aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)AuthorFilesLines
2019-08-29drm/amdgpu/psp: keep TMR in visible vram region for SRIOVTianci.Yin1-2/+8
Fix compute ring test failure in sriov scenario. Reviewed-by: Christian König <[email protected]> Signed-off-by: Tianci.Yin <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: keep the stolen memory in visible vram regionTianci.Yin2-3/+6
stolen memory should be fixed in visible region. Reviewed-by: Christian König <[email protected]> Signed-off-by: Tianci.Yin <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: fix spelling mistake "jumpimng" -> "jumping"Colin Ian King1-1/+1
There is a spelling mistake in a DRM_DEBUG_DRIVER debug message. Fix it. Signed-off-by: Colin Ian King <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu/virtual_dce: drop error message in hw_initAlex Deucher1-10/+1
No need to add new asic cases. This is a sw display implementation, so just drop the error message so when we add new asics, all we have to do is add the virtual dce IP module. Reviewed-by: Xiaojie Yuan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu/si: fix ASIC testsJean Delvare1-3/+3
Comparing adev->family with CHIP constants is not correct. adev->family can only be compared with AMDGPU_FAMILY constants and adev->asic_type is the struct member to compare with CHIP constants. They are separate identification spaces. Signed-off-by: Jean Delvare <[email protected]> Fixes: 62a37553414a ("drm/amdgpu: add si implementation v10") Cc: Ken Wang <[email protected]> Cc: Alex Deucher <[email protected]> Cc: "Christian König" <[email protected]> Cc: "David (ChunMing) Zhou" <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amd/amdgpu: hide voltage and power sensors on SI and KV partsJean Delvare1-1/+9
The driver does not support these sensors yet and there is no point in creating sysfs attributes which will always return an error. Signed-off-by: Jean Delvare <[email protected]> Cc: Alex Deucher <[email protected]> Cc: "Christian König" <[email protected]> Cc: "David (ChunMing) Zhou" <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: introduce vram lost for reset (v2)Monk Liu4-4/+10
for SOC15/vega10 the BACO reset & mode1 would introduce vram lost in high end address range, current kmd's vram lost checking cannot catch it since it only check very ahead visible frame buffer v2: cover NV as well Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: enable athub powergating for navi12Xiaojie Yuan1-1/+2
Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: enable vcn powergating for navi12Xiaojie Yuan1-1/+2
Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-29drm/amdgpu: correct in_suspend setting for navi seriesHawking Zhang1-10/+2
in_suspend flag should be set in amdgpu_device_suspend/resume in pairs, instead of gfx10 ip suspend/resume function. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27drm/amdgpu: fix GFXOFF on Picasso and Raven2Aaron Liu1-7/+7
For picasso(adev->pdev->device == 0x15d8)&raven2(adev->rev_id >= 0x8), firmware is sufficient to support gfxoff. In commit 98f58ada2d37e, for picasso&raven2, return directly and cause gfxoff disabled. Fixes: 98f58ada2d37 ("drm/amdgpu/gfx9: update pg_flags after determining if gfx off is possible") Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27drm/amdgpu: Add APTX quirk for Dell Latitude 5495Kai-Heng Feng1-0/+1
Needs ATPX rather than _PR3 to really turn off the dGPU. This can save ~5W when dGPU is runtime-suspended. Signed-off-by: Kai-Heng Feng <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27drm/amdgpu: Vega20 SMU I2C HW engine controller.Andrey Grodzovsky4-3/+762
Implement HW I2C enigne controller to be used by the RAS EEPROM table manager. This is based on code from ATITOOLs. v2: Rename the file and all function prefixes to smu_v11_0_i2c By Luben's observation always fill the TX fifo to full so we don't have garbadge interpreted by the slave as valid data. v3: Remove preemption disable as the HW I2C controller will not stop the clock on empty TX fifo and so it's not critical to keep not empty queue. Switch to fast mode 400 khz SCL clock for faster read and write. v5: Restore clock gating before releasing I2C bus and fix some style comments. v6: squash in warning fix, fix includes (Alex) Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Luben Tuikov <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27drm/amdgpu: Add RAS EEPROM table.Andrey Grodzovsky4-1/+586
Add RAS EEPROM table manager to eanble RAS errors to be stored upon appearance and retrived on driver load. v2: Fix some prints. v3: Fix checksum calculation. Make table record and header structs packed to do correct byte value sum. Fix record crossing EEPROM page boundry. v4: Fix byte sum val calculation for record - look at sizeof(record). Fix some style comments. v5: Add description to EEPROM_TABLE_RECORD_SIZE and syntax fixes. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Luben Tuikov <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27Revert "drm/amdgpu: free up the first paging queue v2"Gang Ba1-15/+8
