kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c, branch v5.10.185

drm/amdgpu: fix pci device refcount leak

2023-01-14T09:15:31+00:00

[ Upstream commit b85e285e3d6352b02947fc1b72303673dfacb0aa ] As comment of pci_get_domain_bus_and_slot() says, it returns a pci device with refcount increment, when finish using it, the caller must decrement the reference count by calling pci_dev_put(). So before returning from amdgpu_device_resume|suspend_display_audio(), pci_dev_put() is called to avoid refcount leak. Fixes: 3f12acc8d6d4 ("drm/amdgpu: put the audio codec into suspend state before gpu reset V3") Reviewed-by: Evan Quan Signed-off-by: Yang Yingliang Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

Revert "drm/amdgpu: make sure to init common IP before gmc"

2022-10-26T11:25:56+00:00

This reverts commit 7b0db849ea030a70b8fb9c9afec67c81f955482e which is commit a8671493d2074950553da3cf07d1be43185ef6c6 upstream. The patches that this patch depends on were not backported properly and the patch that caused the regression that this patch set fixed was reverted in commit 412b844143e3 ("Revert "PCI/portdrv: Don't disable AER reporting in get_port_device_capability()""). This isn't necessary and causes a regression so drop it. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2216 Cc: Shuah Khan Cc: Sasha Levin Signed-off-by: Alex Deucher Cc: # 5.10 Signed-off-by: Greg Kroah-Hartman

drm/amd/amdgpu: fixing read wrong pf2vf data in SRIOV

2022-09-28T09:10:37+00:00

commit 9a458402fb69bda886aa6cbe067311b6e3d9c52a upstream. [Why] This fixes 892deb48269c ("drm/amdgpu: Separate vf2pf work item init from virt data exchange"). we should read pf2vf data based at mman.fw_vram_usage_va after gmc sw_init. commit 892deb48269c breaks this logic. [How] calling amdgpu_virt_exchange_data in amdgpu_virt_init_data_exchange to set the right base in the right sequence. v2: call amdgpu_virt_init_data_exchange after gmc sw_init to make data exchange workqueue run v3: clean up the code logic v4: add some comment and make the code more readable Fixes: 892deb48269c ("drm/amdgpu: Separate vf2pf work item init from virt data exchange") Signed-off-by: Jingwen Chen Reviewed-by: Horace Chen Signed-off-by: Alex Deucher Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: make sure to init common IP before gmc

2022-09-28T09:10:20+00:00

[ Upstream commit a8671493d2074950553da3cf07d1be43185ef6c6 ] Move common IP init before GMC init so that HDP gets remapped before GMC init which uses it. This fixes the Unsupported Request error reported through AER during driver load. The error happens as a write happens to the remap offset before real remapping is done. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373 The error was unnoticed before and got visible because of the commit referenced below. This doesn't fix anything in the commit below, rather fixes the issue in amdgpu exposed by the commit. The reference is only to associate this commit with below one so that both go together. Fixes: 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()") Acked-by: Christian König Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin

drm/amdgpu: Separate vf2pf work item init from virt data exchange

2022-09-28T09:10:19+00:00

[ Upstream commit 892deb48269c65376f3eeb5b4c032ff2c2979bd7 ] We want to be able to call virt data exchange conditionally after gmc sw init to reserve bad pages as early as possible. Since this is a conditional call, we will need to call it again unconditionally later in the init sequence. Refactor the data exchange function so it can be called multiple times without re-initializing the work item. v2: Cleaned up the code. Kept the original call to init_exchange_data() inside early init to initialize the work item, afterwards call exchange_data() when needed. Signed-off-by: Victor Skvortsov Reviewed By: Shaoyun.liu Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdgpu: indirect register access for nv12 sriov

2022-09-28T09:10:19+00:00

[ Upstream commit 8b8a162da820d48bb94261ae4684f2c839ce148c ] unify host driver and guest driver indirect access control bits names Signed-off-by: Peng Ju Zhou Reviewed-by: Emily.Deng Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: fix boot failure when iommu is disabled in Picasso.

2021-12-14T10:32:36+00:00

commit afd18180c07026f94a80ff024acef5f4159084a4 upstream. When IOMMU disabled in sbios and kfd in iommuv2 path, iommuv2 init will fail. But this failure should not block amdgpu driver init. Reported-by: youling Tested-by: youling Signed-off-by: Yifan Zhang Reviewed-by: James Zhu Signed-off-by: Alex Deucher Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: init iommu after amdkfd device init

2021-12-14T10:32:35+00:00

commit 714d9e4574d54596973ee3b0624ee4a16264d700 upstream. This patch is to fix clinfo failure in Raven/Picasso: Number of platforms: 1 Platform Profile: FULL_PROFILE Platform Version: OpenCL 2.2 AMD-APP (3364.0) Platform Name: AMD Accelerated Parallel Processing Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd cl_amd_event_callback Platform Name: AMD Accelerated Parallel Processing Number of devices: 0 Signed-off-by: Yifan Zhang Reviewed-by: James Zhu Tested-by: James Zhu Acked-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: move iommu_resume before ip init/resume

2021-12-14T10:32:35+00:00

commit f02abeb0779700c308e661a412451b38962b8a0b upstream. Separate iommu_resume from kfd_resume, and move it before other amdgpu ip init/resume. Bug: https://bugzilla.kernel.org/show_bug.cgi?id=211277 Signed-off-by: James Zhu Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: Cancel delayed work when GFXOFF is disabled

2021-09-03T08:09:22+00:00

commit 32bc8f8373d2d6a681c96e4b25dca60d4d1c6016 upstream. schedule_delayed_work does not push back the work if it was already scheduled before, so amdgpu_device_delay_enable_gfx_off ran ~100 ms after the first time GFXOFF was disabled and re-enabled, even if GFXOFF was disabled and re-enabled again during those 100 ms. This resulted in frame drops / stutter with the upcoming mutter 41 release on Navi 14, due to constantly enabling GFXOFF in the HW and disabling it again (for getting the GPU clock counter). To fix this, call cancel_delayed_work_sync when the disable count transitions from 0 to 1, and only schedule the delayed work on the reverse transition, not if the disable count was already 0. This makes sure the delayed work doesn't run at unexpected times, and allows it to be lock-free. v2: * Use cancel_delayed_work_sync & mutex_trylock instead of mod_delayed_work. v3: * Make amdgpu_device_delay_enable_gfx_off lock-free (Christian König) v4: * Fix race condition between amdgpu_gfx_off_ctrl incrementing adev->gfx.gfx_off_req_count and amdgpu_device_delay_enable_gfx_off checking for it to be 0 (Evan Quan) Cc: stable@vger.kernel.org Reviewed-by: Evan Quan Reviewed-by: Lijo Lazar # v3 Acked-by: Christian König # v3 Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher Signed-off-by: Greg Kroah-Hartman