7 This section contains a list of smaller janitorial tasks in the kernel DRM
8 graphics subsystem useful as newbie projects. Or for slow rainy days.
10 Subsystem-wide refactorings
11 ===========================
13 Remove custom dumb_map_offset implementations
14 ---------------------------------------------
16 All GEM based drivers should be using drm_gem_create_mmap_offset() instead.
17 Audit each individual driver, make sure it'll work with the generic
18 implementation (there's lots of outdated locking leftovers in various
19 implementations), and then remove it.
21 Contact: Daniel Vetter, respective driver maintainers
23 Convert existing KMS drivers to atomic modesetting
24 --------------------------------------------------
26 3.19 has the atomic modeset interfaces and helpers, so drivers can now be
27 converted over. Modern compositors like Wayland or Surfaceflinger on Android
28 really want an atomic modeset interface, so this is all about the bright
31 There is a conversion guide for atomic and all you need is a GPU for a
32 non-converted driver (again virtual HW drivers for KVM are still all
35 As part of this drivers also need to convert to universal plane (which means
36 exposing primary & cursor as proper plane objects). But that's much easier to
37 do by directly using the new atomic helper driver callbacks.
39 Contact: Daniel Vetter, respective driver maintainers
41 Clean up the clipped coordination confusion around planes
42 ---------------------------------------------------------
44 We have a helper to get this right with drm_plane_helper_check_update(), but
45 it's not consistently used. This should be fixed, preferrably in the atomic
46 helpers (and drivers then moved over to clipped coordinates). Probably the
47 helper should also be moved from drm_plane_helper.c to the atomic helpers, to
48 avoid confusion - the other helpers in that file are all deprecated legacy
51 Contact: Ville Syrjälä, Daniel Vetter, driver maintainers
53 Convert early atomic drivers to async commit helpers
54 ----------------------------------------------------
56 For the first year the atomic modeset helpers didn't support asynchronous /
57 nonblocking commits, and every driver had to hand-roll them. This is fixed
58 now, but there's still a pile of existing drivers that easily could be
59 converted over to the new infrastructure.
61 One issue with the helpers is that they require that drivers handle completion
62 events for atomic commits correctly. But fixing these bugs is good anyway.
64 Contact: Daniel Vetter, respective driver maintainers
66 Fallout from atomic KMS
67 -----------------------
69 ``drm_atomic_helper.c`` provides a batch of functions which implement legacy
70 IOCTLs on top of the new atomic driver interface. Which is really nice for
71 gradual conversion of drivers, but unfortunately the semantic mismatches are
72 a bit too severe. So there's some follow-up work to adjust the function
73 interfaces to fix these issues:
75 * atomic needs the lock acquire context. At the moment that's passed around
76 implicitly with some horrible hacks, and it's also allocate with
77 ``GFP_NOFAIL`` behind the scenes. All legacy paths need to start allocating
78 the acquire context explicitly on stack and then also pass it down into
79 drivers explicitly so that the legacy-on-atomic functions can use them.
81 Except for some driver code this is done. This task should be finished by
82 adding WARN_ON(!drm_drv_uses_atomic_modeset) in drm_modeset_lock_all().
84 * A bunch of the vtable hooks are now in the wrong place: DRM has a split
85 between core vfunc tables (named ``drm_foo_funcs``), which are used to
86 implement the userspace ABI. And then there's the optional hooks for the
87 helper libraries (name ``drm_foo_helper_funcs``), which are purely for
88 internal use. Some of these hooks should be move from ``_funcs`` to
89 ``_helper_funcs`` since they are not part of the core ABI. There's a
90 ``FIXME`` comment in the kerneldoc for each such case in ``drm_crtc.h``.
92 Contact: Daniel Vetter
94 Get rid of dev->struct_mutex from GEM drivers
95 ---------------------------------------------
97 ``dev->struct_mutex`` is the Big DRM Lock from legacy days and infested
98 everything. Nowadays in modern drivers the only bit where it's mandatory is
99 serializing GEM buffer object destruction. Which unfortunately means drivers
100 have to keep track of that lock and either call ``unreference`` or
101 ``unreference_locked`` depending upon context.
103 Core GEM doesn't have a need for ``struct_mutex`` any more since kernel 4.8,
104 and there's a ``gem_free_object_unlocked`` callback for any drivers which are
105 entirely ``struct_mutex`` free.
107 For drivers that need ``struct_mutex`` it should be replaced with a driver-
108 private lock. The tricky part is the BO free functions, since those can't
109 reliably take that lock any more. Instead state needs to be protected with
110 suitable subordinate locks or some cleanup work pushed to a worker thread. For
111 performance-critical drivers it might also be better to go with a more
112 fine-grained per-buffer object and per-context lockings scheme. Currently only the
113 ``msm`` driver still use ``struct_mutex``.
