drm/amdgpu: support get_cu_info for Arcturus

This change is because SE/SH layout on Arcturus is 8*1, different from
4*2(or 4*1) on Vega ASICs.

Currently the cu bitmap array is 4x4 size, and besides the bitmap is used widely
across SW stack. To mostly reduce the scale of impact, we make the cu bitmap
array compatible with SE/SH layout on Arcturus. Then the store of cu bits of
each shader array for Arcturus will be like below:
    SE0,SH0 --> bitmap[0][0]
    SE1,SH0 --> bitmap[1][0]
    SE2,SH0 --> bitmap[2][0]
    SE3,SH0 --> bitmap[3][0]
    SE4,SH0 --> bitmap[0][1]
    SE5,SH0 --> bitmap[1][1]
    SE6,SH0 --> bitmap[2][1]
    SE7,SH0 --> bitmap[3][1]

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix pcie_bw on Vega20

The registers used for VG20 are different in that certain performance
counters were split off to TXCLK3/4. Vega10/12 doesn't have this, so add
a new vg20_get_pcie_usage to reflect this change.

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

drm/amdgpu: Update NBIO headers to add TXCLK3/4

These are added for VG20, and are needed for PCIe bandwidth.

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

drm/amdgpu: Add amdgpu_asic_funcs.reset_method for Vega20

Fixes GPU reset crash.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Mark KFD VRAM allocations for wipe on release

Memory used by KFD applications can contain sensitive information that
should not be leaked to other processes. The current approach to prevent
leaks is to clear VRAM at allocation time. This is not effective because
memory can be reused in other ways without being cleared. Synchronously
clearing memory on the allocation path also carries a significant
performance penalty.

Stop clearing memory at allocation time. Instead mark the memory for
wipe on release.

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/amdgpu: Implement VRAM wipe on release

Wipe VRAM memory containing sensitive data when moving or releasing
BOs. Clearing the memory is pipelined to minimize any impact on
subsequent memory allocation latency. Use of a poison value should
help debug future use-after-free bugs.

When moving BOs, the existing ttm_bo_pipelined_move ensures that the
memory won't be reused before being wiped.

When releasing BOs, the BO is fenced with the memory fill operation,
which results in queuing the BO for a delayed delete.

v2: Move amdgpu_amdkfd_unreserve_memory_limit into
amdgpu_bo_release_notify so that KFD can use memory that's still
being cleared in the background

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/amdgpu: Add flag to wipe VRAM on release

This memory allocation flag will be used to indicate BOs containing
sensitive data that should not be leaked to other processes.

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/ttm: Add release_notify callback to ttm_bo_driver

This notifies the driver that a BO is about to be released.

Releasing a BO also invokes the move_notify callback from
ttm_bo_cleanup_memtype_use, but that happens too late for anything
that would add fences to the BO and require a delayed delete.

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/amd/display: Use switch table for dc_to_smu_clock_type

Using a static int array will cause errors if the given dm_pp_clk_type
is out-of-bounds. For robustness, use a switch table, with a default
case to handle all invalid values.

v2: 0 is a valid clock type for smu_clk_type. Return SMU_CLK_COUNT
    instead on invalid mapping.

Signed-off-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Use proper enum conversion functions

clang warns:

drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_pp_smu.c:336:8:
warning: implicit conversion from enumeration type 'enum smu_clk_type'
to different enumeration type 'enum amd_pp_clock_type'
[-Wenum-conversion]
                                        dc_to_smu_clock_type(clk_type),
                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_pp_smu.c:421:14:
warning: implicit conversion from enumeration type 'enum
amd_pp_clock_type' to different enumeration type 'enum smu_clk_type'
[-Wenum-conversion]
                                        dc_to_pp_clock_type(clk_type),
                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

There are functions to properly convert between all of these types, use
them so there are no longer any warnings.

Fixes: a43913ea50a5 ("drm/amd/powerplay: add function get_clock_by_type_with_latency for navi10")
Fixes: e5e4e22391c2 ("drm/amd/powerplay: add interface to get clock by type with latency for display (v2)")
Link: https://github.com/ClangBuiltLinux/linux/issues/586
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix double ucode load by PSP(v3)

previously the ucode loading of PSP was repreated, one executed in
phase_1 init/re-init/resume and the other in fw_loading routine

Avoid this double loading by clearing ip_blocks.status.hw in suspend or reset
prior to the FW loading and any block's hw_init/resume

v2:
still do the smu fw loading since it is needed by bare-metal

v3:
drop the change in reinit_early_sriov, just clear all block's status.hw
in the head place and set the status.hw after hw_init done is enough

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix incorrect judge on sos fw version

for SRIOV the SOS fw of PSP is loaded in hypervisor thus
guest won't tell the version of it, and judging feature by
reading the sos fw version in guest side is completely wrong

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup vega10 SRIOV code path

we can simplify all those unnecessary function under
SRIOV for vega10 since:
1) PSP L1 policy is by force enabled in SRIOV
2) original logic always set all flags which make itself
   a dummy step

besides,
1) the ih_doorbell_range set should also be skipped
for VEGA10 SRIOV.
2) the gfx_common registers should also be skipped
for VEGA10 SRIOV.

