- 29 Aug, 2022 15 commits
-
-
Alvin Lee authored
[Description] We only want to commit the SubVP config to DMCUB after the main and phantom pipe programming has completed. Commiting the state early can cause issues such as P-State being allowed by the HW early which causes the SubVP state machine to go into a bad state Reviewed-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
Brian Chang <Brian.Chang@amd.com> Signed-off-by:
Alvin Lee <Alvin.Lee2@amd.com> Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Taimur Hassan authored
[Why & How] In some cases, there are calls to transition from TX_ON to TX_ON. This is expected, so do not assert. However, these are redundant, so return prematurely. Reviewed-by:
Taimur Hassan <Syed.Hassan@amd.com> Tested-by:
-
Wang Fudong authored
[Why] DIG_FIFO_ERROR = 1 caused mst daisy chain 2nd monitor black. [How] We need to set dig fifo read start level = 7 before dig fifo reset during dig fifo enable according to hardware designer's suggestion. If it is zero, it will cause underflow or overflow and DIG_FIFO_ERROR = 1. Reviewed-by:
Aric Cyr <Aric.Cyr@amd.com> Acked-by:
Wang Fudong <Fudong.Wang@amd.com> Tested-by:
-
YuBiao Wang authored
[Why] In amdgpu_cs_ioctl, amdgpu_job_free could be performed ealier if there is -ERESTARTSYS error. In this case, job->hw_fence could be not initialized yet. Putting hw_fence during amdgpu_job_free could lead to a use-after-free warning. [How] Check if drm_sched_job_init is performed before job_free by checking s_fence. v2: Check hw_fence.ops instead since it could be NULL if fence is not initialized. Reverse the condition since !=NULL check is discouraged in kernel. Signed-off-by:
YuBiao Wang <YuBiao.Wang@amd.com> Reviewed-by:
Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by:
-
Duncan Ma authored
[Why] When ODM is enabled, H timing control register reset to 0. Div mode manual field get overwritten causing no display on certain modes for dcn314. [How] Use REG_UPDATE instead of REG_SET to set div_mode field. Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
Duncan Ma <duncan.ma@amd.com> Tested-by:
-
Lewis Huang authored
[Why] LTTPR caps will read fail if aux channel is not active. [How] 1.Perform 600 read upto 10 retry with 1ms delay in between. 2.If fail, return false and trigger another retry detection. 3.If pass, read LTTPR caps in retrieve link caps. Reviewed-by:
Jimmy Kizito <Jimmy.Kizito@amd.com> Acked-by:
Lewis Huang <Lewis.Huang@amd.com> Tested-by:
-
George Shen authored
[Why] Each index in the DPSTREAMCLK_CNTL register phyiscally maps 1-to-1 with HPO stream encoder instance. On the other hand, each index in DTBCLK_P_CNTL physically maps 1-to-1 with OTG instance. Current DCN32 DPSTREAMCLK_CLK programing assumes that OTG instance always maps 1-to-1 with HPO stream encoder instance. This is not always guaranteed and can result in blackscreen. [How] Program the correct dpstreamclk instance with the correct dtbclk_p source. Reviewed-by:
Ariel Bernstein <Eric.Bernstein@amd.com> Acked-by:
George Shen <george.shen@amd.com> Tested-by:
-
Alvin Lee authored
[Description] There can be SubVP scheduling issues if a SubVP display is chosen has ActiveDramClockChangeLatency > 0. Block this case for now, and enable Vactive case (later) to handle this. Reviewed-by:
-
Taimur Hassan authored
[Why & How] Add addtional check in CalculateODMMode for cases where the ODM combine is needed due to number of DSC slices. Reviewed-by:
-
Aric Cyr authored
This version brings along following fixes: - Modify pipe split policy - Fix odm 2:1 policy in 4k144 mode - Correct HDMI ODM combine policy - Change AUX NACK behavior - Change runtime initialization for DCN32/321 - Fix cursor flicker in PSRSU - Fix k1/k2 divider for virtual signal for DCN32 - Free phantom plane after removing the context - Add interface to track PHY state - Add SubVP scaling case - Add log clock table for SMU - Fix atomic_check check - Fix SMU 13.0.0 driver_if header - Fix doorbells allocation Acked-by:
-
Yang Yingliang authored
Add missing pci_disable_device() if amdgpu_device_resume() fails. Fixes: 8e4d5d43 ("drm/amdgpu: Handling of amdgpu_device_resume return value for graceful teardown") Signed-off-by:
Yang Yingliang <yangyingliang@huawei.com> Signed-off-by:
-
sunliming authored
