kernel/linux.git/drivers/gpu/drm/amd/amdkfd, branch v4.19.67

drm/amdkfd: Fix sdma queue map issue

2019-07-31T05:26:58+00:00

[ Upstream commit 065e4bdfa1f3ab2884c110394d8b7e7ebe3b988c ] Previous codes assumes there are two sdma engines. This is not true e.g., Raven only has 1 SDMA engine. Fix the issue by using sdma engine number info in device_info. Signed-off-by: Oak Zeng Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix a potential memory leak

2019-07-31T05:26:58+00:00

[ Upstream commit e73390d181103a19e1111ec2f25559a0570e9fe0 ] Free mqd_mem_obj it GTT buffer allocation for MQD+control stack fails. Signed-off-by: Oak Zeng Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: use init_mqd function to allocate object for hid_mqd (CI)

2019-04-20T07:16:04+00:00

[ Upstream commit cac734c2dbd2514f14c8c6a17caba1990d83bf1d ] if use the legacy method to allocate object, when mqd_hiq need to run uninit code, it will be cause WARNING call trace. eg: (s3 suspend test) [ 34.918944] Call Trace: [ 34.918948] [] dump_stack+0x19/0x1b [ 34.918950] [] __warn+0xd8/0x100 [ 34.918951] [] warn_slowpath_null+0x1d/0x20 [ 34.918991] [] uninit_mqd_hiq_sdma+0x4e/0x50 [amdgpu] [ 34.919028] [] uninitialize+0x37/0xe0 [amdgpu] [ 34.919064] [] kernel_queue_uninit+0x16/0x30 [amdgpu] [ 34.919086] [] pm_uninit+0x12/0x20 [amdgpu] [ 34.919107] [] stop_nocpsch+0x15/0x20 [amdgpu] [ 34.919129] [] kgd2kfd_suspend.part.4+0x2e/0x50 [amdgpu] [ 34.919150] [] kgd2kfd_suspend+0x17/0x20 [amdgpu] [ 34.919171] [] amdgpu_amdkfd_suspend+0x1a/0x20 [amdgpu] [ 34.919187] [] amdgpu_device_suspend+0x88/0x3a0 [amdgpu] [ 34.919189] [] ? enqueue_entity+0x2ef/0xbe0 [ 34.919205] [] amdgpu_pmops_suspend+0x20/0x30 [amdgpu] [ 34.919207] [] pci_pm_suspend+0x6f/0x150 [ 34.919208] [] ? pci_pm_freeze+0xf0/0xf0 [ 34.919210] [] dpm_run_callback+0x46/0x90 [ 34.919212] [] __device_suspend+0xfb/0x2a0 [ 34.919213] [] async_suspend+0x1f/0xa0 [ 34.919214] [] async_run_entry_fn+0x3f/0x130 [ 34.919216] [] process_one_work+0x17f/0x440 [ 34.919217] [] worker_thread+0x126/0x3c0 [ 34.919218] [] ? manage_workers.isra.25+0x2a0/0x2a0 [ 34.919220] [] kthread+0xd1/0xe0 [ 34.919221] [] ? insert_kthread_work+0x40/0x40 [ 34.919222] [] ret_from_fork_nospec_begin+0x7/0x21 [ 34.919224] [] ? insert_kthread_work+0x40/0x40 [ 34.919224] ---[ end trace 38cd9f65c963adad ]--- Signed-off-by: Kevin Wang Reviewed-by: Oak Zeng Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Don't assign dGPUs to APU topology devices

2019-02-27T09:08:55+00:00

[ Upstream commit bbdf514fe5648566b0754476cbcb92ac3422dde2 ] dGPUs need their own topology devices. Don't assign them to APU topology devices with CPU cores. Bug: https://github.com/RadeonOpenCompute/ROCK-Kernel-Driver/issues/66 Signed-off-by: Felix Kuehling Tested-by: Elias Konstantinidis Acked-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: fix interrupt spin lock

