summaryrefslogtreecommitdiff
path: root/include/drm/drmP.h
diff options
context:
space:
mode:
authorDave Airlie <airlied@redhat.com>2017-01-09 09:55:57 +1000
committerDave Airlie <airlied@redhat.com>2017-01-09 09:55:57 +1000
commit3806a271bf4be375f304e492148edb2507181158 (patch)
treecd1b2410e8a98e63e44c4cee058b2beeb1924cf4 /include/drm/drmP.h
parenta121103c922847ba5010819a3f250f1f7fc84ab8 (diff)
parent9b8b75de4cb379187c481a5332a12429f31c0887 (diff)
Merge tag 'drm-misc-next-2016-12-30' of git://anongit.freedesktop.org/git/drm-misc into drm-next
First -misc pull for 4.11: - drm_mm rework + lots of selftests (Chris Wilson) - new connector_list locking+iterators - plenty of kerneldoc updates - format handling rework from Ville - atomic helper changes from Maarten for better plane corner-case handling in drivers, plus the i915 legacy cursor patch that needs this - bridge cleanup from Laurent - plus plenty of small stuff all over - also contains a merge of the 4.10 docs tree so that we could apply the dma-buf kerneldoc patches It's a lot more than usual, but due to the merge window blackout it also covers about 4 weeks, so all in line again on a per-week basis. The more annoying part with no pull request for 4 weeks is managing cross-tree work. The -intel pull request I'll follow up with does conflict quite a bit with -misc here. Longer-term (if drm-misc keeps growing) a drm-next-queued to accept pull request for the next merge window during this time might be useful. I'd also like to backmerge -rc2+this into drm-intel next week, we have quite a pile of patches waiting for the stuff in here. * tag 'drm-misc-next-2016-12-30' of git://anongit.freedesktop.org/git/drm-misc: (126 commits) drm: Add kerneldoc markup for new @scan parameters in drm_mm drm/mm: Document locking rules drm: Use drm_mm_insert_node_in_range_generic() for everyone drm: Apply range restriction after color adjustment when allocation drm: Wrap drm_mm_node.hole_follows drm: Apply tight eviction scanning to color_adjust drm: Simplify drm_mm scan-list manipulation drm: Optimise power-of-two alignments in drm_mm_scan_add_block() drm: Compute tight evictions for drm_mm_scan drm: Fix application of color vs range restriction when scanning drm_mm drm: Unconditionally do the range check in drm_mm_scan_add_block() drm: Rename prev_node to hole in drm_mm_scan_add_block() drm: Fix O= out-of-tree builds for selftests drm: Extract struct drm_mm_scan from struct drm_mm drm: Add asserts to catch overflow in drm_mm_init() and drm_mm_init_scan() drm: Simplify drm_mm_clean() drm: Detect overflow in drm_mm_reserve_node() drm: Fix kerneldoc for drm_mm_scan_remove_block() drm: Promote drm_mm alignment to u64 drm: kselftest for drm_mm and restricted color eviction ...
Diffstat (limited to 'include/drm/drmP.h')
-rw-r--r--include/drm/drmP.h13
1 files changed, 13 insertions, 0 deletions
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 192016e2b518..6acea8cf746d 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -634,6 +634,19 @@ struct drm_device {
int switch_power_state;
};
+/**
+ * drm_drv_uses_atomic_modeset - check if the driver implements
+ * atomic_commit()
+ * @dev: DRM device
+ *
+ * This check is useful if drivers do not have DRIVER_ATOMIC set but
+ * have atomic modesetting internally implemented.
+ */
+static inline bool drm_drv_uses_atomic_modeset(struct drm_device *dev)
+{
+ return dev->mode_config.funcs->atomic_commit != NULL;
+}
+
#include <drm/drm_irq.h>
#define DRM_SWITCH_POWER_ON 0