Chris Wilson [Thu, 8 Mar 2018 14:07:32 +0000 (14:07 +0000)]
drm/i915: Remove the impedance mismatch around intel_engine_enable_signaling
There is some redundancy between dma_fence->ops->enable_signaling (via
i915_fence_enable_signaling) and our backend,
intel_engine_enable_signaling() in that both levels recheck the fence
status multiple times. If we convert intel_engine_enable_signaling() to
return the information desired by dma_fence->ops->enable_signaling, we
can reduce i915_fence_enable_signaling to a simple stub and avoid
trying to reinterpret the same information.
Rodrigo Vivi [Fri, 9 Mar 2018 00:52:18 +0000 (16:52 -0800)]
drm/i915/psr: Display WA 0884 applied broadly for more HW tracking.
WA 0884:bxt:all,cnl:*:A - "When FBC is enabled with eDP PSR,
the CPU host modify writes may not get updated on the Display
as expected.
WA: Write 0x00000000 to CUR_SURFLIVE_A with every CPU
host modify write to trigger PSR exit."
We can also find on spec other cases where they describe
bogus writes to cursor registers to force PSR exit with
HW tracking. And it was confirmed by HW engineers that
this Wa can be safely applied for any frontbuffer activity.
So let's use this more and more here instead of forcibly
disable and re-enable PSR everytime that we have a simple
reliable flush case.
Other commits improve the fbcon/fbdev use a lot, but this
approach is the only when where we can get a fully reliable
console with no slowness or missed frames and PSR still
enabled and active.
v2: - Rebase on drm-tip
- (DK) Add a comment to explain that WA
tells about writing 0 to CUR_SURFLIVE_A but we write to
CUR_SURFLIVE(pipe).
v3: Wa doesn't work on PSR2.
Error state management code was moved into separate .c unit
but we didn't move related definitions into own header.
v2: move also intel_display_error_state forward decl
fix ("Prefer 'unsigned int' to bare use of 'unsigned'")
warnings detected by checkpatch in moved code (Michal)
Definitions in i915_pmu.h header depend on other types and
declarations that were not explicitly included. Fix that by
adding related headers and forward declarations.
While here, change license text to SPDX format.
drm/i915: Change parameters order in i915_gem_batch_pool_init
Function i915_gem_batch_pool_init() failed to follow obj-verb
naming schema. Fix that by swapping function parameters.
While here, change license text to SPDX format.
v2: use intel_engine_init_batch_pool (Chris) as proxy (Michal)
drm/i915: Include i915_reg.h in intel_ringbuffer.h
Header intel_ringbuffer.h is using definitions from i915_reg.h
but forget to include it. Remove this hidden dependency by
explicitly include missing header.
drm/i915/guc: Create common entry points for log register/unregister
We have many functions responsible for allocating different parts of
GuC log runtime called from multiple places. Let's stick with keeping
everything in guc_log_register instead.
v2: Use more generic intel_uc_register name, keep using "misc" suffix (Michał)
s/dev_priv/i915 (Sagar)
Make guc_log_relay_* static (sparse)
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Sagar Arun Kamble <sagar.a.kamble@intel.com> Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180308154707.21716-2-michal.winiarski@intel.com
Chris Wilson [Wed, 7 Mar 2018 13:42:26 +0000 (13:42 +0000)]
drm/i915: Only call tasklet_kill() on the first prepare_reset
tasklet_kill() will spin waiting for the current tasklet to be executed.
However, if tasklet_disable() has been called, then the tasklet is never
executed but permanently put back onto the runlist until
tasklet_enable() is called. Ergo, we cannot use tasklet_kill() inside a
disable/enable pair. This is the case when we call set-wedge from inside
i915_reset(), and another request was submitted to us concurrent to the
reset.
Chris Wilson [Wed, 7 Mar 2018 13:42:25 +0000 (13:42 +0000)]
drm/i915: Wrap engine->schedule in RCU locks for set-wedge protection
Similar to the staging around handling of engine->submit_request, we
need to stop adding to the execlists->queue prior to calling
engine->cancel_requests. cancel_requests will move requests from the
queue onto the timeline, so if we add a request onto the queue after that
point, it will be lost.
Chris Wilson [Wed, 7 Mar 2018 13:42:23 +0000 (13:42 +0000)]
drm/i915: Update ring position from request on retiring
When wedged, we do not update the ring->tail as we submit the requests
causing us to leak the ring->space upon cleaning up the wedged driver.
We can just use the value stored in rq->tail, and keep the submission
backend details away from set-wedge.
Chris Wilson [Wed, 7 Mar 2018 13:42:21 +0000 (13:42 +0000)]
drm/i915: Finish the wait-for-wedge by retiring all the inflight requests
Before we reset the GPU after marking the device as wedged, we wait for
all the remaining requests to be completed (and marked as EIO).
