aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)AuthorFilesLines
2022-09-01drm/amdgpu/vcn: Add vcn/vcn1 in white list to load its firmware under sriovJane Jian1-1/+3
Previously since vcn0/vcn1 are not enabled, loading firmware is skipped. Now add firmware loading back since vcn0/vcn1 has already been enabled on sriov Acked-by: Christian König <[email protected]> Signed-off-by: Jane Jian <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu/vcn: Disable CG/PG for SRIOVJane Jian1-0/+5
For sriov, CG and MG are controlled from hypervisor side, no need to manage them again in ip init Acked-by: Christian König <[email protected]> Signed-off-by: Jane Jian <[email protected]> Reviewed-by: Sonny Jiang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/admgpu: Skip CG/PG on SOC21 under SRIOV VFYifan Zha1-0/+4
[Why] There is no CG(Clock Gating)/PG(Power Gating) requirement on SRIOV VF. For multi VF, VF should not enable any CG/PG features. For one VF, PF will program CG/PG related registers. [How] Do not set any cg/pg flag bit at early init under sriov. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Use RLCG to program GRBM_GFX_CNTL during full access timeYifan Zha1-1/+1
[Why] KIQ register init requires GRBM_GFX_CNTL to select KIQ. [How] As RLCG accessing registers will save the data of GRBM_GFX_CNTL and restore it. Use RLCG indirect accessing register method to select grbm instead of mmio directly access. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Skip program SDMA0_SEM_WAIT_FAIL_TIMER_CNTL under SRIOV VFYifan Zha1-1/+2
[Why] As SDMA0_SEM_WAIT_FAIL_TIMER_CNTL is a PF-only register, L1 would block this register for VF access. [How] VF do not program it. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Skip the VRAM base offset on SRIOVYifan Zha1-1/+4
[Why] As VF cannot read MMMC_VM_FB_OFFSET with L1 Policy(read 0xffffffff). It leads to driver get the incorrect vram base offset. [How] Since SR-IOV is dGPU only, skip reading this register and set the fb_offest to 0. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: skip "Issue additional private vm invalidation to MMHUB" on SRIOVYifan Zha1-1/+2
[Why] vm_l2_bank_select_reserved_cid2 is a PF_only register that cannot be programmed by VF. This feature is only support HDP using GPUVM page tables to access FB memory which should be disabled on SRIOV. [How] Disable the feature on VF. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Skip the program of MMMC_VM_AGP_* in SRIOV on MMHUB v3_0_0Yifan Zha1-5/+5
[Why] VF should not program these registers, the value were defined in the host. [How] Skip writing them in SRIOV environment and program them on host side. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Use PSP program IH_RB_CNTL registers under SRIOVYifan Zha1-3/+23
[Why] With L1 Policy applied, IH_RB_CNTL/RING cannot be accessed by VF. [How] Use PSP program IH_RB_CNTL in VF. Acked-by: Christian König <[email protected]> Signed-off-by: Yifan Zha <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: Support PSP 13.0.10 on SR-IOVHorace Chen3-22/+55
Add support for PSP 13.0.10 for SR-IOV VF Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: sriov remove vcn_4_0 and jpeg_4_0Horace Chen1-2/+4
SRIOV needs to initialize mmsch instead of multimedia engines directly. So currently remove them for SR-IOV until the code and firmwares are ready. Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: refine virtualization psp fw skip checkHorace Chen3-14/+34
SR-IOV may need to load different firmwares for different ASIC inside VF. So create a new function in amdgpu_virt to check whether FW load needs to be skipped. Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: enable WPTR_POLL_ENABLE for sriov on sdma_v6_0Horace Chen1-1/+4
[Why] Under SR-IOV, if VF is switched out then its doorbell will be disabled, SDMA rely on WPTR_POLL to get doorbells which was sent during VF switched-out time. [How] For SR-IOV, set SDMA WPTR_POLL_ENABLE to 1. Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: add a compute pipe reset for RS64Horace Chen1-0/+15
[Why] Under SR-IOV, we are not sure whether pipe status is good or not when doing initialization. The compute engine maybe fail to bringup if pipe status is bad. [How] Do an RS64 pipe reset for MEC before we do initialization. Also apply to bare-metal. Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: add sriov nbio callback structureHorace Chen3-1/+49
