summaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)Author
2018-03-15drm/amdkfd: Add module option for testing large-BAR functionalityFelix Kuehling
Simulate large-BAR system by exporting only visible memory. This limits the amount of available VRAM to the size of the BAR, but enables CPU access to VRAM. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Kmap event page for dGPUsFelix Kuehling
The events page must be accessible in user mode by the GPU and CPU as well as in kernel mode by the CPU. On dGPUs user mode virtual addresses are managed by the Thunk's GPU memory allocation code. Therefore we can't allocate the memory in kernel mode like we do on APUs. But KFD still needs to map the memory for kernel access. To facilitate this, the Thunk provides the buffer handle of the events page to KFD when creating the first event. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Add ioctls for GPUVM memory managementFelix Kuehling
v2: * Fix error handling after kfd_bind_process_to_device in kfd_ioctl_map_memory_to_gpu v3: * Add ioctl to acquire VM from a DRM FD v4: * Return number of successful map/unmap operations in failure cases * Facilitate partial retry after failed map/unmap * Added comments with parameter descriptions to new APIs * Defined AMDKFD_IOC_FREE_MEMORY_OF_GPU write-only Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Add TC flush on VMID deallocation for HawaiiFelix Kuehling
On GFX7 the CP does not perform a TC flush when queues are unmapped. To avoid TC eviction from accessing an invalid VMID, flush it explicitly before releasing a VMID. v2: Fix unnecessary list_for_each_entry_safe v3: Moved allocation to kfd_process_device_init_vm Signed-off-by: Amber Lin <Amber.Lin@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Allocate CWSR trap handler memory for dGPUsFelix Kuehling
Add helpers for allocating GPUVM memory in kernel mode and use them to allocate memory for the CWSR trap handler. v2: Use dev instead of pdd->dev in kfd_process_free_gpuvm v3: * Cleaned up and simplified kfd_process_alloc_gpuvm * Moved allocation for dGPU to kfd_process_device_init_vm Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Add per-process IDR for buffer handlesFelix Kuehling
Also used for cleaning up on process termination. v2: Refactored cleanup on process termination Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Aperture setup for dGPUsFelix Kuehling
Set up the GPUVM aperture for SVM (shared virtual memory) that allows sharing a part of virtual address space between GPUs and CPUs. Report the size of the GPUVM aperture that is supported by KGD accurately. The low part of the GPUVM aperture is reserved for kernel use. This is for kernel-allocated buffers that are only accessed on the GPU: - CWSR trap handler - IB for submitting commands in user-mode context from kernel mode Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Remove limit on number of GPUsFelix Kuehling
Currently the number of GPUs is limited by aperture placement options available on GFX7 and GFX8 hardware. This limitation is not necessary. Scratch and LDS represent per-work-item and per-work-group storage respectively. Different work-items and work-groups use the same virtual address to access their own data. Work running on different GPUs is by definition in different work-groups (different dispatches, in fact). That means the same virtual addresses can be used for these apertures on different GPUs. Add a new AMDKFD_IOC_GET_PROCESS_APERTURES_NEW ioctl that removes the artificial limitation on the number of GPUs that can be supported. The new ioctl allows user mode to query the number of GPUs to allocate enough memory for all GPUs to be reported. This deprecates AMDKFD_IOC_GET_PROCESS_APERTURES. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Populate DRM render device minorOak Zeng
Populate DRM render device minor in kfd topology Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdkfd: Create KFD VMs on demandFelix Kuehling
Instead of creating all VMs on process creation, create them when a process is bound to a device. This will later allow registering an existing VM from a DRM render node FD at runtime, before the process is bound to the device. This way the render node VM can be used for KFD instead of creating our own redundant VM. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Add kfd2kgd interface to acquire an existing VMFelix Kuehling
This allows acquiring an existing VM from a render node FD to use it for a compute process. Such VMs get destroyed when the original file descriptor is released. Added a callback from amdgpu_vm_fini to handle KFD VM destruction correctly in this case. v2: * Removed vm->vm_context check in amdgpu_amdkfd_gpuvm_destroy_cb, check vm->process_info earlier instead Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Add helper to turn an existing VM into a compute VMFelix Kuehling
v2: Removed updating and checking of vm->vm_context v3: Enable amdgpu_vm_clear_bo in amdgpu_vm_make_compute Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Fix initial validation of PD BO for KFD VMsFelix Kuehling