Afterwards, we should flush the request lists so the next batch start
with the driver in an idle state.
Chris Wilson [Fri, 9 Mar 2018 01:08:08 +0000 (01:08 +0000)]
drm/i915: Push irq_shift from gen8_cs_irq_handler() to caller
Originally we were inlining gen8_cs_irq_handler() and so expected the
compiler to constant-fold away the irq_shift (so we had hardcoded it as
opposed to use engine->irq_shift). However, we dropped the inline given
the proliferation of gen8_cs_irq_handler()s. If we pull the shifting
of the iir into the caller, we can shrink the code still further:
add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-34 (-34)
Function old new delta
gen8_cs_irq_handler 123 118 -5
gen8_gt_irq_handler 261 248 -13
gen11_irq_handler 722 706 -16
v2: Drop gen11_cs_irq_handler now that it is a simple
stub around gen8_cs_irq_handler (Daniele)
References: 5d3d69d5c119 ("drm/i915: Stop inlining the execlists IRQ handler") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180309010808.11921-1-chris@chris-wilson.co.uk
Chris Wilson [Thu, 8 Mar 2018 14:26:48 +0000 (14:26 +0000)]
drm/i915: Index the ring frequency table by HW frequency range
When reporting the frequency table stored in the punit, report the full
range and not just the user restricted frequency range. In the process
keep the code to set the frequency table and read it the same.
v3: As we haven't separated the sb_lock from the pcu_lock yet, there's a
cycle between the pcu_lock and intel_runtime_pm_get.
Chris Wilson [Thu, 8 Mar 2018 14:26:47 +0000 (14:26 +0000)]
drm/i915: Kick the rps worker when changing the boost frequency
The boost frequency is only applied from the RPS worker while someone is
waiting on a request and requested a boost. As such, when the user
wishes to change the frequency, we have to kick the worker in order to
re-evaluate whether to apply the boost frequency.
v2: Check num_waiters to decide if we should kick the worker to handle
boosting.
Changes since v1:
- Don't try to apply CRC workaround when enabling pipe, it should already be enabled.
Changes since v2:
- Make crc functions for !DEBUGFS case inline.
- Pass intel_crtc to crc functions.
- Add comments to callsites.
Changes since v3:
- Cache selected source to pipe_crc->source.
- Set pipe_crc->skipped to MIN_INT during disable to close a race condition.
Changes since v4:
- Handle fallout from setting pipe_crc->source in irq handler.
Chris Wilson [Wed, 7 Mar 2018 17:13:03 +0000 (17:13 +0000)]
drm/i915: Only prune fences after wait-for-all
Currently, we only allow ourselves to prune the fences so long as
all the waits completed (i.e. all the fences we checked were signaled),
and that the reservation snapshot did not change across the wait.
However, if we only waited for a subset of the reservation object, i.e.
just waiting for the last writer to complete as opposed to all readers
as well, then we would erroneously conclude we could prune the fences as
indeed although all of our waits were successful, they did not represent
the totality of the reservation object.
v2: We only need to check the shared fences due to construction (i.e.
all of the shared fences will be later than the exclusive fence, if
any).
Fixes: e54ca9774777 ("drm/i915: Remove completed fences after a wait") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307171303.29466-1-chris@chris-wilson.co.uk
Weinan Li [Tue, 6 Mar 2018 02:15:57 +0000 (10:15 +0800)]
drm/i915: add schedule out notification of preempted but completed request
There is one corner case missing schedule out notification of the preempted
request. The preempted request is just completed when preemption happen,
then it will be canceled and won't be resubmitted later, GVT-g will lost
the schedule out notification.
Here add schedule out notification if found the preempted request has been
completed.
v2:
- refine description, add completed check and notification in
execlists_cancel_port_requests. (Chris)
v3:
- use ternary confitional, remove local variable. (Tvrtko)
With the introduction of asymmetric slices in CNL, we cannot rely on
the previous SUBSLICE_MASK getparam to tell userspace what subslices
are available. Here we introduce a more detailed way of querying the
Gen's GPU topology that doesn't aggregate numbers.
This is essential for monitoring parts of the GPU with the OA unit,
because counters need to be normalized to the number of
EUs/subslices/slices. The current aggregated numbers like EU_TOTAL do
not gives us sufficient information.
The Mesa series making use of this API is :
https://patchwork.freedesktop.org/series/38795/
As a bonus we can draw representations of the GPU :
https://imgur.com/a/vuqpa
v2: Rename uapi struct s/_mask/_info/ (Tvrtko)
Report max_slice/subslice/eus_per_subslice rather than strides (Tvrtko)
Add uapi macros to read data from *_info structs (Tvrtko)
v3: Use !!(v & DRM_I915_BIT()) for uapi macros instead of custom shifts (Tvrtko)
There are a number of information that are readable from hardware
registers and that we would like to make accessible to userspace. One
particular example is the topology of the execution units (how are
execution units grouped in subslices and slices and also which ones
have been fused off for die recovery).
