drm: Add support for two-ended allocation, v3

Clients like i915 need to segregate cache domains within the GTT which
can lead to small amounts of fragmentation. By allocating the uncached
buffers from the bottom and the cacheable buffers from the top, we can
reduce the amount of wasted space and also optimize allocation of the
mappable portion of the GTT to only those buffers that require CPU
access through the GTT.

For other drivers, allocating small bos from one end and large ones
from the other helps improve the quality of fragmentation.

Based on drm_mm work by Chris Wilson.

v3: Changed to use a TTM placement flag
v2: Updated kerneldoc

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Christian König <deathsimple@vodafone.de>
Signed-off-by: Lauri Kasanen <cand@gmx.com>
Signed-off-by: David Airlie <airlied@redhat.com>

2 files changed, 31 insertions, 4 deletions

diff --git a/include/drm/drm_mm.h b/include/drm/drm_mm.h
index 8b6981ab3fcf..a24addfdfcec 100644
--- a/include/drm/drm_mm.h
+++ b/include/drm/drm_mm.h
@@ -47,8 +47,17 @@
 enum drm_mm_search_flags {
 	DRM_MM_SEARCH_DEFAULT =		0,
 	DRM_MM_SEARCH_BEST =		1 << 0,
+	DRM_MM_SEARCH_BELOW =		1 << 1,
 };
 
+enum drm_mm_allocator_flags {
+	DRM_MM_CREATE_DEFAULT =		0,
+	DRM_MM_CREATE_TOP =		1 << 0,
+};
+
+#define DRM_MM_BOTTOMUP DRM_MM_SEARCH_DEFAULT, DRM_MM_CREATE_DEFAULT
+#define DRM_MM_TOPDOWN DRM_MM_SEARCH_BELOW, DRM_MM_CREATE_TOP
+
 struct drm_mm_node {
 	struct list_head node_list;
 	struct list_head hole_stack;
@@ -186,6 +195,9 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)
  * Implementation Note:
  * We need to inline list_for_each_entry in order to be able to set hole_start
  * and hole_end on each iteration while keeping the macro sane.
+ *
+ * The __drm_mm_for_each_hole version is similar, but with added support for
+ * going backwards.
  */
 #define drm_mm_for_each_hole(entry, mm, hole_start, hole_end) \
 	for (entry = list_entry((mm)->hole_stack.next, struct drm_mm_node, hole_stack); \
@@ -195,6 +207,14 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)
 	     1 : 0; \
 	     entry = list_entry(entry->hole_stack.next, struct drm_mm_node, hole_stack))
 
+#define __drm_mm_for_each_hole(entry, mm, hole_start, hole_end, backwards) \
+	for (entry = list_entry((backwards) ? (mm)->hole_stack.prev : (mm)->hole_stack.next, struct drm_mm_node, hole_stack); \
+	     &entry->hole_stack != &(mm)->hole_stack ? \
+	     hole_start = drm_mm_hole_node_start(entry), \
+	     hole_end = drm_mm_hole_node_end(entry), \
+	     1 : 0; \
+	     entry = list_entry((backwards) ? entry->hole_stack.prev : entry->hole_stack.next, struct drm_mm_node, hole_stack))
+
 /*
  * Basic range manager support (drm_mm.c)
  */
@@ -205,7 +225,8 @@ int drm_mm_insert_node_generic(struct drm_mm *mm,
 			       unsigned long size,
 			       unsigned alignment,
 			       unsigned long color,
-			       enum drm_mm_search_flags flags);
+			       enum drm_mm_search_flags sflags,
+			       enum drm_mm_allocator_flags aflags);
 /**
  * drm_mm_insert_node - search for space and insert @node
  * @mm: drm_mm to allocate from
@@ -228,7 +249,8 @@ static inline int drm_mm_insert_node(struct drm_mm *mm,
 				     unsigned alignment,
 				     enum drm_mm_search_flags flags)
 {
-	return drm_mm_insert_node_generic(mm, node, size, alignment, 0, flags);
+	return drm_mm_insert_node_generic(mm, node, size, alignment, 0, flags,
+					  DRM_MM_CREATE_DEFAULT);
 }
 
 int drm_mm_insert_node_in_range_generic(struct drm_mm *mm,
@@ -238,7 +260,8 @@ int drm_mm_insert_node_in_range_generic(struct drm_mm *mm,
 					unsigned long color,
 					unsigned long start,
 					unsigned long end,
-					enum drm_mm_search_flags flags);
+					enum drm_mm_search_flags sflags,
+					enum drm_mm_allocator_flags aflags);
 /**
  * drm_mm_insert_node_in_range - ranged search for space and insert @node
  * @mm: drm_mm to allocate from
@@ -266,7 +289,8 @@ static inline int drm_mm_insert_node_in_range(struct drm_mm *mm,
 					      enum drm_mm_search_flags flags)
 {
 	return drm_mm_insert_node_in_range_generic(mm, node, size, alignment,
-						   0, start, end, flags);
+						   0, start, end, flags,
+						   DRM_MM_CREATE_DEFAULT);
 }
 
 void drm_mm_remove_node(struct drm_mm_node *node);
diff --git a/include/drm/ttm/ttm_placement.h b/include/drm/ttm/ttm_placement.h
index c84ff153a564..8ed44f9bbdfb 100644
--- a/include/drm/ttm/ttm_placement.h
+++ b/include/drm/ttm/ttm_placement.h
@@ -65,6 +65,8 @@
  * reference the buffer.
  * TTM_PL_FLAG_NO_EVICT means that the buffer may never
  * be evicted to make room for other buffers.
+ * TTM_PL_FLAG_TOPDOWN requests to be placed from the
+ * top of the memory area, instead of the bottom.
  */
 
 #define TTM_PL_FLAG_CACHED      (1 << 16)
@@ -72,6 +74,7 @@
 #define TTM_PL_FLAG_WC          (1 << 18)
 #define TTM_PL_FLAG_SHARED      (1 << 20)
 #define TTM_PL_FLAG_NO_EVICT    (1 << 21)
+#define TTM_PL_FLAG_TOPDOWN     (1 << 22)
 
 #define TTM_PL_MASK_CACHING     (TTM_PL_FLAG_CACHED | \
 				 TTM_PL_FLAG_UNCACHED | \