2019-01-26T08:32:33+00:00

[ Upstream commit 2383a767c0ca06f96534456d8313909017c6c8d0 ] Vega10 has multiple interrupt rings, so this can be called from multiple calles at the same time resulting in: [ 71.779334] ================================ [ 71.779406] WARNING: inconsistent lock state [ 71.779478] 4.19.0-rc1+ #44 Tainted: G W [ 71.779565] -------------------------------- [ 71.779637] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. [ 71.779740] kworker/6:1/120 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 71.779832] 00000000ad761971 (&(&kfd->interrupt_lock)->rlock){?...}, at: kgd2kfd_interrupt+0x75/0x100 [amdgpu] [ 71.780058] {IN-HARDIRQ-W} state was registered at: [ 71.780115] _raw_spin_lock+0x2c/0x40 [ 71.780180] kgd2kfd_interrupt+0x75/0x100 [amdgpu] [ 71.780248] amdgpu_irq_callback+0x6c/0x150 [amdgpu] [ 71.780315] amdgpu_ih_process+0x88/0x100 [amdgpu] [ 71.780380] amdgpu_irq_handler+0x20/0x40 [amdgpu] [ 71.780409] __handle_irq_event_percpu+0x49/0x2a0 [ 71.780436] handle_irq_event_percpu+0x30/0x70 [ 71.780461] handle_irq_event+0x37/0x60 [ 71.780484] handle_edge_irq+0x83/0x1b0 [ 71.780506] handle_irq+0x1f/0x30 [ 71.780526] do_IRQ+0x53/0x110 [ 71.780544] ret_from_intr+0x0/0x22 [ 71.780566] cpuidle_enter_state+0xaa/0x330 [ 71.780591] do_idle+0x203/0x280 [ 71.780610] cpu_startup_entry+0x6f/0x80 [ 71.780634] start_secondary+0x1b0/0x200 [ 71.780657] secondary_startup_64+0xa4/0xb0 Fix this by always using irq save spin locks. Signed-off-by: Christian König Acked-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: add new vega10 pci ids

2018-12-19T18:19:53+00:00

commit 756e16bf79f2815e7c83a04881b5545b55a99fd3 upstream. New vega10 ids. Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdkfd: Fix incorrect use of process->mm

2018-10-04T15:37:25+00:00

This mm_struct pointer should never be dereferenced. If running in a user thread, just use current->mm. If running in a kernel worker use get_task_mm to get a safe reference to the mm_struct. Reviewed-by: Oded Gabbay Acked-by: Christian König Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Fix ATS capablity was not reported correctly on some APUs

2018-09-20T15:25:23+00:00

Because CRAT_CU_FLAGS_IOMMU_PRESENT was not set in some BIOS crat, we need to workaround this. For future compatibility, we also overwrite the bit in capability according to the value of needs_iommu_device. Acked-by: Alex Deucher Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Change the control stack MTYPE from UC to NC on GFX9

2018-09-20T15:25:17+00:00

CWSR fails on Raven if the control stack is MTYPE_UC, which is used for regular GART mappings. As a workaround we map it using MTYPE_NC. The MEC firmware expects the control stack at one page offset from the start of the MQD so it is part of the MQD allocation on GFXv9. AMDGPU added a memory allocation flag just for this purpose. Acked-by: Alex Deucher Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Call kfd2kgd.set_compute_idle

2018-07-16T23:10:37+00:00

User mode queue submissions don't go through KFD. Therefore we don't know exactly when compute is idle or not idle. We use the existence of user mode queues on a device as an approximation. register_process is called when the first queue of a process is created. Conversely unregister_process is called when the last queue is destroyed. The first process that is registered takes compute out of idle. The last process that is unregisters sets compute back to idle. Signed-off-by: Felix Kuehling Reviewed-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Oded Gabbay