xref: /linux/Documentation/gpu/nova/core/todo.rst (revision 55a42f78ffd386e01a5404419f8c5ded7db70a21)
1.. SPDX-License-Identifier: (GPL-2.0+ OR MIT)
2
3=========
4Task List
5=========
6
7Tasks may have the following fields:
8
9- ``Complexity``: Describes the required familiarity with Rust and / or the
10  corresponding kernel APIs or subsystems. There are four different complexities,
11  ``Beginner``, ``Intermediate``, ``Advanced`` and ``Expert``.
12- ``Reference``: References to other tasks.
13- ``Link``: Links to external resources.
14- ``Contact``: The person that can be contacted for further information about
15  the task.
16
17A task might have `[ABCD]` code after its name. This code can be used to grep
18into the code for `TODO` entries related to it.
19
20Enablement (Rust)
21=================
22
23Tasks that are not directly related to nova-core, but are preconditions in terms
24of required APIs.
25
26FromPrimitive API [FPRI]
27------------------------
28
29Sometimes the need arises to convert a number to a value of an enum or a
30structure.
31
32A good example from nova-core would be the ``Chipset`` enum type, which defines
33the value ``AD102``. When probing the GPU the value ``0x192`` can be read from a
34certain register indication the chipset AD102. Hence, the enum value ``AD102``
35should be derived from the number ``0x192``. Currently, nova-core uses a custom
36implementation (``Chipset::from_u32`` for this.
37
38Instead, it would be desirable to have something like the ``FromPrimitive``
39trait [1] from the num crate.
40
41Having this generalization also helps with implementing a generic macro that
42automatically generates the corresponding mappings between a value and a number.
43
44| Complexity: Beginner
45| Link: https://docs.rs/num/latest/num/trait.FromPrimitive.html
46
47Conversion from byte slices for types implementing FromBytes [TRSM]
48-------------------------------------------------------------------
49
50We retrieve several structures from byte streams coming from the BIOS or loaded
51firmware. At the moment converting the bytes slice into the proper type require
52an inelegant `unsafe` operation; this will go away once `FromBytes` implements
53a proper `from_bytes` method.
54
55| Complexity: Beginner
56
57CoherentAllocation improvements [COHA]
58--------------------------------------
59
60`CoherentAllocation` needs a safe way to write into the allocation, and to
61obtain slices within the allocation.
62
63| Complexity: Beginner
64| Contact: Abdiel Janulgue
65
66Generic register abstraction [REGA]
67-----------------------------------
68
69Work out how register constants and structures can be automatically generated
70through generalized macros.
71
72Example:
73
74.. code-block:: rust
75
76	register!(BOOT0, 0x0, u32, pci::Bar<SIZE>, Fields [
77	   MINOR_REVISION(3:0, RO),
78	   MAJOR_REVISION(7:4, RO),
79	   REVISION(7:0, RO), // Virtual register combining major and minor rev.
80	])
81
82This could expand to something like:
83
84.. code-block:: rust
85
86	const BOOT0_OFFSET: usize = 0x00000000;
87	const BOOT0_MINOR_REVISION_SHIFT: u8 = 0;
88	const BOOT0_MINOR_REVISION_MASK: u32 = 0x0000000f;
89	const BOOT0_MAJOR_REVISION_SHIFT: u8 = 4;
90	const BOOT0_MAJOR_REVISION_MASK: u32 = 0x000000f0;
91	const BOOT0_REVISION_SHIFT: u8 = BOOT0_MINOR_REVISION_SHIFT;
92	const BOOT0_REVISION_MASK: u32 = BOOT0_MINOR_REVISION_MASK | BOOT0_MAJOR_REVISION_MASK;
93
94	struct Boot0(u32);
95
96	impl Boot0 {
97	   #[inline]
98	   fn read(bar: &RevocableGuard<'_, pci::Bar<SIZE>>) -> Self {
99	      Self(bar.readl(BOOT0_OFFSET))
100	   }
101
102	   #[inline]
103	   fn minor_revision(&self) -> u32 {
104	      (self.0 & BOOT0_MINOR_REVISION_MASK) >> BOOT0_MINOR_REVISION_SHIFT
105	   }
106
107	   #[inline]
108	   fn major_revision(&self) -> u32 {
109	      (self.0 & BOOT0_MAJOR_REVISION_MASK) >> BOOT0_MAJOR_REVISION_SHIFT
110	   }
111
112	   #[inline]
113	   fn revision(&self) -> u32 {
114	      (self.0 & BOOT0_REVISION_MASK) >> BOOT0_REVISION_SHIFT
115	   }
116	}
117
118Usage:
119
120.. code-block:: rust
121
122	let bar = bar.try_access().ok_or(ENXIO)?;
123
124	let boot0 = Boot0::read(&bar);
125	pr_info!("Revision: {}\n", boot0.revision());
126
127A work-in-progress implementation currently resides in
128`drivers/gpu/nova-core/regs/macros.rs` and is used in nova-core. It would be
129nice to improve it (possibly using proc macros) and move it to the `kernel`
130crate so it can be used by other components as well.