At the moment the GET_PARAM ioctl covers some basic needs, but
generally is only able to return a single value for each defined
parameter. This is a bit problematic with topology descriptions which
are array/maps of available units.
This change introduces a new ioctl that can deal with requests to fill
structures of potentially variable lengths. The user is expected fill
a query with length fields set at 0 on the first call, the kernel then
sets the length fields to the their expected values. A second call to
the kernel with length fields at their expected values will trigger a
copy of the data to the pointed memory locations.
The scope of this uAPI is only to provide information to userspace,
not to allow configuration of the device.
v3: Rename pad fields into flags (Chris)
Return error on flags field != 0 (Chris)
Only copy length back to userspace in drm_i915_query_item (Chris)
v4: Use array of functions instead of switch (Chris)
v5: More comments in uapi (Tvrtko)
Return query item errors in length field (All)
v6: Tweak uapi comments style to match the coding style (Lionel)
v7: Add i915_query.h (Joonas)
v8: (Lionel) Change the behavior of the item iterator to report
invalid queries into the query item rather than stopping the
iteration. This enables userspace applications to query newer
items on older kernels and only have failure on the items that are
not supported.
While the end goal is to make this information available to userspace
through a new ioctl, there is no reason we can't display it in a human
readable fashion through debugfs.
slice0: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
slice1: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
slice2: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
v2: Reformat debugfs printing (Tvrtko)
Use the new EU mask helper (Tvrtko)
v3: Move printing code to intel_device_info.c to be shared with error
state (Michal)
v4: Bump u8 to u16 when using sseu_get_eus() (Lionel)
Up to now, subslice mask was assumed to be uniform across slices. But
starting with Cannonlake, slices can be asymmetric (for example slice0
has different number of subslices as slice1+). This change stores all
subslices masks for all slices rather than having a single mask that
applies to all slices.
v2: Rework how we store total numbers in sseu_dev_info (Tvrtko)
Fix CHV eu masks, was reading disabled as enabled (Tvrtko)
Readability changes (Tvrtko)
Add EU index helper (Tvrtko)
v3: Turn ALIGN(v, 8) / 8 into DIV_ROUND_UP(v, BITS_PER_BYTE) (Tvrtko)
Reuse sseu_eu_idx() for setting eu_mask on CHV (Tvrtko)
Reformat debug prints for subslices (Tvrtko)
v4: Change eu_mask helper into sseu_set_eus() (Tvrtko)
v5: With Haswell reporting masks & counts, bump sseu_*_eus() functions
to use u16 (Lionel)
v6: Fix sseu_get_eus() for > 8 EUs per subslice (Lionel)
v7: Change debugfs enabels for number of subslices per slice, will
need a small igt/pm_sseu change (Lionel)
Drop subslice_total field from sseu_dev_info, rely on
sseu_subslice_total() to recompute the value instead (Lionel)
v8: Remove unused function compute_subslice_total() (Lionel)
Andrew Morton [Thu, 8 Mar 2018 00:13:33 +0000 (16:13 -0800)]
drm/i915/guc: work around gcc-4.4.4 union initializer issue
gcc-4.4.4 has problems with initalizers of anon unions.
drivers/gpu/drm/i915/intel_guc_log.c: In function 'guc_log_control':
drivers/gpu/drm/i915/intel_guc_log.c:64: error: unknown field 'logging_enabled' specified in initializer
Work around this.
Fixes: 35fe703c3161 ("drm/i915/guc: Change values for i915_guc_log_control") Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Sagar Arun Kamble <sagar.a.kamble@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180308001333.rI2vrNRTY%akpm@linux-foundation.org
The main difference with previous GENs is that starting from Gen11
each VCS and VECS engine has its own power well, which only exist
if the related engine exists in the HW.
The fallback forcewake request workaround is only needed on gen9
according to the HSDES WA entry (1604254524), so we can go back to using
the simpler fw_domains_get/put functions.
BSpec: 18331
v2: fix fwtable, use array to test shadow tables, create new
accessors to avoid check on every access (Tvrtko)
v3 (from Paulo): Rebase.
v4:
- Range 09400-097FF should be FORCEWAKE_ALL (Daniele)
- Use the BIT macro for forcewake domains (Daniele)
- Add a comment about the range ordering (Oscar)
- Updated commit message (Oscar)
v5: Rebased
v6: Use I915_MAX_VCS/VECS (Michal)
v7: translate FORCEWAKE_ALL to available domains
v8: rebase, add clarification on fallback ack in commit message.
