kernel/drivers/gpu/drm/amd/amdkfd/kfd_device.c, branch linux-4.17.y

drm/amdkfd: Implement KFD process eviction/restore

2018-02-07T01:32:45Z

When the TTM memory manager in KGD evicts BOs, all user mode queues potentially accessing these BOs must be evicted temporarily. Once user mode queues are evicted, the eviction fence is signaled, allowing the migration of the BO to proceed. A delayed worker is scheduled to restore all the BOs belonging to the evicted process and restart its queues. During suspend/resume of the GPU we also evict all processes to allow KGD to save BOs in system memory, since VRAM will be lost. v2: * Account for eviction when updating of q->is_active in MQD manager Signed-off-by: Harish Kasiviswanathan Signed-off-by: Felix Kuehling Acked-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Centralize IOMMUv2 code and make it conditional

2017-12-09T00:22:12Z

dGPUs work without IOMMUv2. Make IOMMUv2 initialization dependent on ASIC information. Also allow building KFD without IOMMUv2 support. This is still useful for dGPUs and prepares for enabling KFD on architectures that don't support AMD IOMMUv2. v2: * Centralize IOMMUv2 code to avoid #ifdefs in too many places v3: * Imply AMD_IOMMU_V2 in Kconfig Signed-off-by: Felix Kuehling Acked-by: Christian Konig Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Add dGPU device IDs and device info

2018-01-04T22:17:47Z

v2: remove needs_iommu field as it doesn't exists CC: linux-pci@vger.kernel.org Signed-off-by: Felix Kuehling Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Make sched_policy a per-device setting

2018-01-04T22:17:43Z

Some dGPUs don't support HWS. Allow them to use a per-device sched_policy that may be different from the global default. Signed-off-by: Felix Kuehling Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Conditionally enable PCIe atomics

2018-01-04T22:17:41Z

This will be needed for most dGPUs. CC: linux-pci@vger.kernel.org Signed-off-by: Felix Kuehling Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: map multiple processes to HW scheduler

2017-11-27T23:29:45Z

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: Yong Zhao Signed-off-by: Jay Cornwall Signed-off-by: Felix Kuehling Acked-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Add CWSR support

2017-11-14T21:41:19Z

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: Shaoyun.liu Signed-off-by: Yong Zhao Signed-off-by: Felix Kuehling Signed-off-by: Oded Gabbay

drm/amdkfd: use a high priority workqueue for IH work

2017-10-27T23:35:34Z

In systems under heavy load the IH work may experience significant scheduling delays. Under load + system workqueue: Max Latency: 7.023695 ms Avg Latency: 0.263994 ms Under load + high priority workqueue: Max Latency: 1.162568 ms Avg Latency: 0.163213 ms Further work is required to measure the impact of per-cpu settings on IH performance. Signed-off-by: Andres Rodriguez Signed-off-by: Felix Kuehling Acked-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Use VMID bitmap from KGD v2

2017-09-20T22:10:18Z

The hard-coded values related to VMID were removed in KFD, as those values can be calculated in the KFD initialization function. v2: remove unnecessary local variable Signed-off-by: Yong Zhao Signed-off-by: Felix Kuehling Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

drm/amdkfd: Fix suspend/resume issue on Carrizo v2

2017-09-20T22:10:14Z

When we do suspend/resume through "sudo pm-suspend" while there is HSA activity running, upon resume we will encounter HWS hanging, which is caused by memory read/write failures. The root cause is that when suspend, we neglected to unbind pasid from kfd device. Another major change is that the bind/unbinding is changed to be performed on a per process basis, instead of whether there are queues in dqm. v2: - free IOMMU device if kfd_bind_processes_to_device fails in kfd_resume - add comments to kfd_bind/unbind_processes_to/from_device - minor cleanups Signed-off-by: Yong Zhao Signed-off-by: Felix Kuehling Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay