Compute the minimal required hole during scan and only evict those nodes
that overlap. This enables us to reduce the number of nodes we need to
evict to the bare minimum.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-31-chris@chris-wilson.co.uk
The range restriction should be applied after the color adjustment, or
else we may inadvertently apply the color adjustment to the restricted
hole (and not against its neighbours).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-30-chris@chris-wilson.co.uk
Doing the check is trivial (low cost in comparison to overall eviction)
and helps simplify the code.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-29-chris@chris-wilson.co.uk
Acknowledging that we were building up the hole was more useful to me
when reading the code, than knowing the relationship between this node
and the previous node.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-28-chris@chris-wilson.co.uk
Kbuild really doesn't like non-recursive Makefiles, but they do work
as long as you build without O=
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Fixes: 50f0033d1a ("drm: Add some kselftests for the DRM range manager (struct drm_mm)")
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Christian König <christian.koenig@amd.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1482918077-30027-1-git-send-email-daniel.vetter@ffwll.ch
The scan state occupies a large proportion of the struct drm_mm and is
rarely used and only contains temporary state. That makes it suitable to
moving to its struct and onto the stack of the callers.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
[danvet: Fix up etnaviv to compile, was missing a BUG_ON.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
A simple assert to ensure that we don't overflow start + size when
initialising the drm_mm, or its scanner.
In future, we may want to switch to tracking the value of ranges (rather
than size) so that we can cover the full u64, for example like resource
tracking.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-26-chris@chris-wilson.co.uk
Since commit ea7b1dd448 ("drm: mm: track free areas implicitly"),
to test whether there are any nodes allocated within the range manager,
we merely have to ask whether the node_list is empty.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-25-chris@chris-wilson.co.uk
Protect ourselves from a caller passing in node.start + node.size that
will overflow and trick us into reserving that node.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-24-chris@chris-wilson.co.uk
The nodes must be removed in the *reverse* order. This is correct in the
overview, but backwards in the function description. Whilst here add
Intel's copyright statement and tweak some formatting.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-23-chris@chris-wilson.co.uk
In places (e.g. i915.ko), the alignment is exported to userspace as u64
and there now exists hardware for which we can indeed utilize a u64
alignment. As such, we need to keep 64bit integers throughout when
handling alignment.
Testcase: igt/drm_mm/align64
Testcase: igt/gem_exec_alignment
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-22-chris@chris-wilson.co.uk
Check that after applying the driver's color adjustment, restricted
eviction scanning finds a suitable hole.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-20-chris@chris-wilson.co.uk
Check that after applying the driver's color adjustment, eviction
scanning finds a suitable hole.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-19-chris@chris-wilson.co.uk
Check that after applying the driver's color adjustment, fitting of the
node and its alignment are still correct.
v2: s/no_color_touching/separate_adjacent_colors/
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-18-chris@chris-wilson.co.uk
Check that if we request top-down allocation from drm_mm_insert_node()
we receive the next available hole from the top.
v2: Flip sign on conditional assert.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-17-chris@chris-wilson.co.uk
Check that we add arbitrary blocks to a restrited eviction scanner in
order to find the first minimal hole that matches our request.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-16-chris@chris-wilson.co.uk
Check that we add arbitrary blocks to the eviction scanner in order to
find the first minimal hole that matches our request.
v2: Refactor out some common eviction code for later
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-15-chris@chris-wilson.co.uk
Check that we can request alignment to any power-of-two or prime using a
plain drm_mm_node_insert(), and also handle a reasonable selection of
primes.
v2: Exercise all allocation flags
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-14-chris@chris-wilson.co.uk
Exercise drm_mm_insert_node_in_range(), check that we only allocate from
the specified range.
v2: Use all allocation flags
v3: Don't pass in invalid ranges - these will be asserted later.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-13-chris@chris-wilson.co.uk
Reuse drm_mm_insert_node() with a temporary node to exercise
drm_mm_replace_node(). We use the previous test in order to exercise the
various lists following replacement.
v2: Check that we copy across the important (user) details of the node.
