- 27 Jan, 2022 10 commits
-
-
Tianci.Yin authored
[why] In rmmod procedure, kfd sends cp a dequeue request, but the request does not get response, then an error message "cp queue pipe 4 queue 0 preemption failed" printed. [how] Performing kfd suspending after disabling gfxoff can fix it. Acked-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
Yang Wang <kevinyang.wang@amd.com> Signed-off-by:
Tianci.Yin <tianci.yin@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Evan Quan authored
The sub-routine(amdgpu_gfx_off_ctrl) tried to obtain the lock adev->pm.mutex which was actually hold by amdgpu_dpm_force_performance_level. A deadlock happened then. Signed-off-by:
Evan Quan <evan.quan@amd.com> Reviewed-by:
-
Evan Quan authored
The existing way cannot handle Beige Goby well as a different PPTable data structure(PPTable_beige_goby_t instead of PPTable_t) is used there. Signed-off-by:
-
Fangzhi Zuo authored
[Why] configure_dp_hpo_throttled_vcp_size() was missing promotion before, but it was covered by not calling the missing function hook in the old interface hpo_dp_link_encoder->funcs. Recent refactor replaces with new caller link_hwss->set_throttled_vcp_size which needs that hook, and that causes null ptr hang. Signed-off-by:
Fangzhi Zuo <Jerry.Zuo@amd.com> Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
Philip Yang authored
kfd_process_notifier_release flush svm_range_restore_work which calls svm_range_list_lock_and_flush_work to flush deferred_list work, but if deferred_list work mmput release the last user, it will call exit_mmap -> notifier_release, it is deadlock with below backtrace. Move flush svm_range_restore_work to kfd_process_wq_release to avoid deadlock. Then svm_range_restore_work take task->mm ref to avoid mm is gone while validating and mapping ranges to GPU. Workqueue: events svm_range_deferred_list_work [amdgpu] Call Trace: wait_for_completion+0x94/0x100 __flush_work+0x12a/0x1e0 __cancel_work_timer+0x10e/0x190 cancel_delayed_work_sync+0x13/0x20 kfd_process_notifier_release+0x98/0x2a0 [amdgpu] __mmu_notifier_release+0x74/0x1f0 exit_mmap+0x170/0x200 mmput+0x5d/0x130 svm_range_deferred_list_work+0x104/0x230 [amdgpu] process_one_work+0x220/0x3c0 Signed-off-by:
Philip Yang <Philip.Yang@amd.com> Reported-by:
Ruili Ji <ruili.ji@amd.com> Tested-by:
-
Philip Yang authored
svm_deferred_list work should continue to handle deferred_range_list which maybe split to child range to avoid child range leak, and remove ranges mmu interval notifier to avoid mm mm_count leak. So taking mm reference when adding range to deferred list, to ensure mm is valid in the scheduled deferred_list_work, and drop the mm referrence after range is handled. Signed-off-by:
-
Philip Yang authored
SVM ioctls take proper svms->lock to handle race conditions, don't need take process mutex to serialize ioctls. This also fixes circular locking warning: WARNING: possible circular locking dependency detected Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock((work_completion)(&svms->deferred_list_work)); lock(&process->mutex); lock((work_completion)(&svms->deferred_list_work)); lock(&process->mutex); *** DEADLOCK *** Signed-off-by: -
Bas Nieuwenhuizen authored
Unused. Convert the divisions into asserts on the divisor, to debug why it is zero. The divide by zero is suspected of causing kernel panics. While I have no idea where the zero is coming from I think this patch is a positive either way. Reviewed-by:
Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Signed-off-by:
-
Eric Huang authored
It is to meet the requirement for memory allocation optimization on MI50. Signed-off-by:
Eric Huang <jinhuieric.huang@amd.com> Acked-by:
-
Victor Zhao authored
add determine for passthrough mode under arm64 by reading CurrentEL register v2: squash in warning fix (Alex) Signed-off-by:
Victor Zhao <Victor.Zhao@amd.com> Acked-by:
-
- 25 Jan, 2022 30 commits
-
-
Tim Huang authored
Use IP versions rather than asic_type to differentiate IP version specific features. Signed-off-by:
Tim Huang <tim.huang@amd.com> Reviewed-by:
Aaron Liu <aaron.liu@amd.com> Acked-by:
Huang Rui <ray.huang@amd.com> Signed-off-by:
-
Tim Huang authored
