xref: /openbmc/linux/Documentation/gpu/vkms.rst (revision a44e4f3a)
1.. _vkms:
2
3==========================================
4 drm/vkms Virtual Kernel Modesetting
5==========================================
6
7.. kernel-doc:: drivers/gpu/drm/vkms/vkms_drv.c
8   :doc: vkms (Virtual Kernel Modesetting)
9
10TODO
11====
12
13CRC API Improvements
14--------------------
15
16- Optimize CRC computation ``compute_crc()`` and plane blending ``blend()``
17
18- Use the alpha value to blend vaddr_src with vaddr_dst instead of
19  overwriting it in ``blend()``.
20
21- Add igt test to check cleared alpha value for XRGB plane format.
22
23- Add igt test to check extreme alpha values i.e. fully opaque and fully
24  transparent (intermediate values are affected by hw-specific rounding modes).
25
26Runtime Configuration
27---------------------
28
29We want to be able to reconfigure vkms instance without having to reload the
30module. Use/Test-cases:
31
32- Hotplug/hotremove connectors on the fly (to be able to test DP MST handling of
33  compositors).
34
35- Configure planes/crtcs/connectors (we'd need some code to have more than 1 of
36  them first).
37
38- Change output configuration: Plug/unplug screens, change EDID, allow changing
39  the refresh rate.
40
41The currently proposed solution is to expose vkms configuration through
42configfs.  All existing module options should be supported through configfs too.
43
44Add Plane Features
45------------------
46
47There's lots of plane features we could add support for:
48
49- Real overlay planes, not just cursor.
50
51- Full alpha blending on all planes.
52
53- Rotation, scaling.
54
55- Additional buffer formats, especially YUV formats for video like NV12.
56  Low/high bpp RGB formats would also be interesting.
57
58- Async updates (currently only possible on cursor plane using the legacy cursor
59  api).
60
61For all of these, we also want to review the igt test coverage and make sure all
62relevant igt testcases work on vkms.
63
64Writeback support
65-----------------
66
67Currently vkms only computes a CRC for each frame. Once we have additional plane
68features, we could write back the entire composited frame, and expose it as:
69
70- Writeback connector. This is useful for testing compositors if you don't have
71  hardware with writeback support.
72
73- As a v4l device. This is useful for debugging compositors on special vkms
74  configurations, so that developers see what's really going on.
75
76Prime Buffer Sharing
77--------------------
78
79We already have vgem, which is a gem driver for testing rendering, similar to
80how vkms is for testing the modeset side. Adding buffer sharing support to vkms
81allows us to test them together, to test synchronization and lots of other
82features. Also, this allows compositors to test whether they work correctly on
83SoC chips, where the display and rendering is very often split between 2
84drivers.
85
86Output Features
87---------------
88
89- Variable refresh rate/freesync support. This probably needs prime buffer
90  sharing support, so that we can use vgem fences to simulate rendering in
91  testing. Also needs support to specify the EDID.
92
93- Add support for link status, so that compositors can validate their runtime
94  fallbacks when e.g. a Display Port link goes bad.
95
96- All the hotplug handling describe under "Runtime Configuration".
97
98Atomic Check using eBPF
99-----------------------
100
101Atomic drivers have lots of restrictions which are not exposed to userspace in
102any explicit form through e.g. possible property values. Userspace can only
103inquiry about these limits through the atomic IOCTL, possibly using the
104TEST_ONLY flag. Trying to add configurable code for all these limits, to allow
105compositors to be tested against them, would be rather futile exercise. Instead
106we could add support for eBPF to validate any kind of atomic state, and
107implement a library of different restrictions.
108
109This needs a bunch of features (plane compositing, multiple outputs, ...)
110enabled already to make sense.
111