drm/amdgpu: Add use_xgmi_p2p module parameter

This parameter controls xGMI p2p communication, which is enabled by
default. However, it can be disabled by setting it to 0. In case xGMI
p2p is disabled in a dGPU, PCIe p2p interface will be used instead.
This parameter is ignored in GPUs that do not support xGMI
p2p configuration.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: fix mode2 reset fail for smu 13.0.5

SMU MSG index should be used as parameter.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: config HDP_MISC_CNTL.READ_BUFFER_WATERMARK

To fix applications running across multiple GPU config hang.

Signed-off-by: Xiaogang Chen <xiaogang.chen@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu/pm: Disable managing hwmon sysfs attributes for ONEVF mode

This patch prohibits performing of set commands on all hwmon attributes
through sysfs in ONEVF mode.

Signed-off-by: Danijel Slivka <danijel.slivka@amd.com>
Acked-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix typo in amdgpu_discovery.c

disocvery -> discovery

Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: increasing DRAM BW percent for DCN315

[why]
DML validation fails when we connect two or
more displays with HDR. Need to increase
DRAM BW to make the validation passing.
Following the value from DCN31.

[how]
Change the max DRAM BW DML field to 60%.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Sung Joon Kim <sungkim@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Set compbuf size to min at prep prevent overbook crb

[Why]
Detbuffer size is dynamically set for dcn31x. At certain moment,
compbuf+(def size * num pipes) > config return buffer size causing
flickering. This is easily reproducible when MPO is
enabled with two displays.

[How]
At prepare BW, use the min comp buffer size. When it is to
optimize BW, set compbuf size back to maximum possible size.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Duncan Ma <duncanma@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: revert populating dcn315 clk table based on dcfclk

[Why & How]
Due to how pmfw fills out the table when dcfclk states are disabled,
using dcfclk based clk table would cause a no read situation.
Revert the change to prevent underflow until a better solution is coded.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: limit unbounded requesting to 5k

[WHy & How]
Unbounded requesting is unsupported on pipe split modes
and this change prevents us running into such a situation
with wide modes.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add reset register dump trace on GPU

Dump the list of register values to trace event on GPU reset.

Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add debugfs for reset registers list

List of register populated for dump collection during the GPU reset.

Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: check vm ready by amdgpu_vm->evicting flag

Workstation application ANSA/META v21.1.4 get this error dmesg when
running CI test suite provided by ANSA/META:
[drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)

This is caused by:
1. create a 256MB buffer in invisible VRAM
2. CPU map the buffer and access it causes vm_fault and try to move
   it to visible VRAM
3. force visible VRAM space and traverse all VRAM bos to check if
   evicting this bo is valuable
4. when checking a VM bo (in invisible VRAM), amdgpu_vm_evictable()
   will set amdgpu_vm->evicting, but latter due to not in visible
   VRAM, won't really evict it so not add it to amdgpu_vm->evicted
5. before next CS to clear the amdgpu_vm->evicting, user VM ops
   ioctl will pass amdgpu_vm_ready() (check amdgpu_vm->evicted)
   but fail in amdgpu_vm_bo_update_mapping() (check
   amdgpu_vm->evicting) and get this error log

This error won't affect functionality as next CS will finish the
waiting VM ops. But we'd better clear the error log by checking
the amdgpu_vm->evicting flag in amdgpu_vm_ready() to stop calling
amdgpu_vm_bo_update_mapping() later.

Another reason is amdgpu_vm->evicted list holds all BOs (both
user buffer and page table), but only page table BOs' eviction
prevent VM ops. amdgpu_vm->evicting flag is set only for page
table BOs, so we should use evicting flag instead of evicted list
in amdgpu_vm_ready().

The side effect of this change is: previously blocked VM op (user
buffer in "evicted" list but no page table in it) gets done
immediately.

v2: update commit comments.

Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Qiang Yu <qiang.yu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/nv: set mode2 reset for MP1 13.0.8

Set mode2 reset support for MP1 13.0.8.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/nv: enable gfx10.3.7 clock gating support

This will enable the following gfx clock gating.
- Fine clock gating
- Medium Grain clock gating
- 3D Coarse clock gating
- Coarse Grain clock gating
- RLC/CP light sleep clock gating

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Fix missing prototype warning

Fix below warning
warning: no previous prototype for '__smu_get_enabled_features' [-Wmissing-prototypes]

Fixes: f141e251474d67 ("drm/amd/pm: validate SMU feature enable message for getting feature enabled mask")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add mode2 reset support for smu 13.0.5

This patch adds mode2 reset support for smu 13.0.5.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fixed warning reported by kernel test robot

Fixed warning reported by kernel test robot:
1.warning: variable 'ras_obj' is used uninitialized
  whenever '||' condition is true.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Print bdf in peer map failure message

Print alloc node, peer node and memory domain when peer map fails. This
is more useful

v2: use dev_err instead of pr_err
    use bdf for identify peer gpu

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: 3.2.174

This version brings along following fixes:
- add debug option to bypass ssinfo from bios.
- Refactor fixed VS logic for non-transparent mode
- add cable ID support for usb c connector
- clear remote dc_sink when stop mst
- Ignore Transitional Invalid Link Rate Error Message
- Fix wrong resolution with DP/VGA adapter
- Refactor PSR DPCD caps detection
- Set compbuf size to min at prep prevent overbook crb
- lock/un-lock cursor if odm pipe split used

Acked-by: Solomon Chiu <solomon.chiu@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>

drm/amd/display: add debug option to bypass ssinfo from bios.

[Why&How]
add debug option to bypass ssinfo from bios.

Reviewed-by: Chris Park <Chris.Park@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Refactor fixed VS logic for non-transparent mode

[Why]
All fixed VS/PE link training sequence should be refactored
into a separate function outside of the standard link training
sequence. This includes the sequence for non-transparent
mode.

[How]
Isolate link training sequence for fixed VS/PE non-transparent
mode into a separate function.

Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: George Shen <George.Shen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: add cable ID support for usb c connector

[how]
Call to DMUB to retrieve usb c cable ID data from PD firmware.
If cable id is retrieved from DMUB, skip reading cable ID from RX.

Reviewed-by: George Shen <George.Shen@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@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>

drm/amd/display: clear remote dc_sink when stop mst

[Why]
Currently, we don't have code path to release remote dc_sink when unplug
MST hub from the system. After few times hotplug, we hit the limition of
maximum number of remote dc_sink and can't light up new connected monitor
anymore.

[How]
Releasing all remote dc_sink at dm_helpers_dp_mst_stop_top_mgr() was
removed by previous patch. Restore it.

Reviewed-by: Jerry Zuo <Jerry.Zuo@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Ignore Transitional Invalid Link Rate Error Message

[Why]
When hotplug or unplug happens, each stream disabled one by one, and then
enable any alived streams. Link phy and payload table is cleared when 1st
stream is disabled. That causes the error message pops up when disable 2nd
stream. There is no active stream after link_rate is cleared.
After all streams are disabled, link will be trained again and link rate is
assigned to any alived streams.

Therefore there is no harm for the error message that represents invalid
link rate value in the atomic reset transitional time period.

[How]
Downgrade the log level from ERROR to DEBUG.

Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix wrong resolution with DP/VGA adapter

[Why]
Hotplugging the VGA side of some DP/VGA adapters caused the display to
light up with the wrong (non-native) resolution.

This is caused by the adapter misbehaving by reporting the wrong number
of downstream ports when the VGA side is unplugged (reports 1 instead of
0), but only if the SINK_COUNT DPCD register is read more than once.

[How]
To work around the adapter behavior, remove the sink if link is
detected, but EDID cannot be read.

Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Ilya <Ilya.Bakoulin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Refactor PSR DPCD caps detection

[Why]
To move the PSR DPCD caps detection into detect_edp_sink_caps()

Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Po Ting Chen <robin.chen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Set compbuf size to min at prep prevent overbook crb

[Why]
Detbuffer size is dynamically set for dcn31x. At certain moment,
compbuf+(def size * num pipes) > config return buffer size causing
flickering. This is easily reproducible when MPO is
enabled with two displays.

[How]
At prepare BW, use the min comp buffer size. When it is to
optimize BW, set compbuf size back to maximum possible size.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Duncan Ma <duncanma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: lock/un-lock cursor if odm pipe split used

[Why]
When system resume from sleep, the cursor lock will be reset
to default(lock status). And the cursor programming sequence
doesn't consider about odm pipe split cause cursor can't be
enabled.

[How]
If odm pipe split has been used, lock/un-lock on each pipes.

Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Paul Hsieh <paul.hsieh@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Turn global functions into static functions

Turn previously global functions into static functions to avoid
-Wmissing-prototype warnings, such as:

drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn30/irq_service_dcn30.c:50:20:
warning: no previous prototype for function 'to_dal_irq_source_dcn30'
[-Wmissing-prototypes]
enum dc_irq_source to_dal_irq_source_dcn30(
                   ^
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn30/irq_service_dcn30.c:50:1:
note: declare 'static' if the function is not intended to be used outside
of this translation unit
enum dc_irq_source to_dal_irq_source_dcn30(
^
static
1 warning generated.

drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c:488:6:
warning: no previous prototype for function
'dcn316_clk_mgr_helper_populate_bw_params' [-Wmissing-prototypes]
void dcn316_clk_mgr_helper_populate_bw_params(
     ^
drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c:488:1:
note: declare 'static' if the function is not intended to be used outside
of this translation unit
void dcn316_clk_mgr_helper_populate_bw_params(
^
static
1 warning generated.

v2: drop is_timing_changed hunk (Alex)

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add missing prototypes to dcn201_init

Include the header with the prototype to silence the following clang
warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn201/dcn201_init.c:127:6:
warning: no previous prototype for function 'dcn201_hw_sequencer_construct'
[-Wmissing-prototypes]
    void dcn201_hw_sequencer_construct(struct dc *dc)
         ^

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Remove unused variable

Remove the variable clamshell_closed from the function
dcn10_align_pixel_clocks.

This was pointed by clang with the following warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:2063:7:
warning: variable 'clamshell_closed' set but not used
[-Wunused-but-set-variable]
    bool clamshell_closed = false;
         ^

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Remove unused dmub_outbox_irq_info_funcs variable

Remove the unused struct irq_source_info_funcs
dmub_outbox_irq_info_funcs from the file, which was declared but never
hooked up.

This was pointed by clang with the following warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn201/irq_service_dcn201.c:141:43:
warning: unused variable 'dmub_outbox_irq_info_funcs'
[-Wunused-const-variable]
static const struct irq_source_info_funcs dmub_outbox_irq_info_funcs = {
                                          ^

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Remove vupdate_int_entry definition

Remove the vupdate_int_entry definition and utilization to avoid the
following warning by Clang:

drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:410:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(0),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:404:2:
note: previous initialization is here
    vupdate_int_entry(0),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:411:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(1),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:405:2:
note: previous initialization is here
    vupdate_int_entry(1),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:412:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(2),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:406:2:
note: previous initialization is here
    vupdate_int_entry(2),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:413:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(3),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:407:2:
note: previous initialization is here
    vupdate_int_entry(3),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:414:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(4),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:408:2:
note: previous initialization is here
    vupdate_int_entry(4),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:415:2:
warning: initializer overrides prior initialization of this subobject
[-Winitializer-overrides]
    vupdate_no_lock_int_entry(5),
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:280:39:
note: expanded from macro 'vupdate_no_lock_int_entry'
    [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
    ^~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:409:2:
note: previous initialization is here
    vupdate_int_entry(5),
    ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/irq/dcn21/irq_service_dcn21.c:269:39:
note: expanded from macro 'vupdate_int_entry'
        [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
        ^~
6 warnings generated.

