summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)Author
2020-01-07drm/amdgpu: amalgamated PSP TA invoke functionsJohn Clements
reduce duplicate code Reviewed-by: Guchun Chen <guchun.chen@amd.com> Signed-off-by: John Clements <john.clements@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: amalgamate PSP TA load/unload functionsJohn Clements
reduce duplicate code Reviewed-by: Guchun Chen <guchun.chen@amd.com> Signed-off-by: John Clements <john.clements@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: by default output PSP ret status in event of cmd failureJohn Clements
update log level from DRM_DEBUG_DRIVER to DRM_WARN Reviewed-by: Guchun Chen <guchun.chen@amd.com> Signed-off-by: John Clements <john.clements@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: correct RLC firmwares loading sequenceEvan Quan
Per confirmation with RLC firmware team, the RLC should be unhalted after all RLC related firmwares uploaded. However, in fact the RLC is unhalted immediately after RLCG firmware uploaded. And that may causes unexpected PSP hang on loading the succeeding RLC save restore list related firmwares. So, we correct the firmware loading sequence to load RLC save restore list related firmwares before RLCG ucode. That will help to get around this issue. 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>
2020-01-07drm/amd/powerplay: add check for baco support on ArcturusEvan Quan
This is used to determine whether runtime pm can be supported or not. 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>
2020-01-07Revert "drm/amdgpu: simplify ATPX detection"Alex Deucher
This reverts commit f5fda6d89afe6e9cedaa1c3303903c905262f6e8. You can't use BASE_CLASS in pci_get_class. Bug: https://gitlab.freedesktop.org/drm/amd/issues/995 Acked-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: simplify function return logicGuchun Chen
Former return logic is redundant. Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: add DRIVER_SYNCOBJ_TIMELINE to amdgpuChunming Zhou
Can expose it now that the khronos has exposed the vlk extension. Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdgpu: add DRIVER_SYNCOBJ_TIMELINE to amdgpuChunming Zhou
Can expose it now that the khronos has exposed the vlk extension. Signed-off-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2020-01-07Revert "drm/amdgpu: Set no-retry as default."Alex Deucher
This reverts commit 51bfac71cade386966791a8db87a5912781d249f. This causes stability issues on some raven boards. Revert for now until a proper fix is completed. Bug: https://gitlab.freedesktop.org/drm/amd/issues/934 Bug: https://bugzilla.kernel.org/show_bug.cgi?id=206017 Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2020-01-01drm/amdgpu: correct RLC firmwares loading sequenceEvan Quan
Per confirmation with RLC firmware team, the RLC should be unhalted after all RLC related firmwares uploaded. However, in fact the RLC is unhalted immediately after RLCG firmware uploaded. And that may causes unexpected PSP hang on loading the succeeding RLC save restore list related firmwares. So, we correct the firmware loading sequence to load RLC save restore list related firmwares before RLCG ucode. That will help to get around this issue. 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> Cc: stable@vger.kernel.org
2019-12-29drm/amdgpu: enable gfxoff for raven1 refreshchangzhu
When smu version is larger than 0x41e2b, it will load raven_kicker_rlc.bin.To enable gfxoff for raven_kicker_rlc.bin,it needs to avoid adev->pm.pp_feature &= ~PP_GFXOFF_MASK when it loads raven_kicker_rlc.bin. Signed-off-by: changzhu <Changfeng.Zhu@amd.com> Reviewed-by: Huang Rui <ray.huang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-29Revert "drm/amdgpu: simplify ATPX detection"Alex Deucher
This reverts commit f5fda6d89afe6e9cedaa1c3303903c905262f6e8. You can't use BASE_CLASS in pci_get_class. Bug: https://gitlab.freedesktop.org/drm/amd/issues/995 Acked-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: use true, false for bool variable in amdgpu_psp.czhengbin
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:674:2-26: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:794:1-25: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:897:2-36: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:1016:1-35: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:1087:2-34: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c:1177:1-33: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: use true, false for bool variable in amdgpu_debugfs.czhengbin
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c:132:2-10: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c:140:2-10: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c:142:13-21: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: use true, false for bool variable in amdgpu_device.czhengbin
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3961:1-19: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3981:1-19: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: use true, false for bool variable in mxgpu_nv.czhengbin