The internal details (such as lists and hole tracking) we hope to detect
errors by exercise.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-12-chris@chris-wilson.co.uk
Exercise drm_mm_insert_node(), check that we can't overfill a range and
that the lists are correct after reserving/removing.
v2: Extract helpers for the repeated tests
v3: Iterate over all allocation flags
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-11-chris@chris-wilson.co.uk
Exercise drm_mm_reserve_node(), check that we can't reserve an already
occupied range and that the lists are correct after reserving/removing.
v2: Check for invalid node reservation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-10-chris@chris-wilson.co.uk
First we introduce a smattering of infrastructure for writing selftests.
The idea is that we have a test module that exercises a particular
portion of the exported API, and that module provides a set of tests
that can either be run as an ensemble via kselftest or individually via
an igt harness (in this case igt/drm_mm). To accommodate selecting
individual tests, we export a boolean parameter to control selection of
each test - that is hidden inside a bunch of reusable boilerplate macros
to keep writing the tests simple.
v2: Choose a random random_seed unless one is specified by the user.
v3: More parameters to control max_iterations and max_prime of the
tests.
Testcase: igt/drm_mm
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-7-chris@chris-wilson.co.uk
When testing, we want a random but yet reproducible order in which to
process elements. Here we create an array which is a random (using the
Tausworthe PRNG) permutation of the order in which to execute.
Note these are simple helpers intended to be merged upstream in lib/
v2: Tidier code by David Herrmann
v3: Add reminder that this code is intended to be temporary, with at
least the bulk of the prandom changes going to lib/
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-6-chris@chris-wilson.co.uk
Prime numbers are interesting for testing components that use multiplies
and divides, such as testing DRM's struct drm_mm alignment computations.
v2: Move to lib/, add selftest
v3: Fix initial constants (exclude 0/1 from being primes)
v4: More RCU markup to keep 0day/sparse happy
v5: Fix RCU unwind on module exit, add to kselftests
v6: Tidy computation of bitmap size
v7: for_each_prime_number_from()
v8: Compose small-primes using BIT() for easier verification
v9: Move rcu dance entirely into callers.
v10: Improve quote for Betrand's Postulate (aka Chebyshev's theorem)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222144514.3911-1-chris@chris-wilson.co.uk
Fairly commonly we want to inspect the node list on the struct drm_mm,
which is buried within an embedded node. Bring it to the surface with a
bit of syntatic sugar.
Note this was intended to be split from commit ad579002c8 ("drm: Add
drm_mm_for_each_node_safe()") before being applied, but my timing sucks.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-3-chris@chris-wilson.co.uk
i915 does not set DRIVER_ATOMIC by default yet but uses atomic_check and
atomic_commit. drm_object_property_get_value() does not read the correct
value of atomic properties if DRIVER_ATOMIC is not set. Checking whether
the driver uses atomic modeset is a better check instead as the property
values are tracked in the state structures.
v2: Included header
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1482396643-32456-2-git-send-email-dhinakaran.pandiyan@intel.com
This check is useful for drivers that do not have DRIVER_ATOMIC set but
have atomic modesetting internally implemented. Wrap the check into a
function since this is used in many places and as a bonus, the function
name helps to document what the check is for.
v2:
Change return type to bool (Ville)
Move the function drm_atomic.h (Daniel)
Fixed comment marker for documentation
Suggested-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
[danvet: Move back to drmP.h because include hell.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1482396643-32456-1-git-send-email-dhinakaran.pandiyan@intel.com
Do something similar to vc4, only allow updating the cursor state
in-place through a fastpath when the watermarks are unaffected. This
will allow cursor movement to be smooth, but changing cursor size or
showing/hiding cursor will still fall back so watermarks can be updated.
Only moving and changing fb is allowed.
Changes since v1:
- Set page flip to always_unused for trybot.