Fixes: 688f97ed3f5e ("drm/amd/display: Add vupdate_no_lock interrupts for DCN2.1")
Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Remove unused dcn316_smu_set_voltage_via_phyclk function

Remove dcn316_smu_set_voltage_via_phyclk function, which is not used in the
codebase.

This was pointed by clang with the following warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn316/dcn316_smu.c:171:5:
warning: no previous prototype for function
'dcn316_smu_set_voltage_via_phyclk' [-Wmissing-prototypes]
int dcn316_smu_set_voltage_via_phyclk(struct clk_mgr_internal *clk_mgr, int
requested_phyclk_khz)
    ^

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Remove unused temp variable

Remove unused temp variable from the dmub_rb_flush_pending function by
using arithmetic to remove the loop.

The -Wunused-but-set-variable warning was pointed out by Clang with the
following warning:

drivers/gpu/drm/amd/amdgpu/../display/dmub/inc/dmub_cmd.h:2921:12: warning:
variable 'temp' set but not used [-Wunused-but-set-variable]
    uint64_t temp;
             ^

v2: squash in revert and comment update (Alex)

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove unused get_umc_v8_7_channel_index function

Remove get_umc_v8_7_channel_index function, which is not used
in the codebase.

This was pointed by clang with the following warning:

drivers/gpu/drm/amd/amdgpu/umc_v8_7.c:50:24: warning: unused function
'get_umc_v8_7_channel_index' [-Wunused-function]
static inline uint32_t get_umc_v8_7_channel_index(struct amdgpu_device *adev,
                       ^

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Change amdgpu_ras_block_late_init_default function scope

Turn previously global function into a static function to avoid the
following Clang warning:

drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c:2459:5: warning: no previous prototype
for function 'amdgpu_ras_block_late_init_default' [-Wmissing-prototypes]
int amdgpu_ras_block_late_init_default(struct amdgpu_device *adev,
    ^
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c:2459:1: note: declare 'static' if the
function is not intended to be used outside of this translation unit
int amdgpu_ras_block_late_init_default(struct amdgpu_device *adev,
^
static

Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use ktime rather than jiffies for benchmark results

To protect against wraparounds.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use kernel BO API for benchmark buffer management

Simplifies the code quite a bit.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: derive GTT display support from DM

Rather than duplicating the logic in two places,
consolidate the logic in the display manager.

Acked-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/display: split dmcu and gpuvm handling logic

Separate the logic for each of these features to make the
code easier to understand and update in the future.

Acked-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop testing module parameter

This test is not particularly useful now that GTT and GART
are decoupled in the driver.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop benchmark module parameter

Now that we expose the benchmarks via debugfs, there is no
longer a need for the module parameter.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: expose benchmarks via debugfs

They provide a nice smoke test of transfer performance
using SDMA.  Allow the user to run these at runtime
rather than only at init time.

v2: fix permissions (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Use real device for messages

kfd_chardev() doesn't provide much useful information in dev_... messages
on multi-GPU systems because there is only one KFD device, which doesn't
correspond to any particular GPU. Use the actual GPU device to indicate
the GPU that caused a message.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix for possible integer overflow

Fix for possible integer overflow when doing addition.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David Yat Sin <david.yatsin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add a benchmark mutex

To avoid multiple runs in parallel to avoid mixing results.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: print the selected benchmark test in the log

So you can tell which benchmark was run.

v2: print the test description as well

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: plumb error handling though amdgpu_benchmark()

So we can tell when this function fails.

v2: squash in error handling fix (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/benchmark: use dev_info rather than DRM macros for logging

So we can tell which output goes to which device when multiple GPUs
are present.  Also while we are here, convert DRM_ERROR to dev_info.
The error cases are not critical.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: make CRAT table missing message informational only

