Since it's inception in 2012 it has been understood that the DRM GEM CMA
helpers do not depend on CMA as the backend allocator. In fact the first
bug fix to ensure the cma-helpers work correctly with an IOMMU backend
appeared in 2014. However currently the documentation for
drm_gem_cma_create() talks about "a contiguous chunk of memory" without
making clear which address space it will be a contiguous part of.
Additionally the CMA introduction is actively misleading because it only
contemplates the CMA backend.
This matters because when the device accesses the bus through an IOMMU
(and don't use the CMA backend) then the allocated memory is contiguous
only in the IOVA space. This is a significant difference compared to the
CMA backend and the behaviour can be a surprise even to someone who does
a reasonable level of code browsing (but doesn't find all the relevant
function pointers ;-) ).
Improve the kernel doc comments accordingly.
Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220608135821.1153346-1-daniel.thompson@linaro.org
A new PVC+DG2 workaround has appeared recently:
- Wa_16015675438
And a couple existing DG2 workarounds have been extended to PVC:
- Wa_14015795083
- Wa_18018781329
Note that Wa_16015675438 asks us to program a register that is in the
0x2xxx range typically associated with the RCS engine, even though PVC
does not have an RCS. By default the GuC will think we've made a
mistake and throw an exception when it sees this register on a CCS
engine's save/restore list, so we need to pass an extra GuC control flag
to tell it that this is expected and not a problem.
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Signed-off-by: Prathap Kumar Valsan <prathap.kumar.valsan@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Anshuman Gupta <anshuman.gupta@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220608005108.3717895-1-matthew.d.roper@intel.com
We're not parsing the 5.4 Gbps value for the old eDP fast link
training link rate, nor are we parsing the new fast link training
link rate field. Remedy both.
Also we'll now use the actual link rate instead of the DPCD BW
register value.
Note that we're not even using this information for anything
currently, so should perhaps just nuke it all unless someone
is planning on implementing fast link training finally...
v2: Stop using the DPCD BW values (Jani)
*20 instead of *2 to get the rate in correct units (Jani)
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220602205649.11283-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
All other chips, from gfx6-gfx10, now include the MODE register at the
end of the wave debug state. This appears to have been missed in gfx11,
so this patch adds in MODE to the debug state for gfx11.
Signed-off-by: Joseph Greathouse <Joseph.Greathouse@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Following commit 17e822f759 ("drm/msm: fix unbalanced
pm_runtime_enable in adreno_gpu_{init, cleanup}"), any call to
adreno_unbind() will disable runtime PM twice, as indicated by the call
trees below:
adreno_unbind()
-> pm_runtime_force_suspend()
-> pm_runtime_disable()
adreno_unbind()
-> gpu->funcs->destroy() [= aNxx_destroy()]
-> adreno_gpu_cleanup()
-> pm_runtime_disable()
Note that pm_runtime_force_suspend() is called right before
gpu->funcs->destroy() and both functions are called unconditionally.
With recent addition of the eDP AUX bus code, this problem manifests
itself when the eDP panel cannot be found yet and probing is deferred.
On the first probe attempt, we disable runtime PM twice as described
above. This then causes any later probe attempt to fail with
[drm:adreno_load_gpu [msm]] *ERROR* Couldn't power up the GPU: -13
preventing the driver from loading.
As there seem to be scenarios where the aNxx_destroy() functions are not
called from adreno_unbind(), simply removing pm_runtime_disable() from
inside adreno_unbind() does not seem to be the proper fix. This is what
commit 17e822f759 ("drm/msm: fix unbalanced pm_runtime_enable in
adreno_gpu_{init, cleanup}") intended to fix. Therefore, instead check
whether runtime PM is still enabled, and only disable it in that case.
Fixes: 17e822f759 ("drm/msm: fix unbalanced pm_runtime_enable in adreno_gpu_{init, cleanup}")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Link: https://lore.kernel.org/r/20220606211305.189585-1-luzmaximilian@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>
[Why]
In cases where there are multiple eDP instances, DMUB needs to know
which instance the command is for. Today, the field for specifying the
panel_inst exists in both dmub_cmd_update_dirty_rect_data and
dmub_cmd_update_cursor_info_data.
For cursor updates, we already specify the panel_inst, but that's not
the case for dirty_rect updates. Today, a value of '0' is used (due
to initial memsetting of the cmd struct to 0)
[how]
In dc_dmub_update_dirty_rect(), Call dc_get_edp_link_panel_inst() to get
the panel_inst, and fill it in the DMUB cmd struct.
v2: Update commit message for clarity.
Signed-off-by: Mikita Lipski <mikita.lipski@amd.com>
Signed-off-by: David Zhang <dingchen.zhang@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
PSR-SU Rate Control - or PSR-SU-RC - enables PSR-SU panels to work with
variable refresh rate to allow for more power savings. Lowering the
refresh rate can increase PSR residency by expanding the eDP main link
shut down duration. It can also lower panel power consumption.
There is a complication with PSR, since the eDP main link can be shut
down. Therefore, the timing controller (TCON) on the eDP sink nees to be
able to scan out its remote buffer independent of the main link. To
allow the eDP source to specify the sink's refresh rate while the link
is off, vendor-specific DPCD registers are used. This allows the eDP
source to then "Rate Control" the panel during PSR active.
[How]
Add DC support to communicate with PSR-SU-RC supported eDP sinks. The
sink will need to know the desired VTotal during PSR active.
