aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)AuthorFilesLines
2024-03-07drm/amdgpu/vpe: add VPE 6.1.1 supportLang Yu2-106/+207
Add initial support for VPE 6.1.1. v2: squash in updates (Alex) Signed-off-by: Lang Yu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu/vpe: don't emit cond exec command under collaborate modeLang Yu1-0/+3
Not ready now. Signed-off-by: Lang Yu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu/vpe: add collaborate mode support for VPELang Yu2-0/+21
Under clollaborate mode, multiple VPE instances share a ring buferr and work together to finish a job. Signed-off-by: Lang Yu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu/vpe: add PRED_EXE and COLLAB_SYNC OPCODELang Yu1-1/+2
To support multi VPE collaborate mode. Signed-off-by: Lang Yu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu/vpe: add multi instance VPE supportLang Yu2-0/+13
Add support for multi instance VPE processing. Signed-off-by: Lang Yu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu/discovery: add nbif v6_3_1 ip blockLikun Gao1-0/+5
Add nbif v6_3_1 ip block. Signed-off-by: Likun Gao <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-07drm/amdgpu: Add nbif v6_3_1 ip block supportHawking Zhang3-1/+529
Add nbif v6_3_1 ip block support. Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Likun Gao <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-06drm/amdgpu: add ring timeout information in devcoredumpSunil Khatri2-0/+15
Add ring timeout related information in the amdgpu devcoredump file for debugging purposes. During the gpu recovery process the registered call is triggered and add the debug information in data file created by devcoredump framework under the directory /sys/class/devcoredump/devcdx/ Signed-off-by: Sunil Khatri <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-06drm/amdgpu: add dcn3.5.1 supportYifan Zhang1-0/+1
This patch to add dcn3.5.1 support. Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-06drm/amdgpu: disable ring_muxer if mcbp is offPierre-Eric Pelloux-Prayer2-11/+14
Using the ring_muxer without preemption adds overhead for no reason since mcbp cannot be triggered. Moving back to a single queue in this case also helps when high priority app are used: in this case the gpu_scheduler priority handling will work as expected - much better than ring_muxer with its 2 independant schedulers competing for the same hardware queue. This change requires moving amdgpu_device_set_mcbp above amdgpu_device_ip_early_init because we use adev->gfx.mcbp. Signed-off-by: Pierre-Eric Pelloux-Prayer <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Acked-by: Christian König <[email protected]> Acked-by: Jiadong Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-06drm/amdgpu: remove unused codeJesse Zhang2-47/+15
Remove the unused function - amdgpu_vm_pt_is_root_clean and remove the impossible condition v1: entries == 0 is not possible any more, so this condition could probably be removed (Felix) Signed-off-by: Jesse Zhang <[email protected]> Suggested-by：Felix Kuehling <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-04drm/amdgpu: workaround to avoid SET_Q_MODE packets v2Christian König2-16/+93
It turned out that executing the SET_Q_MODE packet on every submission creates to much overhead. Implement a workaround which allows skipping the SET_Q_MODE packet if subsequent submissions all use the same parameters. v2: add a NULL check for ring_obj Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-04drm/amdgpu: cleanup conditional executionChristian König11-184/+99
First of all calculating the number of dw to patch into a conditional execution is not something HW generation specific. This is just standard ring buffer calculations. While at it also reduce the BUG_ON() into WARN_ON(). Then instead of a random bit pattern use 0 as default value for the number of dw skipped, this way it's not mandatory any more to patch the conditional execution. And last make the address to check a parameter of the conditional execution instead of getting this from the ring. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-04drm/amdgpu: Use rpm_mode flag instead of checking it again for rpmMa Jun1-12/+12
Because the rpm_mode flag is already set when the driver is initialized, we use it directly for runtime suspend/resume instead of checking it again Signed-off-by: Ma Jun <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-04drm/amdgpu: change vm->task_info handlingShashank Sharma13-119/+249