Fixes the following smatch warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dc_dmub_srv.c:311 dc_dmub_srv_p_state_delegate() warn: variable dereferenced before check 'dc' (see line 309) Reported-by:
kernel test robot <lkp@intel.com> Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
sunliming <sunliming@kylinos.cn> Signed-off-by:
-
ye xingchen authored
Return the value sdma_v5_2_start() directly instead of storing it in another redundant variable. Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
ye xingchen <ye.xingchen@zte.com.cn> Signed-off-by:
-
Graham Sider authored
New GFX11 MES FW adds the trap_en bit. For now hardcode to 1 (traps enabled). Signed-off-by:
Graham Sider <Graham.Sider@amd.com> Acked-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by:
-
Guchun Chen authored
Below driver load error will be printed, not friendly to end user. amdgpu: ATOM BIOS: 113-D603GLXE-077 [drm] FRU: Failed to get size field [drm:amdgpu_fru_get_product_info [amdgpu]] *ERROR* Failed to read FRU Manufacturer, ret:-5 Signed-off-by:
Guchun Chen <guchun.chen@amd.com> Reviewed-by:
Kent Russell <kent.russell@amd.com> Signed-off-by:
-
- 25 Aug, 2022 25 commits
-
-
Bernard Zhao authored
This patch fix cocci warning: drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c:1816:6-8: WARNING: possible condition with no effect (if == else). Signed-off-by:
Bernard Zhao <bernard@vivo.com> Signed-off-by:
-
Bernard Zhao authored
This patch fix cocci warning: drivers/gpu/drm/amd/display/dc/core/dc.c:3335:2-4: WARNING: possible condition with no effect (if == else). Signed-off-by:
-
Bernard Zhao authored
This patch fix potential memory leak (clk_src) when function run into last return NULL. Signed-off-by:
-
Bernard Zhao authored
This patch fix potential memory leak (clk_src) when function run into last return NULL. s/free/kfree/ - Alex Signed-off-by:
-
Qu Huang authored
The mmVM_L2_CNTL3 register is not assigned an initial value Signed-off-by:
Qu Huang <jinsdb@126.com> Signed-off-by:
-
Rafael J. Wysocki authored
According to the ACPI specification [1], the ACPI_FADT_LOW_POWER_S0 flag merely means that it is better to use low-power S0 idle on the given platform than S3 (provided that the latter is supported) and it doesn't preclude using either of them (which of them will be used depends on the choices made by user space). However, on some systems that flag is used to indicate whether or not to enable special firmware mechanics allowing the system to save more energy when suspended to idle. If that flag is unset, doing so is generally risky. Accordingly, add a comment to explain the ACPI_FADT_LOW_POWER_S0 check in amdgpu_acpi_is_s0ix_active(), the purpose of which is otherwise somewhat unclear. Link: https://uefi.org/specs/ACPI/6.4/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#fixed-acpi-description-table-fadt # [1] Reviewed-by:
Mario Limonciello <mario.limonciello@amd.com> Signed-off-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by:
-
Yu Zhe authored
time_after() deals with timer wrapping correctly. Signed-off-by:
Yu Zhe <yuzhe@nfschina.com> Signed-off-by:
-
Guchun Chen authored
'free PSP TMR buffer' happens in suspend, but sometimes in mGPU config, it mixes with PSP resume log printing from another GPU, which is confusing. So use dev_info instead of DRM_INFO for printing. [drm] PSP is resuming... [drm] reserve 0xa00000 from 0x877e000000 for PSP TMR amdgpu 0000:e3:00.0: amdgpu: GECC is enabled amdgpu 0000:e3:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available amdgpu 0000:e3:00.0: amdgpu: SMU is resuming... amdgpu 0000:e3:00.0: amdgpu: smu driver if version = 0x00000040, smu fw if version = 0x00000041, smu fw program = 0, version = 0x003a5400 (58.84.0) amdgpu 0000:e3:00.0: amdgpu: SMU driver if version not matched amdgpu 0000:e3:00.0: amdgpu: dpm has been enabled amdgpu 0000:e3:00.0: amdgpu: SMU is resumed successfully! [drm] DMUB hardware initialized: version=0x02020014 [drm] free PSP TMR buffer [drm] kiq ring mec 2 pipe 1 q 0 Signed-off-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Guchun Chen authored
adev_to_drm is used everywhere in amdgpu code, so modify it to keep consistency. Signed-off-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Likun Gao authored