This change only adds support to DC, support in amdgpu_dm is still
pending to enable this fully.
Signed-off-by: David Zhang <dingchen.zhang@amd.com>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why & how]
Expose vendor specific DPCD registers for rate controlling the eDP sink
TCON's refresh rate during PSR active. When used in combination with
PSR-SU and Freesync, it is called PSR-SU Rate Contorol, or PSR-SU-RC for
short.
v2: Add all DPCD registers required
Signed-off-by: David Zhang <dingchen.zhang@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This version brings along the following:
- DPP DTO fix
- Transient encoder fix
- Restrict the reading of LTTPR capabilities in LTTPR mode
- Increase maximum stages for BB
- Distinguish HDMI DTO from DP DTO
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
For Pixel Rate control, when on HDMI, HDMI DTO
should be selected instead of DP DTO.
[How]
Add HDMI member to dtbclk_dto_params, so it can be used tell apart HDMI
and DP DTO in the future.
Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Chris Park <chris.park@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
In some scenarios it is possible for the encoder assignment module to be
set to "transient" mode even though there are no new encoder
assignments.
This can lead to incorrect results when querying encoder assignment,
which in turn can cause incorrect displays to be manipulated.
[How]
Only allow encoder assignment to be in transient mode of operation when
there are valid new encoder assignments.
Reviewed-by: Meenakshikumar Somasundaram <Meenakshikumar.Somasundaram@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Jimmy Kizito <Jimmy.Kizito@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
When switching from 1 pipe to 4to1 mpc combine,
DppDtoClk aren't enabled for the disabled pipes
pior to programming the pipes. Upon optimizing
bandwidth, DppDto are enabled causing intermittent
underflow.
[How]
Update dppclk dto whenever pipe are flagged to
enable.
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Reviewed-by: Hansen Dsouza <Hansen.Dsouza@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Duncan Ma <duncan.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This version brings along the following:
- Optimized blank calculations
- More robust DP MST hotplug support
- eDP bug fix relating to ODM
- Revert a patch that caused a regression with DP
- min comp buffer size fix
- Make DP easier to debug
- Calculate the maximum OLED brightness correctly
- 3 plane MPO.
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
For OLED eDP the Display Manager uses max_cll value as a limit
for brightness control.
max_cll defines the content light luminance for individual pixel.
Whereas max_fall defines frame-average level luminance.
The user may not observe the difference in brightness in between
max_fall and max_cll.
That negatively impacts the user experience.
[How]
Use max_fall value instead of max_cll as a limit for brightness control.
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
1. When HPD deassertion is pulled in the middle of
enabe stream link training, we will abort current training
and turn off PHY. This causes current link settings
to be zeroed this causes later stream enablement
sequence to fail as we prefer to carry on enablement
process despite of link training failure for SST.
2. When HPD is toggled after detection before before
the enable stream sequence as a result. There could be
a race condition where we could end up enable stream based
on the previous link even though the link is updated
after the HPD toggle. This causes an issue where our link
bandwidth is no longer enough to accommodate the timing
therefore causes us to oversubscribe MST payload time
slots. As discussed we decided to add basic sanity check
to make sure that our code can handle the oversubscription
failure silently without system hang.
[how]
1. Keep PHY powered on when HPD is deasserted during
enable stream and wait for the detection sequence to power
it off later.
2. Do not allocate payload if the required timeslot for
current timing is greater than 64 timeslots.
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Reviewed-by: George Shen <George.Shen@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Wenjing Liu <wenjing.liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
In 3-way mpo pipes, there is a case that we
overbook the CRB buffer size. At rare instances,
overbooking the crb will cause underflow. This only
happens when det_size changes dynamically
based on pipe_cnt.
[How]
Set min compbuff size to 1 segment when preparing BW.
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Duncan Ma <duncan.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
For panels with pixel clock > 1200MHz that require ODM
in pre-OS, when driver is disabled in OS, odm is enabled.
Upon driver enablement, corruption is seen if
odm was originally enabled. DP_PIXEL_COMBINE and
pixelclk must be programmed prior to programming the
optc-odm registers. However, eDP displays aren't blanked
prior to initializing odm in this case.
[How]
Upon driver enablement, check whether odm is enabled,
if so, blank eDP prior to programming optc-odm
registers.
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Duncan Ma <duncan.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
LTTPR capable devices on the DisplayPort path may assume that
extended LTTPR AUX timeouts will be used after LTTPR capabilities
are read.
When DPTX operates in non-LTTPR mode, AUX timeouts are not
extended and this can result in AUX transactions timing out.
[How]
Use shared helper function to determine LTTPR mode and do not
read LTTPR capabilities in non-LTTPR mode.
Reviewed-by: Mustapha Ghaddar <Mustapha.Ghaddar@amd.com>
Reviewed-by: Meenakshikumar Somasundaram <Meenakshikumar.Somasundaram@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Jimmy Kizito <Jimmy.Kizito@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The intel_dp_dsc_get_output_bpp() function outputs two lines of
unconditional logs, which was okay when it was called only once. But
now, we also call this function from intel_dp_mode_valid(), which is
in turn called for every mode we need to validate. This causes a lot
of useless noise.
Remove the unconditional prints to avoid spamming the logs. Also
remove one more print that is not unconditional, but is related.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220607074433.1202917-1-luca@coelho.fi
