summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2019-07-05drm/amdgpu: add mode1 (psp) reset for navi asicKevin Wang1-2/+36
add mode1 (by psp) reset for navi asic. Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-05drm/amdgpu: Disable ras features on all IPs before gpu resetxinhui pan1-0/+4
Perform a ras_suspend to disable ras on all IPs to workaround some ROCm stability issue. Signed-off-by: xinhui pan <xinhui.pan@amd.com> Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-05drm/amd/powerplay: add baco smu reset function for smu11Kevin Wang7-3/+158
add baco reset support for smu11. it can help gpu do asic reset when gpu recovery. Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-05drm/amd/powerplay: increase waiting time for smu responseJack Xiao1-3/+3
We observed some SMU commands take more time for execution, so increase waiting time for response. Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-05drm/amd/powerplay/smu7_hwmgr: replace blocking delay with non-blockingYrjan Skrimstad1-1/+1
This driver currently contains a repeated 500ms blocking delay call which causes frequent major buffer underruns in PulseAudio. This patch fixes this issue by replacing the blocking delay with a non-blocking sleep call. Signed-off-by: Yrjan Skrimstad <yrjan@skrimstad.net> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-05drm/amdgpu: Use kmemdup rather than duplicating its implementationFuqian Huang4-13/+7
kmemdup is introduced to duplicate a region of memory in a neat way. Rather than kmalloc/kzalloc + memcpy, which the programmer needs to write the size twice (sometimes lead to mistakes), kmemdup improves readability, leads to smaller code and also reduce the chances of mistakes. Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy. Reviewed-by: Christian König <Christian.Koenig@amd.com> Reviewed-by: Emil Velikov <emil.velikov@collabora.com> Signed-off-by: Fuqian Huang <huangfq.daxian@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-03drm/amdkfd: Disable idle optimization for chained runlistFelix Kuehling1-0/+1
This works around difficult-to-reproduce soft hangs on oversubscribed runlists. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-03drm/amdkfd: Add chained_runlist_idle_disable flag to pm4_mes_runlistFelix Kuehling1-1/+1
New flag to disable an idle runlist optimization that is causing soft hangs with some diffult-to-reproduce customer workloads. This will serve as a workaround until the problem can be reproduced and the root-cause determined. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-03drm/amdgpu: Fix tracking of invalid userptrsFelix Kuehling1-0/+6
Restore the code that resets mem->invalid. Othewise so mapping userptrs after they got an MMU notifiers would always be skipped. This also avoids unnecessarily calling get_user_pages on BOs that have not been invalidated since the last try. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-03drm/amdgpu: Use FENCE_OWNER_KFD in process_sync_pds_resvFelix Kuehling1-1/+1
We don't want eviction fences to trigger when waiting for page table updates to complete during restore. In theory there shouldn't be any unsignaled eviction fences in the PD reservation object, but I'm seeing them in instrumented code for reasons not fully understood. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-03drm/amdkfd: Print a warning when the runlist becomes oversubscribedFelix Kuehling2-1/+6
Oversubscription of queues or processes results in poor performance mostly because HWS blinbly schedules busy and idle queues, resulting in poor occupancy if many queues are idle. Let users know with a warning message when transitioning from a non-oversubscribed to an oversubscribed runlist. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-02drm/amdkfd: remove an unused variableJack Xiao1-1/+0
Just for cleanup. Reviewed-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-02drm/amdgpu/display: fix interrupt client id for naviAlex Deucher1-4/+1
All asics newer than vega10 use client ids, so simplify the check. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdkfd: remove duplicated PCIE atomics requestJack Xiao3-6/+12
Since amdgpu has always requested PCIE atomics, kfd don't need duplicated PCIE atomics enablement. Referring to amdgpu request result is enough. Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu: enable PCIE atomics ops supportJack Xiao1-0/+11
GPU atomics operation depends on PCIE atomics support. Always enable PCIE atomics ops support in case that it hasn't been enabled. Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu: add field indicating if has PCIE atomics supportJack Xiao1-0/+1
The new field in amdgpu device is used to record whether the system has PCIE atomics support. The field can be exposed to UMD or kfd whether PCIE atomics have supported. Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amd/powerplay: use hardware fan control if no powerplay fan tableEvan Quan3-1/+8
Otherwise, you may get divided-by-zero error or corrput the SMU fan control feature. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Tested-by: Slava Abramov <slava.abramov@amd.com> Acked-by: Slava Abramov <slava.abramov@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu: fix MGPU fan boost enablement for XGMI resetEvan Quan3-2/+19
MGPU fan boost feature should not be enabled until all the devices from the same hive are all back from reset. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu: handle AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID on gfx10Marek Olšák1-2/+19