This reverts commit 4f8bc72fbf10f2dc8bca74d5da08b3a981b2e5cd. It turned out that a single reserved queue wouldn't be sufficient for page fault handling. Signed-off-by: Gang Ba <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27drm/amdgpu: add dummy read for some GCVM status registersXiaojie Yuan2-0/+31
The GRBM register interface is now capable of bursting 1 cycle per register wr->wr, wr->rd much faster than previous muticycle per transaction done interface. This has caused a problem where status registers requiring HW to update have a 1 cycle delay, due to the register update having to go through GRBM. SW may operate on an incorrect value if they write a register and immediately check the corresponding status register. Registers requiring HW to clear or set fields may be delayed by 1 cycle. For example, 1. write VM_INVALIDATE_ENG0_REQ mask = 5a 2. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask = 5a a. HW will reset VM_INVALIDATE_ENG0_ACK = 0 until invalidation is complete 3. write VM_INVALIDATE_ENG0_REQ mask = 5a 4. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask = 5a a. First read of VM_INVALIDATE_ENG0_ACK = 5a instead of 0 b. Second read of VM_INVALIDATE_ENG0_ACK = 0 because the remote GRBM h/w register takes one extra cycle to be cleared c. In this case, SW will see a false ACK if they exit on first read Affected registers (only GC variant) | Recommended Dummy Read --------------------------------------+---------------------------- VM_INVALIDATE_ENG*_ACK | VM_INVALIDATE_ENG*_REQ VM_L2_STATUS | VM_L2_STATUS VM_L2_PROTECTION_FAULT_STATUS | VM_L2_PROTECTION_FAULT_STATUS VM_L2_PROTECTION_FAULT_ADDR_HI/LO32 | VM_L2_PROTECTION_FAULT_ADDR_HI/LO32 VM_L2_IH_LOG_BUSY | VM_L2_IH_LOG_BUSY MC_VM_L2_PERFCOUNTER_HI/LO | MC_VM_L2_PERFCOUNTER_HI/LO ATC_L2_PERFCOUNTER_HI/LO | ATC_L2_PERFCOUNTER_HI/LO ATC_L2_PERFCOUNTER2_HI/LO | ATC_L2_PERFCOUNTER2_HI/LO Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Jack Xiao <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-27Merge tag 'drm-next-5.4-2019-08-23' of ↵Dave Airlie57-477/+1703
git://people.freedesktop.org/~agd5f/linux into drm-next drm-next-5.4-2019-08-23: amdgpu: - Enable power features on Navi12 - Enable power features on Arcturus - RAS updates - Initial Renoir APU support - Enable power featyres on Renoir - DC gamma fixes - DCN2 fixes - GPU reset support for Picasso - Misc cleanups and fixes scheduler: - Possible race fix Signed-off-by: Dave Airlie <[email protected]> From: Alex Deucher <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2019-08-23drm/amdgpu: set adev->num_vmhubs for gmc6,7,8Alex Deucher3-0/+6
So that we properly handle them on older asics. Fixes: 3ff985485b29 ("drm/amdgpu: Export function to flush TLB of specific vm hub") Tested-by: Tom St Denis <[email protected]> Reviewed-by: Xiaojie Yuan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-23drm/amdgpu: correct ras error count typeGuchun Chen4-7/+7
Use unsigned long type for the same ras count variable. This will avoid overflow on 64 bit system. Signed-off-by: Guchun Chen <[email protected]> Reviewed-by: Tao Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-23drm: drop resource_id parameter from ↵Gerd Hoffmann1-1/+1
drm_fb_helper_remove_conflicting_pci_framebuffers Not needed any more for remove_conflicting_pci_framebuffers calls. Signed-off-by: Gerd Hoffmann <[email protected]> Reviewed-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2019-08-22drm/amdgpu: enable VCN DPG for RenoirThong Thai1-1/+3
This will enable indirect SRAM loading for VCN DPG mode initialization. Signed-off-by: Thong Thai <[email protected]> Reviewed-by: Leo Liu <[email protected]> Reviewed-by: Prike Liang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22Revert "drm/amdgpu: use direct loading on renoir vcn for the moment"Thong Thai2-12/+6
This reverts commit 444a0fea5107e9ad7e3cbbafed78678489e31713. We are ready to enable it now. Signed-off-by: Thong Thai <[email protected]> Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: update gc/sdma goldensetting for rnAaron Liu2-6/+5
This patch updates gc/sdma goldensetting for renoir Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable SDMA power gating for rnPrike Liang1-1/+1
Enable SDMA PG flag during device ip early init. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Prike Liang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu/sdma4: set sdma clock gating for rnPrike Liang1-0/+1
Add support for SDMA clockgating on RN. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu/mmhub1: set mmhub clock gating for rnPrike Liang1-0/+1
setup mmhub clockgating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable DF clock gating for rnPrike Liang1-1/+2
Enable DF clock gating during DF IP early init. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable athub clock gating for rnPrike Liang1-1/+3
