[Why]
Current FPU code for DCN2x is located under dml/dcn2x.
This is not aligned with DC's general source tree
structure.

[How]
Move FPU code for DCN2x to dml/dcn20.
Reviewed-by: NRodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NQingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The display core files rely on FPU, which requires to be compiled with
special flags. Ideally, we don't want these FPU operations spread around
the DC code; nevertheless, it happens in the current source. This commit
introduces a new directory inside DML for centralizing shared DCN
functions that require FPU and have been used outside DML. For
illustrating this process of transferring FPU functions to the DML
folder, this commit moves one of the functions
dcn20_populate_dml_writeback_from_context) that require FPU access to a
single shared file. Notice that this is the first part of the work, and
it does not fix the FPU issue yet; we still need other patches for
achieving the complete FPU isolation.

Changes since V3:
- Jun: Instead of creating a new directory to keep the FPU code, let's
make the DML folder the only part that requires FPU access. Drop
fpu_operation folder.
- Christian: Fix function code style.

Changes since V2:
- Christian: Remove unnecessary wrapper.
- lkp: Add missing prototype.
- Only compile the FPU operations if the DCN option is enabled.

Change since V1:
- Update documentation and rebase.

Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Anson Jacob <Anson.Jacob@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Hersen Wu <hersenxs.wu@amd.com>
Cc: Aric Cyr <aric.cyr@amd.com>
Cc: Jun Lei <jun.lei@amd.com>
Cc: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Cc: Qingqing Zhuo <qingqing.zhuo@amd.com>
Reported-by: Nkernel test robot <lkp@intel.com>
Signed-off-by: NRodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[why]
IP parameter min_meta_chunk_size_bytes is read for bandwidth
calculations but it was never defined.

[how]
Define min_meta_chunk_size_bytes and initialize value to 256.
Reviewed-by: NLaktyushkin Dmytro <dmytro.laktyushkin@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NVictor Lu <victorchengchi.lu@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[why]
IP parameter min_meta_chunk_size_bytes is read for bandwidth
calculations but it was never defined.

[how]
Define min_meta_chunk_size_bytes and initialize value to 256.
Reviewed-by: NLaktyushkin Dmytro <dmytro.laktyushkin@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NVictor Lu <victorchengchi.lu@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
Enable feature for 21.40
Reviewed-by: NSung Lee <Sung.Lee@amd.com>
Acked-by: NRodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: NMichael Strauss <michael.strauss@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
HW has handed down a new sequence that requires access to these
registers.

v2: squash in DCN3.1 fixes (Alex)
Signed-off-by: NWesley Chalmers <Wesley.Chalmers@amd.com>
Reviewed-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: NStylon Wang <stylon.wang@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
This change was found to break some high-refresh modes. Reverting
to unblock mainline.
Signed-off-by: NIlya Bakoulin <Ilya.Bakoulin@amd.com>
Reviewed-by: NSung Lee <Sung.Lee@amd.com>
Acked-by: NStylon Wang <stylon.wang@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
Currently, the code that fills the clock table can miss filling
information about some of the higher voltage states advertised
by the SMU. This, in turn, may cause some of the higher pixel clock
modes (e.g. 8k60) to fail validation.

[How]
Fill the table with one entry per DCFCLK level instead of one entry
per FCLK level. This is needed because the maximum FCLK does not
necessarily need maximum voltage, whereas DCFCLK values from SMU
cover the full voltage range.
Signed-off-by: NIlya Bakoulin <Ilya.Bakoulin@amd.com>
Reviewed-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: NStylon Wang <stylon.wang@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Some hardware revisions do have a max number of lines limitation
not honouring which can cause pstate switch underflow.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: NEric Bernstein <Eric.Bernstein@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The pmfw structs are specific to the asic and should not be
present in base clk_mgr struct

v2: squash in SI fix (Alex)
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: NEric Yang <eric.yang2@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Not all ASICs have same plane capabilities so need to split them
out for proper support handling.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAtufa Khan <Atufa.Khan@amd.com>
Reviewed-by: NAric Cyr <Aric.Cyr@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable max memory lb config to improve stutter efficiency and
latency hiding. Also increase max number of lb lines to be
used by dml since experiments have shown that there isnt a hard max
beyond what fits in lb.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: NEric Bernstein <Eric.Bernstein@amd.com>
Acked-by: NSolomon Chiu <solomon.chiu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