This patch changes the handling and lifecycle of vm->task_info object. The major changes are: - vm->task_info is a dynamically allocated ptr now, and its uasge is reference counted. - introducing two new helper funcs for task_info lifecycle management - amdgpu_vm_get_task_info: reference counts up task_info before returning this info - amdgpu_vm_put_task_info: reference counts down task_info - last put to task_info() frees task_info from the vm. This patch also does logistical changes required for existing usage of vm->task_info. V2: Do not block all the prints when task_info not found (Felix) V3: Fixed review comments from Felix - Fix wrong indentation - No debug message for -ENOMEM - Add NULL check for task_info - Do not duplicate the debug messages (ti vs no ti) - Get first reference of task_info in vm_init(), put last in vm_fini() V4: Fixed review comments from Felix - fix double reference increment in create_task_info - change amdgpu_vm_get_task_info_pasid - additional changes in amdgpu_gem.c while porting Cc: Christian Koenig <[email protected]> Cc: Alex Deucher <[email protected]> Cc: Felix Kuehling <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Shashank Sharma <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-04Revert "drm/amdgpu: remove vm sanity check from amdgpu_vm_make_compute" for ↵Jesse Zhang3-81/+1
Raven fix the issue: "amdgpu: Failed to create process VM object". [Why]when amdgpu initialized, seq64 do mampping and update bo mapping in vm page table. But when clifo run. It also initializes a vm for a process device through the function kfd_process_device_init_vm and ensure the root PD is clean through the function amdgpu_vm_pt_is_root_clean. So they have a conflict, and clinfo always failed. v1: - remove all the pte_supports_ats stuff from the amdgpu_vm code (Felix) Signed-off-by: Jesse Zhang <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-03-01drm/amdgpu: use GTT only as fallback for VRAM|GTTChristian König1-0/+6
Try to fill up VRAM as well by setting the busy flag on GTT allocations. This fixes the issue that when VRAM was evacuated for suspend it's never filled up again unless the application is restarted. Signed-off-by: Christian König <[email protected]> Reviewed-by: Zack Rusin <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2024-02-29drm/amdgpu: remove misleading amdgpu_pmops_runtime_idle() commentBjorn Helgaas1-2/+1
After 4020c2280233 ("drm/amdgpu: don't runtime suspend if there are displays attached (v3)"), "ret" is unconditionally set later before being used, so there's point in initializing it and the associated comment is no longer meaningful. Remove the comment and the unnecessary initialization. Signed-off-by: Bjorn Helgaas <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-29Revert "drm/amd: Remove freesync video mode amdgpu parameter"Alex Deucher2-0/+28
This reverts commit e94e787e37b99645e7c02d20d0a1ba0f8a18a82a. This conflicts with how compositors want to handle VRR. Now that compositors actually handle VRR, we probably don't need freesync video. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/2985 Acked-by: Hamza Mahfooz <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-29drm/amdgpu: add deferred error check for UMC v12 address queryTao Zhou1-1/+2
Both RAS UE and deferred errors need page retirement. Signed-off-by: Tao Zhou <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-28amd/amdkfd: remove unused parameterEric Huang2-3/+2
The adev can be found from bo by amdgpu_ttm_adev(bo->tbo.bdev), and adev is also not used in the function amdgpu_amdkfd_map_gtt_bo_to_gart(). Signed-off-by: Eric Huang <[email protected]> Reviewed-by: Harish Kasiviswanathan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-27drm/amd/amdgpu: Fix potential ioremap() memory leaks in amdgpu_device_init()Srinivasan Shanmugam1-6/+10
This ensures that the memory mapped by ioremap for adev->rmmio, is properly handled in amdgpu_device_init(). If the function exits early due to an error, the memory is unmapped. If the function completes successfully, the memory remains mapped. Reported by smatch: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:4337 amdgpu_device_init() warn: 'adev->rmmio' from ioremap() not released on lines: 4035,4045,4051,4058,4068,4337 Cc: Christian König <[email protected]> Cc: Alex Deucher <[email protected]> Signed-off-by: Srinivasan Shanmugam <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-27drm/amdgpu: Fix missing break in ATOM_ARG_IMM Case of atom_get_src_int()Srinivasan Shanmugam1-1/+1
Missing break statement in the ATOM_ARG_IMM case of a switch statement, adds the missing break statement, ensuring that the program's control flow is as intended. Fixes the below: drivers/gpu/drm/amd/amdgpu/atom.c:323 atom_get_src_int() warn: ignoring unreachable code. Fixes: d38ceaf99ed0 ("drm/amdgpu: add core driver (v4)") Cc: Jammy Zhou <[email protected]> Cc: Christian König <[email protected]> Cc: Alex Deucher <[email protected]> Signed-off-by: Srinivasan Shanmugam <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-27drm/amd: Drop abm_level propertyMario Limonciello2-10/+0