Use IP versions rather than asic_type to differentiate IP version specific features. Signed-off-by:
-
Tim Huang authored
Use IP versions rather than asic_type to differentiate IP version specific features. Signed-off-by:
-
Randy Dunlap authored
Change a static function's comment from "/**" (indicating kernel-doc notation) to "/*" (indicating a regular C language comment). This prevents multiple kernel-doc warnings: drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link_dp.c:4343: warning: Function parameter or member 'max_supported_frl_bw_in_kbps' not described in 'intersect_frl_link_bw_support' drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link_dp.c:4343: warning: Function parameter or member 'hdmi_encoded_link_bw' not described in 'intersect_frl_link_bw_support' drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link_dp.c:4343: warning: expecting prototype for Return PCON's post FRL link training supported BW if its non(). Prototype was for intersect_frl_link_bw_support() instead Fixes: c022375a ("drm/amd/display: Add DP-HDMI FRL PCON Support in DC") Reviewed-by:
Randy Dunlap <rdunlap@infradead.org> Reported-by:
kernel test robot <lkp@intel.com> Cc: Fangzhi Zuo <Jerry.Zuo@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Leo Li <sunpeng.li@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by:
-
Colin Ian King authored
The pointer reg is being assigned a value that is not read, the exit path via label 'out' never accesses it. The assignment is redundant and can be removed. Cleans up clang scan build warning: drivers/gpu/drm/radeon/radeon_object.c:570:3: warning: Value stored to 'reg' is never read [deadcode.DeadStores] Reviewed-by:
Nick Desaulniers <ndesaulniers@google.com> Signed-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
-
Tom Rix authored
clang static analysis reports this represenative problem amdgpu_smu.c:144:18: warning: The left operand of '*' is a garbage value return clk_freq * 100; ~~~~~~~~ ^ If there is no get_dpm_ultimate_freq function, smu_get_dpm_freq_range returns success without setting the output min,max parameters. So return an -ENOTSUPP error. Fixes: e5ef784b ("drm/amd/powerplay: revise calling chain on retrieving frequency range") Signed-off-by:Tom Rix <trix@redhat.com> Signed-off-by:
-
Bas Nieuwenhuizen authored
Mirrors the logic for dcn30. Cue lots of WARNs and some kernel panics without this fix. Signed-off-by:
-
Bas Nieuwenhuizen authored
It calls populate_dml_pipes which uses doubles to initialize the scale_ratio_depth params. Mirrors the dcn20 logic. Signed-off-by:
-
Zhou Qingyang authored
In amdgpu_dm_connector_add_common_modes(), amdgpu_dm_create_common_mode() is assigned to mode and is passed to drm_mode_probed_add() directly after that. drm_mode_probed_add() passes &mode->head to list_add_tail(), and there is a dereference of it in list_add_tail() without recoveries, which could lead to NULL pointer dereference on failure of amdgpu_dm_create_common_mode(). Fix this by adding a NULL check of mode. This bug was found by a static analyzer. Builds with 'make allyesconfig' show no new warnings, and our static analyzer no longer warns about this code. Fixes: e7b07cee ("drm/amd/display: Merge amdgpu_dm_types and amdgpu_dm") Signed-off-by:
Zhou Qingyang <zhou1615@umn.edu> Signed-off-by:
-
Zhou Qingyang authored
In calculate_bandwidth(), the tag free_sclk and free_yclk are reversed, which could lead to a memory leak of yclk. Fix this bug by changing the location of free_sclk and free_yclk. This bug was found by a static analyzer. Builds with 'make allyesconfig' show no new warnings, and our static analyzer no longer warns about this code. Fixes: 2be8989d ("drm/amd/display/dc/calcs/dce_calcs: Move some large variables from the stack to the heap") Signed-off-by:
-
Alex Deucher authored
Check IP versions rather than asic types. Acked-by:
-
Alex Deucher authored
Extend secondary function handling for runtime pm beyond audio to USB and UCSI. Reviewed-by:
-
Alex Deucher authored
Seems more logical to enable runtime pm at the end of the init sequence so we don't end up entering runtime suspend before init is finished. Reviewed-by:
-
Alex Deucher authored