[Why] under SR-IOV, the nbio doorbell range will be defined by PF. So VF nbio doorbell range registers will be blocked. It will cause violation if VF access those registers directly. [How] create an nbio_v4_3_sriov_funcs for sriov nbio_v4_3 initialization to skip the setting for the doorbell range registers. Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu: add CHIP_IP_DISCOVERY support for virtualizationHorace Chen1-0/+2
For further chips we will use CHIP_IP_DISCOVERY, so add this support for virtualization Acked-by: Christian König <[email protected]> Signed-off-by: Horace Chen <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1Yifan Zhang1-0/+3
there is only one SDMA engine in SDMA 6.0.1, the sdma_hqd_mask has to be zeroed for the 2nd engine, otherwise MES scheduler will consider 2nd engine exists and map/unmap SDMA queues to the non-existent engine. Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: Tim Huang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-09-01drm/amd/amdgpu: skip ucode loading if ucode_size == 0Chengming Gui1-1/+1
Restrict the ucode loading check to avoid frontdoor loading error. Signed-off-by: Chengming Gui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-31drm/amd/amdgpu: skip ucode loading if ucode_size == 0Chengming Gui1-1/+1
Restrict the ucode loading check to avoid frontdoor loading error. Signed-off-by: Chengming Gui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: only init tap_delay ucode when it's included in ucode binaryHawking Zhang1-25/+35
Not all the gfx10 variants need to integrate global tap_delay and per se tap_delay firmwares Only init tap_delay ucode when it does include in rlc ucode binary so driver doesn't send a null buffer to psp for firmware loading Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Jack Gui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: ensure no PCIe peer access for CPU XGMI iolinksAlex Sierra1-1/+2
[Why] Devices with CPU XGMI iolink do not support PCIe peer access. Signed-off-by: Alex Sierra <[email protected]> Acked-by: Alex Deucher <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: Fix use-after-free in amdgpu_cs_ioctlYuBiao Wang1-1/+4
[Why] In amdgpu_cs_ioctl, amdgpu_job_free could be performed ealier if there is -ERESTARTSYS error. In this case, job->hw_fence could be not initialized yet. Putting hw_fence during amdgpu_job_free could lead to a use-after-free warning. [How] Check if drm_sched_job_init is performed before job_free by checking s_fence. v2: Check hw_fence.ops instead since it could be NULL if fence is not initialized. Reverse the condition since !=NULL check is discouraged in kernel. Signed-off-by: YuBiao Wang <[email protected]> Reviewed-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: Update mes_v11_api_def.hGraham Sider1-0/+1
New GFX11 MES FW adds the trap_en bit. For now hardcode to 1 (traps enabled). Signed-off-by: Graham Sider <[email protected]> Acked-by: Felix Kuehling <[email protected]> Reviewed-by: Harish Kasiviswanathan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: disable FRU access on special SIENNA CICHLID cardGuchun Chen1-2/+7
Below driver load error will be printed, not friendly to end user. amdgpu: ATOM BIOS: 113-D603GLXE-077 [drm] FRU: Failed to get size field [drm:amdgpu_fru_get_product_info [amdgpu]] *ERROR* Failed to read FRU Manufacturer, ret:-5 Signed-off-by: Guchun Chen <[email protected]> Reviewed-by: Kent Russell <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: Remove the unneeded result variable 'r'ye xingchen1-4/+1
Return the value sdma_v4_0_start() directly instead of storing it in another redundant variable. Reported-by: Zeal Robot <[email protected]> Signed-off-by: ye xingchen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: add new ip block for MES 11.0.3Hawking Zhang1-0/+1
Add ip block support for mes v11_0_3. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: add new ip block for GFX 11.0Hawking Zhang1-0/+1
Add ip block support for gfx v11_0_3. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: Set GC family for GC 11.0.3Hawking Zhang1-0/+1
Set AMDGPU_FAMILY_GC_11_0_0. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: enable imu_rlc_ram programming for v11_0_3Hawking Zhang4-1/+181
All gc v11_0_3 registers in gcvml2 range have different register offset from the ones in gc v11_0_0. v11_0_3 imu_rlc_ram programming has to be separated from v11_0_0 implementation v2: fix checkpatch errors (Alex) Signed-off-by: Hawking Zhang <[email protected]> Signed-off-by: Yang Wang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: init gfx config for gfx v11_0_3Hawking Zhang1-0/+2