- Copy fence correctly, ignore plane_state->state, should be NULL.
- Check crtc_state for !active and modeset, go to slowpath if the case.
Changes since v2:
- Make error handling work correctly. (Matthew Auld)
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/a8e4cb00-5171-14e5-bbe3-dadb654ff296@linux.intel.com
There's 2 reasons for doing a vblank wait:
- To fulfill uabi expectations, but the legacy ioctls are ill-defined
enough that we really only need this when we do send out an event.
- To make sure we don't tear down mappings before the scanout engine
stops accessing it.
The later is problematic with the current code since e.g. rotation
might need a different mapping than normal orientation. And rotation
is a plane property, and not on the fb. Hence we need to remove this
optimization.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
[danvet: Completely new commit message.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1481204729-9058-5-git-send-email-maarten.lankhorst@linux.intel.com
Stop relying on a per crtc_state last_vblank_count, we shouldn't touch
crtc_state after commit. Move it to atomic_state->crtcs.
Also stop re-using new_crtc_state->enable, we can now simply set a
bitmask with crtc_crtc_mask.
Changes since v1:
- Keep last_vblank_count in __drm_crtc_state.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/8e4759a4-24d3-3f80-bd1a-1e7a9c83b612@linux.intel.com
Atomic drivers may set properties like rotation on the same fb, which
may require a call to prepare_fb even when framebuffer stays identical.
Instead of handling all the special cases in the core, let the driver
decide when prepare_fb and cleanup_fb are noops.
This is a revert of:
commit fcc60b413d
Author: Keith Packard <keithp@keithp.com>
Date: Sat Jun 4 01:16:22 2016 -0700
drm: Don't prepare or cleanup unchanging frame buffers [v3]
The original commit mentions that this prevents waiting in i915 on all
previous rendering during cursor updates, but there are better ways to
fix this.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/6d82f9b6-9d16-91d1-d176-4a37b09afc44@linux.intel.com
Prepare to mark sensitive kernel structures for randomization by making
sure they're using designated initializers. These were identified during
allyesconfig builds of x86, arm, and arm64, with most initializer fixes
extracted from grsecurity.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161217010442.GA140619@beast
Prepare to mark sensitive kernel structures for randomization by making
sure they're using designated initializers. These were identified during
allyesconfig builds of x86, arm, and arm64, with most initializer fixes
extracted from grsecurity.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161217010402.GA140546@beast
Prepare to mark sensitive kernel structures for randomization by making
sure they're using designated initializers. These were identified during
allyesconfig builds of x86, arm, and arm64, with most initializer fixes
extracted from grsecurity.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161217010011.GA140300@beast
Prepare to mark sensitive kernel structures for randomization by making
sure they're using designated initializers. These were identified during
allyesconfig builds of x86, arm, and arm64, with most initializer fixes
extracted from grsecurity.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161217005929.GA140260@beast
- Modeset state needs mode_config->connection mutex, that covers
figuring out the encoder, and reading properties (since in the
atomic case those need to look at connector->state).
- Don't hold any locks for stuff that's invariant (i.e. possible
connectors).
- Same for connector lookup and unref, those don't need any locks.
- And finally the probe stuff is only protected by mode_config->mutex.
While at it updated the kerneldoc for these fields in drm_connector
and add docs explaining what's protected by which locks.
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161213230814.19598-10-daniel.vetter@ffwll.ch
If we're unlucky then the registration from a hotplugged connector
might race with the final registration step on driver load. And since
MST topology discover is asynchronous that's even somewhat likely.
v2: Also update the kerneldoc for @registered!
v3: Review from Chris:
- Improve kerneldoc for late_register/early_unregister callbacks.
- Use mutex_destroy.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161218133545.2106-1-daniel.vetter@ffwll.ch
Only static connectors should be left at this point, and we should be
able to clean them out by simply dropping that last reference still
around from drm_connector_init.
If that leaves anything behind then we have a driver bug.