Enable GFX11 MGCG perfmon setting. V2: set rlc to saft mode before setting. Signed-off-by:
Likun Gao <Likun.Gao@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
Chengming Gui authored
Avoid soft reset, even ip hang check (ring/ib test) when gpu recovery disabled. v2: add missing "}" Signed-off-by:
Chengming Gui <Jack.Gui@amd.com> Reviewed-by:
-
Prike Liang authored
Correct the isa version for handling KFD test. Fixes: 7c4f4f19 ("drm/amdkfd: Add GC 10.3.6 and 10.3.7 KFD definitions") Signed-off-by:
Prike Liang <Prike.Liang@amd.com> Reviewed-by:
Aaron Liu <aaron.liu@amd.com> Signed-off-by:
-
Mukul Joshi authored
When translate_further is enabled, page table depth needs to be updated. This was missing on Arcturus MMHUB init. This was causing address translations to fail for SDMA user-mode queues. Fixes: 352e683b ("drm/amdgpu: Enable translate_further to extend UTCL2 reach") Reviewed-by:
Mukul Joshi <mukul.joshi@amd.com> Signed-off-by:
-
Felix Kuehling authored
Only allocate doorbells when the first queue is created on a GPU or the doorbells need to be mapped into CPU or GPU virtual address space. This avoids allocating doorbells unnecessarily and can allow more processes to use KFD on multi-GPU systems. Signed-off-by:
-
Evan Quan authored
To fit the latest 78.53 PMFW. Signed-off-by:
Feifei Xu <Feifei.Xu@amd.com> Signed-off-by:
-
Roman Li authored
[Why] We have 2 back-to-back checks for skipping connectors. Logically one of them will do the job. [How] Remove redundant check. Reviewed-by:
Hersen Wu <hersenxs.wu@amd.com>> Acked-by:
Roman Li <roman.li@amd.com> Tested-by:
-
Leo Chen authored
[Why & How] Adding log for clock table from SMU helps with the debugging process. Implemented using DC_LOG_SMU to output log. Reviewed-by:
Leo Chen <sancchen@amd.com> Tested-by:
-
Alvin Lee authored
[Description] Uncomment scaling cmd assignment since FW headers are now promoted. Reviewed-by:
Martin Leung <Martin.Leung@amd.com> Acked-by:
-
Alvin Lee authored
[Why] Sometimes pixel clock needs to remain active after transmitter disable. [How] Use update_phy_state to track PHY state after stream enable/disable and program pixel clock as needed. Reviewed-by:
-
Alvin Lee authored
[Description] Refcount is incremented on allocation and when adding to the context. Therefore we must release the phantom plane and stream after removing from the context. Reviewed-by:
-
Aurabindo Pillai authored
[Why&How] When using IGT, kms_bw multi display tests trigger an assert since we ignore virtual signal type. k1/k2 dividers should be correctly programmed if VSYNC needs to be correct. Add the appropriate condition to the if arm to fix this. Reviewed-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Acked-by:
Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by:
-
Robin Chen authored
[Why] The DAL driver may transmit the wrong cursor position to PSRSU DMUB driver when there are multiple planes. [How] Currently the driver apply the HW cursor on the top plane. So we should only transmit the cursor position on the top plane to PSRSU DMUB driver. Reviewed-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
Robin Chen <po-tchen@amd.com> Tested-by:
-
Aurabindo Pillai authored
DC was using compile time initialization of register addresses using SR_* macros and their variants. These have been converted to use runtime initialization. The REG_STRUCT macro is a definition that is added to SR_* macros. During initialization, this must be defined before SR_* macros are invoked, which are in turn invoked through various IP initialization macros. Reviewed-by:
Harry Wentland <Harry.Wentland@amd.com> Acked-by:
-
Aurabindo Pillai authored
DC was using compile time initialization of register addresses using SR_* macros and their variants. These have been converted to use runtime initialization. The REG_STRUCT macro is a definition that is added to SR_* macros. During initialization, this must be defined before SR_* macros are invoked, which are in turn invoked through various IP initialization macros. Reviewed-by:
-
Ilya Bakoulin authored
[Why] Retrying on receiving a NACK can result in long overall EDID read times in some cases. [How] Retry only on DEFER and return immediately on NACK. Reviewed-by:
Ilya Bakoulin <Ilya.Bakoulin@amd.com> Tested-by:
-