This vendor specific property has never been used by userspace software and conflicts with the panel_power_savings sysfs file. That is a compositor and user could fight over the same data. Fixes: 63d0b87213a0 ("drm/amd/display: add panel_power_savings sysfs entry to eDP connectors") Suggested-by: Harry Wentland <[email protected]> Cc: Hamza Mahfooz <[email protected]> Cc: "Sun peng Li (Leo)" <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Mario Limonciello <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: reserve more memory for MES runtime DRAMTim Huang1-3/+10
This patch fixes a MES firmware boot failure issue when backdoor loading the MES firmware. MES firmware runtime DRAM size is changed to 512k, the driver needs to reserve this amount of memory in FB, otherwise adjacent memory will be overwritten by the MES firmware startup code. Signed-off-by: Tim Huang <[email protected]> Reviewed-by: Yifan Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: Enable gpu reset for S3 abort cases on Raven seriesPrike Liang1-20/+25
Currently, GPU resets can now be performed successfully on the Raven series. While GPU reset is required for the S3 suspend abort case. So now can enable gpu reset for S3 abort cases on the Raven series. Signed-off-by: Prike Liang <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: Do not program SQ_TIMEOUT_CONFIG in SRIOVVictor Lu1-0/+3
VF should not program this register. Signed-off-by: Victor Lu <[email protected]> Reviewed-by: Zhigang Luo <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: Fix ineffective ras_mask settingsStanley.Yang1-0/+1
Check amdgpu_ras_mask to fix ineffective ras_mask setting due to special asic without sram ecc enable but with poison supported. Signed-off-by: Stanley.Yang <[email protected]> Reviewed-by: Tao Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdkfd: Skip packet submission on fatal errorLijo Lazar2-0/+6
If fatal error is detected, packet submission won't go through. Return error in such cases. Also, avoid waiting for fence when fatal error is detected. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Asad Kamal <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: Add fatal error detected flagLijo Lazar3-0/+39
For a RAS error that needs a full reset to recover, set the fatal error status. Clear the status once the device is reset. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Asad Kamal <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26drm/amdgpu: Fix the runtime resume failure issueMa Jun1-0/+3
Don't set power state flag when system enter runtime suspend, or it may cause runtime resume failure issue. Fixes: 3a9626c816db ("drm/amd: Stop evicting resources on APUs in suspend") Signed-off-by: Ma Jun <[email protected]> Reviewed-by: Mario Limonciello <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-26Merge v6.8-rc6 into drm-nextDaniel Vetter1-0/+3
Thomas Zimmermann asked to backmerge -rc6 for drm-misc branches, there's a few same-area-changed conflicts (xe and amdgpu mostly) that are getting a bit too annoying. Signed-off-by: Daniel Vetter <[email protected]>
2024-02-26Merge tag 'drm-misc-next-2024-02-22' of ↵Daniel Vetter3-0/+21
git://anongit.freedesktop.org/drm/drm-misc into drm-next drm-misc-next for v6.9: UAPI Changes: - changes to fdinfo stats Cross-subsystem Changes: agp: - remove unused type field from struct agp_bridge_data Core Changes: ci: - update test names - cleanups gem: - add stats for shared buffers plus updates to amdgpu, i915, xe Documentation: - fixes syncobj: - fixes to waiting and sleeping Driver Changes: bridge: - adv7511: fix crash on irq during probe - dw_hdmi: set bridge type host1x: - cleanups ivpu: - updates to firmware API - refactor BO allocation meson: - fix error handling in probe panel: - revert "drm/panel-edp: Add auo_b116xa3_mode" - add Himax HX83112A plus DT bindings - ltk500hd1829: add support for ltk101b4029w and admatec 9904370 - simple: add BOE BP082WX1-100 8.2" panel plus DT bindungs renesas: - add RZ/G2L DU support plus DT bindings Signed-off-by: Daniel Vetter <[email protected]> From: Thomas Zimmermann <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2024-02-22treewide: update LLVM Bugzilla linksNathan Chancellor1-1/+1
LLVM moved their issue tracker from their own Bugzilla instance to GitHub issues. While all of the links are still valid, they may not necessarily show the most up to date information around the issues, as all updates will occur on GitHub, not Bugzilla. Another complication is that the Bugzilla issue number is not always the same as the GitHub issue number. Thankfully, LLVM maintains this mapping through two shortlinks: https://llvm.org/bz<num> -> https://bugs.llvm.org/show_bug.cgi?id=<num> https://llvm.org/pr<num> -> https://github.com/llvm/llvm-project/issues/<mapped_num> Switch all "https://bugs.llvm.org/show_bug.cgi?id=<num>" links to the "https://llvm.org/pr<num>" shortlink so that the links show the most up to date information. Each migrated issue links back to the Bugzilla entry, so there should be no loss of fidelity of information here. Link: https://lkml.kernel.org/r/[email protected] Signed-off-by: Nathan Chancellor <[email protected]> Reviewed-by: Kees Cook <[email protected]> Acked-by: Fangrui Song <[email protected]> Cc: Alexei Starovoitov <[email protected]> Cc: Andrii Nakryiko <[email protected]> Cc: Daniel Borkmann <[email protected]> Cc: Mykola Lysenko <[email protected]> Signed-off-by: Andrew Morton <[email protected]>