'dce110_resource.h' included in 'dcn21_resource.c' is duplicated.
'hw_gpio.h' included in 'hw_factory_dce110.c' is duplicated.
Signed-off-by: NZhang Yunkai <zhang.yunkai@zte.com.cn>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY & HOW]
Using values provided by DF for latency may cause hangs in
multi display configurations. Revert change to previous value.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NHaonan Wang <Haonan.Wang2@amd.com>
Acked-by: NEryk Brol <eryk.brol@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
With certain fclock overclocks, state 1 may be chosen
as the closest clock level. This may result in this state
being empty if not populated beforehand, resulting in
black screens and screen corruption.

[How]
Copy over all soc states to clock_limits before bounding
box creation to avoid any cases with empty states.

Fixes: f2459c52 ("drm/amd/display: Add Bounding Box State for Low DF PState but High Voltage State")
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1514Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Reviewed-by: NYongqiang Sun <yongqiang.sun@amd.com>
Acked-by: NQingqing Zhuo <Qingqing.Zhuo@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

[WHY & HOW]
Using values provided by DF for latency may cause hangs in
multi display configurations. Revert change to previous value.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NHaonan Wang <Haonan.Wang2@amd.com>
Acked-by: NEryk Brol <eryk.brol@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

After fixing nested FPU contexts caused by 41401ac6 we're still seeing
complaints about spurious kernel_fpu_end(). As it turns out this was
already fixed for dcn20 in commit f41ed88c ("drm/amdgpu/display:
use GFP_ATOMIC in dcn20_validate_bandwidth_internal") but never moved
forward to dcn21.
Signed-off-by: NHolger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Commit 41401ac6 added FPU wrappers to dcn21_validate_bandwidth(),
which was correct. Unfortunately a nested function alredy contained
DC_FP_START()/DC_FP_END() calls, which results in nested FPU context
enter/exit and complaints by kernel_fpu_begin_mask().
This can be observed e.g. with 5.10.20, which backported 41401ac6
and now emits the following warning on boot:

WARNING: CPU: 6 PID: 858 at arch/x86/kernel/fpu/core.c:129 kernel_fpu_begin_mask+0xa5/0xc0
Call Trace:
 dcn21_calculate_wm+0x47/0xa90 [amdgpu]
 dcn21_validate_bandwidth_fp+0x15d/0x2b0 [amdgpu]
 dcn21_validate_bandwidth+0x29/0x40 [amdgpu]
 dc_validate_global_state+0x3c7/0x4c0 [amdgpu]

The warning is emitted due to the additional DC_FP_START/END calls in
patch_bounding_box(), which is inlined into dcn21_calculate_wm(),
its only caller. Removing the calls brings the code in line with
dcn20 and makes the warning disappear.

Fixes: 41401ac6 ("drm/amd/display: Add FPU wrappers to dcn21_validate_bandwidth()")
Signed-off-by: NHolger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

After fixing nested FPU contexts caused by 41401ac6 we're still seeing
complaints about spurious kernel_fpu_end(). As it turns out this was
already fixed for dcn20 in commit f41ed88c ("drm/amdgpu/display:
use GFP_ATOMIC in dcn20_validate_bandwidth_internal") but never moved
forward to dcn21.
Signed-off-by: NHolger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Commit 41401ac6 added FPU wrappers to dcn21_validate_bandwidth(),
which was correct. Unfortunately a nested function alredy contained
DC_FP_START()/DC_FP_END() calls, which results in nested FPU context
enter/exit and complaints by kernel_fpu_begin_mask().
This can be observed e.g. with 5.10.20, which backported 41401ac6
and now emits the following warning on boot:

WARNING: CPU: 6 PID: 858 at arch/x86/kernel/fpu/core.c:129 kernel_fpu_begin_mask+0xa5/0xc0
Call Trace:
 dcn21_calculate_wm+0x47/0xa90 [amdgpu]
 dcn21_validate_bandwidth_fp+0x15d/0x2b0 [amdgpu]
 dcn21_validate_bandwidth+0x29/0x40 [amdgpu]
 dc_validate_global_state+0x3c7/0x4c0 [amdgpu]

The warning is emitted due to the additional DC_FP_START/END calls in
patch_bounding_box(), which is inlined into dcn21_calculate_wm(),
its only caller. Removing the calls brings the code in line with
dcn20 and makes the warning disappear.

Fixes: 41401ac6 ("drm/amd/display: Add FPU wrappers to dcn21_validate_bandwidth()")
Signed-off-by: NHolger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
With certain fclock overclocks, state 1 may be chosen
as the closest clock level. This may result in this state
being empty if not populated beforehand, resulting in
black screens and screen corruption.

[How]
Copy over all soc states to clock_limits before bounding
box creation to avoid any cases with empty states.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Reviewed-by: NYongqiang Sun <yongqiang.sun@amd.com>
Acked-by: NQingqing Zhuo <Qingqing.Zhuo@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
If system is overclocked, only 1 bounding box state will
be sent by SMU. This results in an empty state being copied
for DML calculations causing black screens and corruption.

[How]
Fully populate bounding box before duplicating last state.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NQingqing Zhuo <Qingqing.Zhuo@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

dcn21_validate_bandwidth() calls functions that use floating point math.
On my machine this sometimes results in simd exceptions when there are
other FPU users such as KVM virtual machines running. The screen freezes
completely in this case.

Wrapping the function with DC_FP_START()/DC_FP_END() seems to solve the
problem. This mirrors the approach used for dcn20_validate_bandwidth.

Tested on a AMD Ryzen 7 PRO 4750U (Renoir).

Bug: https://bugzilla.kernel.org/show_bug.cgi?id=206987Signed-off-by: NJan Kokemüller <jan.kokemueller@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

[WHY]
When enabling HDMI on ComboPHY, there are not
enough clock sources to complete display detection.

[HOW]
Initialize more clock sources.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NAnson Jacob <Anson.Jacob@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
When enabling HDMI on ComboPHY, there are not
enough clock sources to complete display detection.

[HOW]
Initialize more clock sources.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NAnson Jacob <Anson.Jacob@amd.com>
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
dram clock change latencies get updated using ddr4 latency table, but
that update does not happen before validation. This value
should not be the default and should be number received from
df for better mode support.
This may cause a PState hang on high refresh panels with short vblanks
such as on 1080p 360hz or 300hz panels.

[HOW]
Update latency from 23.84 to 11.72.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NJake Wang <haonan.wang2@amd.com>
Reviewed-by: NSung Lee <Sung.Lee@amd.com>
Acked-by: NAnson Jacob <anson.jacob@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
dram clock change latencies get updated using ddr4 latency table, but
that update does not happen before validation. This value
should not be the default and should be number received from
df for better mode support.
This may cause a PState hang on high refresh panels with short vblanks
such as on 1080p 360hz or 300hz panels.

[HOW]
Update latency from 23.84 to 11.72.
Tested-by: NDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: NJake Wang <haonan.wang2@amd.com>
Reviewed-by: NSung Lee <Sung.Lee@amd.com>
Acked-by: NAnson Jacob <anson.jacob@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
New value breaks VSR on high refresh panels, reverting until a fix is developed
Signed-off-by: NMichael Strauss <michael.strauss@amd.com>
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NYongqiang Sun <yongqiang.sun@amd.com>
Acked-by: NEryk Brol <eryk.brol@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
dram clock change latencies get updated using ddr4 latency table, but
does that update does not happen before validation. This value
should not be the default and should be number received from
df for better mode support.
This may cause a PState hang on high refresh panels with short vblanks
such as on 1080p 360hz or 300hz panels.

