Loading...
1/**************************************************************************
2 *
3 * Copyright 2006 Tungsten Graphics, Inc., Bismarck, ND., USA.
4 * Copyright 2016 Intel Corporation
5 * All Rights Reserved.
6 *
7 * Permission is hereby granted, free of charge, to any person obtaining a
8 * copy of this software and associated documentation files (the
9 * "Software"), to deal in the Software without restriction, including
10 * without limitation the rights to use, copy, modify, merge, publish,
11 * distribute, sub license, and/or sell copies of the Software, and to
12 * permit persons to whom the Software is furnished to do so, subject to
13 * the following conditions:
14 *
15 * The above copyright notice and this permission notice (including the
16 * next paragraph) shall be included in all copies or substantial portions
17 * of the Software.
18 *
19 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
20 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
21 * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
22 * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
23 * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
24 * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
25 * USE OR OTHER DEALINGS IN THE SOFTWARE.
26 *
27 *
28 **************************************************************************/
29
30/*
31 * Generic simple memory manager implementation. Intended to be used as a base
32 * class implementation for more advanced memory managers.
33 *
34 * Note that the algorithm used is quite simple and there might be substantial
35 * performance gains if a smarter free list is implemented. Currently it is
36 * just an unordered stack of free regions. This could easily be improved if
37 * an RB-tree is used instead. At least if we expect heavy fragmentation.
38 *
39 * Aligned allocations can also see improvement.
40 *
41 * Authors:
42 * Thomas Hellström <thomas-at-tungstengraphics-dot-com>
43 */
44
45#include <linux/export.h>
46#include <linux/interval_tree_generic.h>
47#include <linux/seq_file.h>
48#include <linux/slab.h>
49#include <linux/stacktrace.h>
50
51#include <drm/drm_mm.h>
52
53/**
54 * DOC: Overview
55 *
56 * drm_mm provides a simple range allocator. The drivers are free to use the
57 * resource allocator from the linux core if it suits them, the upside of drm_mm
58 * is that it's in the DRM core. Which means that it's easier to extend for
59 * some of the crazier special purpose needs of gpus.
60 *
61 * The main data struct is &drm_mm, allocations are tracked in &drm_mm_node.
62 * Drivers are free to embed either of them into their own suitable
63 * datastructures. drm_mm itself will not do any memory allocations of its own,
64 * so if drivers choose not to embed nodes they need to still allocate them
65 * themselves.
66 *
67 * The range allocator also supports reservation of preallocated blocks. This is
68 * useful for taking over initial mode setting configurations from the firmware,
69 * where an object needs to be created which exactly matches the firmware's
70 * scanout target. As long as the range is still free it can be inserted anytime
71 * after the allocator is initialized, which helps with avoiding looped
72 * dependencies in the driver load sequence.
73 *
74 * drm_mm maintains a stack of most recently freed holes, which of all
75 * simplistic datastructures seems to be a fairly decent approach to clustering
76 * allocations and avoiding too much fragmentation. This means free space
77 * searches are O(num_holes). Given that all the fancy features drm_mm supports
78 * something better would be fairly complex and since gfx thrashing is a fairly
79 * steep cliff not a real concern. Removing a node again is O(1).
80 *
81 * drm_mm supports a few features: Alignment and range restrictions can be
82 * supplied. Furthermore every &drm_mm_node has a color value (which is just an
83 * opaque unsigned long) which in conjunction with a driver callback can be used
84 * to implement sophisticated placement restrictions. The i915 DRM driver uses
85 * this to implement guard pages between incompatible caching domains in the
86 * graphics TT.
87 *
88 * Two behaviors are supported for searching and allocating: bottom-up and
89 * top-down. The default is bottom-up. Top-down allocation can be used if the
90 * memory area has different restrictions, or just to reduce fragmentation.
91 *
92 * Finally iteration helpers to walk all nodes and all holes are provided as are
93 * some basic allocator dumpers for debugging.
