linux/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c, branch v5.14

drm/amdgpu: Use delayed work to collect RAS error counters

2021-05-27T16:23:06Z

On Context Query2 IOCTL return the correctable and uncorrectable errors in O(1) fashion, from cached values, and schedule a delayed work function to calculate and cache them for the next such IOCTL. v2: Cancel pending delayed work at ras_fini(). v3: Remove conditionals when dealing with delayed work manipulation as they're inherently racy. Cc: Alexander Deucher Cc: Christian König Cc: John Clements Cc: Hawking Zhang Signed-off-by: Luben Tuikov Reviewed-by: Alexander Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: Don't query CE and UE errors

2021-05-27T16:22:19Z

On QUERY2 IOCTL don't query counts of correctable and uncorrectable errors, since when RAS is enabled and supported on Vega20 server boards, this takes insurmountably long time, in O(n^3), which slows the system down to the point of it being unusable when we have GUI up. Fixes: ae363a212b14 ("drm/amdgpu: Add a new flag to AMDGPU_CTX_OP_QUERY_STATE2") Cc: Alexander Deucher Cc: stable@vger.kernel.org Signed-off-by: Luben Tuikov Reviewed-by: Alexander Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: fix fence calculation (v2)

2021-05-13T18:09:12Z

The proper metric for fence utilization over several contexts is an harmonic mean, but such calculation is prohibitive in kernel space, so the code approximates it. Because the approximation diverges when one context has a very small ratio compared with the other context, this change filter out ratios smaller that 0.01% v2: make the fence calculation static and initialize variables within that function v3: Fix warnings (Alex) Reviewed-by: Alex Deucher Signed-off-by: David M Nieto Signed-off-by: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20210513174539.27409-2-david.nieto@amd.com

drm/amdgpu: free resources on fence usage query

2021-05-13T18:02:24Z

Free the resources if the fence needs to be ignored during the ratio calculation Reviewed-by: Alex Deucher Signed-off-by: David M Nieto Signed-off-by: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20210513174539.27409-1-david.nieto@amd.com

drm/amdgpu: Add show_fdinfo() interface

2021-05-05T07:26:53Z

Tracking devices, process info and fence info using /proc/pid/fdinfo Signed-off-by: David M Nieto Signed-off-by: Roy Sun Reviewed-by: Christian König Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20210426062701.39732-2-Roy.Sun@amd.com

drm/amdgpu/amdgpu: use "*" adjacent to data name

2020-11-02T20:35:50Z

When declaring pointer data, the "*" symbol should be used adjacent to the data name as per the coding standards. This resolves following issues reported by checkpatch script: ERROR: "foo * bar" should be "foo *bar" ERROR: "foo * bar" should be "foo *bar" ERROR: "foo* bar" should be "foo *bar" ERROR: "(foo*)" should be "(foo *)" Signed-off-by: Deepak R Varma Signed-off-by: Alex Deucher

amd/amdgpu_ctx: Use struct_size() helper and kmalloc() (v2)

2020-10-09T18:44:39Z

Make use of the new struct_size() helper instead of the offsetof() idiom. Also, use kmalloc() instead of kcalloc(). v2: squash in kzalloc fix Signed-off-by: Gustavo A. R. Silva Signed-off-by: Alex Deucher

drm/amdgpu: disable gpu-sched load balance for uvd

2020-09-03T18:46:54Z

On hardware with multiple uvd instances, dependent uvd jobs may get scheduled to different uvd instances. Because uvd_enc jobs retain hw context, dependent jobs should always run on the same uvd instance. This patch disables GPU scheduler's load balancer for a context that binds jobs from the same context to a uvd instance. v2: Squash in uvd_enc fix Signed-off-by: Nirmoy Das Reviewed-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: drm_device to amdgpu_device by inline-f (v2)

2020-08-24T17:06:06Z

Get the amdgpu_device from the DRM device by use of an inline function, drm_to_adev(). The inline function resolves a pointer to struct drm_device to a pointer to struct amdgpu_device. v2: Use a typed visible static inline function instead of an invisible macro. Signed-off-by: Luben Tuikov Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/scheduler: Remove priority macro INVALID (v2)

2020-08-18T22:20:26Z

Remove DRM_SCHED_PRIORITY_INVALID. We no longer carry around an invalid priority and cut it off at the source. Backwards compatibility behaviour of AMDGPU CTX IOCTL passing in garbage for context priority from user space and then mapping that to DRM_SCHED_PRIORITY_NORMAL is preserved. v2: Revert "res" --> "r" and "prio" --> "priority". Signed-off-by: Luben Tuikov Reviewed-by: Christian König Signed-off-by: Alex Deucher