- 25 Mar, 2022 40 commits
-
-
Christian König authored
Instead of trying to figure out if a TLB flush is necessary or not use the information provided by the VM subsystem now. Signed-off-by:
Christian König <christian.koenig@amd.com> Reviewed-by: Philip Yang<Philip.Yang@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Christian König authored
Instead of tracking the VM updates through the dependencies just use a sequence counter for page table updates which indicates the need to flush the TLB. This reduces the need to flush the TLB drastically. v2: squash in NULL check fix (Christian) Signed-off-by:
-
Christian König authored
Store the 64bit sequence directly. Makes it simpler to use and saves a bit of fence reference counting overhead. Signed-off-by:
-
Christian König authored
Separate the VM page table backend operations from the state machine since the amdgpu_vm.c file is becoming to complex. The allocating, freeing and updating page tables and page directories can easily be moved into a separate file. While at it cleanup everything checkpatch.pl reported and rename the functions a bit to make more clear that they belong together. Signed-off-by:
-
Christian König authored
Move the page tables to the idle list after updating the PDEs. We have gone back and forth with that a couple of times because of problems with the inter PD dependencies, but it should work now that we have the state handling cleanly separated. Signed-off-by:
-
Melissa Wen authored
Moves FPU-related structs and dcn316_update_bw_bounding_box from dcn316 driver to dml/dcn31 that centralize FPU operations for DCN 3.1x Signed-off-by:
Melissa Wen <mwen@igalia.com> Reviewed-by:
Alex Hung <alex.hung@amd.com> Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by:
-
Melissa Wen authored
Moves related structs and dcn315_update_bw_bounding_box from dcn315 driver code to dml/dcn31_fpu that centralizes FPU code for DCN 3.1x. Signed-off-by:
-
Melissa Wen authored
Creates FPU files in dml/dcn31 folder to centralize FPU operations from 3.1x drivers and moves all FPU-associated code from dcn31 driver to there. It includes the struct _vcs_dpi_ip_params_st and _vcs_dpi_soc_bounding_box_st and functions: - dcn31_calculate_wm_and_dlg_fp() - dcn31_update_bw_bounding_box() adding dc_assert_fp_enabled to them and drop DC_FP_START/END inside functions that was moved to dml folder, as required. Signed-off-by:
-
Aric Cyr authored
This version brings along following fixes: - HDCP SEND AKI INIT error - fix audio format not updated after edid updated - Reduce stack size - FEC check in timing validation - Add fSMC_MSG_SetDtbClk support - Update VTEM Infopacket definition - [FW Promotion] Release 0.0.109.0 Acked-by:
Aric Cyr <aric.cyr@amd.com> Tested-by:
-
Anthony Koo authored
Reviewed-by:
Anthony Koo <Anthony.Koo@amd.com> Tested-by:
-
Tao Zhou authored
Print the status out when it passes, and also tell user gpu reset is triggered when we fall back to legacy way. v2: make the message more explicit. v3: change succeeds to succeeded. replace pr_warn with dev_warn. Signed-off-by:Tao Zhou <tao.zhou1@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
Ahmad Othman authored
[why] HDCP sends AKI INIT error in case of multiple display on dock [how] Add new checks and method to handle display adjustment for multiple display cases Reviewed-by:
Wenjing Liu <Wenjing.Liu@amd.com> Acked-by:
Ahmad Othman <ahmad.othman@amd.com> Tested-by:
-
Guchun Chen authored
Harvest bit setting in IP data structure promises this, so no need to set it explicitly. Signed-off-by:
Guchun Chen <guchun.chen@amd.com> Reviewed-by:
-
Colin Ian King authored
There is a spelling mistake in a dev_error error message. Fix it. Signed-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
-
Tao Zhou authored
Do RAS page retirement and use gpu reset as fallback in UTCL2 fault handler. v2: replace vm fault event with posion consumed event in UTCL2 poison consumption. Signed-off-by:
-
Tao Zhou authored
Add help functions to query and reset RAS UTCL2 poison status. v2: implement it on amdgpu side and kfd only calls it. Signed-off-by:
-
Tao Zhou authored
Client ID is more accruate here and we can deal with more different cases with client ID. Signed-off-by:
-
Tao Zhou authored
Combine reading and setting poison flag as one atomic operation and add print message for the function. Signed-off-by:
-
Alex Deucher authored
Unused outside of amdgpu_display.c. Reviewed-by:
Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by:
-
Alex Deucher authored
Unused. Reviewed-by:
-
Alex Deucher authored
It's not used outside of amdgpu_display.c. Reviewed-by:
-
Prike Liang authored
Enable gfx1037 clock counter retrieval function for KFDPerfCountersTest.ClockCountersBasicTest. Signed-off-by:
Prike Liang <Prike.Liang@amd.com> Reviewed-by:
-
Prike Liang authored
Disable xnack on the gfx10.3.7 for the KFD test. Signed-off-by:
-
Felix Kuehling authored
Retry faults are not supported on GFX 10.3.4. Signed-off-by:
-
Yifan Zhang authored
this patch to set noretry=1 for gc 10.3.6. Signed-off-by:
Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by:
-
Alex Deucher authored
Port current list from amd-staging-drm-next. Signed-off-by: -
QintaoShen authored
As the kmalloc_array() may return null, the 'event_waiters[i].wait' would lead to null-pointer dereference. Therefore, it is better to check the return value of kmalloc_array() to avoid this confusion. Signed-off-by:
QintaoShen <unSimple1993@163.com> Signed-off-by:
-
Tianci Yin authored
Prior to disabling dpg, VCN need unpausing dpg mode, or VCN will hang in S3 resuming. Reviewed-by:
James Zhu <James.Zhu@amd.com> Signed-off-by:
Tianci Yin <tianci.yin@amd.com> Signed-off-by:
-
Divya Shikre authored
Recently introduced commit 158a05a0 ("drm/amdgpu: Add use_xgmi_p2p module parameter") did not update XGMI iolinks when use_xgmi_p2p is disabled. Add fix to not create XGMI iolinks in KFD topology when this parameter is disabled. Fixes: 158a05a0 ("drm/amdgpu: Add use_xgmi_p2p module parameter") Signed-off-by:
Divya Shikre <DivyaUday.Shikre@amd.com> Reviewed-by:
-
Tushar Patel authored
Compute-only GPUs have more than 8 VMIDs allocated to KFD. Fix this by passing correct number of VMIDs to HWS v2: squash in warning fix (Alex) Signed-off-by:
Tushar Patel <tushar.patel@amd.com> Reviewed-by:
-
Charlene Liu authored
[why and how] TMDS does not need destructive verify link Reviewed-by:
Alan Liu <HaoPing.Liu@amd.com> Acked-by:
Charlene Liu <Charlene.Liu@amd.com> Tested-by:
-
Gabe Teeger authored
[why] When we enter FREESYNC_STATE_VIDEO, we want to use the extra vblank portion to enter zstate if possible. [how] When we enter freesync, a full update is triggered and the new vtotal with extra lines is passed to dml in a stream update. The time gained from extra vblank lines is calculated in microseconds. We allow zstate entry if the time gained is greater than 5 ms, which is the current policy. Furthermore, an optimized value for min_dst_y_next_start is calculated and written to its register. When exiting freesync, another full update is triggered and default values are restored. Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
Gabe Teeger <gabe.teeger@amd.com> Tested-by:
-
Emily Deng authored
Correct the code error for setting register UVD_GFX10_ADDR_CONFIG. Need to use inst_idx, or it only will set VCN0. Signed-off-by:
Emily Deng <Emily.Deng@amd.com> Reviewed-by:
-
Leo (Hanghong) Ma authored
[Why & How] The latest HDMI SPEC has updated the VTEM packet structure, so change the VTEM Infopacket defined in the driver side to align with the SPEC. Reviewed-by:
Chris Park <Chris.Park@amd.com> Acked-by:
Leo (Hanghong) Ma <hanghong.ma@amd.com> Tested-by:
-
Oliver Logush authored
[why] Needed to support dcn315 Reviewed-by:
Oliver Logush <oliver.logush@amd.com> Tested-by:
-
Chiawen Huang authored
[Why] disable/enable leads FEC mismatch between hw/sw FEC state. [How] check FEC status to fastboot on/off. Reviewed-by:
Chiawen Huang <chiawen.huang@amd.com> Tested-by:
-
Rodrigo Siqueira authored
Linux kernel enabled more compilation restrictions related to the stack size, which caused compilation failures in our code. This commit reduces the allocation size by allocating the required memory dynamically. Reviewed-by:
Harry Wentland <Harry.Wentland@amd.com> Reviewed-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Tested-by:
-
Charlene Liu authored
[why] for the case edid change only changed audio format. driver still need to update stream. Reviewed-by:
Alvin Lee <Alvin.Lee2@amd.com> Reviewed-by:
-
Nicholas Kazlauskas authored
[Why] It changed since dcn30 but the hubbub31 constructor hasn't been modified to reflect this. [How] Update the value in the constructor to 0x6 so we're checking the right bits for p-state allow. It worked before by accident, but can falsely assert 0 depending on HW state transitions. The most frequent of which appears to be when all pipes turn off during IGT tests. Cc: Harry Wentland <harry.wentland@amd.com> Fixes: e7031d82 ("drm/amd/display: Add pstate verification and recovery for DCN31") Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Acked-by:
-
Stanley.Yang authored
It must check asic whether support smu before call smu powerplay function, otherwise it may cause null point on no support smu asic. Signed-off-by:
Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