131
132Features desired before this happens:
133
134* Make I/O optional I/O (for field values that are not registers),
135* Support other sizes than `u32`,
136* Allow visibility control for registers and individual fields,
137* Use Rust slice syntax to express fields ranges.
138
139| Complexity: Advanced
140| Contact: Alexandre Courbot
141
142Numerical operations [NUMM]
143---------------------------
144
145Nova uses integer operations that are not part of the standard library (or not
146implemented in an optimized way for the kernel). These include:
147
148- The "Find Last Set Bit" (`fls` function of the C part of the kernel)
149  operation.
150
151A `num` core kernel module is being designed to provide these operations.
152
153| Complexity: Intermediate
154| Contact: Alexandre Courbot
155
156Delay / Sleep abstractions [DLAY]
157---------------------------------
158
159Rust abstractions for the kernel's delay() and sleep() functions.
160
161FUJITA Tomonori plans to work on abstractions for read_poll_timeout_atomic()
162(and friends) [1].
163
164| Complexity: Beginner
165| Link: https://lore.kernel.org/netdev/20250228.080550.354359820929821928.fujita.tomonori@gmail.com/ [1]
166
167IRQ abstractions
168----------------
169
170Rust abstractions for IRQ handling.
171
172There is active ongoing work from Daniel Almeida [1] for the "core" abstractions
173to request IRQs.
174
175Besides optional review and testing work, the required ``pci::Device`` code
176around those core abstractions needs to be worked out.
177
178| Complexity: Intermediate
179| Link: https://lore.kernel.org/lkml/20250122163932.46697-1-daniel.almeida@collabora.com/ [1]
180| Contact: Daniel Almeida
181
182Page abstraction for foreign pages
183----------------------------------
184
185Rust abstractions for pages not created by the Rust page abstraction without
186direct ownership.
187
188There is active onging work from Abdiel Janulgue [1] and Lina [2].
189
190| Complexity: Advanced
191| Link: https://lore.kernel.org/linux-mm/20241119112408.779243-1-abdiel.janulgue@gmail.com/ [1]
192| Link: https://lore.kernel.org/rust-for-linux/20250202-rust-page-v1-0-e3170d7fe55e@asahilina.net/ [2]
193
194Scatterlist / sg_table abstractions
195-----------------------------------
196
197Rust abstractions for scatterlist / sg_table.
198
199There is preceding work from Abdiel Janulgue, which hasn't made it to the
200mailing list yet.
201
202| Complexity: Intermediate
203| Contact: Abdiel Janulgue
204
205PCI MISC APIs
206-------------
207
208Extend the existing PCI device / driver abstractions by SR-IOV, config space,
209capability, MSI API abstractions.
210
211| Complexity: Beginner
212
213XArray bindings [XARR]
214----------------------
215
216We need bindings for `xa_alloc`/`xa_alloc_cyclic` in order to generate the
217auxiliary device IDs.
218
219| Complexity: Intermediate
220
221Debugfs abstractions
222--------------------
223
224Rust abstraction for debugfs APIs.
225
226| Reference: Export GSP log buffers
227| Complexity: Intermediate
228
229GPU (general)
230=============
231
232Initial Devinit support
233-----------------------
234
235Implement BIOS Device Initialization, i.e. memory sizing, waiting, PLL
236configuration.
237
238| Contact: Dave Airlie
239| Complexity: Beginner
240
241MMU / PT management
242-------------------
243
244Work out the architecture for MMU / page table management.
245
246We need to consider that nova-drm will need rather fine-grained control,
247especially in terms of locking, in order to be able to implement asynchronous
248Vulkan queues.
249
250While generally sharing the corresponding code is desirable, it needs to be
251evaluated how (and if at all) sharing the corresponding code is expedient.
252
253| Complexity: Expert
254
255VRAM memory allocator
256---------------------
257
258Investigate options for a VRAM memory allocator.
