aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)AuthorFilesLines
2019-08-21drm/amdgpu/psp: move TMR to cpu invisible vram regionTianci.Yin3-5/+5
so that more visible vram can be available for umd. Reviewed-by: Christian König <[email protected]>. Signed-off-by: Tianci.Yin <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callbackXiaojie Yuan6-9/+3
Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: Set no-retry as default.Feifei Xu1-2/+2
This is to improve performance. Signed-off-by: Feifei Xu <[email protected]> Tested-by: Candice Li <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: add firmware header printing for psp fw loading (v2)Xiaojie Yuan1-0/+56
firmware header information is printed for direct fw loading but not added for psp fw loading yet v2: squash in warning fix (Alex) Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: fix debug level for ppt offset/sizeXiaojie Yuan1-2/+2
Signed-off-by: Xiaojie Yuan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu: remove special autoload handling for navi12Xiaojie Yuan1-2/+1
s/r list in rlc firmware is ready, so remove the special autoload handling Signed-off-by: Xiaojie Yuan <[email protected]> Acked-by: Alex Deucher <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-21drm/amdgpu/gfx9: update pg_flags after determining if gfx off is possibleAlex Deucher2-5/+4
We need to set certain power gating flags after we determine if the firmware version is sufficient to support gfxoff. Previously we set the pg flags in early init, but we later we might have disabled gfxoff if the firmware versions didn't support it. Move adding the additional pg flags after we determine whether or not to support gfxoff. Fixes: 005440066f92 ("drm/amdgpu: enable gfxoff again on raven series (v2)") Tested-by: Kai-Heng Feng <[email protected]> Tested-by: Tom St Denis <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: Kai-Heng Feng <[email protected]>
2019-08-21Merge branch 'odp_fixes' into hmm.gitJason Gunthorpe13-74/+196
From rdma.git Jason Gunthorpe says: ==================== This is a collection of general cleanups for ODP to clarify some of the flows around umem creation and use of the interval tree. ==================== The branch is based on v5.3-rc5 due to dependencies, and is being taken into hmm.git due to dependencies in the next patches. * odp_fixes: RDMA/mlx5: Use odp instead of mr->umem in pagefault_mr RDMA/mlx5: Use ib_umem_start instead of umem.address RDMA/core: Make invalidate_range a device operation RDMA/odp: Use kvcalloc for the dma_list and page_list RDMA/odp: Check for overflow when computing the umem_odp end RDMA/odp: Provide ib_umem_odp_release() to undo the allocs RDMA/odp: Split creating a umem_odp from ib_umem_get RDMA/odp: Make the three ways to create a umem_odp clear RMDA/odp: Consolidate umem_odp initialization RDMA/odp: Make it clearer when a umem is an implicit ODP umem RDMA/odp: Iterate over the whole rbtree directly RDMA/odp: Use the common interval tree library instead of generic RDMA/mlx5: Fix MR npages calculation for IB_ACCESS_HUGETLB Signed-off-by: Jason Gunthorpe <[email protected]>
2019-08-21Merge tag 'drm-misc-next-2019-08-19' of ↵Dave Airlie17-69/+69
git://anongit.freedesktop.org/drm/drm-misc into drm-next drm-misc-next for 5.4: UAPI Changes: Cross-subsystem Changes: Core Changes: - dma-buf: add reservation_object_fences helper, relax reservation_object_add_shared_fence, remove reservation_object seq number (and then restored) - dma-fence: Shrinkage of the dma_fence structure, Merge dma_fence_signal and dma_fence_signal_locked, Store the timestamp in struct dma_fence in a union with cb_list Driver Changes: - More dt-bindings YAML conversions - More removal of drmP.h includes - dw-hdmi: Support get_eld and various i2s improvements - gm12u320: Few fixes - meson: Global cleanup - panfrost: Few refactors, Support for GPU heap allocations - sun4i: Support for DDC enable GPIO - New panels: TI nspire, NEC NL8048HL11, LG Philips LB035Q02, Sharp LS037V7DW01, Sony ACX565AKM, Toppoly TD028TTEC1 Toppoly TD043MTEA1 Signed-off-by: Dave Airlie <[email protected]> [airlied: fixup dma_resv rename fallout] From: Maxime Ripard <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/20190819141923.7l2adietcr2pioct@flea
2019-08-20hmm: use mmu_notifier_get/put for 'struct hmm'Jason Gunthorpe1-0/+2
This is a significant simplification, it eliminates all the remaining 'hmm' stuff in mm_struct, eliminates krefing along the critical notifier paths, and takes away all the ugly locking and abuse of page_table_lock. mmu_notifier_get() provides the single struct hmm per struct mm which eliminates mm->hmm. It also directly guarantees that no mmu_notifier op callback is callable while concurrent free is possible, this eliminates all the krefs inside the mmu_notifier callbacks. The remaining krefs in the range code were overly cautious, drivers are already not permitted to free the mirror while a range exists. Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Ralph Campbell <[email protected]> Tested-by: Ralph Campbell <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2019-08-16dma-buf: Restore seqlock around dma_resv updatesChris Wilson1-1/+6