115 Contact: Daniel Vetter, respective driver maintainers
117 Convert instances of dev_info/dev_err/dev_warn to their DRM_DEV_* equivalent
118 ----------------------------------------------------------------------------
120 For drivers which could have multiple instances, it is necessary to
121 differentiate between which is which in the logs. Since DRM_INFO/WARN/ERROR
122 don't do this, drivers used dev_info/warn/err to make this differentiation. We
123 now have DRM_DEV_* variants of the drm print macros, so we can start to convert
124 those drivers back to using drm-formwatted specific log messages.
126 Before you start this conversion please contact the relevant maintainers to make
127 sure your work will be merged - not everyone agrees that the DRM dmesg macros
130 Contact: Sean Paul, Maintainer of the driver you plan to convert
132 Convert drivers to use simple modeset suspend/resume
133 ----------------------------------------------------
135 Most drivers (except i915 and nouveau) that use
136 drm_atomic_helper_suspend/resume() can probably be converted to use
137 drm_mode_config_helper_suspend/resume(). Also there's still open-coded version
138 of the atomic suspend/resume code in older atomic modeset drivers.
140 Contact: Maintainer of the driver you plan to convert
142 Convert drivers to use drm_fb_helper_fbdev_setup/teardown()
143 -----------------------------------------------------------
145 Most drivers can use drm_fb_helper_fbdev_setup() except maybe:
147 - amdgpu which has special logic to decide whether to call
148 drm_helper_disable_unused_functions()
150 - armada which isn't atomic and doesn't call
151 drm_helper_disable_unused_functions()
153 - i915 which calls drm_fb_helper_initial_config() in a worker
155 Drivers that use drm_framebuffer_remove() to clean up the fbdev framebuffer can
156 probably use drm_fb_helper_fbdev_teardown().
158 Contact: Maintainer of the driver you plan to convert
160 Clean up mmap forwarding
161 ------------------------
163 A lot of drivers forward gem mmap calls to dma-buf mmap for imported buffers.
164 And also a lot of them forward dma-buf mmap to the gem mmap implementations.
165 There's drm_gem_prime_mmap() for this now, but still needs to be rolled out.
167 Contact: Daniel Vetter
169 Generic fbdev defio support
170 ---------------------------
172 The defio support code in the fbdev core has some very specific requirements,
173 which means drivers need to have a special framebuffer for fbdev. Which prevents
174 us from using the generic fbdev emulation code everywhere. The main issue is
175 that it uses some fields in struct page itself, which breaks shmem gem objects
178 Possible solution would be to write our own defio mmap code in the drm fbdev
179 emulation. It would need to fully wrap the existing mmap ops, forwarding
180 everything after it has done the write-protect/mkwrite trickery:
182 - In the drm_fbdev_fb_mmap helper, if we need defio, change the
183 default page prots to write-protected with something like this::
185 vma->vm_page_prot = pgprot_wrprotect(vma->vm_page_prot);
187 - Set the mkwrite and fsync callbacks with similar implementions to the core
188 fbdev defio stuff. These should all work on plain ptes, they don't actually
189 require a struct page. uff. These should all work on plain ptes, they don't
190 actually require a struct page.
192 - Track the dirty pages in a separate structure (bitfield with one bit per page
193 should work) to avoid clobbering struct page.
195 Might be good to also have some igt testcases for this.
197 Contact: Daniel Vetter, Noralf Tronnes
202 DRM core&drivers uses a lot of idr (integer lookup directories) for mapping
203 userspace IDs to internal objects, and in most places ID=0 means NULL and hence
204 is never used. Switching to idr_init_base() for these would make the idr more
207 Contact: Daniel Vetter
209 struct drm_gem_object_funcs
210 ---------------------------
212 GEM objects can now have a function table instead of having the callbacks on the
213 DRM driver struct. This is now the preferred way and drivers can be moved over.
215 We also need a 2nd version of the CMA define that doesn't require the
216 vmapping to be present (different hook for prime importing). Plus this needs to
217 be rolled out to all drivers using their own implementations, too.
219 Use DRM_MODESET_LOCK_ALL_* helpers instead of boilerplate
220 ---------------------------------------------------------
222 For cases where drivers are attempting to grab the modeset locks with a local
223 acquire context. Replace the boilerplate code surrounding
224 drm_modeset_lock_all_ctx() with DRM_MODESET_LOCK_ALL_BEGIN() and
225 DRM_MODESET_LOCK_ALL_END() instead.
227 This should also be done for all places where drm_modest_lock_all() is still
230 As a reference, take a look at the conversions already completed in drm core.
232 Contact: Sean Paul, respective driver maintainers
234 Rename CMA helpers to DMA helpers
235 ---------------------------------
237 CMA (standing for contiguous memory allocator) is really a bit an accident of
238 what these were used for first, a much better name would be DMA helpers. In the
239 text these should even be called coherent DMA memory helpers (so maybe CDM, but
240 no one knows what that means) since underneath they just use dma_alloc_coherent.