94 *
95 * Note that this range allocator is not thread-safe, drivers need to protect
96 * modifications with their own locking. The idea behind this is that for a full
97 * memory manager additional data needs to be protected anyway, hence internal
98 * locking would be fully redundant.
99 */
100
101#ifdef CONFIG_DRM_DEBUG_MM
102#include <linux/stackdepot.h>
103
104#define STACKDEPTH 32
105#define BUFSZ 4096
106
107static noinline void save_stack(struct drm_mm_node *node)
108{
109 unsigned long entries[STACKDEPTH];
110 unsigned int n;
111
112 n = stack_trace_save(entries, ARRAY_SIZE(entries), 1);
113
114 /* May be called under spinlock, so avoid sleeping */
115 node->stack = stack_depot_save(entries, n, GFP_NOWAIT);
116}
117
118static void show_leaks(struct drm_mm *mm)
119{
120 struct drm_mm_node *node;
121 char *buf;
122
123 buf = kmalloc(BUFSZ, GFP_KERNEL);
124 if (!buf)
125 return;
126
127 list_for_each_entry(node, drm_mm_nodes(mm), node_list) {
128 if (!node->stack) {
129 DRM_ERROR("node [%08llx + %08llx]: unknown owner\n",
130 node->start, node->size);
131 continue;
132 }
133
134 stack_depot_snprint(node->stack, buf, BUFSZ, 0);
135 DRM_ERROR("node [%08llx + %08llx]: inserted at\n%s",
136 node->start, node->size, buf);
137 }
138
139 kfree(buf);
140}
141
142#undef STACKDEPTH
143#undef BUFSZ
144#else
145static void save_stack(struct drm_mm_node *node) { }
146static void show_leaks(struct drm_mm *mm) { }
147#endif
148
149#define START(node) ((node)->start)
150#define LAST(node) ((node)->start + (node)->size - 1)
151
152INTERVAL_TREE_DEFINE(struct drm_mm_node, rb,
153 u64, __subtree_last,
154 START, LAST, static inline __maybe_unused, drm_mm_interval_tree)
155
156struct drm_mm_node *
157__drm_mm_interval_first(const struct drm_mm *mm, u64 start, u64 last)
158{
159 return drm_mm_interval_tree_iter_first((struct rb_root_cached *)&mm->interval_tree,
160 start, last) ?: (struct drm_mm_node *)&mm->head_node;
161}
162EXPORT_SYMBOL(__drm_mm_interval_first);
163
164static void drm_mm_interval_tree_add_node(struct drm_mm_node *hole_node,
165 struct drm_mm_node *node)
166{
167 struct drm_mm *mm = hole_node->mm;
168 struct rb_node **link, *rb;
169 struct drm_mm_node *parent;
170 bool leftmost;
171
172 node->__subtree_last = LAST(node);
173
174 if (drm_mm_node_allocated(hole_node)) {
175 rb = &hole_node->rb;
176 while (rb) {
177 parent = rb_entry(rb, struct drm_mm_node, rb);
178 if (parent->__subtree_last >= node->__subtree_last)
179 break;
180
181 parent->__subtree_last = node->__subtree_last;
182 rb = rb_parent(rb);
183 }
184
185 rb = &hole_node->rb;
186 link = &hole_node->rb.rb_right;
187 leftmost = false;
188 } else {
189 rb = NULL;
190 link = &mm->interval_tree.rb_root.rb_node;
191 leftmost = true;
192 }
193
194 while (*link) {
195 rb = *link;
196 parent = rb_entry(rb, struct drm_mm_node, rb);
197 if (parent->__subtree_last < node->__subtree_last)
198 parent->__subtree_last = node->__subtree_last;
199 if (node->start < parent->start) {
200 link = &parent->rb.rb_left;
201 } else {
202 link = &parent->rb.rb_right;
203 leftmost = false;
204 }
205 }
206
207 rb_link_node(&node->rb, rb, link);
208 rb_insert_augmented_cached(&node->rb, &mm->interval_tree, leftmost,
209 &drm_mm_interval_tree_augment);
210}
211
212#define HOLE_SIZE(NODE) ((NODE)->hole_size)
213#define HOLE_ADDR(NODE) (__drm_mm_hole_node_start(NODE))