Make sure the PD BO is valid and attach the eviction fence during VM creation. This ensures that the pd_phys_address is actually valid and an eviction that would invalidate it triggers a KFD process eviction like it should. v2: Use uninterruptible waiting in initial PD validation Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Move KFD-specific fields into struct amdgpu_vmFelix Kuehling
Remove struct amdkfd_vm and move the fields into struct amdgpu_vm. This will allow turning a VM created by a DRM render node into a KFD VM. v2: Removed vm_context field Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Improve documentation of bo_ptr in amdgpu_bo_create_kernelAndrey Grodzovsky
and amdgpu_bo_create_reserved. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/tegra: fb: Properly support linear modifierThierry Reding
Instead of relying on the tiling attached to a buffer object, make sure to set the proper tiling for linear buffers. Signed-off-by: Thierry Reding <treding@nvidia.com>
2018-03-15drm/tegra: plane: Support format modifiersThierry Reding
Pass the list of valid format modifiers to planes upon initialization and implement the ->format_mod_supported() callback so that userspace can query for the valid combinations of formats and modifiers. Signed-off-by: Thierry Reding <treding@nvidia.com>
2018-03-15drm/imx: move arming of the vblank event to atomic_flushLucas Stach
Right now the vblank event completion is racing with the atomic update, which is especially bad when the PRE is in use, as one of the hardware issue workaround might extend the atomic commit for quite some time. If the vblank IRQ happens to trigger during that time, we will prematurely signal the atomic commit completion to userspace, which causes tearing when userspace re-uses a framebuffer we haven't managed to flip away from yet. Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
2018-03-15drm/imx: ipuv3-plane: Include "imx-drm.h" header fileFabio Estevam
ipu_planes_assign_pre() prototype is in "imx-drm.h" header file, so include it to fix the following sparse warning: drivers/gpu/drm/imx/ipuv3-plane.c:729:5: warning: symbol 'ipu_planes_assign_pre' was not declared. Should it be static? Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
2018-03-15drm/imx: ipuv3-plane: Make functions static when possibleFabio Estevam
ipu_plane_state_reset(), ipu_plane_duplicate_state() and ipu_plane_destroy_state() are only used in this file, so make them static. This fixes the following sparse warnings: drivers/gpu/drm/imx/ipuv3-plane.c:275:6: warning: symbol 'ipu_plane_state_reset' was not declared. Should it be static? drivers/gpu/drm/imx/ipuv3-plane.c:295:24: warning: symbol 'ipu_plane_duplicate_state' was not declared. Should it be static? drivers/gpu/drm/imx/ipuv3-plane.c:309:6: warning: symbol 'ipu_plane_destroy_state' was not declared. Should it be static? Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
2018-03-15gpu: ipu-v3: prg: avoid possible array underflowArnd Bergmann
gcc-8 reports that we access an array with a negative index in an error case: drivers/gpu/ipu-v3/ipu-prg.c: In function 'ipu_prg_channel_disable': drivers/gpu/ipu-v3/ipu-prg.c:252:43: error: array subscript -22 is below array bounds of 'struct ipu_prg_channel[3]' [-Werror=array-bounds] This moves the range check in front of the first time that variable gets used. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
2018-03-15drm/amdkfd: fix uninitialized variable useArnd Bergmann
When CONFIG_ACPI is disabled, we never initialize the acpi_table structure in kfd_create_crat_image_virtual: drivers/gpu/drm/amd/amdkfd/kfd_crat.c: In function 'kfd_create_crat_image_virtual': drivers/gpu/drm/amd/amdkfd/kfd_crat.c:888:40: error: 'acpi_table' may be used uninitialized in this function [-Werror=maybe-uninitialized] The undefined behavior also happens for any other acpi_get_table() failure, but then the compiler can't warn about it. This adds an error check that prevents the structure from being used in error, avoiding both the undefined behavior and the warning about it. Fixes: 520b8fb755cc ("drm/amdkfd: Add topology support for CPUs") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
2018-03-15drm/amdgpu: Use atomic function to disable crtcs with dc enabledMikita Lipski
This change fixes the deadlock when unloading the driver with displays connected. Signed-off-by: Mikita Lipski <mikita.lipski@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/radeon: Don't turn off DP sink when disconnectedMichel Dänzer
Turning off the sink in this case causes various issues, because userspace expects it to stay on until it turns it off explicitly. Instead, turn the sink off and back on when a display is connected again. This dance seems necessary for link training to work correctly. Bugzilla: https://bugs.freedesktop.org/105308 Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/radeon: Don't turn off DP sink when disconnectedMichel Dänzer