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: sort feature status index by asic feature id for smu

before this change, the pp_feature sysfs show feature enable state by
logic feature id, it is not easy to read.
this change will sort pp_features show index by asic feature id.

before:
features high: 0x00000623 low: 0xb3cdaffb
00. DPM_PREFETCHER       ( 0) : enabeld
01. DPM_GFXCLK           ( 1) : enabeld
02. DPM_UCLK             ( 3) : enabeld
03. DPM_SOCCLK           ( 4) : enabeld
04. DPM_MP0CLK           ( 5) : enabeld
05. DPM_LINK             ( 6) : enabeld
06. DPM_DCEFCLK          ( 7) : enabeld
07. DS_GFXCLK            (10) : enabeld
08. DS_SOCCLK            (11) : enabeld
09. DS_LCLK              (12) : disabled
10. PPT                  (23) : enabeld
11. TDC                  (24) : enabeld
12. THERMAL              (33) : enabeld
13. RM                   (35) : disabled
......

after:
features high: 0x00000623 low: 0xb3cdaffb
00. DPM_PREFETCHER       ( 0) : enabeld
01. DPM_GFXCLK           ( 1) : enabeld
02. DPM_GFX_PACE         ( 2) : disabled
03. DPM_UCLK             ( 3) : enabeld
04. DPM_SOCCLK           ( 4) : enabeld
05. DPM_MP0CLK           ( 5) : enabeld
06. DPM_LINK             ( 6) : enabeld
07. DPM_DCEFCLK          ( 7) : enabeld
08. MEM_VDDCI_SCALING    ( 8) : enabeld
09. MEM_MVDD_SCALING     ( 9) : enabeld
10. DS_GFXCLK            (10) : enabeld
11. DS_SOCCLK            (11) : enabeld
12. DS_LCLK              (12) : disabled
13. DS_DCEFCLK           (13) : enabeld
......

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: enable KFD support for navi14

Same as navi10.

Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: disable inject for failed subblocks of gfx

some subblocks of gfx fail in inject test, disable them

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: support gfx ras error injection and err_cnt query

check gfx error count in both ras querry function and
ras interrupt handler.

gfx ras is still disabled by default due to known stability
issue found in gpu reset.

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add RAS callback for gfx

Add functions for RAS error inject and query error counter

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add define for gfx ras subblock

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/include: add define of TCP_EDC_CNT_NEW

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/include: add bitfield define for EDC registers

Add EDC registers to support VEGA20 RAS

Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove ras_reserve_vram in ras injection

error injection address is not in gpu address space

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add check for ras error type

only ue and ce errors are supported

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: update interrupt callback for all ras clients

add err_data parameter in interrupt cb for ras clients

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: allow ras interrupt callback to return error data

add error data as parameter for ras interrupt cb and process it

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: query umc ras error address

query umc ras error address, translate it to gpu 4k page view
and save it.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add structures for umc error address translation

add related registers, callback function and channel index table

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

drm/amdgpu: add support for recording ras error address

more than one error address may be recorded in one query

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: update algorithm of umc uncorrectable error counting

remove the check of ErrorCodeExt

v2: refine the if condition for ue counting

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: switch to amdgpu_umc structure

create new amdgpu_umc structure to for more umc
settings in future and switch to the new structure

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use 64bit operation macros for umc

replace some 32bit macros with 64bit operations to simplify code

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add RREG64/WREG64(_PCIE) operations

add 64 bits register access functions

v2: implement 64 bit functions in low level

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add ras error count after each query (v2)

v1: increase ras ce/ue error count
v2: log the number of correctable and uncorrectable errors

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: querry umc error count

check umc error count in both ras querry function and
ras interrupt handler

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: init umc v6_1 functions for vega20

init umc callback function for vega20 in sw early init phase

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add umc v6_1 query error count support

Implement umc query_ras_error_count function to support querry
both correctable and uncorrectable error

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add umc v6_1_1 IP headers

the change introduces IP headers for unified memory controller (umc)

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add rsmu v_0_0_2 ip headers

remote smu (rsmu) is a sub-block used as ip register interface,
error handling, reset generation.etc

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add amdgpu_umc_functions structure

This is common structure as UMC callback function

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: init RSMU and UMC ip base address for vega20

the driver needs to program RSMU and UMC registers to
support vega20 RAS feature

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move some ras data structure to amdgpu_ras.h

These are common structures that can be included by IP specific
source files

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from vcn_v2_5.c

Unused.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from vcn_v2_0.c

And fix the fallout.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from sdma_v5_0.c

And fix the fallout.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from nv.c

And fix up the fallout.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from navi10_ih.c

And fix the fallout.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h in gfx_v10_0.c

And fix the fallout.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h from amdgpu_amdkfd_gfx_v10.c

Unused.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop drmP.h in amdgpu_amdkfd_arcturus.c

Unused.

Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: determine the features to enable by pptable only

Per current logics, the features to enable are determined together
by driver and pptable. This is not efficient in co-debug with
firmware team.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: correct irq type used for sdma ecc

we should pass irq type, instead of irq client id,
to irq_get/put interface

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: make power limit retrieval as asic specific

The power limit retrieval should be done per asic. Since we may
need to lookup in the pptable and that's really asic specific.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: correct arcturus current clock level calculation

There may be 1Mhz delta between target and actual frequency. That
should be taken into consideration for current level check.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support UMD PSTATE settings on arcturus

Enable arcturus UMD PSTATE support.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: fix arcturus real-time clock frequency retrieval

Make sure we can still get the accurate gfxclk/uclk/socclk frequency
even on dpm disabled.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: remove redundancy debug log in smu

remove redundacy debug log in smu.
eg:
[ 6897.969447] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6899.024114] amdgpu: [powerplay] Unsupported SMU message: 38
[ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6899.024152] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6900.078296] amdgpu: [powerplay] Unsupported SMU message: 38
[ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6900.078333] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled
[ 6901.133230] amdgpu: [powerplay] Unsupported SMU message: 38

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: correct the bitmask used in arcturus

Those bitmask prefixed by "SMU_" should be used.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add missing arcturus feature maps

Add missing feature maps for arcturus.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support fan speed retrieval on arcturus

Support arcturus fan speed retrieval.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support real-time clock retrieval on arcturus

Enable arcturus real-time clock retrieval.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support sensor reading on arcturus

Support sensor reading for gpu loading, power and
temperatures.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: init arcturus SMU metrics table on bootup

Initialize arcturus SMU metrics table.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: correct UVD/VCE/VCN power status retrieval

VCN should be used for Vega20 later ASICs while UVD and VCE
are for previous ASICs.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: correct Navi10 VCN powergate control (v2)

No VCN DPM bit check as that's different from VCN PG. Also
no extra check for possible double enablement/disablement
as that's already done by VCN.

v2: check return value of smu_feature_set_enabled

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support VCN powergate status retrieval for SW SMU

Commonly used for VCN powergate status retrieval for SW SMU.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: support VCN powergate status retrieval on Raven

Enable VCN powergate status report on Raven.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add new sensor type for VCN powergate status

VCN is widely used in new ASICs and different from tranditional
UVD and VCE.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: update more sdma instances irq support

Update for sdma ras ecc_irq and other minors.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/include: adjust base offset of SMUIO and THM for Arcturus

Arcturus has different _BASE_IDX value in some HWIP_offset.h. To make source
files like smu_v11_0.c and soc15.c that include HWIP_offset.h of Vega20
reusable for Arcturus, align this base offset with Vega20.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: hold on the arcturus gfx dpm support in driver

As for now, only "Prefetcher" is guarded to be working from
SMU firmware.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: correct VCN powergate routine for acturus

Arcturus VCN should powergate in the way as Navi.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: enable arcturus powerplay

Arcturus powerplay is ready to use.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: initialize arcturus MP1 and THM base address

Initialize base address for those IPs which are used in powerplay.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: enable SW SMU routine support for arcturus

Enable arcturus SW SMU routines.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: update arcturus_ppt.c/h V3

Arcturus ASIC specific powerplay interfaces.

V2: correct SMU msg naming
    drop unnecessary debugs

V3: rebase (Alex)

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: update arcturus_ppsmc.h

Correct header and fix typo.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: update smu11_driver_if_arcturus.h

It guides how driver should interface with SMU in arcturus.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add SW SMU interface for dumping pptable out (v2)

This is especially useful in early bring up phase.

v2: disabled by default (Alex)

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add smcdpminfo table v4_6 support

New smcdpminfo table used in arcturus.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Save/restore vcc on gfx10

VCC moved out of user SGPR allocation in gfx10. It's now stored
in SGPRs 106-107.

Also fixes incorrect SGPR read offsets.

Cc: Shaoyun Liu <shaoyun.liu@amd.com>
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: shaoyunl <shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Save/restore flat_scratch_lo/hi on gfx10

These moved from SGPRs in gfx9 to HWREG in gfx10.

Cc: Shaoyun Liu <shaoyun.liu@amd.com>
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: shaoyunl <shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix gfx10 wave64 VGPR context restore

Copy/paste error, first 4 VGPRs are separated by 64 dwords (256 bytes).