v9: fix rebase issue, change check in fw_domains_init from IS_GEN11
to GEN >= 11
v10: Generate is_genX_shadowed with a macro (Daniele)
Include gen11_fw_ranges in the selftest (Michel)
v11: Simplify FORCEWAKE_ALL, new line between NEEDS_FORCEWAKEs (Tvrtko)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Paulo Zanoni <paulo.r.zanoni@intel.com> Acked-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Signed-off-by: Oscar Mateo <oscar.mateo@intel.com> Signed-off-by: Michel Thierry <michel.thierry@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180302161501.28594-6-mika.kuoppala@linux.intel.com Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Thomas Daniel [Fri, 2 Mar 2018 16:14:59 +0000 (18:14 +0200)]
drm/i915/icl: Enhanced execution list support
Enhanced Execlists is an upgraded version of execlists which supports
up to 8 ports. The lrcs to be submitted are written to a submit queue
(the ExecLists Submission Queue - ELSQ), which is then loaded on the
HW. When writing to the ELSP register, the lrcs are written cyclically
in the queue from position 0 to position 7. Alternatively, it is
possible to write directly in the individual positions of the queue
using the ELSQC registers. To be able to re-use all the existing code
we're using the latter method and we're currently limiting ourself to
only using 2 elements.
v2: Rebase.
v3: Switch from !IS_GEN11 to GEN < 11 (Daniele Ceraolo Spurio).
v4: Use the elsq registers instead of elsp. (Daniele Ceraolo Spurio)
v5: Reword commit, rename regs to be closer to specs, turn off
preemption (Daniele), reuse engine->execlists.elsp (Chris)
v6: use has_logical_ring_elsq to differentiate the new paths
v7: add preemption support, rename els to submit_reg (Chris)
v8: save the ctrl register inside the execlists struct, drop CSB
handling updates (superseded by preempt_complete_status) (Chris)
v9: s/drm_i915_gem_request/i915_request (Mika)
v10: resolved conflict in inject_preempt_context (Mika)
Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Signed-off-by: Thomas Daniel <thomas.daniel@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180302161501.28594-4-mika.kuoppala@linux.intel.com
Starting from Gen11 the context descriptor format has been updated in
the HW. The hw_id field has been considerably reduced in size and engine
class and instance fields have been added.
There is a slight name clashing issue because the field that we call
hw_id is actually called SW Context ID in the specs for Gen11+.
With the current size of the hw_id field we can have a maximum of 2k
contexts at any time, but we could use the sw_counter field (which is sw
defined) to increase that because the HW requirement is that
engine_id + sw id + sw_counter is a unique number.
GuC uses a similar method to support more contexts but does its tracking
at lrc level. To avoid doing an implementation that will need to be
reworked once GuC support lands, defer it for now and mark it as TODO.
v2: rebased, add documentation, fix GEN11_ENGINE_INSTANCE_SHIFT
v3: rebased, bring back lost code from i915_gem_context.c
v4: make TODO comment more generic
v5: be consistent with bit ordering, add extra checks (Chris)
Cc: Oscar Mateo <oscar.mateo@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Oscar Mateo <oscar.mateo@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180302161501.28594-3-mika.kuoppala@linux.intel.com Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Oscar Mateo [Fri, 2 Mar 2018 16:14:57 +0000 (18:14 +0200)]
drm/i915/icl: Correctly initialize the Gen11 engines
Gen11 has up to 4 VCS and up to 2 VECS engines, this patch adds mmio
base definitions for all of them.
Bspec: 20944
Bspec: 7021
v2: Set the correct mmio_base in intel_engines_init_mmio; updating the
base mmio values any later would cause incorrect reads in
i915_gem_sanitize (Michel).
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Ceraolo Spurio, Daniele <daniele.ceraolospurio@intel.com> Signed-off-by: Oscar Mateo <oscar.mateo@intel.com> Signed-off-by: Michel Thierry <michel.thierry@intel.com> Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180302161501.28594-2-mika.kuoppala@linux.intel.com Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Ville Syrjälä [Wed, 17 Jan 2018 19:21:49 +0000 (21:21 +0200)]
drm/i915: Track whether the DP link is trained or not
LSPCON likes to throw short HPDs during the enable seqeunce prior to the
link being trained. These obviously result in the channel CR/EQ check
failing and thus we schedule a pointless hotplug work to retrain the
link. Avoid that by ignoring the bad CR/EQ status until we've actually
initially trained the link.
I've not actually investigated to see what LSPCON is trying to signal
with the short pulse. But as long as it signals anything I think we're
supposed to check the link status anyway, so I don't really see other
good ways to solve this. I've not seen these short pulses being
generated by normal DP sinks.
Ville Syrjälä [Wed, 17 Jan 2018 19:21:48 +0000 (21:21 +0200)]
drm/i915: Nuke intel_dp->channel_eq_status
intel_dp->channel_eq_status is used in exactly one function, and we
don't need it to persist between calls. So just go back to using a
local variable instead.