Turning off the sink in this case causes various issues, because userspace expects it to stay on until it turns it off explicitly. Instead, turn the sink off and back on when a display is connected again. This dance seems necessary for link training to work correctly. Bugzilla: https://bugs.freedesktop.org/105308 Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Rename file name cz_* to smu8_*Rex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Replace function/struct name cz_* with smu8_*Rex Zhu
hw ip smu8 was used on CZ/ST, so use smu8 as the prefix of the function/struct name in powerplay. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Remove unneeded void * casts in cz_hwmgr.c/cz_smumgr.cRex Zhu
Removes unneeded void * casts for the following pointers: hwmgr->backend hwmgr->smu_backend Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Mv cz uvd/vce pg/dpm functions to cz_hwmgr.cRex Zhu
1. delete cz_clockpowergating.c/.h files 2. mark uvd/vce dpm/pg functions static Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Remove dead header file pp_asicblocks.hRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Delete dead code on cz_clockpowergating.cRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amdgpu: Call amdgpu_ucode_fini_bo in amd_powerplay.cRex Zhu
make it symmetric with amdgpu_ucode_init_bo in amd_powerplay.c refine the "commit b22558bb4ff8fc9fe925222f90297d7a03a5fb20" Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amdgpu: Remove wrapper layer of smu ip functionsRex Zhu
1. delete amdgpu_powerplay.c used for wrapping smu ip functions 2. delete struct pp_instance, 3. make struct hwmgr as the smu hw handle. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amdgpu: Don't compared ip_block_type with ip_block_indexRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amdgpu: Plus NULL function pointer checkRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Move helper functions to smu_help.cRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Replace rv_* with smu10_*Rex Zhu
Powerplay is for the hw ip smu, for RV, smu10 is used, so use smu10 as the prefix of the files name/function name. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Fix function parameter not correctRex Zhu
caused by "commit dcefb7668e5b4fb56099b16d1790cd3056322b03" Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Add rv_copy_table_from/to_smc to smu backend function tableRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/tegra: plane: Correct legacy blendingDmitry Osipenko
Keep old 'dependent' state of unaffected planes, this way new state takes into account current state of unaffected planes. Fixes: ebae8d07435a ("drm/tegra: dc: Implement legacy blending") Signed-off-by: Dmitry Osipenko <digetx@gmail.com> Signed-off-by: Thierry Reding <treding@nvidia.com>
2018-03-15drm/amd/pp: Add new smu backend function smc_table_managerRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Mark internal functions as static in rv_smumgr.cRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Add rv_read_arg_from_smc to smu backend function tableRex Zhu
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Remove meanless return value check in RVRex Zhu
In send_message_to_smu helper functions, Print out the error code for debug if smu failed to response. The helper functions always return true, so no need to check their return value. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Delete is_smc_ram_running function on RVRex Zhu
1. There is a race condition when another ip also use same register pairs 2. check once at boot up by GetDriverIfVersion message is sufficient to check SMU health. so delete is_smc_ram_running check. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Clean up header file includeRex Zhu
smu7_smumgr.h should not be included in rv and vega, The common functions for all smu7 asics are put in smu_smumgr.c. Not include cgs interface in smumgr.c. the code used cgs interface has been deleted. Not include smu_ucode_xfer_vi.h in rv/vega Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Fix memory leak in error path in smumgrRex Zhu
Free the backend structure if we fail to allocate device memory. Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/pp: Simplified the avfs btc state on smu7Rex Zhu
AVFS feature support/not support is enough to driver. so remove the complex define of the avfs btc state. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/amd/powerplay: Fix KASAN user after free on driver unload.Andrey Grodzovsky
Reusing local handle to initialize BO without resetting it to NULL is wrong since it causes amdgpu_bo_create_reserved to skip new BO creation and just reuse the given pointer for pinning. Reviewed-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-03-15drm/tegra: dc: Dedicate overlay plane to cursor on older Tegra'sDmitry Osipenko
Older Tegra's do not support RGBA format for the cursor, but instead overlay plane could be used for it. Since there is no much use for the overlays on a regular desktop and HW-accelerated cursor is much better than a SW cursor, let's dedicate one overlay plane to the mouse cursor. Signed-off-by: Dmitry Osipenko <digetx@gmail.com> Signed-off-by: Thierry Reding <treding@nvidia.com>
generated by cgit (git 2.34.1) at 2025-09-15 11:08:29 +0000