The driver has a fallback so make the message informational
rather than a warning. The driver has a fallback if the
Component Resource Association Table (CRAT) is missing, so
make this informational now.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1906
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix typo in *whether* in comment

Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: read harvest bit per IP data on legacy GPUs

Based on firmware team's input, harvest table in VBIOS does
not apply well to legacy products like Navi1x, so seperate
harvest mask configuration retrieve from different places.
On legacy GPUs, scan harvest bit per IP data stuctures,
while for newer ones, still read IP harvest info from harvest
table.

v2: squash in fix to limit it to specific skus (Guchun)

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: validate SMU feature enable message for getting feature enabled mask

There's always miss the SMU feature enabled checked in the NPI phase,
so let validate the SMU feature enable message directly rather than
add more and more MP1 version check.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Signed-off-by: Lijo Lazar <Lijo.Lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable TMZ option for onwards asic

The TMZ is disabled by default and enable TMZ option
for the IP discovery based asic will help on the TMZ
function verification.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: bypass tiling flag check in virtual display case (v2)

vkms leverages common amdgpu framebuffer creation, and
also as it does not support FB modifier, there is no need
to check tiling flags when initing framebuffer when virtual
display is enabled.

This can fix below calltrace:

amdgpu 0000:00:08.0: GFX9+ requires FB check based on format modifier
WARNING: CPU: 0 PID: 1023 at drivers/gpu/drm/amd/amdgpu/amdgpu_display.c:1150 amdgpu_display_framebuffer_init+0x8e7/0xb40 [amdgpu]

v2: check adev->enable_virtual_display instead as vkms can be
enabled in bare metal as well.

Signed-off-by: Leslie Shi <Yuliang.Shi@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amdgpu: add modifiers in amdgpu_vkms_plane_init()"

This reverts commit 4046afcebfc3c8c0dd5666c2671b2c192b344f78.

No need to support modifier in virtual kms, otherwise, in SRIOV
mode, when lanuching X server, set crtc will fail due to mismatch
between primary plane modifier and framebuffer modifier.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix criu_restore_bo error handling

Clang static analysis reports this problem
kfd_chardev.c:2327:2: warning: 1st function call argument
  is an uninitialized value
  kvfree(bo_privs);
  ^~~~~~~~~~~~~~~~

Make sure bo_buckets and bo_privs are initialized so freeing them in the
error handling code path will never result in undefined behaviour.

Fixes: 73fa13b6a511 ("drm/amdkfd: CRIU Implement KFD restore ioctl")
Reported-by: Tom Rix <trix@redhat.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Drop IH ring overflow message to dbg

When this was first implemented, overflows weren't expected in regular
operations, and tests weren't in place to cause said overflow. Now there
are cases where overflows occur with real workloads, but we know that
the kernel can handle this robustly, so move the message to a debug
message.

Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add dm ip block for dcn 3.1.5

this patch adds dm ip block for dcn 3.1.5.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 DM Support

Add display manager support for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 CORE

DC core changes for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 blocks to Makefile

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 Command Table Helper

Add vbios table support for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 Resource

Add display resource management for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 DMUB

Add support for the DMUB firmware for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 IRQ

Add irq interfaces for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 GPIO

Add GPIO interfaces for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 CLK_MGR

Clk and SMU interfaces for DCN 3.1.5.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add DCN315 family information

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Acked-by: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/include: add DCN 3.1.5 registers

Add DCN 3.1.5 and DPCS 4.2.2 register headers.

Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Change-Id: I5588a1c422ae384cc76aa42380545dfc1aad1948
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add mp 13.0.5 header files

This patch is to add mp 13.0.5 header files.

v2: update headers

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: add smu_v13_0_5_ppt implementation

this patch adds smu_v13_0_5_ppt implementation.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: update smc message sequence for smu 13.0.5

this patch updates smc message sequence for smu 13.0.5.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/admgpu/pm: add smu 13.0.5 driver interface headers

This patch is to add smu 13.0.5 driver interface headers.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for psp 13.0.5

Enabl psp support for psp 13.0.5.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add smuio support for smuio 13.0.10

this patch adds smuio support for smuio 13.0.10.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for nbio 7.3.0

this patch adds support for nbio 7.3.0.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable vcn pg and cg for vcn 3.1.2

Enable PG and CG for VCN/JPEG

Signed-off-by: Boyuan Zhang <boyuan.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn: add vcn support for vcn 3.1.2

Load VCN FW, set caps.