2024-02-22drm/amdgpu: Fix the runtime resume failure issueMa Jun1-0/+3
Don't set power state flag when system enter runtime suspend, or it may cause runtime resume failure issue. Fixes: 3a9626c816db ("drm/amd: Stop evicting resources on APUs in suspend") Signed-off-by: Ma Jun <[email protected]> Reviewed-by: Mario Limonciello <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2024-02-22drm/amdgpu: add vcn 4.0.6 discovery supportYifan Zhang1-0/+1
This patch is to add vcn 4.0.6 support Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Yifan Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Use RMW accessors for changing LNKCTL2Ilpo Järvinen2-52/+30
Convert open coded RMW accesses for LNKCTL2 to use pcie_capability_clear_and_set_word() which makes its easier to understand what the code tries to do. LNKCTL2 is not really owned by any driver because it is a collection of control bits that PCI core might need to touch. RMW accessors already have support for proper locking for a selected set of registers (LNKCTL2 is not yet among them but likely will be in the future) to avoid losing concurrent updates. Acked-by: Alex Deucher <[email protected]> Suggested-by: Lukas Wunner <[email protected]> Signed-off-by: Ilpo Järvinen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Simplify the allocation of sync slab cachesKunwu Chan1-3/+1
Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. Reviewed-by: Christian König <[email protected]> Signed-off-by: Kunwu Chan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu/soc21: Enabling PG and CG flags for VCN 4.0.6Veerabadhran Gopalakrishnan1-2/+7
Enabled the VCN Power Gating and Clock Gating flags for VCN 4.0.6. Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Veerabadhran Gopalakrishnan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Simplify the allocation of mux_chunk slab cachesKunwu Chan1-3/+1
Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. Reviewed-by: Christian König <[email protected]> Signed-off-by: Kunwu Chan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Simplify the allocation of fence slab cachesKunwu Chan1-3/+1
Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. Reviewed-by: Christian König <[email protected]> Signed-off-by: Kunwu Chan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu/soc21: Added Video Capabilities for VCN 406Veerabadhran Gopalakrishnan1-0/+6
Updated Query Video codecs for VCN 406 Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Veerabadhran Gopalakrishnan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu/vcn: Enable VCN 4.0.6 SupportVeerabadhran Gopalakrishnan2-0/+5
Modified driver to use the appropriate FW files and instance. v2: squash in fixes (Alex) Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Veerabadhran Gopalakrishnan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu/jpeg: add support for jpeg multi instanceSaleemkhan Jamadar1-54/+123
Enable support for multi instance on JPEG 4.0.6. v2: squash in fixes (Alex) Signed-off-by: Saleemkhan Jamadar <[email protected]> Reviewed-by: Veerabadhran Gopalakrishnan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Use correct SRIOV macro for gmc_v9_0_vm_fault_interrupt_stateVictor Lu1-4/+4
Under SRIOV, programming to VM_CONTEXT*_CNTL regs failed because the current macro does not pass through the correct xcc instance. Use the *REG32_XCC macro in this case. The behaviour without SRIOV is the same without this patch. Signed-off-by: Victor Lu <[email protected]> Reviewed-by: Zhigang Luo <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Do not program IH_CHICKEN in vega20_ih.c under SRIOVVictor Lu1-18/+20
IH_CHICKEN is blocked for VF writes; this access should be skipped. Signed-off-by: Victor Lu <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: Improve error checking in amdgpu_virt_rlcg_reg_rw (v2)Victor Lu2-2/+4
The current error detection only looks for a timeout. This should be changed to also check scratch_reg1 for any errors returned from RLCG. v2: remove new error value Signed-off-by: Victor Lu <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: enable MES discovery for GC 11.5.1Yifan Zhang1-0/+1
This patch to enable MES for GC 11.5.1 Reviewed-by: shaoyun.liu <[email protected]> Signed-off-by: Yifan Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: add GC 11.5.1 discovery supportYifan Zhang1-0/+1
This patch to add GC 11.5.1 support Signed-off-by: Yifan Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2024-02-22drm/amdgpu: enable CGPG for GFX ip v11.5.1Tim Huang1-2/+21
Enable CGPG support for GFX ip v11.5.1 Signed-off-by: Tim Huang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.