$ size i915-inline.ko
text data bss dec hex filename
2455195 105890 10272 2571357 273c5d i915-inline.ko
Kudos to Stan for reporting the raw-wakeref WARNs this issue caused. His
config has CONFIG_DRM_I915_DEBUG_RUNTIME_PM=n, which I didn't retest
after v1, and we are also not testing this config in CI.
Now tested both with CONFIG_DRM_I915_DEBUG_RUNTIME_PM=y/n on ICL,
connecting both Chamelium and regular DP, HDMI sinks.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjala <ville.syrjala@linux.intel.com>
Cc: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190513192533.12586-1-imre.deak@intel.com
There is no reason why we couldn't verify the power domains state during
suspend in all cases, so do that. I overlooked this when originally
adding the check.
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: https://patchwork.freedesktop.org/patch/msgid/20190509173446.31095-4-imre.deak@intel.com