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c:255:2-20: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c:267:2-20: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: use true, false for bool variable in mxgpu_ai.czhengbin
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c:253:2-20: WARNING: Assignment of 0/1 to bool variable drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c:265:2-20: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: add missed return value set for error caseGuchun Chen
Return value should be set when going to error handle tag for error case, this can avoid potential invalid array access by upper caller. Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: remove FB location config for sriovFrank.Min
FB location is already programmed by HV driver for arcutus so remove this part Signed-off-by: Frank.Min <Frank.Min@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: enable xgmi init for sriov use caseFrank.Min
1. enable xgmi ta initialization for sriov 2. enable xgmi initialization for sriov Signed-off-by: Frank.Min <Frank.Min@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: remove memory training p2c buffer reservation(V2)Tianci.Yin
IP discovery TMR(occupied the top VRAM with size DISCOVERY_TMR_SIZE) has been reserved, and the p2c buffer is in the range of this TMR, so the p2c buffer reservation is unnecessary. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com> Signed-off-by: Tianci.Yin <tianci.yin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: update the method to get fb_loc of memory training(V4)Tianci.Yin
The method of getting fb_loc changed from parsing VBIOS to taking certain offset from top of VRAM Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Tianci.Yin <tianci.yin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: Remove unneeded variable 'ret' in navi10_ih.cMa Feng
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/navi10_ih.c:113:5-8: Unneeded variable: "ret". Return "0" on line 182 Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Ma Feng <mafeng.ma@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: Remove unneeded variable 'ret' in amdgpu_device.cMa Feng
Fixes coccicheck warning: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:1036:5-8: Unneeded variable: "ret". Return "0" on line 1079 Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Ma Feng <mafeng.ma@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu/gfx: Add mmSDMA2-7_EDC_COUNTER to support ArcturusJames Zhu
Add mmSDMA2-7_EDC_COUNTER to support Arcturus Signed-off-by: James Zhu <James.Zhu@amd.com> Reviewed-by: Yong Zhao <Yong.Zhao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu/gfx: Add mmCOMPUTE_STATIC_THREAD_MGMT_SE4-7 to support ArcturusJames Zhu
Add mmCOMPUTE_STATIC_THREAD_MGMT_SE4-7 to support Arcturus Signed-off-by: James Zhu <James.Zhu@amd.com> Reviewed-by: Yong Zhao <Yong.Zhao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu/gfx: Replace ARRAY_SIZE with size variableJames Zhu
Replace ARRAY_SIZE with size variables to support different ASICs. Signed-off-by: James Zhu <James.Zhu@amd.com> Reviewed-by: Yong Zhao <Yong.Zhao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: Added ASIC specific check in gmc v9.0 ECC interrupt programming ↵John Clements
sequence Devices newer then VEGA10/12 shall have these programming sequences performed by PSP BL Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: John Clements <john.clements@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: enlarge agp_start address into 48bitFrank.Min
max range of the agp aperture is 48 bits, so enlarge agp_start address into 48bit with all bits set Signed-off-by: Frank.Min <Frank.Min@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: disable VCN2.5 ib test for Arcturus sriovJane Jian
currently using TMR loading VCN fw MMSCH would fail to init after FLR, just disable ib test for temporarily daily testing, continuing debug with mm team. Signed-off-by: Jane Jian <Jane.Jian@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: fix ctx init failure for asics without gfx ringLe Ma
This workaround does not affect other asics because amdgpu only need expose one gfx sched to user for now. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Nirmoy Das <nirmoy.das@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: attempt xgmi perfmon re-arm on failed armJonathan Kim
The DF routines to arm xGMI performance will attempt to re-arm both on performance monitoring start and read on initial failure to arm. Signed-off-by: Jonathan Kim <Jonathan.Kim@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-23drm/amdgpu: simplify padding calculations (v2)Luben Tuikov
Simplify padding calculations. v2: Comment update and spacing. Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-19drm/amdgpu: enable VCN0 and VCN1 sriov instances support for ArcturusJane Jian
