- 09 Dec, 2017 12 commits
-
-
Harish Kasiviswanathan authored
Reorganize and rename kfd_topology_get_crat_acpi function. In this way acpi_get_table(..) needs to be called only once. This will also aid in dGPU topology implementation. Signed-off-by:
Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by:
Kent Russell <kent.russell@amd.com> Signed-off-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by:
-
Felix Kuehling authored
Take CRAT related functions out of kfd_topology.c and place them in kfd_crat.c. This is the initial step of supporting more CRAT features, i.e. creating virtual CRAT table for KFD devices without CRAT. v2: Minor cleanup that was missed previously because code moved around Signed-off-by:
Amber Lin <Amber.Lin@amd.com> Signed-off-by:
-
Yong Zhao authored
Kobject created using kobject_create_and_add() can be freed using kobject_put() when there is no referenece any more. However, kobject memory allocated with kzalloc() has to set up a release callback in order to free it when the counter decreases to 0. Otherwise it causes memory leak. Signed-off-by:
Yong Zhao <yong.zhao@amd.com> Signed-off-by:
-
Harish Kasiviswanathan authored
Fix location_id format to match Thunk specification. Signed-off-by:
-
Flora Cui authored
Overwrite the active simd_count from KGD at driver loading time. This is based on assumption that register GC_USER_SHADER_ARRAY_CONFIG won’t get changed. V2: remove the incorrect simd_count reported at loading module. Signed-off-by:
Flora Cui <flora.cui@amd.com> Reviewed by: Yair Shachar< yair.shachar@amd.com> Signed-off-by:
-
Harish Kasiviswanathan authored
Signed-off-by:
-
Harish Kasiviswanathan authored
Signed-off-by:
-
Harish Kasiviswanathan authored
get_vmem_size() is deprecated. Instead use get_local_mem_info(). Signed-off-by:
Ben Goz <ben.goz@amd.com> Signed-off-by:
-
Harish Kasiviswanathan authored
Implement new kgd-kfd interface function get_local_mem_info. Signed-off-by:
-
Harish Kasiviswanathan authored
Add get_local_mem_info which provides more information about local memory than get_vmem_size: - public and private framebuffer size - memory clock Signed-off-by:
-
Flora Cui authored
Signed-off-by:
-
Flora Cui authored
Signed-off-by:
-
- 27 Nov, 2017 12 commits
-
-
Yong Zhao authored
Also fixes error handling if kfd_process_init_cwsr fails. Signed-off-by:
-
Felix Kuehling authored
Signed-off-by:
-
Felix Kuehling authored
Signed-off-by:
-
Yong Zhao authored
If no matching process is found, return NULL instead of a pointer to the last process in the kfd_processes_table. Signed-off-by:
-
Felix Kuehling authored
Use a reference counter instead of a lock to prevent process destruction while functions running out of process context are using the kfd_process structure. In many cases these functions don't need the structure to be locked. In the few cases that really do need the process lock, take it explicitly. This helps simplify lock dependencies between the process lock and other locks, particularly amdgpu and mm_struct locks. This will be important when amdgpu calls back to amdkfd for memory evictions. Signed-off-by:
Christian König <christian.koenig@amd.com> Reviewed-by:
-
Felix Kuehling authored
This will be used to elliminate the use of the process lock for preventing concurrent process destruction. This will simplify lock dependencies between KFD and KGD. This also simplifies the process destruction in a few ways: * Don't allocate work struct dynamically * Remove unnecessary hack that increments mm reference counter * Remove unnecessary process locking during destruction Signed-off-by:
-
Felix Kuehling authored
Increment the kfd_process.lead_thread's reference counter to make it safe to dereference. This is needed for getting a safe reference to the process' mm_struct. Signed-off-by:
-
Felix Kuehling authored
This commit adds several debugfs entries for kfd: kfd/hqds: dumps all HQDs on all GPUs for KFD-controlled compute and SDMA RLC queues kfd/mqds: dumps all MQDs of all KFD processes on all GPUs kfd/rls: dumps HWS runlists on all GPUs Signed-off-by: -
Felix Kuehling authored
This can be used by KFD for debugging features, such as dumping HQDs in debugfs. Signed-off-by:
-
Felix Kuehling authored
This counts the queue offset in register index, not register address. Signed-off-by:
-
Felix Kuehling authored
Don't count SDMA queues towards compute HQD oversubscription when deciding whether to create a chained runlist. Signed-off-by:
Jay Cornwall <Jay.Cornwall@amd.com> Signed-off-by:
-
Felix Kuehling authored