Ville Syrjälä [Wed, 17 Jan 2018 19:21:47 +0000 (21:21 +0200)]
drm/i915: Move SST DP link retraining into the ->post_hotplug() hook
Doing link retraining from the short pulse handler is problematic since
that might introduce deadlocks with MST sideband processing. Currently
we don't retrain MST links from this code, but we want to change that.
So better to move the entire thing to the hotplug work. We can utilize
the new encoder->hotplug() hook for this.
The only thing we leave in the short pulse handler is the link status
check. That one still depends on the link parameters stored under
intel_dp, so no locking around that but races should be mostly harmless
as the actual retraining code will recheck the link state if we
end up there by mistake.
v2: Rebase due to ->post_hotplug() now being just ->hotplug()
Check the connector type to figure out if we should do
the HDMI thing or the DP think for DDI
[pushed with whitespace changes for sparse] Cc: Manasi Navare <manasi.d.navare@intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Acked-by: Manasi Navare <manasi.d.navare@intel.com> Signed-off-by: Lyude Paul <lyude@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180117192149.17760-3-ville.syrjala@linux.intel.com
Ville Syrjälä [Wed, 17 Jan 2018 19:21:46 +0000 (21:21 +0200)]
drm/i915: Reinitialize sink scrambling/TMDS clock ratio on HPD
The LG 4k TV I have doesn't deassert HPD when I turn the TV off, but
when I turn it back on it will pulse the HPD line. By that time it has
forgotten everything we told it about scrambling and the clock ratio.
Hence if we want to get a picture out if it again we have to tell it
whether we're currently sending scrambled data or not. Implement
that via the encoder->hotplug() hook.
v2: Force a full modeset to not follow the HDMI 2.0 spec more
closely (Shashank)
Ville Syrjälä [Wed, 17 Jan 2018 19:21:45 +0000 (21:21 +0200)]
drm/i915: Convert intel_hpd_irq_event() into an encoder hotplug hook
Allow encoders to customize their hotplug processing by moving the
intel_hpd_irq_event() code into an encoder hotplug vfunc. Currently
only SDVO needs this to re-enable hotplug signalling in the SDVO
chip. We'll use this same hook for DP/HDMI link management later.
No functional change since WA is already applied.
But since it has different names on different databases,
let's document it here to avoid future confusion.
Rodrigo Vivi [Tue, 6 Mar 2018 01:20:00 +0000 (17:20 -0800)]
drm/i915/cnl: document WaVFUnitClockGatingDisable
No functional change. WA is already properly applied.
but in different databases it has different names.
Let's document all of them to avoid future confusion.
drm/i915/psr: Update PSR2 resolution check for Cannonlake
In fact, apply the Cannonlake resolution check for all >= Gen-10 platforms
to be safe.
v3: Update GLK too. (Ville)
Longer variable names.
if-else in place of ternary operator.
v2: Use local variables for resolution limits and print them (Ville)
Chris Wilson [Tue, 6 Mar 2018 13:01:43 +0000 (13:01 +0000)]
drm/i915: Flush waiters on seqno wraparound
Previously, we would spin waiting for all waiters to wake up and notice
their request had completed before we would reset the seqno upon
wraparound. However, we can mark their waits as complete and wake them
up directly using the existing machinery for handling the flushing of
missed wakeups when idling.
Chris Wilson [Tue, 6 Mar 2018 13:01:42 +0000 (13:01 +0000)]
drm/i915: Stop kicking the signaling thread on seqno wraparound
Since commit fd10e2ce9905 ("drm/i915/breadcrumbs: Ignore unsubmitted
signalers"), we cancel the signaler when retiring the request and so
upon wraparound, where we wait for all requests to be retired, we no
longer need to spin waiting for the signaling thread to release its
references to the in-flight requests, and so we can assert that the
signaler is idle.
Chris Wilson [Thu, 22 Feb 2018 09:25:44 +0000 (09:25 +0000)]
drm/i915/breadcrumbs: Reduce signaler rbtree to a sorted list
The goal here is to try and reduce the latency of signaling additional
requests following the wakeup from interrupt by reducing the list of
to-be-signaled requests from an rbtree to a sorted linked list. The
original choice of using an rbtree was to facilitate random insertions
of request into the signaler while maintaining a sorted list. However,
if we assume that most new requests are added when they are submitted,
we see those new requests in execution order making a insertion sort
fast, and the reduction in overhead of each signaler iteration
significant.
Since commit 56299fb7d904 ("drm/i915: Signal first fence from irq handler
if complete"), we signal most fences directly from notify_ring() in the
interrupt handler greatly reducing the amount of work that actually
needs to be done by the signaler kthread. All the thread is then
required to do is operate as the bottom-half, cleaning up after the
interrupt handler and preparing the next waiter. This includes signaling
all later completed fences in a saturated system, but on a mostly idle
system we only have to rebuild the wait rbtree in time for the next
interrupt. With this de-emphasis of the signaler's role, we want to
rejig it's datastructures to reduce the amount of work we require to
both setup the signal tree and maintain it on every interrupt.