259
260Some possible options:
261  - Rust abstractions for
262    - RB tree (interval tree) / drm_mm
263    - maple_tree
264  - native Rust collections
265
266| Complexity: Advanced
267
268Instance Memory
269---------------
270
271Implement support for instmem (bar2) used to store page tables.
272
273| Complexity: Intermediate
274| Contact: Dave Airlie
275
276GPU System Processor (GSP)
277==========================
278
279Export GSP log buffers
280----------------------
281
282Recent patches from Timur Tabi [1] added support to expose GSP-RM log buffers
283(even after failure to probe the driver) through debugfs.
284
285This is also an interesting feature for nova-core, especially in the early days.
286
287| Link: https://lore.kernel.org/nouveau/20241030202952.694055-2-ttabi@nvidia.com/ [1]
288| Reference: Debugfs abstractions
289| Complexity: Intermediate
290
291GSP firmware abstraction
292------------------------
293
294The GSP-RM firmware API is unstable and may incompatibly change from version to
295version, in terms of data structures and semantics.
296
297This problem is one of the big motivations for using Rust for nova-core, since
298it turns out that Rust's procedural macro feature provides a rather elegant way
299to address this issue:
300
3011. generate Rust structures from the C headers in a separate namespace per version
3022. build abstraction structures (within a generic namespace) that implement the
303   firmware interfaces; annotate the differences in implementation with version
304   identifiers
3053. use a procedural macro to generate the actual per version implementation out
306   of this abstraction
3074. instantiate the correct version type one on runtime (can be sure that all
308   have the same interface because it's defined by a common trait)
309
310There is a PoC implementation of this pattern, in the context of the nova-core
311PoC driver.
312
313This task aims at refining the feature and ideally generalize it, to be usable
314by other drivers as well.
315
316| Complexity: Expert
317
318GSP message queue
319-----------------
320
321Implement low level GSP message queue (command, status) for communication
322between the kernel driver and GSP.
323
324| Complexity: Advanced
325| Contact: Dave Airlie
326
327Bootstrap GSP
328-------------
329
330Call the boot firmware to boot the GSP processor; execute initial control
331messages.
332
333| Complexity: Intermediate
334| Contact: Dave Airlie
335
336Client / Device APIs
337--------------------
338
339Implement the GSP message interface for client / device allocation and the
340corresponding client and device allocation APIs.
341
342| Complexity: Intermediate
343| Contact: Dave Airlie
344
345Bar PDE handling
346----------------
347
348Synchronize page table handling for BARs between the kernel driver and GSP.
349
350| Complexity: Beginner
351| Contact: Dave Airlie
352
353FIFO engine
354-----------
355
356Implement support for the FIFO engine, i.e. the corresponding GSP message
357interface and provide an API for chid allocation and channel handling.
358
359| Complexity: Advanced
360| Contact: Dave Airlie
361
362GR engine
363---------
364
365Implement support for the graphics engine, i.e. the corresponding GSP message
366interface and provide an API for (golden) context creation and promotion.
367
368| Complexity: Advanced
369| Contact: Dave Airlie
370
371CE engine
372---------
373
374Implement support for the copy engine, i.e. the corresponding GSP message
375interface.
376
377| Complexity: Intermediate
378| Contact: Dave Airlie
379
380VFN IRQ controller
381------------------
382
383Support for the VFN interrupt controller.
384
385| Complexity: Intermediate
386| Contact: Dave Airlie
387
388External APIs
389=============
390
391nova-core base API
392------------------
393
394Work out the common pieces of the API to connect 2nd level drivers, i.e. vGPU
395manager and nova-drm.
396
397| Complexity: Advanced
398
399vGPU manager API
400----------------
401
402Work out the API parts required by the vGPU manager, which are not covered by
403the base API.
404
405| Complexity: Advanced
406
407nova-core C API
408---------------
409
410Implement a C wrapper for the APIs required by the vGPU manager driver.
411
412| Complexity: Intermediate
413
414Testing
415=======
416
417CI pipeline
418-----------
419
420Investigate option for continuous integration testing.
421
422This can go from as simple as running KUnit tests over running (graphics) CTS to
423booting up (multiple) guest VMs to test VFIO use-cases.
424
425It might also be worth to consider the introduction of a new test suite directly
426sitting on top of the uAPI for more targeted testing and debugging. There may be
427options for collaboration / shared code with the Mesa project.
428
429| Complexity: Advanced
430