214
215static u64 rb_to_hole_size(struct rb_node *rb)
216{
217 return rb_entry(rb, struct drm_mm_node, rb_hole_size)->hole_size;
218}
219
220static void insert_hole_size(struct rb_root_cached *root,
221 struct drm_mm_node *node)
222{
223 struct rb_node **link = &root->rb_root.rb_node, *rb = NULL;
224 u64 x = node->hole_size;
225 bool first = true;
226
227 while (*link) {
228 rb = *link;
229 if (x > rb_to_hole_size(rb)) {
230 link = &rb->rb_left;
231 } else {
232 link = &rb->rb_right;
233 first = false;
234 }
235 }
236
237 rb_link_node(&node->rb_hole_size, rb, link);
238 rb_insert_color_cached(&node->rb_hole_size, root, first);
239}
240
241RB_DECLARE_CALLBACKS_MAX(static, augment_callbacks,
242 struct drm_mm_node, rb_hole_addr,
243 u64, subtree_max_hole, HOLE_SIZE)
244
245static void insert_hole_addr(struct rb_root *root, struct drm_mm_node *node)
246{
247 struct rb_node **link = &root->rb_node, *rb_parent = NULL;
248 u64 start = HOLE_ADDR(node), subtree_max_hole = node->subtree_max_hole;
249 struct drm_mm_node *parent;
250
251 while (*link) {
252 rb_parent = *link;
253 parent = rb_entry(rb_parent, struct drm_mm_node, rb_hole_addr);
254 if (parent->subtree_max_hole < subtree_max_hole)
255 parent->subtree_max_hole = subtree_max_hole;
256 if (start < HOLE_ADDR(parent))
257 link = &parent->rb_hole_addr.rb_left;
258 else
259 link = &parent->rb_hole_addr.rb_right;
260 }
261
262 rb_link_node(&node->rb_hole_addr, rb_parent, link);
263 rb_insert_augmented(&node->rb_hole_addr, root, &augment_callbacks);
264}
265
266static void add_hole(struct drm_mm_node *node)
267{
268 struct drm_mm *mm = node->mm;
269
270 node->hole_size =
271 __drm_mm_hole_node_end(node) - __drm_mm_hole_node_start(node);
272 node->subtree_max_hole = node->hole_size;
273 DRM_MM_BUG_ON(!drm_mm_hole_follows(node));
274
275 insert_hole_size(&mm->holes_size, node);
276 insert_hole_addr(&mm->holes_addr, node);
277
278 list_add(&node->hole_stack, &mm->hole_stack);
279}
280
281static void rm_hole(struct drm_mm_node *node)
282{
283 DRM_MM_BUG_ON(!drm_mm_hole_follows(node));
284
285 list_del(&node->hole_stack);
286 rb_erase_cached(&node->rb_hole_size, &node->mm->holes_size);
287 rb_erase_augmented(&node->rb_hole_addr, &node->mm->holes_addr,
288 &augment_callbacks);
289 node->hole_size = 0;
290 node->subtree_max_hole = 0;
291
292 DRM_MM_BUG_ON(drm_mm_hole_follows(node));
293}
294
295static inline struct drm_mm_node *rb_hole_size_to_node(struct rb_node *rb)
296{
297 return rb_entry_safe(rb, struct drm_mm_node, rb_hole_size);
298}
299
300static inline struct drm_mm_node *rb_hole_addr_to_node(struct rb_node *rb)
301{
302 return rb_entry_safe(rb, struct drm_mm_node, rb_hole_addr);
303}
304
305static struct drm_mm_node *best_hole(struct drm_mm *mm, u64 size)
306{
307 struct rb_node *rb = mm->holes_size.rb_root.rb_node;
308 struct drm_mm_node *best = NULL;
309
310 do {
311 struct drm_mm_node *node =
312 rb_entry(rb, struct drm_mm_node, rb_hole_size);
313
314 if (size <= node->hole_size) {
315 best = node;
316 rb = rb->rb_right;
317 } else {
318 rb = rb->rb_left;
319 }
320 } while (rb);
321
322 return best;
323}
324
325static bool usable_hole_addr(struct rb_node *rb, u64 size)
326{
327 return rb && rb_hole_addr_to_node(rb)->subtree_max_hole >= size;
328}
329
330static struct drm_mm_node *find_hole_addr(struct drm_mm *mm, u64 addr, u64 size)
331{
