xref: /linux/Documentation/gpu/nova/core/todo.rst (revision 2330437da0994321020777c605a2a8cb0ecb7001)
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- Aligning up and down to a power of two,
149- The "Find Last Set Bit" (`fls` function of the C part of the kernel)
150  operation.
151
152A `num` core kernel module is being designed to provide these operations.
153
154| Complexity: Intermediate
155| Contact: Alexandre Courbot
156
157Delay / Sleep abstractions [DLAY]
158---------------------------------
159
160Rust abstractions for the kernel's delay() and sleep() functions.
161
162FUJITA Tomonori plans to work on abstractions for read_poll_timeout_atomic()
163(and friends) [1].
164
165| Complexity: Beginner
166| Link: https://lore.kernel.org/netdev/20250228.080550.354359820929821928.fujita.tomonori@gmail.com/ [1]
167
168IRQ abstractions
169----------------
170
171Rust abstractions for IRQ handling.
172
173There is active ongoing work from Daniel Almeida [1] for the "core" abstractions
174to request IRQs.
175
176Besides optional review and testing work, the required ``pci::Device`` code
177around those core abstractions needs to be worked out.
178
179| Complexity: Intermediate
180| Link: https://lore.kernel.org/lkml/20250122163932.46697-1-daniel.almeida@collabora.com/ [1]
181| Contact: Daniel Almeida
182
183Page abstraction for foreign pages
184----------------------------------
185
186Rust abstractions for pages not created by the Rust page abstraction without
187direct ownership.
188
189There is active onging work from Abdiel Janulgue [1] and Lina [2].
190
191| Complexity: Advanced
192| Link: https://lore.kernel.org/linux-mm/20241119112408.779243-1-abdiel.janulgue@gmail.com/ [1]
193| Link: https://lore.kernel.org/rust-for-linux/20250202-rust-page-v1-0-e3170d7fe55e@asahilina.net/ [2]
194
195Scatterlist / sg_table abstractions
196-----------------------------------
197
198Rust abstractions for scatterlist / sg_table.
199
200There is preceding work from Abdiel Janulgue, which hasn't made it to the
201mailing list yet.
202
203| Complexity: Intermediate
204| Contact: Abdiel Janulgue
205
206PCI MISC APIs
207-------------
208
209Extend the existing PCI device / driver abstractions by SR-IOV, config space,
210capability, MSI API abstractions.
211
212| Complexity: Beginner
213
214XArray bindings [XARR]
215----------------------
216
217We need bindings for `xa_alloc`/`xa_alloc_cyclic` in order to generate the
218auxiliary device IDs.
219
220| Complexity: Intermediate
221
222Debugfs abstractions
223--------------------
224
225Rust abstraction for debugfs APIs.
226
227| Reference: Export GSP log buffers
228| Complexity: Intermediate
229
230GPU (general)
231=============
232
233Initial Devinit support
234-----------------------
235
236Implement BIOS Device Initialization, i.e. memory sizing, waiting, PLL
237configuration.
238
239| Contact: Dave Airlie
240| Complexity: Beginner
241
242MMU / PT management
243-------------------
244
245Work out the architecture for MMU / page table management.
246
247We need to consider that nova-drm will need rather fine-grained control,
248especially in terms of locking, in order to be able to implement asynchronous
249Vulkan queues.
250
251While generally sharing the corresponding code is desirable, it needs to be
252evaluated how (and if at all) sharing the corresponding code is expedient.
253
254| Complexity: Expert
255
256VRAM memory allocator
257---------------------
258
259Investigate options for a VRAM memory allocator.
260
261Some possible options:
262  - Rust abstractions for
263    - RB tree (interval tree) / drm_mm
264    - maple_tree
265  - native Rust collections
266
267| Complexity: Advanced
268
269Instance Memory
270---------------
271
272Implement support for instmem (bar2) used to store page tables.