References: 56299fb7d904 ("drm/i915: Signal first fence from irq handler if complete") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180222092545.17216-1-chris@chris-wilson.co.uk
drm/i915/error: standardize function style in error capture
some of the static functions used from capture() have the "i915_"
prefix while other don't; most of them take i915 as a parameter, but one
of them derives it internally from error->i915. Let's be consistent by
avoiding prefix for static functions and by getting i915 from
error->i915. While at it, s/dev_priv/i915 in functions that don't
perform register reads.
v2: take i915 from error->i915 (Michal), s/dev_priv/i915,
update commit message
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Michel Thierry <michel.thierry@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180305222122.3547-2-daniele.ceraolospurio@intel.com Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Xiong Zhang [Mon, 5 Mar 2018 22:07:28 +0000 (06:07 +0800)]
drm/i915/gvt: Return error at the failure of finding page_track
In XenGT, ioreq copy is used to trap mmio write and ppgtt write. Both
of them are memory write, ioreq handler couldn't distinguish them. So
ioreq handler probe the ppgtt write handler, if it is succuess, this
ioreq is ppgtt write, otherwise it is mmio write.
So ppgtt write handler should return an error at the failure of finding
page track, it is fatal to implement ioreq handler in XenGT.
Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Thu, 1 Mar 2018 07:49:59 +0000 (15:49 +0800)]
drm/i915/gvt: Fix guest vGPU hang caused by very high dma setup overhead
The implementation of current kvmgt implicitly setup dma mapping at MPT
API gfn_to_mfn. First this design against the API's original purpose.
Second, there is no unmap hit in this design. The result is that the
dma mapping keep growing larger and larger. For mutl-vm case, they will
consume IOMMU IOVA low 4GB address space quickly and so tons of rbtree
entries crated in the IOMMU IOVA allocator. Finally, single IOVA
allocation can take as long as ~70ms. Such latency is intolerable.
To address both above issues, this patch introduced two new MPT API:
o dma_map_guest_page - setup dma map for guest page
o dma_unmap_guest_page - cancel dma map for guest page
The kvmgt implements these 2 API. And to reduce dma setup overhead for
duplicated pages (eg. scratch pages), two caches are used: one is for
mapping gfn to struct gvt_dma, another is for mapping dma addr to
struct gvt_dma.
With these 2 new API, the gtt now is able to cancel dma mapping when page
table is invalidated. The dma mapping is not in a gradual increase now.
v2: follow the old logic for VFIO_IOMMU_NOTIFY_DMA_UNMAP at this point.
Cc: Hang Yuan <hang.yuan@intel.com> Cc: Xiong Zhang <xiong.y.zhang@intel.com> Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Zhenyu Wang [Thu, 22 Feb 2018 07:16:14 +0000 (15:16 +0800)]
drm/i915/gvt: Fix check error of vgpu create failure message
Fix check error at
CHECK drivers/gpu/drm/i915//gvt/kvmgt.c
drivers/gpu/drm/i915//gvt/kvmgt.c:455 intel_vgpu_create() error: we previously assumed 'vgpu' could be null (see line 454)
For failed vgpu create, just show error return in failure message.
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Weinan Li [Fri, 23 Feb 2018 06:46:45 +0000 (14:46 +0800)]
drm/i915/gvt: init mmio by lri command in vgpu inhibit context
There is one issue relates to Coarse Power Gating(CPG) on KBL NUC in GVT-g,
vgpu can't get the correct default context by updating the registers before
inhibit context submission. It always get back the hardware default value
unless the inhibit context submission happened before the 1st time
forcewake put. With this wrong default context, vgpu will run with
incorrect state and meet unknown issues.
The solution is initialize these mmios by adding lri command in ring buffer
of the inhibit context, then gpu hardware has no chance to go down RC6 when
lri commands are right being executed, and then vgpu can get correct
default context for further use.
v3:
- fix code fault, use 'for' to loop through mmio render list(Zhenyu)
v4:
- save the count of engine mmio need to be restored for inhibit context and
refine some comments. (Kevin)
v5:
- code rebase
Cc: Kevin Tian <kevin.tian@intel.com> Cc: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Weinan Li <weinan.z.li@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:54 +0000 (19:19 +0800)]
drm/i915/gvt: Manage shadow pages with radix tree
We don't know how many page tables will be shadowed. It varies
considerably corresponding to guest load. Radix tree is a better
choice for us. Since Page Frame Number is used as key so most of
the bits are common.