332 struct rb_node *rb = mm->holes_addr.rb_node;
333 struct drm_mm_node *node = NULL;
334
335 while (rb) {
336 u64 hole_start;
337
338 if (!usable_hole_addr(rb, size))
339 break;
340
341 node = rb_hole_addr_to_node(rb);
342 hole_start = __drm_mm_hole_node_start(node);
343
344 if (addr < hole_start)
345 rb = node->rb_hole_addr.rb_left;
346 else if (addr > hole_start + node->hole_size)
347 rb = node->rb_hole_addr.rb_right;
348 else
349 break;
350 }
351
352 return node;
353}
354
355static struct drm_mm_node *
356first_hole(struct drm_mm *mm,
357 u64 start, u64 end, u64 size,
358 enum drm_mm_insert_mode mode)
359{
360 switch (mode) {
361 default:
362 case DRM_MM_INSERT_BEST:
363 return best_hole(mm, size);
364
365 case DRM_MM_INSERT_LOW:
366 return find_hole_addr(mm, start, size);
367
368 case DRM_MM_INSERT_HIGH:
369 return find_hole_addr(mm, end, size);
370
371 case DRM_MM_INSERT_EVICT:
372 return list_first_entry_or_null(&mm->hole_stack,
373 struct drm_mm_node,
374 hole_stack);
375 }
376}
377
378/**
379 * DECLARE_NEXT_HOLE_ADDR - macro to declare next hole functions
380 * @name: name of function to declare
381 * @first: first rb member to traverse (either rb_left or rb_right).
382 * @last: last rb member to traverse (either rb_right or rb_left).
383 *
384 * This macro declares a function to return the next hole of the addr rb tree.
385 * While traversing the tree we take the searched size into account and only
386 * visit branches with potential big enough holes.
387 */
388
389#define DECLARE_NEXT_HOLE_ADDR(name, first, last) \
390static struct drm_mm_node *name(struct drm_mm_node *entry, u64 size) \
391{ \
392 struct rb_node *parent, *node = &entry->rb_hole_addr; \
393 \
394 if (!entry || RB_EMPTY_NODE(node)) \
395 return NULL; \
396 \
397 if (usable_hole_addr(node->first, size)) { \
398 node = node->first; \
399 while (usable_hole_addr(node->last, size)) \
400 node = node->last; \
401 return rb_hole_addr_to_node(node); \
402 } \
403 \
404 while ((parent = rb_parent(node)) && node == parent->first) \
405 node = parent; \
406 \
407 return rb_hole_addr_to_node(parent); \
408}
409
410DECLARE_NEXT_HOLE_ADDR(next_hole_high_addr, rb_left, rb_right)
411DECLARE_NEXT_HOLE_ADDR(next_hole_low_addr, rb_right, rb_left)
412
413static struct drm_mm_node *
414next_hole(struct drm_mm *mm,
415 struct drm_mm_node *node,
416 u64 size,
417 enum drm_mm_insert_mode mode)
418{
419 switch (mode) {
420 default:
421 case DRM_MM_INSERT_BEST:
422 return rb_hole_size_to_node(rb_prev(&node->rb_hole_size));
423
424 case DRM_MM_INSERT_LOW:
425 return next_hole_low_addr(node, size);
426
427 case DRM_MM_INSERT_HIGH:
428 return next_hole_high_addr(node, size);
429
430 case DRM_MM_INSERT_EVICT:
431 node = list_next_entry(node, hole_stack);
432 return &node->hole_stack == &mm->hole_stack ? NULL : node;
433 }
434}
435
436/**
437 * drm_mm_reserve_node - insert an pre-initialized node
438 * @mm: drm_mm allocator to insert @node into
439 * @node: drm_mm_node to insert
440 *
441 * This functions inserts an already set-up &drm_mm_node into the allocator,
442 * meaning that start, size and color must be set by the caller. All other
443 * fields must be cleared to 0. This is useful to initialize the allocator with
444 * preallocated objects which must be set-up before the range allocator can be
445 * set-up, e.g. when taking over a firmware framebuffer.