Add the gfx10 equivalent of the gfx9 code. Signed-off-by: Marek Olšák <marek.olsak@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu: fix transform feedback GDS hang on gfx10 (v2)Marek Olšák2-4/+13
v2: update emit_ib_size (though it's still wrong because it was wrong before) Signed-off-by: Marek Olšák <marek.olsak@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu/gfx10: use reset default for PA_SC_FIFO_SIZEAlex Deucher1-18/+0
Recommended by the hw team. Reviewed-and-Tested-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-07-01drm/amdgpu/gfx9: use reset default for PA_SC_FIFO_SIZEAlex Deucher1-19/+0
Recommended by the hw team. Reviewed-and-Tested-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-28drm/amdgpu: Don't skip display settings in hwmgr_resume()Lyude Paul1-1/+1
I'm not entirely sure why this is, but for some reason: 921935dc6404 ("drm/amd/powerplay: enforce display related settings only on needed") Breaks runtime PM resume on the Radeon PRO WX 3100 (Lexa) in one the pre-production laptops I have. The issue manifests as the following messages in dmesg: [drm] UVD and UVD ENC initialized successfully. amdgpu 0000:3b:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring vce1 test failed (-110) [drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block <vce_v3_0> failed -110 [drm:amdgpu_device_resume [amdgpu]] *ERROR* amdgpu_device_ip_resume failed (-110). And happens after about 6-10 runtime PM suspend/resume cycles (sometimes sooner, if you're lucky!). Unfortunately I can't seem to pin down precisely which part in psm_adjust_power_state_dynamic that is causing the issue, but not skipping the display setting setup seems to fix it. Hopefully if there is a better fix for this, this patch will spark discussion around it. Fixes: 921935dc6404 ("drm/amd/powerplay: enforce display related settings only on needed") Cc: Evan Quan <evan.quan@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Huang Rui <ray.huang@amd.com> Cc: Rex Zhu <Rex.Zhu@amd.com> Cc: Likun Gao <Likun.Gao@amd.com> Cc: <stable@vger.kernel.org> # v5.1+ Signed-off-by: Lyude Paul <lyude@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-28drm/amd/powerplay: update smu11_driver_if_navi10.htiancyin2-4/+4
update the smu11_driver_if_navi10.h since navi10 smu fw update to 42.28 Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: tiancyin <tianci.yin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amd/display: fix a couple of spelling mistakesColin Ian King2-5/+5
There are a couple of spelling mistakes in dm_error messages and a comment. Fix these. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amd/powerplay: no memory activity support on Vega10Evan Quan1-2/+4
Make mem_busy_percent sysfs interface invisible on Vega10. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amdkfd: remove unnecessary warning message on gpu resetshaoyunl1-1/+0
In XGMI configuration, more than one asic can be reset at same time, kfd is able to handle this and no need to trigger the warning Signed-off-by: shaoyunl <shaoyun.liu@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amdgpu: Set queue_preemption_timeout_ms default valueOak Zeng1-1/+1
Set default value of this kernel parameter to 9000 Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amd/powerplay: add missing smu_get_clk_info_from_vbios() callEvan Quan1-0/+4
This seems a merge error. 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>
2019-06-27drm/amd/powerplay: support runtime ppfeatures setting on Navi10Evan Quan1-0/+165
Implement Navi10 backend for runtime ppfeatures status retrieving and setting support. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amd/powerplay: check prerequisite for VCN power gatingEvan Quan1-6/+10
VCN DPM is a necessary prerequisite for VCN power gating. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Kevin Wang <kevin1.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/ttm: return -EBUSY if waiting for busy BO failsFelix Kuehling1-1/+1
Returning -EAGAIN prevents ttm_bo_mem_space from trying alternate placements and can lead to live-locks in amdgpu_cs, retrying indefinitely and never succeeding. Fixes: d367bd2a5e2b12 ("drm/ttm: fix busy memory to fail other user v10") 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>
2019-06-27drm/amdgpu: drop copy/paste leftover to fix big endianAlex Deucher1-3/+0
The buf swap field doesn't exist on RB1. Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27drm/amdgpu: fix warning on 32 bitAlex Deucher1-2/+2
Properly cast pointer to int. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-27Merge tag 'drm-next-5.3-2019-06-25' of ↵Dave Airlie358-2943/+414197
git://people.freedesktop.org/~agd5f/linux into drm-next drm-next-5.3-2019-06-25: Merge drm-next amdgpu: - SR-IOV L1 policy fixes - Removed no longer needed vram_page_split module parameter - Add module parameter to override default ABM level - Gamma fixes - No need to check return values for debugfs - Improve HMM error handling - Avoid possible OOM situations when lots of thread are submitting with memory contention - Improve hw i2c access abritration - DSC (Display Stream Compression) support in DC - Initial navi10 support * DC support * GFX/Compute support * SDMA support * Power Management support * VCN support - Static checker fixes - Misc cleanups - fix long udelay on arm amdkfd: - Implement priority controls for gfx9 - Enable VEGAM - Rework mqd allocation and init - Circular locking fix - Fix SDMA queue allocation race condition - No need to check return values for debugfs - Add proc style process information - Initial navi10 support radeon: - No need to check return values for debugfs UAPI changes: - GDDR6 added to vram type query - New Navi10 details added gpu info query - Navi family added to asic family query Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190625195520.3817-1-alexander.deucher@amd.com