Cc: Shaoyun Liu <shaoyun.liu@amd.com>
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: shaoyunl <shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Support uclk switching for DCN2

[Why]
We were previously forcing the uclk for every state to max and reducing
the switch time to prevent uclk switching from occuring. This workaround
was previously needed in order to avoid hangs + underflow under certain
display configurations.

Now that DC has the proper fix complete we can drop the hacks and
improve power for most display configurations.

[How]
We still need the function pointers hooked up to grab the real uclk
states from pplib. The rest of the prior hack can be reverted.

The key requirements here are really just DC support, updated firmware,
and support for disabling p-state support when needed in pplib/smu.

When these requirements are met uclk switching works without underflow
or hangs.

Fixes: 02316e963a5a ("drm/amd/display: Force uclk to max for every state")
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Embed DCN2 SOC bounding box

[Why]
In order to support uclk switching on NV10 the SOC bounding box
needs to be updated.

[How]
We currently read the constants from the gpu info FW, but supporting
workarounds in DC for different versions of the FW adds additional
complexity to the codebase.

NV10 has been released so it's cleanest to keep the bounding box and
source code in sync by embedding the bounding box like we do for
other ASICs.

Fixes: 02316e963a5a ("drm/amd/display: Force uclk to max for every state")
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix a potential information leaking bug

Coccinelle reports a path that the array "data" is never initialized.
The path skips the checks in the conditional branches when either
of callback functions, read_wave_vgprs and read_wave_sgprs, is not
registered. Later, the uninitialized "data" array is read
in the while-loop below and passed to put_user().

Fix the path by allocating the array with kcalloc().

The patch is simplier than adding a fall-back branch that explicitly
calls memset(data, 0, ...). Also it does not need the multiplication
1024*sizeof(*data) as the size parameter for memset() though there is
no risk of integer overflow.

Signed-off-by: Wang Xiayang <xywang.sjtu@sjtu.edu.cn>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix error handling in amdgpu_cs_process_fence_dep

We always need to drop the ctx reference and should check
for errors first and then dereference the fence pointer.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/powerplay: provide the interface to disable uclk switch for DAL

provide the interface for DAL to disable uclk switch on navi10.
in this case, the uclk will be fixed to maximum.
this is a workaround when display configuration causes underflow issue.

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu/vcn_v2_0: Move VCN 2.0 specific dec ring test to vcn_v2_0

VCN 2.0 firmware now requires a packet start command to be sent before
any other decode ring buffer command.

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx10: update golden settings for navi14

Updated settings for hw team.

Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: implment sysfs feature status function in smu

1. Unified feature enable status format in sysfs
2. Rename ppfeature to pp_features to adapt other pp sysfs node name
3. this function support all asic, not asic related function.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Acked-by: Rui Huang <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: move smu_feature_update_enable_state to up level

this function is not ip or asic related function,
so move it to top level as public api in smu.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add smu feature name support

add smu_get_feature_name support in smu.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add smu message name support

add smu_get_message_name support in smu.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: move smu types to smu_types.h

move some enum type (message, feature, clock) to smu_types.h.
these types is too long in amdgpu_smu.h, and not clearly.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Default disable GDS for compute+gfx

Units in the GDS block default to allowing all VMIDs access to all
entries. Disable shader access to the GDS, GWS, and OA blocks from all
compute and gfx VMIDs by default. For compute, HWS firmware will set
up the access bits for the appropriate VMID when a compute queue
requires access to these blocks.
The driver will handle enabling access on-demand for graphics VMIDs.

Leaving VMID0 with full access because otherwise HWS cannot save or
restore values during task switch.

v2: Fixed code and comment styling.

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

drm/amd/amdgpu/vcn_v2_0: Mark RB commands as KMD commands

Sets the CMD_SOURCE bit for VCN 2.0 Decoder Ring Buffer commands. This
bit was previously set by the RBC HW on older firmware. Newer firmware
uses a SW RBC and this bit has to be set by the driver.

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/display: fix the build without CONFIG_DRM_AMD_DC_DSC_SUPPORT

Some code was missing the CONFIG_DRM_AMD_DC_DSC_SUPPORT guard.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix amdgpu_display_supported_domains logic.

Add restriction to dissallow GTT domain if the relevant BO
doesn't have USWC flag set to avoid the APU hang scenario.

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

drm/amdgpu/powerplay: use proper revision id for navi

The PCI revision id determines the sku.

Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: put the SMC into the proper state on reset/unload

When doing a GPU reset or unloading the driver, we need to
put the SMU into the apprpriate state for the re-init after
the reset or unload to reliably work.

I don't think this is necessary for BACO because the SMU actually
controls the BACO state to it needs to be active.

For suspend (S3), the asic is put into D3 so the SMU would be
powered down so I don't think we need to put the SMU into
any special state.

Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>