Allow HWS to to execute multiple processes on the hardware concurrently. The number of concurrent processes is limited by the number of VMIDs allocated to the HWS. A module parameter can be used for limiting this further or turn it off altogether (mainly for debugging purposes). Signed-off-by:
-
- 04 Dec, 2017 1 commit
-
-
Kent Russell authored
Just print a pointer instead of casting v2: Remove the 0x prefix, since %p prints that automatically, and remove it from one other spot as well Signed-off-by:
-
- 27 Nov, 2017 2 commits
-
-
Philip Yang authored
After debugger is registered, the pqm_destroy_queue fails because is_debug is true, the queue should not be removed from process_queue_list since the count is not reduced. Test application calls debugger unregister without register debugger, add null pointer check protection to avoid crash for this case Signed-off-by:
Philip Yang <Philip.Yang@amd.com> Signed-off-by:
-
Felix Kuehling authored
Use proper powerplay function. This fixes OpenCL initialization problems. Signed-off-by:
-
- 24 Nov, 2017 1 commit
-
-
Yong Zhao authored
Signed-off-by:
-
- 14 Nov, 2017 4 commits
-
-
Felix Kuehling authored
A second-level user mode trap handler can be installed. The CWSR trap handler jumps to the secondary trap handler conditionally for any conditions not handled by it. This can be used e.g. for debugging or catching math exceptions. When CWSR is disabled, the user mode trap handler is installed as first level trap handler. Signed-off-by:
Shaoyun.liu <shaoyun.liu@amd.com> Signed-off-by:
-
Felix Kuehling authored
This hardware feature allows the GPU to preempt shader execution in the middle of a compute wave, save the state and restore it later to resume execution. Memory for saving the state is allocated per queue in user mode and the address and size passed to the create_queue ioctl. The size depends on the number of waves that can be in flight simultaneously on a given ASIC. Signed-off-by:
-
Felix Kuehling authored
The trap handler is like an interrupt handler running on the GPU compute unit. It is needed for supporting CWSR (compute wave save/restore). This file defines an array with the pre-compiled GFXv8 shader ISA. The assembly code is included for reference in #if 0 ... #endif. Signed-off-by:
-
Felix Kuehling authored
The PQM doesn't change after process creation. So initialize it in kfd_create_process_device_data. Signed-off-by:
-
- 06 Nov, 2017 2 commits
-
-
Felix Kuehling authored
Replace (ffs(size) - 1) with order_base_2(size) as a more straight forward way to get log2 of buffer sizes. Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Felix Kuehling authored
Don't use sizeof(uint32_t) or similar types for hardware or firmware DWORD size. The hardware and firmware don't care about Linux types. Signed-off-by:
-
- 01 Nov, 2017 5 commits
-
-
Philip Cox authored
Signed-off-by:
Philip Cox <Philip.Cox@amd.com> Signed-off-by:
-
Felix Kuehling authored
Signed-off-by:
-
Philip Cox authored
Signed-off-by:
-
Felix Kuehling authored
Save wptr in hqd_sdma_destroy, restore it in hqd_sdma_load. Also read updated wptr from user mode when resuming an SDMA queue. Signed-off-by:
-
Felix Kuehling authored
Add wptr and mm parameters to hqd_sdma_load and pass these parameters from device_queue_manager through the mqd_manager. SDMA doesn't support polling while the engine believes it's idle. The driver must update the wptr. The new parameters will be used for looking up the updated value from the specified mm when SDMA queues are resumed after being disabled. Signed-off-by:
-
- 21 Dec, 2017 1 commit
-
-
git://people.freedesktop.org/~agd5f/linuxDave Airlie authored
* 'drm-next-4.16' of git://people.freedesktop.org/~agd5f/linux: (171 commits) drm/amdgpu: fix test for shadow page tables drm/amd/display: Expose dpp1_set_cursor_attributes drm/amd/display: Update FMT and OPPBUF functions drm/amd/display: check for null before calling is_blanked drm/amd/display: dal 3.1.27 drm/amd/display: Fix unused variable warnings. drm/amd/display: Only blank DCN when we have set_blank implementation drm/amd/display: Put dcn_mi_registers with other structs drm/amd/display: hubp refactor drm/amd/display: integrating optc pseudocode drm/amd/display: Call validate_fbc should_enable_fbc drm/amd/display: Clean up DCN cursor code drm/amd/display: fix 180 full screen pipe split drm/amd/display: reprogram surface config on scaling change drm/amd/display: Remove dwbc from pipe_ctx drm/amd/display: Use the maximum link setting which EDP reported. drm/amd/display: Add hdr_supported flag drm/amd/display: fix global sync param retrieval when not pipe splitting drm/amd/display: Update HUBP drm/amd/display: fix rotated surface scaling ...
-