Here is some performance data (duration in us) of looking up a
element:
Before: (aka. ppgtt_find_shadow_page)
0.308 0.292 0.246 0.432 0.143 ... 0.311 0.225 0.382 0.199 0.325
After: (aka. intel_vgpu_find_spt_by_mfn)
0.106 0.106 0.107 0.106 0.105 0.107 ... 0.107 0.109 0.105 0.108
This time I didn't get the early data of hash table. The data is
measured when desktop is shown.
As last change, the overall benchmark almost is not changed, but
we get better scalability.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:53 +0000 (19:19 +0800)]
drm/i915/gvt: Provide generic page_track infrastructure for write-protected page
This patch provide generic page_track infrastructure for write-protected
guest page. The old page_track logic gets rewrote and now stays in a new
standalone page_track.c. This page track infrastructure can be both used
by vGUC and GTT shadowing.
The important change is that it uses radix tree instead of hash table.
We don't have a predictable number of pages that will be tracked.
Here is some performance data (duration in us) of looking up a element:
Before: (aka. intel_vgpu_find_tracked_page)
0.091 0.089 0.090 ... 0.093 0.091 0.087 ... 0.292 0.285 0.292 0.291
After: (aka. intel_vgpu_find_page_track)
0.104 0.105 0.100 0.102 0.102 0.100 ... 0.101 0.101 0.105 0.105
The hash table has good performance at beginning, but turns bad with
more pages being tracked even no 3D applications are running. As
expected, radix tree has stable duration and very quick.
The overall benchmark (tested with Heaven Benchmark) marginally improved
since this is not the bottleneck. What we benefit more from this change
is scalability.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:51 +0000 (19:19 +0800)]
drm/i915/gvt: Rename mpt api {set, unset}_wp_page to {enable, disable}_page_track
The kvmgt's implementation of mpt api {set,unset}_wp_page is not real
write-protection - the data get written before invoke this two api.
As discussed, change the mpt api to match the real behavior.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:50 +0000 (19:19 +0800)]
drm/i915/gvt: Rename shadow_page to short name spt
The target structure of some functions is struct intel_vgpu_ppgtt_spt and
their names are xxx_shadow_page. It should be xxx_shadow_page_table. Let's
use short name 'spt' instead to reduce the length. As well as the hash
table name.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:49 +0000 (19:19 +0800)]
drm/i915/gvt: Rework shadow page management code
This is a another big one and the GVT shadow page management code is
heavily refined.
The new code only use struct intel_vgpu_ppgtt_spt to represent a vgpu
shadow page table - w/ or wo/ a guest page associated with. A pure shadow
page (no guest page associated) will be used to shadow splited 2M huge
gtt. In this case, the spt.guest_page.gfn should be a zero.
To search a existed shadow page table, we have two new interfaces:
- intel_vgpu_find_spt_by_gfn(), find a spt by guest gfn. It must not
be a pure spt.
- intel_vgpu_find_spt_by_mfn, Find the spt using shadow page mfn in
shadowed PTE.
The oos_page management is remained as what is was.
v2: Split some changes into small standalone patches.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:44 +0000 (19:19 +0800)]
drm/i915/gvt: Add verbose gtt shadow logs
This add a new macro gvt_vdbg_mm() to print more verbose logs for
gtt shadowing. The added verbose logs are very useful for debugging.
gvt_vdbg_mm() only comes into effect if VERBOSE_DEBUG is defined by
the developer.
Signed-off-by: Changbin Du <changbin.du@intel.com> Reviewed-by: Zhi Wang <zhi.a.wang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Changbin Du [Tue, 30 Jan 2018 11:19:41 +0000 (19:19 +0800)]
drm/i915/gvt: Refine the intel_vgpu_mm reference management
If we manage an object with a reference count, then its life cycle
must flow the reference count operations. Meanwhile, change the
operation functions to generic name *put* and *get*.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This is a big one and the GVT shadow graphic memory management code is
heavily refined. The new code is more straightforward with less code.
The struct intel_vgpu_mm is restructured to be clearly defined, use
accurate names and some of the original fields are removed which are
really redundant.
Now we only manage ppgtt mm object with mm->ppgtt_mm.lru_list. No need
to mix ppgtt and ggtt together, since one vGPU only has one ggtt object.
v4: Don't invoke ppgtt_free_all_shadow_page before intel_vgpu_destroy_all_ppgtt_mm.
v3: Add GVT_RING_CTX_NR_PDPS to avoid confusing about the PDPs.
v2: Split some changes into small standalone patches.
Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Chris Wilson [Mon, 5 Mar 2018 10:33:12 +0000 (10:33 +0000)]
drm/i915: Unwind vma pinning for intel_pin_and_fence_fb_obj error path
If we fail to acquire a fence when we must, we must unwind before
reporting the error. Otherwise, we lose tracking of the vma pinning and
eventually hit a bug like
Mahesh Kumar [Tue, 6 Feb 2018 06:08:55 +0000 (11:38 +0530)]
drm/i915/icl: remove port A/E lane sharing limitation.
Platforms before Gen11 were sharing lanes between port-A & port-E.
This limitation is no more there.
Changes since V1:
- optimize the code (Shashank/Jani)
- create helper function to get max lanes (ville)
Changes since V2:
- Include BIOS fail fix-up in same helper function (ville)
Changes since V3:
- remove confusing if/else (jani)
- group intel_encoder initialization
Chris Wilson [Fri, 2 Mar 2018 14:32:45 +0000 (14:32 +0000)]
drm/i915/execlists: Split spinlock from its irq disabling side-effect
During reset/wedging, we have to clean up the requests on the timeline
and flush the pending interrupt state. Currently, we are abusing the irq
disabling of the timeline spinlock to protect the irq state in
conjunction to the engine's timeline requests, but this is accidental
and conflates the spinlock with the irq state. A baffling state of
affairs for the reader.
Instead, explicitly disable irqs over the critical section, and separate
modifying the irq state from the timeline's requests.
Suggested-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Michel Thierry <michel.thierry@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180302143246.2579-4-chris@chris-wilson.co.uk Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Chris Wilson [Fri, 2 Mar 2018 13:12:46 +0000 (13:12 +0000)]
drm/i915/execlists: Move irq state manipulation inside irq disabled region
Although this state (execlists->active and engine->irq_posted) itself is
not protected by the engine->timeline spinlock, it does conveniently
ensure that irqs are disabled. We can use this to protect our
manipulation of the state and so ensure that the next IRQ to arrive sees
consistent state and (hopefully) ignores the reset engine.
the conclusion is that the only place where the ports are reset to zero,
is from engine->cancel_requests called during i915_gem_set_wedged().
The race is horrible as it results from calling set-wedged on active HW
(the GPU reset failed) and as such we need to be careful as the HW state
changes beneath us. Fortunately, it's the same scary conditions as
affect normal reset, so we can reuse the same machinery to disable state
tracking as we clobber it.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104945 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Michel Thierry <michel.thierry@intel.com> Fixes: af7a8ffad9c5 ("drm/i915: Use rcu instead of stop_machine in set_wedged") Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180302113324.23189-2-chris@chris-wilson.co.uk
Imre Deak [Thu, 1 Mar 2018 13:44:57 +0000 (15:44 +0200)]
drm/i915/gen9, gen10: Disable FBC on planes with a misaligned Y-offset
Enabling FBC on a plane having a Y-offset that isn't divisible by 4 may
cause pipe FIFO underruns and flickers, so disable FBC on such a config.
I tried the followings to work around the issue:
- enable each HW work around in ILK_DPFC_CHICKEN
- disable each compression algorithm in ILK_DPFC_CONTROL
- disable low-power watermarks
None of the above got rid of the problem. I haven't found this issue in
the Bspec/WA database either.
Besides the igt testcase below (yet to be merged) an easy way to
reproduce the issue is to enable a plane with FBC and a plane Y-offset
not aligned to 4 and then just enable/disable FBC in a loop, keeping the
plane enabled.
I could trigger the problem on BXT/GLK/SKL/CNL, so assume for now that it's
only present on GEN9 and GEN10.
v2: (Ville)
- Run the test/apply the WA on CNL as well.
- Use IS_GEN() instead of INTEL_GEN().
- Fix spelling.
drm/i915/uc: Make GuC/HuC fw fetch and loading functions/file structure symmetric
GuC load function is named intel_guc_fw_upload() and HuC load function is
named intel_huc_init_hw(). Make them consistent intel_*_fw_upload. Also
move HuC fw loading functions and declarations to separate files
intel_huc_fw.c|h like GuC.
While at this, do below changes
1. Update kernel-doc comment for intel_*_fw_upload() functions
2. s/huc_ucode_xfer/huc_fw_xfer
3. Introduce intel_huc_fw_init_early()
v2: Changed patch to update HuC functions instead of changing
guc_fw_upload and update file structure. (Michal Wajdeczko)
v3: Added SPDX License identifier to huc_fw.c|h. (Michal Wajdeczko)
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com> Cc: Michal Winiarski <michal.winiarski@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Anusha Srivatsa <anusha.srivatsa@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/1519922745-25441-1-git-send-email-sagar.a.kamble@intel.com
drm/i915: Check for I915_MODE_FLAG_INHERITED before drm_atomic_helper_check_modeset
Moving the check upwards will mean we we no longer have to add planes
and connectors manually, because everything is handled correctly by
drm_atomic_helper_check_modeset() as intended.
[applied with whitespace changes to make sparse happy] Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Lyude Paul <lyude@redhat.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Lyude Paul <lyude@redhat.com> Signed-off-by: Lyude Paul <lyude@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180221092808.30060-1-maarten.lankhorst@linux.intel.com
projects / mirror_ubuntu-hirsute-kernel.git / log