446 *
447 * Returns:
448 * 0 on success, -ENOSPC if there's no hole where @node is.
449 */
450int drm_mm_reserve_node(struct drm_mm *mm, struct drm_mm_node *node)
451{
452 struct drm_mm_node *hole;
453 u64 hole_start, hole_end;
454 u64 adj_start, adj_end;
455 u64 end;
456
457 end = node->start + node->size;
458 if (unlikely(end <= node->start))
459 return -ENOSPC;
460
461 /* Find the relevant hole to add our node to */
462 hole = find_hole_addr(mm, node->start, 0);
463 if (!hole)
464 return -ENOSPC;
465
466 adj_start = hole_start = __drm_mm_hole_node_start(hole);
467 adj_end = hole_end = hole_start + hole->hole_size;
468
469 if (mm->color_adjust)
470 mm->color_adjust(hole, node->color, &adj_start, &adj_end);
471
472 if (adj_start > node->start || adj_end < end)
473 return -ENOSPC;
474
475 node->mm = mm;
476
477 __set_bit(DRM_MM_NODE_ALLOCATED_BIT, &node->flags);
478 list_add(&node->node_list, &hole->node_list);
479 drm_mm_interval_tree_add_node(hole, node);
480 node->hole_size = 0;
481
482 rm_hole(hole);
483 if (node->start > hole_start)
484 add_hole(hole);
485 if (end < hole_end)
486 add_hole(node);
487
488 save_stack(node);
489 return 0;
490}
491EXPORT_SYMBOL(drm_mm_reserve_node);
492
493static u64 rb_to_hole_size_or_zero(struct rb_node *rb)
494{
495 return rb ? rb_to_hole_size(rb) : 0;
496}
497
498/**
499 * drm_mm_insert_node_in_range - ranged search for space and insert @node
500 * @mm: drm_mm to allocate from
501 * @node: preallocate node to insert
502 * @size: size of the allocation
503 * @alignment: alignment of the allocation
504 * @color: opaque tag value to use for this node
505 * @range_start: start of the allowed range for this node
506 * @range_end: end of the allowed range for this node
507 * @mode: fine-tune the allocation search and placement
508 *
509 * The preallocated @node must be cleared to 0.
510 *
511 * Returns:
512 * 0 on success, -ENOSPC if there's no suitable hole.