2019-06-27Merge tag 'for-airlie-tda998x' of git://git.armlinux.org.uk/~rmk/linux-arm ↵Dave Airlie1-148/+302
into drm-next TDA998x updates: - improve the driver's approach to audio, adding support for more I2S based formats, particularly other justifications, and preparing the driver to support other bclk ratios. - add support for pixel repeated modes, tested with a Panasonic TV. - correct the quantisation range handling; in particular, do not send full range RGB to the sink when the sink does not support full range RGB. - Send the HDMI vendor info frame when required. Signed-off-by: Dave Airlie <airlied@redhat.com> From: Russell King <rmk@armlinux.org.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20190625125005.GA31503@rmk-PC.armlinux.org.uk
2019-06-25drm/amd/powerplay: make athub pg bit configured by pg_flagsHuang Rui3-3/+7
The athub pg features enabling should be indicated by pg_flags. Reported-by: Lijo Lazar <Lijo.Lazar@amd.com> Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay: make mmhub pg bit configured by pg_flagsHuang Rui2-2/+5
The mmhub pg features enabling should be indicated by pg_flags. Reported-by: Lijo Lazar <Lijo.Lazar@amd.com> Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay: add feature check in unforce_dpm_levels function (v2)Kevin Wang1-6/+13
if not check dpm feature is enabled, it will cause show smc send message failed log in dmesg log. eg: echo "auto" > power_dpm_force_performance_level v2: whitespace fix (Alex) Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Rui Teng <rui.teng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay: check gfxclk dpm enablement before proceedingEvan Quan1-0/+1
Support gfxclk dpm disablement. Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Feifei Xu <Feifei.Xu@amd.com> Reviewed-by: Kevin Wang <kevin1.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay: print smu versions only if version mismatchKevin Wang1-4/+4
only printf smu version when if version not matched. Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay: fix fan speed show error (for hwmon pwm)Kevin Wang1-2/+20
the navi asic can't get current rpm by MSG_GetCurrentRpm, it will cause show fan rpm error, use Metrics_t to replace it. Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/display: update infoframe after dig fe is turned on (v2)Nicholas Kazlauskas1-0/+1
[Why] The AVI infoframe is incorrectly programmed on DCN1/2 when enabling a stream - causing the wrong pixel encoding to be used for display. This is because the AVI infoframe is programmed before the DIG BE is connected to the FE and turned on, so enabling the AFMT block doesn't actually work and the registers subsequently can't be written to. [How] Program the infoframe *after* turning on the DIG FE. This was the behavior previously used but it was incorrectly reverted when adding the DCN2 HW sequencer code. v2: Don't call update_info_frame twice Fixes: 7ed4e6352c16fe01 ("drm/amd/display: Add DCN2 HW Sequencer and Resource") Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/powerplay:clean up the residual mutex for smu_hw_initPrike Liang1-1/+0
The mutex for procting SMU during hw_init was removed as system will be deadlock when smu_populate_umd_state_clk try get SMU mutex. Therefore need remove the residual mutex from failed path. Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: sdma_v4_0_start: initialize rErnst Sjöstrand1-1/+1
Reported by smatch: drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c:1167 sdma_v4_0_start() error: uninitialized symbol 'r'. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: Fix style issues in dcn20_resource.cErnst Sjöstrand1-5/+5
Inconsistent indentation and mixed use of brackets. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: Check stream in amdgpu_dm_commit_planesErnst Sjöstrand1-1/+2
Reported by smatch: amdgpu_dm.c:5637 amdgpu_dm_commit_planes() error: we previously assumed 'acrtc_state->stream' could be null This seems to be checked for null pretty consistently elsewhere. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: amdgpu_hwmon_show_temp: initialize tempErnst Sjöstrand1-1/+1
Reported by smatch: drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c:1496 amdgpu_hwmon_show_temp() error: uninitialized symbol 'temp'. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: Fix amdgpu_set_pp_od_clk_voltage error checkErnst Sjöstrand1-4/+4
Reported by smatch: drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c:693 amdgpu_set_pp_od_clk_voltage() error: uninitialized symbol 'ret'. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-06-25drm/amd/amdgpu: Indent AMD_IS_APU properlyErnst Sjöstrand2-4/+4
Reported by smatch: drivers/gpu/drm/amd/amdgpu/soc15.c:715 soc15_get_pcie_usage() warn: inconsistent indenting And a similar one in si.c. Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>