We need to set the APU flag from IP discovery before we evaluate this code. Acked-by:
-
Alex Deucher authored
Use the IP versions to set the APU flag when necessary. Reviewed-by:
-
Alex Deucher authored
Earlier chips only had two performance levels, but newer ones potentially had more. The message is harmless. Drop the message to avoid spamming the log. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1874Reviewed-by:
-
Nicholas Kazlauskas authored
[Why] A porting error resulted in the stream assignment for the link being retained without being released - a memory leak. [How] Fix the porting error by adding back the dc_stream_release() intended as part of the original patch. Fixes: 0bb24555 ("drm/amd/display: retain/release at proper places in link_enc assignment") Signed-off-by:
Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by:
-
Eric Yang authored
[Why] To help triage issues and coordinate driver/bios release dependency [How] Only enable the new Z9 interface when debug option is set, otherwise treat Z10 only support case as Zstate disallowed. Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Reviewed-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Signed-off-by:
-
Eric Yang authored
[Why] psr_feature_enabled flag is dynamically updated, and sometimes when zstate allow status is determined the flag has not been set to true yet even on PSR enabled config, lid off/on is such a case, which will result in zstate disabled even though PSR is supported. [How] Check the supported PSR version and the PSR disable status instead. Tested-by:
-
David Galiffi authored
[How & Why] Disable physym clock when it's not in use. Tested-by:
David Galiffi <David.Galiffi@amd.com> Signed-off-by:
-
David Galiffi authored
[How & Why] Updated procedure to match hardware programming guide. Tested-by:
-
Wenjing Liu authored
[why] There is a change previously to disallow DM to set dp drive setings when stream is not present. The logic might not work well with DP PHY complaince scenario with a PHY test fixture attachment. We need to make the method allow DP link drive settings changes even without stream attached to it. [how] revert back to previous code in set drive setting function then add an empty link_resource structure, then assign link resource based on current link resource if link resource is allocated to the current pipe. Tested-by:
Wayne Lin <Wayne.Lin@amd.com> Acked-by:
Wenjing Liu <wenjing.liu@amd.com> Signed-off-by:
-
Aric Cyr authored
Tested-by:
Aric Cyr <aric.cyr@amd.com> Signed-off-by:
-
Aric Cyr authored
This version brings along the following fixes: - Z9 improvements - Clocks management adjustments - Code cleanup - Improve DSC and MST code Tested-by:
-
Wayne Lin authored
[Why & How] In order to easily test ilr by immediately reset the preferred training settings, fix the code to disable skip_immediate_retrain. Tested-by:
Solomon Chiu <solomon.chiu@amd.com> Acked-by:
-
Anthony Koo authored
- Add Scr8 for GPINT messaging between driver and fw Tested-by:
Anthony Koo <Anthony.Koo@amd.com> Signed-off-by:
-
Reza Amini authored
[why] OnSetSourceContentAttribute it does not trigger an update for the VSC with TF change. [how] In this call, create a new VSC infoPacket based on the new config, and allow DisplayTarget decide if an update and pursuant passive flip is necessary Tested-by:
Krunoslav Kovac <Krunoslav.Kovac@amd.com> Acked-by:
Reza Amini <Reza.Amini@amd.com> Signed-off-by:
-
Sung Joon Kim authored
[why] Due to bad hardware, the PHY repeater count in LTTPR cap is read as 0xFF in some monitors while the LTTPR is actually present. [how] Remove PHY repeater counter check when configuring LTTPR mode. Tested-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
Sung Joon Kim <sungkim@amd.com> Signed-off-by:
-
Roy Chan authored
[Why] indirect register index/data pair may be used by multi-threads. when it happens, it would cause register access issue that is hard to trace. [How] Using cgs service, which provide a sync indirect reg access api. Tested-by:
Roy Chan <roy.chan@amd.com> Signed-off-by:
-
Fangzhi Zuo authored
DSC sequence for non virtual dpcd synaptics hub Tested-by:
Hersen Wu <hersenxs.wu@amd.com> Acked-by:
-