513 */
514int drm_mm_insert_node_in_range(struct drm_mm * const mm,
515 struct drm_mm_node * const node,
516 u64 size, u64 alignment,
517 unsigned long color,
518 u64 range_start, u64 range_end,
519 enum drm_mm_insert_mode mode)
520{
521 struct drm_mm_node *hole;
522 u64 remainder_mask;
523 bool once;
524
525 DRM_MM_BUG_ON(range_start > range_end);
526
527 if (unlikely(size == 0 || range_end - range_start < size))
528 return -ENOSPC;
529
530 if (rb_to_hole_size_or_zero(rb_first_cached(&mm->holes_size)) < size)
531 return -ENOSPC;
532
533 if (alignment <= 1)
534 alignment = 0;
535
536 once = mode & DRM_MM_INSERT_ONCE;
537 mode &= ~DRM_MM_INSERT_ONCE;
538
539 remainder_mask = is_power_of_2(alignment) ? alignment - 1 : 0;
540 for (hole = first_hole(mm, range_start, range_end, size, mode);
541 hole;
542 hole = once ? NULL : next_hole(mm, hole, size, mode)) {
543 u64 hole_start = __drm_mm_hole_node_start(hole);
544 u64 hole_end = hole_start + hole->hole_size;
545 u64 adj_start, adj_end;
546 u64 col_start, col_end;
547
548 if (mode == DRM_MM_INSERT_LOW && hole_start >= range_end)
549 break;
550
551 if (mode == DRM_MM_INSERT_HIGH && hole_end <= range_start)
552 break;
553
554 col_start = hole_start;
555 col_end = hole_end;
556 if (mm->color_adjust)
557 mm->color_adjust(hole, color, &col_start, &col_end);
558
559 adj_start = max(col_start, range_start);
560 adj_end = min(col_end, range_end);
561
562 if (adj_end <= adj_start || adj_end - adj_start < size)
563 continue;
564
565 if (mode == DRM_MM_INSERT_HIGH)
566 adj_start = adj_end - size;
567
568 if (alignment) {
569 u64 rem;
570
571 if (likely(remainder_mask))
572 rem = adj_start & remainder_mask;
573 else
574 div64_u64_rem(adj_start, alignment, &rem);
575 if (rem) {
576 adj_start -= rem;
577 if (mode != DRM_MM_INSERT_HIGH)
578 adj_start += alignment;
579
580 if (adj_start < max(col_start, range_start) ||
581 min(col_end, range_end) - adj_start < size)
582 continue;
583
584 if (adj_end <= adj_start ||
585 adj_end - adj_start < size)
586 continue;
587 }
588 }
589
590 node->mm = mm;
591 node->size = size;
592 node->start = adj_start;
593 node->color = color;
594 node->hole_size = 0;
595
596 __set_bit(DRM_MM_NODE_ALLOCATED_BIT, &node->flags);
597 list_add(&node->node_list, &hole->node_list);
598 drm_mm_interval_tree_add_node(hole, node);
599
600 rm_hole(hole);
601 if (adj_start > hole_start)
602 add_hole(hole);
603 if (adj_start + size < hole_end)
604 add_hole(node);
605
606 save_stack(node);
607 return 0;
608 }
609
610 return -ENOSPC;
611}
612EXPORT_SYMBOL(drm_mm_insert_node_in_range);
613
614static inline __maybe_unused bool drm_mm_node_scanned_block(const struct drm_mm_node *node)
615{
616 return test_bit(DRM_MM_NODE_SCANNED_BIT, &node->flags);
617}
618
619/**
620 * drm_mm_remove_node - Remove a memory node from the allocator.
621 * @node: drm_mm_node to remove
622 *
623 * This just removes a node from its drm_mm allocator. The node does not need to
624 * be cleared again before it can be re-inserted into this or any other drm_mm
625 * allocator. It is a bug to call this function on a unallocated node.
626 */
627void drm_mm_remove_node(struct drm_mm_node *node)
628{
629 struct drm_mm *mm = node->mm;
630 struct drm_mm_node *prev_node;
631
632 DRM_MM_BUG_ON(!drm_mm_node_allocated(node));
633 DRM_MM_BUG_ON(drm_mm_node_scanned_block(node));
634
635 prev_node = list_prev_entry(node, node_list);
636
637 if (drm_mm_hole_follows(node))
638 rm_hole(node);
639
640 drm_mm_interval_tree_remove(node, &mm->interval_tree);
641 list_del(&node->node_list);
642
643 if (drm_mm_hole_follows(prev_node))
644 rm_hole(prev_node);
645 add_hole(prev_node);
646
647 clear_bit_unlock(DRM_MM_NODE_ALLOCATED_BIT, &node->flags);
648}
649EXPORT_SYMBOL(drm_mm_remove_node);
650
651/**
652 * DOC: lru scan roster
653 *
654 * Very often GPUs need to have continuous allocations for a given object. When
655 * evicting objects to make space for a new one it is therefore not most
656 * efficient when we simply start to select all objects from the tail of an LRU
657 * until there's a suitable hole: Especially for big objects or nodes that
658 * otherwise have special allocation constraints there's a good chance we evict
659 * lots of (smaller) objects unnecessarily.
