Felix Kuehling [Tue, 9 Jul 2019 00:09:21 +0000 (20:09 -0400)]
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>
Felix Kuehling [Tue, 9 Jul 2019 23:09:42 +0000 (19:09 -0400)]
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>
Leo Li [Thu, 25 Jul 2019 17:12:24 +0000 (13:12 -0400)]
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>
Nathan Chancellor [Thu, 4 Jul 2019 05:52:16 +0000 (22:52 -0700)]
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>
Monk Liu [Wed, 31 Jul 2019 08:47:56 +0000 (16:47 +0800)]
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>
Monk Liu [Tue, 30 Jul 2019 09:32:27 +0000 (17:32 +0800)]
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>
Monk Liu [Tue, 30 Jul 2019 09:21:19 +0000 (17:21 +0800)]
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>
Kevin Wang [Wed, 31 Jul 2019 07:37:07 +0000 (15:37 +0800)]
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>
Alex Deucher [Fri, 26 Jul 2019 19:15:12 +0000 (14:15 -0500)]
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>
Dennis Li [Tue, 23 Jul 2019 10:23:44 +0000 (18:23 +0800)]
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>
Dennis Li [Wed, 31 Jul 2019 12:45:50 +0000 (20:45 +0800)]
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>
Dennis Li [Wed, 31 Jul 2019 12:42:15 +0000 (20:42 +0800)]
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>
Dennis Li [Fri, 19 Jul 2019 07:22:29 +0000 (15:22 +0800)]
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>
Dennis Li [Fri, 19 Jul 2019 06:50:25 +0000 (14:50 +0800)]
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>
Dennis Li [Fri, 19 Jul 2019 06:42:49 +0000 (14:42 +0800)]
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>
Tao Zhou [Wed, 24 Jul 2019 03:19:56 +0000 (11:19 +0800)]
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>
Tao Zhou [Tue, 23 Jul 2019 05:07:24 +0000 (13:07 +0800)]
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>
Tao Zhou [Mon, 22 Jul 2019 12:33:39 +0000 (20:33 +0800)]
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>
Tao Zhou [Mon, 22 Jul 2019 12:27:25 +0000 (20:27 +0800)]
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>
Tao Zhou [Wed, 24 Jul 2019 13:43:45 +0000 (21:43 +0800)]
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>
Tao Zhou [Mon, 22 Jul 2019 10:30:59 +0000 (18:30 +0800)]
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>
Tao Zhou [Mon, 22 Jul 2019 11:20:29 +0000 (19:20 +0800)]
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>
Tao Zhou [Tue, 23 Jul 2019 04:25:16 +0000 (12:25 +0800)]
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>
Tao Zhou [Tue, 23 Jul 2019 04:18:39 +0000 (12:18 +0800)]
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>
Tao Zhou [Tue, 23 Jul 2019 03:57:15 +0000 (11:57 +0800)]
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>
Tao Zhou [Wed, 24 Jul 2019 07:13:27 +0000 (15:13 +0800)]
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>
Tao Zhou [Wed, 31 Jul 2019 12:28:13 +0000 (20:28 +0800)]
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>
Hawking Zhang [Wed, 17 Jul 2019 13:49:53 +0000 (21:49 +0800)]
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>
Hawking Zhang [Wed, 17 Jul 2019 13:47:44 +0000 (21:47 +0800)]
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>
Hawking Zhang [Wed, 31 Jul 2019 12:23:01 +0000 (20:23 +0800)]
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>
Hawking Zhang [Wed, 24 Jul 2019 06:36:49 +0000 (14:36 +0800)]
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>
Hawking Zhang [Wed, 24 Jul 2019 06:13:53 +0000 (14:13 +0800)]
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>
Hawking Zhang [Tue, 23 Jul 2019 11:42:03 +0000 (19:42 +0800)]
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>
Hawking Zhang [Wed, 17 Jul 2019 09:52:28 +0000 (17:52 +0800)]
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>
Hawking Zhang [Wed, 17 Jul 2019 09:34:46 +0000 (17:34 +0800)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:47:26 +0000 (10:47 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:45:52 +0000 (10:45 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:43:40 +0000 (10:43 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:39:40 +0000 (10:39 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:34:39 +0000 (10:34 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:31:44 +0000 (10:31 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:27:57 +0000 (10:27 -0500)]
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>
Alex Deucher [Wed, 31 Jul 2019 15:26:39 +0000 (10:26 -0500)]
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>
Evan Quan [Thu, 25 Jul 2019 08:40:51 +0000 (16:40 +0800)]
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>
Hawking Zhang [Thu, 25 Jul 2019 09:22:01 +0000 (17:22 +0800)]
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>
Evan Quan [Wed, 31 Jul 2019 03:52:37 +0000 (22:52 -0500)]
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>
Evan Quan [Tue, 23 Jul 2019 12:28:14 +0000 (20:28 +0800)]
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>
Evan Quan [Tue, 23 Jul 2019 09:30:35 +0000 (17:30 +0800)]
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>
Evan Quan [Tue, 23 Jul 2019 03:42:24 +0000 (11:42 +0800)]
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>
Kevin Wang [Fri, 19 Jul 2019 08:06:29 +0000 (16:06 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 09:03:02 +0000 (17:03 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 08:26:04 +0000 (16:26 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 04:09:38 +0000 (12:09 +0800)]
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>
Evan Quan [Fri, 19 Jul 2019 09:18:34 +0000 (17:18 +0800)]
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>
Evan Quan [Mon, 29 Jul 2019 18:18:37 +0000 (13:18 -0500)]
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>