v1: compared to bare-metal: sriov support psp loading VCN firmware; only one encoding ring would be used in each instance. v2: keep unchange for bare-metal VCN2.5 hw_init, just add a flag with sriov and also remove multiple lines. v3: squash in warning fix Signed-off-by: Jane Jian <Jane.Jian@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: skip VCN2.5 power gating and clock gating for sriov ArcturusJane Jian
v1: skip gating in serveral called functions by power gating and clock gating v2: from suggestion, skip setting gate in both set function, which is where it being called. Signed-off-by: Jane Jian <Jane.Jian@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: update VCN1(dual instances) fw types ID and VCN ip block typeJane Jian
Previously there is no VCN1 type ID in psp gfx interface. Also add VCN ip block type unless the reinit after FLR for sriov would fail. Signed-off-by: Jane Jian <Jane.Jian@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: add VCN2.5 sriov start for ArctrusJane Jian
Use MMSCH V1 to finish Memory Controller programming as well as start MMSCH to do VCN2.5 initialization. Signed-off-by: Jane Jian <Jane.Jian@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: add VCN2.5 MMSCH start for ArcturusJane Jian
Use MMSCH to do the initialization since MMSCH manages VCN2.5 instances and its world switch. Signed-off-by: Jane Jian <Jane.Jian@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: move umc offset to one new header file for ArcturusGuchun Chen
Code refactor and no functional change. Fixes: 4cf781c24c3b ("drm/amdgpu: Added RAS UMC error query support for Arcturus") Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: wait for all rings to drain before runtime suspendingAlex Deucher
Add a safety check to runtime suspend to make sure all outstanding fences have signaled before we suspend. Doesn't fix any known issue. We already do this via the fence driver suspend function, but we just force completion rather than bailing. This bails on runtime suspend so we can try again later once the fences are signaled to avoid missing any outstanding work. Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: Switch from system_highpri_wq to system_unbound_wqAndrey Grodzovsky
This is to avoid queueing jobs to same CPU during XGMI hive reset because there is a strict timeline for when the reset commands must reach all the GPUs in the hive. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: Redo XGMI reset synchronization.Andrey Grodzovsky
Use task barrier in XGMI hive to synchronize ASIC resets across devices in XGMI hive. v2: Return right away with a warning if no xgmi hive, update doc. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: Add task barrier to XGMI hive.Andrey Grodzovsky
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: reverts commit ce316fa55ef0f1751276b846a54fb3b835bd5e64.Andrey Grodzovsky
In preparation for doing XGMI reset synchronization using task barrier. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu/vcn: remove unnecessary included headersLeo Liu
Esp. VCN1.0 headers should not be here v2: add back the <linux/module.h> to keep consistent. Signed-off-by: Leo Liu <leo.liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: fix KIQ ring test fail in TDR of SRIOVMonk Liu
issues: MEC is ruined by the amdkfd_pre_reset after VF FLR done fix: amdkfd_pre_reset() would ruin MEC after hypervisor finished the VF FLR, the correct sequence is do amdkfd_pre_reset before VF FLR but there is a limitation to block this sequence: if we do pre_reset() before VF FLR, it would go KIQ way to do register access and stuck there, because KIQ probably won't work by that time (e.g. you already made GFX hang) so the best way right now is to simply remove it. 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>
2019-12-18drm/amdgpu: fix double gpu_recovery for NV of SRIOVMonk Liu
issues: gpu_recover() is re-entered by the mailbox interrupt handler mxgpu_nv.c fix: we need to bypass the gpu_recover() invoke in mailbox interrupt as long as the timeout is not infinite (thus the TDR will be triggered automatically after time out, no need to invoke gpu_recover() through mailbox interrupt. 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>
2019-12-18amd/amdgpu: add sched array to IPs with multiple run-queuesNirmoy Das
This sched array can be passed on to entity creation routine instead of manually creating such sched array on every context creation. v2: squash in missing break fix Signed-off-by: Nirmoy Das <nirmoy.das@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdgpu: replace vm_pte's run-queue list with drm gpu scheds listNirmoy Das
drm_sched_entity_init() takes drm gpu scheduler list instead of drm_sched_rq list. This makes conversion of drm_sched_rq list to drm gpu scheduler list unnecessary Signed-off-by: Nirmoy Das <nirmoy.das@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.3 (git 2.53.0) at 2026-11-13 02:34:41 +0000