[HOW]
Update latency from 23.84 to 11.72.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
If the system does not have hostvm enabled, disabling it for
DML validation will allow more modes to pass at lower voltage
levels.

[How]
When initializing HostVM save state to hubbub and read back
when populating DML pipes from context.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Acked-by: NBindu Ramamurthy <bindu.r@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why & How]
Program modulo with ref dpp clk Mhz/10.
Program phase with pipe dpp clk Mhz /10.
DMUB FW could use these value to determine optimization clk
for PSR power saving.
Signed-off-by: NYongqiang Sun <yongqiang.sun@amd.com>
Acked-by: NBindu Ramamurthy <bindu.r@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
When mpo enabled, video comes is 709. Desktop use sRGB.
So color change easily noticeable especially when switch between
mpo/non-mpo.

[How]
Force use sRGB for video TF is sRGB or BT709.
DCN1/DCN2 use predefined type with YUV.
DCN3 use distributed points type with YUV.
Signed-off-by: NJing Zhou <Jing.Zhou@amd.com>
Acked-by: NBindu Ramamurthy <bindu.r@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
Due to the freesync power optimization increasing vtotal, the driver
believes that MPO is supported. MPO is turned on. Freesync then
exits, MPO can no longer be supported and immediate flip fails. This
causes corruption on the panel.

[How]
Avoid the freesync optimization when doing validation so in the case
freesync exits, driver can still support the configuration.
Signed-off-by: NIsabel Zhang <isabel.zhang@amd.com>
Acked-by: NBindu Ramamurthy <bindu.r@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
In some cases, prefetch mode 0 is unsupported but prefetch 1 is
supported. Due to previous change always forcing prefetch mode to 0, we
are failing bandwidth validation in cases where we should not.

[How]
By default try prefetch mode 0 but in the case validation fails, attempt
to do prefetch mode 1 to see if it is supported.
Signed-off-by: NIsabel Zhang <isabel.zhang@amd.com>
Acked-by: NBindu Ramamurthy <bindu.r@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
On APU should be always using prefetch mode 0.
Currently, sometimes prefetch mode 1 is being
used causing system to hard hang due to
minTTUVBlank being too low.

[How]
Any ASIC running DCN21 will by default allow
self refresh and mclk switch. This sets both
min and max prefetch mode to 0 by default.
Signed-off-by: NIsabel Zhang <isabel.zhang@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NQingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
Some sink devices wish to have access to the minimum
HBlank supported by the ASIC.

[How]
Make the ASIC minimum HBlank available in Source
Device information address 0x340.
Signed-off-by: NAshley Thomas <Ashley.Thomas2@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY & HOW]
Enable ODM Combine + Fullscreen MPO on DCN2.1
For lower power consumption in video use cases.
Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Signed-off-by: NSung Lee <sung.lee@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why&How]
Create a separate dcn21_fast_validate_bw function for dcn21.
Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
DF PState and Voltage State are coupled such that one cannot be
raised without raising the other. This uses more power than
is necessary in high bandwidth scenarios.

[HOW]
Add logic to create a new bounding box state that allows for
DF PState to be low while Voltage State is high. Watermarks
vlevel calculation logic was also udpated to assume
state 1 contains the new optimized state.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[WHY]
Currently construction of clock limits gets skipped for diags.
This logic would be useful to get tested in diags.

[HOW]
Copy existing states to clk_table such that update_bw_bounding_box
logic gets used to fill the table.
Signed-off-by: NSung Lee <sung.lee@amd.com>
Reviewed-by: NTony Cheng <Tony.Cheng@amd.com>
Acked-by: NEryk Brol <eryk.brol@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>