273
274| Complexity: Intermediate
275| Contact: Dave Airlie
276
277GPU System Processor (GSP)
278==========================
279
280Export GSP log buffers
281----------------------
282
283Recent patches from Timur Tabi [1] added support to expose GSP-RM log buffers
284(even after failure to probe the driver) through debugfs.
285
286This is also an interesting feature for nova-core, especially in the early days.
287
288| Link: https://lore.kernel.org/nouveau/20241030202952.694055-2-ttabi@nvidia.com/ [1]
289| Reference: Debugfs abstractions
290| Complexity: Intermediate
291
292GSP firmware abstraction
293------------------------
294
295The GSP-RM firmware API is unstable and may incompatibly change from version to
296version, in terms of data structures and semantics.
297
298This problem is one of the big motivations for using Rust for nova-core, since
299it turns out that Rust's procedural macro feature provides a rather elegant way
300to address this issue:
301
3021. generate Rust structures from the C headers in a separate namespace per version
3032. build abstraction structures (within a generic namespace) that implement the
304   firmware interfaces; annotate the differences in implementation with version
305   identifiers
3063. use a procedural macro to generate the actual per version implementation out
307   of this abstraction
3084. instantiate the correct version type one on runtime (can be sure that all
309   have the same interface because it's defined by a common trait)
310
311There is a PoC implementation of this pattern, in the context of the nova-core
312PoC driver.
313
314This task aims at refining the feature and ideally generalize it, to be usable
315by other drivers as well.
316
317| Complexity: Expert
318
319GSP message queue
320-----------------
321
322Implement low level GSP message queue (command, status) for communication
323between the kernel driver and GSP.
324
325| Complexity: Advanced
326| Contact: Dave Airlie
327
328Bootstrap GSP
329-------------
330
331Call the boot firmware to boot the GSP processor; execute initial control
332messages.
333
334| Complexity: Intermediate
335| Contact: Dave Airlie
336
337Client / Device APIs
338--------------------
339
340Implement the GSP message interface for client / device allocation and the
341corresponding client and device allocation APIs.
342
343| Complexity: Intermediate
344| Contact: Dave Airlie
345
346Bar PDE handling
347----------------
348
349Synchronize page table handling for BARs between the kernel driver and GSP.
350
351| Complexity: Beginner
352| Contact: Dave Airlie
353
354FIFO engine
355-----------
356
357Implement support for the FIFO engine, i.e. the corresponding GSP message
358interface and provide an API for chid allocation and channel handling.
359
360| Complexity: Advanced
361| Contact: Dave Airlie
362
363GR engine
364---------
365
366Implement support for the graphics engine, i.e. the corresponding GSP message
367interface and provide an API for (golden) context creation and promotion.
368
369| Complexity: Advanced
370| Contact: Dave Airlie
371
372CE engine
373---------
374
375Implement support for the copy engine, i.e. the corresponding GSP message
376interface.
377
378| Complexity: Intermediate
379| Contact: Dave Airlie
380
381VFN IRQ controller
382------------------
383
384Support for the VFN interrupt controller.
385
386| Complexity: Intermediate
387| Contact: Dave Airlie
388
389External APIs
390=============
391
392nova-core base API
393------------------
394
395Work out the common pieces of the API to connect 2nd level drivers, i.e. vGPU
396manager and nova-drm.
397
398| Complexity: Advanced
399
400vGPU manager API
401----------------
402
403Work out the API parts required by the vGPU manager, which are not covered by
404the base API.
405
406| Complexity: Advanced
407
408nova-core C API
409---------------
410
411Implement a C wrapper for the APIs required by the vGPU manager driver.
412
413| Complexity: Intermediate
414
415Testing
416=======
417
418CI pipeline
419-----------
420
421Investigate option for continuous integration testing.
422
423This can go from as simple as running KUnit tests over running (graphics) CTS to
424booting up (multiple) guest VMs to test VFIO use-cases.
425
426It might also be worth to consider the introduction of a new test suite directly
427sitting on top of the uAPI for more targeted testing and debugging. There may be
428options for collaboration / shared code with the Mesa project.
429
430| Complexity: Advanced
431