initialize some gfx config for gfx v11_0_3 Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: declare firmware for new MES 11.0.3Hawking Zhang1-0/+2
To support new mes ip block Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: declare firmware for new GC 11.0.3Hawking Zhang1-0/+4
To support new gfx ip block Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: add new ip block for GMC 11.0Hawking Zhang1-0/+1
Add ip block support for gmc v11_0_3. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: initialize gmc sw config for v11_0_3Hawking Zhang1-0/+1
initialize gmc sw config for v11_0_3 Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: add gfxhub_v3_0_3 supportYang Wang4-2/+550
add gfxhub_v3_0_3 support Signed-off-by: Yang Wang <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-30drm/amdgpu: only init tap_delay ucode when it's included in ucode binaryHawking Zhang1-25/+35
Not all the gfx10 variants need to integrate global tap_delay and per se tap_delay firmwares Only init tap_delay ucode when it does include in rlc ucode binary so driver doesn't send a null buffer to psp for firmware loading Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Jack Gui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: add new ip block for LSDMA 6.0Hawking Zhang1-0/+1
Add ip block support for lsdma v6_0_3. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: add new ip block for sdma 6.0Hawking Zhang1-0/+1
Add ip block support for sdma v6_0_3. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: declare firmware for new SDMA 6.0.3Hawking Zhang1-0/+1
To support new sdma ip block Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: enable smu block for smu 13.0.10John Clements1-0/+1
Force to enable smu block for SMU v13.0.10 Signed-off-by: John Clements <[email protected]> Signed-off-by: Likun Gao <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: add new ip block for PSP 13.0Frank Min1-0/+1
Add ip block support for psp v13_0_10. Signed-off-by: Frank Min <[email protected]> Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: added firmware module for psp 13.0.10John Clements1-0/+1
added missing firmware module Signed-off-by: John Clements <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: support psp v13_0_10 ip blockFrank Min2-0/+2
Add psp v13_0_10 ip block, initialize firmware and psp functions Signed-off-by: Frank Min <[email protected]> Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: add new ip block for SOC21Hawking Zhang1-0/+1
Add ip block support for soc21_common. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: Enable pg/cg flags on GC11_0_3 for VCNSonny Jiang1-2/+5
This enable VCN PG, CG, DPG and JPEG PG, CG Signed-off-by: Sonny Jiang <[email protected]> Reviewed-by: Leo Liu <[email protected]> Reviewed-by: John Clements <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: initialize common sw config for v11_0_3Hawking Zhang1-0/+5
init cp/pg_flags and extenal_rev_id Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Frank Min <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: drop gc 11_0_0 golden settingsHawking Zhang1-41/+0
driver doesn't need to program any gc 11_0_0 golden Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Likun Gao <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: ensure no PCIe peer access for CPU XGMI iolinksAlex Sierra1-1/+2
[Why] Devices with CPU XGMI iolink do not support PCIe peer access. Signed-off-by: Alex Sierra <[email protected]> Acked-by: Alex Deucher <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: Fix use-after-free in amdgpu_cs_ioctlYuBiao Wang1-1/+4
[Why] In amdgpu_cs_ioctl, amdgpu_job_free could be performed ealier if there is -ERESTARTSYS error. In this case, job->hw_fence could be not initialized yet. Putting hw_fence during amdgpu_job_free could lead to a use-after-free warning. [How] Check if drm_sched_job_init is performed before job_free by checking s_fence. v2: Check hw_fence.ops instead since it could be NULL if fence is not initialized. Reverse the condition since !=NULL check is discouraged in kernel. Signed-off-by: YuBiao Wang <[email protected]> Reviewed-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2022-08-29drm/amdgpu: add missing pci_disable_device() in amdgpu_pmops_runtime_resume()Yang Yingliang1-1/+4
Add missing pci_disable_device() if amdgpu_device_resume() fails. Fixes: 8e4d5d43cc6c ("drm/amdgpu: Handling of amdgpu_device_resume return value for graceful teardown") Signed-off-by: Yang Yingliang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.