Evan Quan [Mon, 22 Jul 2019 07:55:52 +0000 (15:55 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 02:42:29 +0000 (10:42 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 02:27:21 +0000 (10:27 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 01:57:27 +0000 (09:57 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 01:55:36 +0000 (09:55 +0800)]
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>
Evan Quan [Mon, 22 Jul 2019 01:51:59 +0000 (09:51 +0800)]
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>
Le Ma [Tue, 16 Jul 2019 07:21:54 +0000 (15:21 +0800)]
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>
Le Ma [Mon, 15 Jul 2019 10:00:50 +0000 (18:00 +0800)]
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>
Evan Quan [Wed, 17 Jul 2019 01:34:13 +0000 (09:34 +0800)]
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>
Evan Quan [Tue, 16 Jul 2019 03:03:10 +0000 (11:03 +0800)]
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>
Evan Quan [Fri, 12 Jul 2019 08:53:28 +0000 (16:53 +0800)]
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>
Evan Quan [Fri, 12 Jul 2019 08:50:52 +0000 (16:50 +0800)]
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>
Evan Quan [Wed, 31 Jul 2019 04:30:07 +0000 (23:30 -0500)]
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>
Evan Quan [Mon, 29 Jul 2019 17:43:28 +0000 (12:43 -0500)]
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>
Evan Quan [Fri, 12 Jul 2019 08:28:02 +0000 (16:28 +0800)]
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>
Evan Quan [Fri, 12 Jul 2019 08:24:34 +0000 (16:24 +0800)]
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>
Evan Quan [Wed, 31 Jul 2019 03:50:14 +0000 (22:50 -0500)]
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>
Evan Quan [Wed, 10 Jul 2019 01:29:57 +0000 (09:29 +0800)]
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>
Jay Cornwall [Sun, 28 Jul 2019 21:00:59 +0000 (16:00 -0500)]
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>
Jay Cornwall [Sun, 28 Jul 2019 20:25:05 +0000 (15:25 -0500)]
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>
Jay Cornwall [Sun, 28 Jul 2019 20:24:40 +0000 (15:24 -0500)]
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>
Nicholas Kazlauskas [Tue, 30 Jul 2019 13:45:33 +0000 (09:45 -0400)]
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>
Nicholas Kazlauskas [Tue, 30 Jul 2019 13:08:34 +0000 (09:08 -0400)]
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>
Wang Xiayang [Sat, 27 Jul 2019 09:30:30 +0000 (17:30 +0800)]
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>
Christian König [Tue, 30 Jul 2019 09:17:03 +0000 (11:17 +0200)]
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>
Kenneth Feng [Mon, 29 Jul 2019 09:51:55 +0000 (17:51 +0800)]
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>
Thong Thai [Thu, 25 Jul 2019 15:26:56 +0000 (11:26 -0400)]
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>
Alex Deucher [Fri, 26 Jul 2019 19:14:15 +0000 (14:14 -0500)]
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>
Kevin Wang [Thu, 25 Jul 2019 03:47:44 +0000 (11:47 +0800)]
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>
Kevin Wang [Thu, 25 Jul 2019 03:57:25 +0000 (11:57 +0800)]
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>
Kevin Wang [Thu, 25 Jul 2019 03:08:42 +0000 (11:08 +0800)]
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>
Kevin Wang [Thu, 25 Jul 2019 02:32:48 +0000 (10:32 +0800)]
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>
Kevin Wang [Thu, 25 Jul 2019 01:59:46 +0000 (09:59 +0800)]
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>
Joseph Greathouse [Fri, 26 Jul 2019 20:52:05 +0000 (15:52 -0500)]
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>
Thong Thai [Thu, 25 Jul 2019 15:21:58 +0000 (11:21 -0400)]
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>
Alex Deucher [Fri, 26 Jul 2019 16:04:39 +0000 (11:04 -0500)]
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>
Andrey Grodzovsky [Fri, 26 Jul 2019 13:24:35 +0000 (09:24 -0400)]
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>
Alex Deucher [Fri, 26 Jul 2019 03:28:58 +0000 (22:28 -0500)]
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>
Alex Deucher [Thu, 25 Jul 2019 16:44:59 +0000 (11:44 -0500)]
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>
Alex Deucher [Thu, 25 Jul 2019 15:55:57 +0000 (10:55 -0500)]
drm/amdgpu/powerplay: add set_mp1_state for vega12
This sets the SMU into the proper state for various
operations (shutdown, unload, GPU reset, etc.).
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 25 Jul 2019 15:55:25 +0000 (10:55 -0500)]
drm/amdgpu/powerplay: add set_mp1_state for vega10
This sets the SMU into the proper state for various
operations (shutdown, unload, GPU reset, etc.).
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 25 Jul 2019 15:43:28 +0000 (10:43 -0500)]
drm/amdgpu/powerplay: add set_mp1_state for vega20
This sets the SMU into the proper state for various
operations (shutdown, unload, GPU reset, etc.).
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 31 Jul 2019 02:27:03 +0000 (21:27 -0500)]
drm/amdgpu/powerplay: return success if set_mp1_state is not set
Some asics (APUs) don't have this callback so we want to return
success. Avoids spurious error messages on APUs.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 25 Jul 2019 14:41:53 +0000 (09:41 -0500)]
drm/amdgpu/powerplay: add a new interface to set the mp1 state
This is required for certain cases such as various GPU resets
(mode1, mode2), BACO, shutdown, unload, etc. to put the SMU into
the appropriate state for when the hw is re-initialized.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 24 Jul 2019 04:48:21 +0000 (23:48 -0500)]
drm/amdgpu: add reset_method asic callback for navi
Navi uses either mode1 or baco depending on various
conditions.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
projects / platform / kernel / linux-starfive.git / log