This reverts 67c97fb79a7f ("dma-buf: add reservation_object_fences helper") dd7a7d1ff2f1 ("drm/i915: use new reservation_object_fences helper") 0e1d8083bddb ("dma-buf: further relax reservation_object_add_shared_fence") 5d344f58da76 ("dma-buf: nuke reservation_object seq number") The scenario that defeats simply grabbing a set of shared/exclusive fences and using them blissfully under RCU is that any of those fences may be reallocated by a SLAB_TYPESAFE_BY_RCU fence slab cache. In this scenario, while keeping the rcu_read_lock we need to establish that no fence was changed in the dma_resv after a read (or full) memory barrier. Signed-off-by: Chris Wilson <[email protected]> Cc: Chris Wilson <[email protected]> Cc: Daniel Vetter <[email protected]> Acked-by: Christian König <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2019-08-15drm/amdgpu: Use new mode2 reset interface for RV.Andrey Grodzovsky2-8/+18
Integrate the mode2 reset into rest sequence. v2: Check ppfuncs pointer for NULL Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15dmr/amdgpu: Fix compile error with CONFIG_DRM_AMDGPU_GART_DEBUGFSAndrey Grodzovsky1-1/+1
Double defintion of 'i' Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amd/amdgpu: Update VM function pointerGang Ba1-0/+7
When VM state changed and system in large bar mode, make sure to use CPU update function, otherwise use SDMA function. Reviewed-by: Christian König <[email protected]> Signed-off-by: Gang Ba <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: Set VM_L2_CNTL.PDE_FAULT_CLASSIFICATION to 0 for GFX10Yong Zhao2-2/+2
We have done this for pre-GFX10 asics, but GFX10 did not pick up the new change. The below is the commit message for that change. This is recommended by HW designers. Previously when it was set to 1, the PDE walk error in VM fault will be treated as PERMISSION_OR_INVALID_PAGE_FAULT rather than usually expected OTHER_FAULT. As a result, the retry control in VM_CONTEXT*_CNTL will change accordingly. The above behavior is kind of abnormal. Furthermore, the PDE_FAULT_CLASSIFICATION == 1 feature was targeted for very old ASICs and it never made it way to production. Therefore, we should set it to 0. Signed-off-by: Yong Zhao <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: Add more page fault info printing for GFX10Yong Zhao1-5/+28
The printing we did for GFX9 was not propogated to GFX10 somehow, so fix it now. Signed-off-by: Yong Zhao <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: Add printing for RW extracted from VM_L2_PROTECTION_FAULT_STATUSYong Zhao1-0/+3
RW is also useful in most cases. Signed-off-by: Yong Zhao <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdkfd/gfx10: Calling amdgpu functions to invalidate TLBOak Zeng1-38/+3
Calling amdgpu function to invalidate TLB, instead of using a kfd implementation. Delete the kfd local TLB invalidation implementation. Signed-off-by: Oak Zeng <[email protected]> Reviewed-by: Christian Konig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: Export function to flush TLB of specific vm hubOak Zeng9-57/+78
This is for kfd to reuse amdgpu TLB invalidation function. On gfx10, kfd only needs to flush TLB on gfx hub but not on mm hub. So export a function for KFD flush TLB only on specific hub. Signed-off-by: Oak Zeng <[email protected]> Reviewed-by: Christian Konig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: MODULE_FIRMWARE requires linux/module.hStephen Rothwell1-0/+1
Fixes: 6a7a0bdbfa0c ("drm/amdgpu: add psp_v12_0 for renoir (v2)") Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Stephen Rothwell <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: implement querying ras error count for mmhubTao Zhou1-0/+55
get mmhub ea ras error count by accessing EDC_CNT register Signed-off-by: Tao Zhou <[email protected]> Reviewed-by: Guchun Chen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: use exiting amdgpu_ctx_total_num_entities functionKevin Wang1-4/+1
simplify driver code. Signed-off-by: Kevin Wang <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: fix typo error amdgput -> amdgpuKevin Wang1-6/+6
fix typo error: change function name from "amdgput_ctx_total_num_entities" to "amdgpu_ctx_total_num_entities". Signed-off-by: Kevin Wang <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: simplify and cleanup setting the dma maskChristoph Hellwig6-83/+14
Use dma_set_mask_and_coherent to set both masks in one go, and remove the no longer required fallback, as the kernel now always accepts larger than required DMA masks. Fail the driver probe if we can't set the DMA mask, as that means the system can only support a larger mask. Reviewed-by: Christian König <[email protected]> Signed-off-by: Christoph Hellwig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-15drm/amdgpu: handle PCIe root ports with addressing limitationsChristoph Hellwig1-1/+1
