linux.git/include/uapi/drm/msm_drm.h, branch v5.8

drm/msm: Add syncobj support.

2020-05-18T16:26:32+00:00

This

1) Enables core DRM syncobj support.
2) Adds options to the submission ioctl to wait/signal syncobjs.

Just like the wait fence fd, this does inline waits. Using the
scheduler would be nice but I believe it is out of scope for
this work.

Support for timeline syncobjs is implemented and the interface
is ready for it, but I'm not enabling it yet until there is
some code for turnip to use it.

The reset is mostly in there because in the presence of waiting
and signalling the same semaphores, resetting them after
signalling can become very annoying.

v2:
  - Fixed style issues
  - Removed a cleanup issue in a failure case
  - Moved to a copy_from_user per syncobj

v3:
 - Fixed a missing declaration introduced in v2
 - Reworked to use ERR_PTR/PTR_ERR
 - Simplified failure gotos.

Used by: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2769

Signed-off-by: Bas Nieuwenhuizen 
Reviewed-by: Jordan Crouse 
Signed-off-by: Rob Clark

drm/msm/gpu: Add submit queue queries

2019-04-19T18:50:06+00:00

Add the capability to query information from a submit queue.
The first available parameter is for querying the number of GPU faults
(hangs) that can be attributed to the queue.

This is useful for implementing context robustness. A user context can
regularly query the number of faults to see if it is responsible for any
and if so it can invalidate itself.

This is also helpful for testing by confirming to the user  driver if a
particular command stream caused a fault (or not as the case may be).

Signed-off-by: Jordan Crouse 
Signed-off-by: Rob Clark

drm/msm: add param to retrieve # of GPU faults (global)

2019-04-19T18:50:00+00:00

For KHR_robustness, userspace wants to know two things, the count of GPU
faults globally, and the count of faults attributed to a given context.
This patch providees the former, and the next patch provides the latter.

Signed-off-by: Rob Clark 
Reviewed-by: Jordan Crouse

drm/msm/gpu: add per-process pagetables param

2019-04-19T18:49:42+00:00

For now it always returns '0' (false), but once the iommu work is in
place to enable per-process pagetables we can update the value returned.

Userspace needs to know this to make an informed decision about exposing
KHR_robustness.

Signed-off-by: Rob Clark 
Reviewed-by: Jordan Crouse

drm/msm: add uapi to get/set debug name

2018-12-11T18:07:05+00:00

Add UAPI to get/set GEM objects' debug name.

Signed-off-by: Rob Clark

drm/msm: rework GEM_INFO ioctl

2018-12-11T18:07:04+00:00

Prep work to add a way to get/set the GEM objects debug name.

Signed-off-by: Rob Clark

drm/msm/gpu: add submit flag to hint which buffers should be dumped

2018-12-11T18:07:04+00:00

To lower CPU  overhead, future userspace will be switching to pinning
iova and avoiding the use of relocs, and only include cmds table entries
for IB1 level cmdstream (but not IB2 or state-groups).

This leaves the kernel unsure what to dump for rd/hangrd cmdstream
dumping.  So add a MSM_SUBMIT_BO_DUMP flag so userspace can indicate
buffers that contain cmdstream (or are otherwise important to dump).

Signed-off-by: Rob Clark

drm/msm: add sudo flag to submit ioctl

2018-02-20T15:41:20+00:00

This flags cause cmdstream to be executed from the ringbuffer (RB)
instead of IB1.  Normally not something you'd ever want to do, but
it is super useful for firmware debugging.

Hidden behind CAP_SYS_RAWIO and a default=n kconfig option which
depends on EXPERT (and has a suitably scary warning), to prevent
it from being used on accident.

Signed-off-by: Rob Clark

drm/msm: Add a parameter query for the number of ringbuffers

2017-10-28T15:01:37+00:00

In order to manage ringbuffer priority to its fullest userspace
should know how many ringbuffers it has to work with. Add a
parameter to return the number of active rings.

Signed-off-by: Jordan Crouse 
Signed-off-by: Rob Clark

drm/msm: Support multiple ringbuffers

2017-10-28T15:01:36+00:00

Add the infrastructure to support the idea of multiple ringbuffers.
Assign each ringbuffer an id and use that as an index for the various
ring specific operations.

The biggest delta is to support legacy fences. Each fence gets its own
sequence number but the legacy functions expect to use a unique integer.
To handle this we return a unique identifier for each submission but
map it to a specific ring/sequence under the covers. Newer users use
a dma_fence pointer anyway so they don't care about the actual sequence
ID or ring.

The actual mechanics for multiple ringbuffers are very target specific
so this code just allows for the possibility but still only defines
one ringbuffer for each target family.

Signed-off-by: Jordan Crouse 
Signed-off-by: Rob Clark