Signed-off-by: Boyuan Zhang <boyuan.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for sdma 5.2.6

This patch adds support for sdma 5.2.6.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: do not enable asic reset for raven2

The GPU reset function of raven2 is not maintained or tested, so it should be
very unstable.

Now the amdgpu_asic_reset function is added to amdgpu_pmops_suspend, which
causes the S3 test of raven2 to fail, so the asic_reset of raven2 is ignored
here.

Fixes: daf8de0874ab5b ("drm/amdgpu: always reset the asic in suspend (v2)")
Signed-off-by: Chen Gong <curry.gong@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Use proper enum in pm_unmap_queues_v9()

Clang warns:

  drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_packet_manager_v9.c:267:3:
  error: implicit conversion from enumeration type 'enum
  mes_map_queues_extended_engine_sel_enum' to different enumeration type
  'enum mes_unmap_queues_extended_engine_sel_enum'
  [-Werror,-Wenum-conversion]
                  extended_engine_sel__mes_map_queues__sdma0_to_7_sel :
                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  1 error generated.

Use 'extended_engine_sel__mes_unmap_queues__sdma0_to_7_sel' to eliminate
the warning, which is the same numeric value of the proper type.

Fixes: 009e9a158505 ("drm/amdkfd: navi2x requires extended engines to map and unmap sdma queues")
Link: https://github.com/ClangBuiltLinux/linux/issues/1596
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gc 10.3.6 support

this patch adds gc 10.3.6 support.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for gmc10 for gc 10.3.6

this patch adds support for gmc10.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add Clock and Power Gating support for gc 10.3.6

Add below supports:
GFX Coarse Grain Clock Gating(CGCG)
GFX Coarse grain light sleep/deep sleep(CGLS)
GFX Medium Grain Clock Gating(MGCG)
GFX Medium Grain light sleep/deep sleep(MGLS)
GFX Fine Grain Clock Gating(FGCG)
RLC MGLS
CP  MGLS
MMHUB Clock Gating
SDMA Clock Gating
HDP Clock Gating
ATHUB Clock Gating
IH Clock Gating
GFX Power Gating

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add nv common init for gc 10.3.6

This patch adds add nv common init for gc 10.3.6.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/smu11.5: restore cclks in vangogh_set_performance_level

When we disable manual clock setting, we need to restore the cclks
as well as the gfxclk.

Acked-by: Huang Rui <ray.huang@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix amdgpu_ras_block_late_init error handler

Clang build fails with
amdgpu_ras.c:2416:7: error: variable 'ras_obj' is used uninitialized
  whenever 'if' condition is true
  if (adev->in_suspend || amdgpu_in_reset(adev)) {
  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

amdgpu_ras.c:2453:6: note: uninitialized use occurs here
 if (ras_obj->ras_cb)
     ^~~~~~~

There is a logic error in the error handler's labels.
ex/ The sysfs: is the last goto label in the normal code but
is the middle of error handler.  Rework the error handler.

cleanup: is the first error, so it's handler should be last.

interrupt: is the second error, it's handler is next.  interrupt:
handles the failure of amdgpu_ras_interrupt_add_hander() by
calling amdgpu_ras_interrupt_remove_handler().  This is wrong,
remove() assumes the interrupt has been setup, not torn down by
add().  Change the goto label to cleanup.

sysfs is the last error, it's handler should be first.  sysfs:
handles the failure of amdgpu_ras_sysfs_create() by calling
amdgpu_ras_sysfs_remove().  But when the create() fails there
is nothing added so there is nothing to remove.  This error
handler is not needed. Remove the error handler and change
goto label to interrupt.

Fixes: b293e891b057 ("drm/amdgpu: add helper function to do common ras_late_init/fini (v3)")
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Dynamically initialize IP instance attributes

Dynamically initialize IP instance attributes. This eliminates bugs
stemming from adding new attributes to an IP instance.

Cc: Alex Deucher <Alexander.Deucher@amd.com>
Reported-by: Tom StDenis <tom.stdenis@amd.com>
Fixes: 4d7ba312dd1f ("drm/amdgpu: Add "harvest" to IP discovery sysfs")
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu: Add APU flag to gca_config debugfs data (v3)

Needed by umr to detect if ip discovered ASIC is an APU or not.

(v2): Remove asic type from packet it's not strictly needed
(v3): Correct comment

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Use amdgpu_device_should_use_aspm on navi umd pstate switching

The `program_aspm` callback is already guarded for aspm, but the
`enable_aspm` callback doesn't follow the module parameter.

Update it to use the helper `amdgpu_device_should_use_aspm`.

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Refactor `amdgpu_aspm` to be evaluated per device

Evaluating `pcie_aspm_enabled` as part of driver probe has the implication
that if one PCIe bridge with an AMD GPU connected doesn't support ASPM
then none of them do.  This is an invalid assumption as the PCIe core will
configure ASPM for individual PCIe bridges.

Create a new helper function that can be called by individual dGPUs to
react to the `amdgpu_aspm` module parameter without having negative results
for other dGPUs on the PCIe bus.

Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix ARM compilation warning

Fix this ARM warning:

drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:664:35: warning: format '%ld'
expects argument of type 'long int', but argument 4 has type 'size_t' {aka
'unsigned int'} [-Wformat=]

Cc: Alex Deucher <Alexander.Deucher@amd.com>
Cc: kbuild-all@lists.01.org
Cc: linux-kernel@vger.kernel.org
Reported-by: kernel test robot <lkp@intel.com>
Fixes: a6c40b178092 ("drm/amdgpu: Show IP discovery in sysfs")
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Check if ASPM is enabled from PCIe subsystem

commit 0064b0ce85bb ("drm/amd/pm: enable ASPM by default") enabled ASPM
by default but a variety of hardware configurations it turns out that this
caused a regression.

* PPC64LE hardware does not support ASPM at a hardware level.
  CONFIG_PCIEASPM is often disabled on these architectures.
* Some dGPUs on ALD platforms don't work with ASPM enabled and PCIe subsystem
  disables it

Check with the PCIe subsystem to see that ASPM has been enabled
or not.

Fixes: 0064b0ce85bb ("drm/amd/pm: enable ASPM by default")
Link: https://wiki.raptorcs.com/w/images/a/ad/P9_PHB_version1.0_27July2018_pub.pdf
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1723
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1739
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1885
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1907
Tested-by: koba.ko@canonical.com
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant .ras_late_init initialization in some ras blocks

1. Define amdgpu_ras_block_late_init_default in amdgpu_ras.c as
   .ras_late_init common function, which is called when
   .ras_late_init of ras block isn't initialized.
2. Remove the code of using amdgpu_ras_block_late_init to
   initialize .ras_late_init in ras blocks.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: define amdgpu_ras_late_init to call all ras blocks' .ras_late_init

Define amdgpu_ras_late_init to call all ras blocks' .ras_late_init.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Optimize xxx_ras_late_init function of each ras block

1. Move calling ras block instance members from module internal
   function to the top calling xxx_ras_late_init.
2. Module internal function calls can only use parameter variables
   of xxx_ras_late_init instead of ras block instance members.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant calls of ras_late_init in mca ras block

Remove redundant calls of ras_late_init in mca ras block.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant calls of ras_late_init in mmhub ras block

Remove redundant calls of ras_late_init in mmhub ras block.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>