242 Contact: Laurent Pinchart, Daniel Vetter
244 Convert direct mode.vrefresh accesses to use drm_mode_vrefresh()
245 ----------------------------------------------------------------
247 drm_display_mode.vrefresh isn't guaranteed to be populated. As such, using it
248 is risky and has been known to cause div-by-zero bugs. Fortunately, drm core
249 has helper which will use mode.vrefresh if it's !0 and will calculate it from
250 the timings when it's 0.
252 Use simple search/replace, or (more fun) cocci to replace instances of direct
253 vrefresh access with a call to the helper. Check out
254 https://lists.freedesktop.org/archives/dri-devel/2019-January/205186.html for
257 Once all instances of vrefresh have been converted, remove vrefresh from
258 drm_display_mode to avoid future use.
262 Remove drm_display_mode.hsync
263 -----------------------------
265 We have drm_mode_hsync() to calculate this from hsync_start/end, since drivers
266 shouldn't/don't use this, remove this member to avoid any temptations to use it
267 in the future. If there is any debug code using drm_display_mode.hsync, convert
268 it to use drm_mode_hsync() instead.
275 - drm_fb_helper_restore_fbdev_mode_unlocked() should call restore_fbdev_mode()
276 not the _force variant so it can bail out if there is a master. But first
277 these igt tests need to be fixed: kms_fbcon_fbt@psr and
278 kms_fbcon_fbt@psr-suspend.
280 - The max connector argument for drm_fb_helper_init() and
281 drm_fb_helper_fbdev_setup() isn't used anymore and can be removed.
283 - The helper doesn't keep an array of connectors anymore so these can be
284 removed: drm_fb_helper_single_add_all_connectors(),
285 drm_fb_helper_add_one_connector() and drm_fb_helper_remove_one_connector().
287 connector register/unregister fixes
288 -----------------------------------
290 - For most connectors it's a no-op to call drm_connector_register/unregister
291 directly from driver code, drm_dev_register/unregister take care of this
292 already. We can remove all of them.
294 - For dp drivers it's a bit more a mess, since we need the connector to be
295 registered when calling drm_dp_aux_register. Fix this by instead calling
296 drm_dp_aux_init, and moving the actual registering into a late_register
297 callback as recommended in the kerneldoc.
302 Make panic handling work
303 ------------------------
305 This is a really varied tasks with lots of little bits and pieces:
307 * The panic path can't be tested currently, leading to constant breaking. The
308 main issue here is that panics can be triggered from hardirq contexts and
309 hence all panic related callback can run in hardirq context. It would be
310 awesome if we could test at least the fbdev helper code and driver code by
311 e.g. trigger calls through drm debugfs files. hardirq context could be
312 achieved by using an IPI to the local processor.
314 * There's a massive confusion of different panic handlers. DRM fbdev emulation
315 helpers have one, but on top of that the fbcon code itself also has one. We
316 need to make sure that they stop fighting over each another.
318 * ``drm_can_sleep()`` is a mess. It hides real bugs in normal operations and
319 isn't a full solution for panic paths. We need to make sure that it only
320 returns true if there's a panic going on for real, and fix up all the
323 * The panic handler must never sleep, which also means it can't ever
324 ``mutex_lock()``. Also it can't grab any other lock unconditionally, not
325 even spinlocks (because NMI and hardirq can panic too). We need to either
326 make sure to not call such paths, or trylock everything. Really tricky.
328 * For the above locking troubles reasons it's pretty much impossible to
329 attempt a synchronous modeset from panic handlers. The only thing we could
330 try to achive is an atomic ``set_base`` of the primary plane, and hope that
331 it shows up. Everything else probably needs to be delayed to some worker or
332 something else which happens later on. Otherwise it just kills the box
333 harder, prevent the panic from going out on e.g. netconsole.
335 * There's also proposal for a simplied DRM console instead of the full-blown
336 fbcon and DRM fbdev emulation. Any kind of panic handling tricks should
337 obviously work for both console, in case we ever get kmslog merged.
339 Contact: Daniel Vetter
341 Clean up the debugfs support
342 ----------------------------
344 There's a bunch of issues with it:
346 - The drm_info_list ->show() function doesn't even bother to cast to the drm
347 structure for you. This is lazy.
349 - We probably want to have some support for debugfs files on crtc/connectors and
350 maybe other kms objects directly in core. There's even drm_print support in
351 the funcs for these objects to dump kms state, so it's all there. And then the
352 ->show() functions should obviously give you a pointer to the right object.