Enable athub MG and LS clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable IH clock gating for rnPrike Liang1-1/+2
Enable IH clock gating during IH block initialized. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable vcn clock gating for rnPrike Liang1-1/+2
Enable VCN middle grain clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable rom clock gating for rnPrike Liang1-1/+2
Enable rom light sleep clock gating. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Prike Liang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable HDP clock gating for rnPrike Liang1-1/+2
Enable HDP light sleep clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable BIF clock gating for rnPrike Liang1-1/+2
Enable BIF light sleep clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable sdma clock gating for rnPrike Liang1-1/+3
Enable sdma middle grain and light sleep clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable mmhub clock gating for rnPrike Liang1-1/+3
Enable mmhub midle grain and light sleep clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: enable gfx clock gating for rnPrike Liang1-1/+7
Enable gfx cg/mg/cp etc clock gating. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: add and enable gfxoff featureAaron Liu1-0/+5
This patch updates gfxoff feature. Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Evan Quan <[email protected]> Reviewed-by: Kevin Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: add set_gfx_cgpg implement (v2)Aaron Liu1-0/+5
add set_gfx_cgpg implement v2: check if using sw_smu (Alex) Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu/powerplay: add power up/down SDMA interfaces for renoirAaron Liu1-2/+2
1.Implement PowerUpSDMA/PowerDownSDMA interfaces in the swSMU for renoir 2.adjust smu ip block ahead of gfx&sdma ip block Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu/powerplay: add smu ip block for renoir (v2)Aaron Liu1-0/+2
add swSMU [smu_v12_0] for renoir v2: whitespace fixes (Alex) Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu/sdma5: fix number of sdma5 trap irq types for navi1xXiaojie Yuan1-1/+2
v2: set num_types based on num_instances navi1x has 2 sdma engines but commit "e7b58d03b678 drm/amdgpu: reorganize sdma v4 code to support more instances" changes the max number of sdma irq types (AMDGPU_SDMA_IRQ_LAST) from 2 to 8 which causes amdgpu_irq_gpu_reset_resume_helper() to recover irq of sdma engines with following logic: (enable irq for sdma0) * 1 time (enable irq for sdma1) * 1 time (disable irq for sdma1) * 6 times as a result, after gpu reset, interrupt for sdma1 is lost. Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: fix dma_fence_wait without referenceChristian König1-12/+15
We need to grab a reference to the fence we wait for. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22amd/amdgpu: add Arcturus vf DID supportFrank.Min1-0/+1
Reviewed-by: Feifei Xu <[email protected]> Signed-off-by: Frank.Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: unity mc base address for arcturusFrank.Min1-6/+6
arcturus for sriov would use the unified mc base address Reviewed-by: Christian König <[email protected]> Signed-off-by: Frank.Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-22drm/amdgpu: disable agp for sriovFrank.Min2-2/+9
Since agp is not used for sriov, just disable it Reviewed-by: Christian König <[email protected]> Signed-off-by: Frank.Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: remove duplicated include from gfx_v9_0.cYueHaibing1-1/+0
Remove duplicated include. Signed-off-by: YueHaibing <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: remove set but not used variable 'psp_enabled'YueHaibing1-5/+0
Fixes gcc '-Wunused-but-set-variable' warning: drivers/gpu/drm/amd/amdgpu/nv.c: In function 'nv_common_early_init': drivers/gpu/drm/amd/amdgpu/nv.c:471:7: warning: variable 'psp_enabled' set but not used [-Wunused-but-set-variable] It's not used since inroduction in commit c6b6a42175f5 ("drm/amdgpu: add navi10 common ip block (v3)") Signed-off-by: YueHaibing <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: prevent memory leaks in AMDGPU_CS ioctlNicolai Hähnle1-2/+7
Error out if the AMDGPU_CS ioctl is called with multiple SYNCOBJ_OUT and/or TIMELINE_SIGNAL chunks, since otherwise the last chunk wins while the allocated array as well as the reference counts of sync objects are leaked. Signed-off-by: Nicolai Hähnle <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amd/amdgpu: disable MMHUB PG for navi10Kenneth Feng1-1/+0
Disable MMHUB PG for navi10 according to the production requirement. Signed-off-by: Kenneth Feng <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Reviewed-by: Kevin Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amd/powerplay: expose supported clock domains only through sysfsEvan Quan1-10/+16
Do not expose those unsupported clock domains through sysfs on Arcturus. Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Kevin Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.