amdgpu uses a need_dma32 flag to indicate to the drm core that some allocations need to be done using GFP_DMA32, but it only checks the device addressing capabilities to make that decision. Unfortunately PCIe root ports that have limited addressing exist as well. Use the dma_addressing_limited instead to also take those into account. Reviewed-by: Christian König <[email protected]> Signed-off-by: Christoph Hellwig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-13dma-buf: rename reservation_object to dma_resvChristian König17-67/+67
Be more consistent with the naming of the other DMA-buf objects. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chris Wilson <[email protected]> Link: https://patchwork.freedesktop.org/patch/323401/
2019-08-12drm/amdgpu: fix gfx9 soft recoveryPierre-Eric Pelloux-Prayer1-1/+1
The SOC15_REG_OFFSET() macro wasn't used, making the soft recovery fail. v2: use WREG32_SOC15 instead of WREG32 + SOC15_REG_OFFSET Signed-off-by: Pierre-Eric Pelloux-Prayer <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2019-08-12drm/amdgpu: flag renoir as experimental for nowAlex Deucher1-1/+1
The current code won't likely work on production hw when it ships so leave it as experimental until it's ready. Acked-by: Huang Rui <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: skip mec2 jump table loading for renoirHuang Rui1-0/+4
Renoir need not load mec2 jump table with psp. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Aaron Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: use direct loading on renoir vcn for the momentHuang Rui2-6/+12
PSP has issue for renoir, that will cause VCN fw failed to be loaded. So use direct loading for the moment till the issue is addressed. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Aaron Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: set fw default loading by psp for renoirAaron Liu1-5/+1
By default, set amdgpu ucode type to AMDGPU_FW_LOAD_PSP. Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: update lbpw for renoirAaron Liu1-0/+1
enable gfx_v9_0_init_lbpw for renoir Acked-by: Huang Rui <[email protected]> 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]>
2019-08-12drm/amdgpu: enable power gating for renoirAaron Liu1-0/+1
enable gfx power gating for renoir Acked-by: Huang Rui <[email protected]> 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]>
2019-08-12drm/amdgpu: enable clock gating for renoirAaron Liu2-0/+2
enable gfx&common clock gating for renoir Acked-by: Huang Rui <[email protected]> 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]>
2019-08-12drm/amdgpu: add VCN2.0 to Renoir IP blocksLeo Liu1-0/+2
Thus enable VCN2.0 for Renoir Acked-by: Huang Rui <[email protected]> Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: enable Doorbell support for Renoir (v2)Leo Liu2-0/+25
Add VCN range aperture to NBIO 7.0 v2: rebase (Alex) Acked-by: Huang Rui <[email protected]> Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: enable Renoir VCN firmware loadingLeo Liu1-0/+8
By adding new Renoir VCN firmware Acked-by: Huang Rui <[email protected]> Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add sdma golden settings for renoirHuang Rui1-0/+17
This patch adds sdma golden settings for renoir asic. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add gfx golden settings for renoir (v2)Huang Rui1-0/+26
This patch adds gfx golden settings for renoir real asic. v2: update settings (Alex) Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add psp_v12_0 for renoir (v2)Aaron Liu7-2/+619
1. Add psp ip block 2. Use direct loading type by default and it can also config psp loading type. 3. Bypass sos fw loading and xgmi&ras interface v2: drop TA loading Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: set rlc funcs for renoirAaron Liu1-0/+1
add gfx_v9_0_rlc_funcs for renoir Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add asic funcs for renoirAaron Liu1-0/+1
add asic funcs for renoir, init soc15_asic_funcs Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: enable dce virtual ip module for RenoirAaron Liu2-0/+3
Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: fix no interrupt issue for renoir emuAaron Liu1-1/+1
In renoir's ih model, there's a change in mmIH_CHICKEN register, that limits IH to use physical address directly. Those chicken bits need to be programmed first. Acked-by: Huang Rui <[email protected]> Signed-off-by: Aaron Liu <[email protected]> Reviewed-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add renoir pci idHuang Rui1-0/+3
Add Renoir PCI id support. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: set ip blocks for renoirHuang Rui1-0/+7
Enable ip blocks for renoir. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add sdma support for renoirHuang Rui1-1/+5
Add renoir checks to appropriate places. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add gfx support for renoirHuang Rui1-2/+24
Add Renoir checks to gfx9 code. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: set fw load type for renoirHuang Rui1-0/+1
This patch sets fw load type as direct for renoir for the moment. Will switch to psp when psp is ready. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2019-08-12drm/amdgpu: add gmc v9 supports for renoirHuang Rui1-0/+7
Add gfx memory controller support for renoir. Acked-by: Huang Rui <[email protected]> Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.