Doing the final cleanup this way also allows us to use
drm_connector_iter, removing the very last place where we walk
connector_list explicitly in drm core&helpers.
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161213230814.19598-8-daniel.vetter@ffwll.ch
Mostly nothing special (except making sure that really all error paths
and friends call iter_put).
v2: Don't forget the raw connector_list walking in
drm_helper_move_panel_connectors_to_head. That one unfortunately can't
be converted to the iterator helpers, but since it's just some list
splicing best to just wrap the entire thing up in one critical
section.
v3: Bail out after iter_put (Harry).
Cc: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161215155843.13408-1-daniel.vetter@ffwll.ch
The requirements for connector_list locking are a bit tricky:
- We need to be able to jump over zombie conectors (i.e. with refcount
== 0, but not yet removed from the list). If instead we require that
there's no zombies on the list then the final kref_put must happen
under the list protection lock, which means that locking context
leaks all over the place. Not pretty - better to deal with zombies
and wrap the locking just around the list_del in the destructor.
- When we walk the list we must _not_ hold the connector list lock. We
walk the connector list at an absolutely massive amounts of places,
if all those places can't ever call drm_connector_unreference the
code would get unecessarily complicated.
- connector_list needs it own lock, again too many places that walk it
that we could reuse e.g. mode_config.mutex without resulting in
inversions.
- Lots of code uses these loops to look-up a connector, i.e. they want
to be able to call drm_connector_reference. But on the other hand we
want connectors to stay on that list until they're dead (i.e.
connector_list can't hold a full reference), which means despite the
"can't hold lock for the loop body" rule we need to make sure a
connector doesn't suddenly become a zombie.
At first Dave&I discussed various horror-show approaches using srcu,
but turns out it's fairly easy:
- For the loop body we always hold an additional reference to the
current connector. That means it can't zombify, and it also means
it'll stay on the list, which means we can use it as our iterator to
find the next connector.
- When we try to find the next connector we only have to jump over
zombies. To make sure we don't chase bad pointers that entire loop
is protected with the new connect_list_lock spinlock. And because we
know that we're starting out with a non-zombie (need to drop our
reference for the old connector only after we have our new one),
we're guranteed to still be on the connector_list and either find
the next non-zombie or complete the iteration.
- Only downside is that we need to make sure that the temporary
reference for the loop body doesn't leak. iter_get/put() functions +
lockdep make sure that's the case.
- To avoid a flag day the new iterator macro has an _iter postfix. We
can rename it back once all the users of the unsafe version are gone
(there's about 100 list walkers for the connector_list).
For now this patch only converts all the list walking in the core,
leaving helpers and drivers for later patches. The nice thing is that
we can now finally remove 2 FIXME comments from the
register/unregister functions.
v2:
- use irqsafe spinlocks, so that we can use this in drm_state_dump
too.
- nuke drm_modeset_lock_all from drm_connector_init, now entirely
cargo-culted nonsense.
v3:
- do {} while (!kref_get_unless_zero), makes for a tidier loop (Dave).
- pretty kerneldoc
- add EXPORT_SYMBOL, helpers&drivers are supposed to use this.
v4: Change lockdep annotations to only check whether we release the
iter fake lock again (i.e. make sure that iter_put is called), but
not check any locking dependecies itself. That seams to require a
recursive read lock in trylock mode.
Cc: Dave Airlie <airlied@gmail.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161213230814.19598-6-daniel.vetter@ffwll.ch
This is single-threaded setup code, no need for locks. And anyway,
all properties need to be set up before the driver is registered
anyway, they can't be hot-added.
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161213230814.19598-5-daniel.vetter@ffwll.ch
This is not driver interface stuff.
Fixes: 6559c901cb ("drm/atomic: add debugfs file to dump out atomic state")
Cc: Rob Clark <robdclark@gmail.com>
Cc: Sean Paul <seanpaul@chromium.org>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161213230814.19598-3-daniel.vetter@ffwll.ch