354 - The drm_info_list stuff is centered on drm_minor instead of drm_device. For
355 anything we want to print drm_device (or maybe drm_file) is the right thing.
357 - The drm_driver->debugfs_init hooks we have is just an artifact of the old
358 midlayered load sequence. DRM debugfs should work more like sysfs, where you
359 can create properties/files for an object anytime you want, and the core
360 takes care of publishing/unpuplishing all the files at register/unregister
361 time. Drivers shouldn't need to worry about these technicalities, and fixing
362 this (together with the drm_minor->drm_device move) would allow us to remove
365 - Drop the return code and error checking from all debugfs functions. Greg KH is
366 working on this already.
368 Contact: Daniel Vetter
373 Some of these date from the very introduction of KMS in 2008 ...
375 - Make ->funcs and ->helper_private vtables optional. There's a bunch of empty
376 function tables in drivers, but before we can remove them we need to make sure
377 that all the users in helpers and drivers do correctly check for a NULL
380 - Cleanup up the various ->destroy callbacks. A lot of them just wrapt the
381 drm_*_cleanup implementations and can be removed. Some tack a kfree() at the
382 end, for which we could add drm_*_cleanup_kfree(). And then there's the (for
383 historical reasons) misnamed drm_primary_helper_destroy() function.
388 Enable trinity for DRM
389 ----------------------
391 And fix up the fallout. Should be really interesting ...
393 Make KMS tests in i-g-t generic
394 -------------------------------
396 The i915 driver team maintains an extensive testsuite for the i915 DRM driver,
397 including tons of testcases for corner-cases in the modesetting API. It would
398 be awesome if those tests (at least the ones not relying on Intel-specific GEM
399 features) could be made to run on any KMS driver.
401 Basic work to run i-g-t tests on non-i915 is done, what's now missing is mass-
402 converting things over. For modeset tests we also first need a bit of
403 infrastructure to use dumb buffers for untiled buffers, to be able to run all
404 the non-i915 specific modeset tests.
406 Extend virtual test driver (VKMS)
407 ---------------------------------
409 See the documentation of :ref:`VKMS <vkms>` for more details. This is an ideal
410 internship task, since it only requires a virtual machine and can be sized to
411 fit the available time.
413 Contact: Daniel Vetter
415 Backlight Refactoring
416 ---------------------
418 Backlight drivers have a triple enable/disable state, which is a bit overkill.
421 1. Roll out backlight_enable() and backlight_disable() helpers everywhere. This
423 2. In all, only look at one of the three status bits set by the above helpers.
424 3. Remove the other two status bits.
426 Contact: Daniel Vetter
431 AMD DC Display Driver
432 ---------------------
434 AMD DC is the display driver for AMD devices starting with Vega. There has been
435 a bunch of progress cleaning it up but there's still plenty of work to be done.
437 See drivers/gpu/drm/amd/display/TODO for tasks.
439 Contact: Harry Wentland, Alex Deucher
444 There is support in place now for writing internal DRM clients making it
445 possible to pick up the bootsplash work that was rejected because it was written
448 - [v6,8/8] drm/client: Hack: Add bootsplash example
449 https://patchwork.freedesktop.org/patch/306579/
451 - [RFC PATCH v2 00/13] Kernel based bootsplash
452 https://lkml.org/lkml/2017/12/13/764
454 Contact: Sam Ravnborg
459 Convert fbdev drivers to DRM
460 ----------------------------
462 There are plenty of fbdev drivers for older hardware. Some hwardware has
463 become obsolete, but some still provides good(-enough) framebuffers. The
464 drivers that are still useful should be converted to DRM and afterwards
467 Very simple fbdev drivers can best be converted by starting with a new
468 DRM driver. Simple KMS helpers and SHMEM should be able to handle any
469 existing hardware. The new driver's call-back functions are filled from
472 More complex fbdev drivers can be refactored step-by-step into a DRM
473 driver with the help of the DRM fbconv helpers. [1] These helpers provide
474 the transition layer between the DRM core infrastructure and the fbdev
475 driver interface. Create a new DRM driver on top of the fbconv helpers,
476 copy over the fbdev driver, and hook it up to the DRM code. Examples for
477 several fbdev drivers are available at [1] and a tutorial of this process
478 available at [2]. The result is a primitive DRM driver that can run X11
481 - [1] https://gitlab.freedesktop.org/tzimmermann/linux/tree/fbconv
482 - [2] https://gitlab.freedesktop.org/tzimmermann/linux/blob/fbconv/drivers/gpu/drm/drm_fbconv_helper.c
484 Contact: Thomas Zimmermann <tzimmermann@suse.de>