660 *
661 * The DRM range allocator supports this use-case through the scanning
662 * interfaces. First a scan operation needs to be initialized with
663 * drm_mm_scan_init() or drm_mm_scan_init_with_range(). The driver adds
664 * objects to the roster, probably by walking an LRU list, but this can be
665 * freely implemented. Eviction candidates are added using
666 * drm_mm_scan_add_block() until a suitable hole is found or there are no
667 * further evictable objects. Eviction roster metadata is tracked in &struct
668 * drm_mm_scan.
669 *
670 * The driver must walk through all objects again in exactly the reverse
671 * order to restore the allocator state. Note that while the allocator is used
672 * in the scan mode no other operation is allowed.
673 *
674 * Finally the driver evicts all objects selected (drm_mm_scan_remove_block()
675 * reported true) in the scan, and any overlapping nodes after color adjustment
676 * (drm_mm_scan_color_evict()). Adding and removing an object is O(1), and
677 * since freeing a node is also O(1) the overall complexity is
678 * O(scanned_objects). So like the free stack which needs to be walked before a
679 * scan operation even begins this is linear in the number of objects. It
680 * doesn't seem to hurt too badly.
681 */
682
683/**
684 * drm_mm_scan_init_with_range - initialize range-restricted lru scanning
685 * @scan: scan state
686 * @mm: drm_mm to scan
687 * @size: size of the allocation
688 * @alignment: alignment of the allocation
689 * @color: opaque tag value to use for the allocation
690 * @start: start of the allowed range for the allocation
691 * @end: end of the allowed range for the allocation
692 * @mode: fine-tune the allocation search and placement
693 *
694 * This simply sets up the scanning routines with the parameters for the desired
695 * hole.
696 *
697 * Warning:
698 * As long as the scan list is non-empty, no other operations than
699 * adding/removing nodes to/from the scan list are allowed.
700 */
701void drm_mm_scan_init_with_range(struct drm_mm_scan *scan,
702 struct drm_mm *mm,
703 u64 size,
704 u64 alignment,
705 unsigned long color,
706 u64 start,
707 u64 end,
708 enum drm_mm_insert_mode mode)
709{
710 DRM_MM_BUG_ON(start >= end);
711 DRM_MM_BUG_ON(!size || size > end - start);
712 DRM_MM_BUG_ON(mm->scan_active);
713
714 scan->mm = mm;
715
716 if (alignment <= 1)
717 alignment = 0;
718
719 scan->color = color;
720 scan->alignment = alignment;
721 scan->remainder_mask = is_power_of_2(alignment) ? alignment - 1 : 0;
722 scan->size = size;
723 scan->mode = mode;
724
725 DRM_MM_BUG_ON(end <= start);
726 scan->range_start = start;
727 scan->range_end = end;
728
729 scan->hit_start = U64_MAX;
730 scan->hit_end = 0;
731}
732EXPORT_SYMBOL(drm_mm_scan_init_with_range);
733
734/**
735 * drm_mm_scan_add_block - add a node to the scan list
736 * @scan: the active drm_mm scanner
737 * @node: drm_mm_node to add
738 *
739 * Add a node to the scan list that might be freed to make space for the desired
740 * hole.
741 *
742 * Returns:
743 * True if a hole has been found, false otherwise.
744 */
745bool drm_mm_scan_add_block(struct drm_mm_scan *scan,
746 struct drm_mm_node *node)
747{
748 struct drm_mm *mm = scan->mm;
749 struct drm_mm_node *hole;
750 u64 hole_start, hole_end;
751 u64 col_start, col_end;
752 u64 adj_start, adj_end;
753
754 DRM_MM_BUG_ON(node->mm != mm);
755 DRM_MM_BUG_ON(!drm_mm_node_allocated(node));
756 DRM_MM_BUG_ON(drm_mm_node_scanned_block(node));
757 __set_bit(DRM_MM_NODE_SCANNED_BIT, &node->flags);
758 mm->scan_active++;
759
760 /* Remove this block from the node_list so that we enlarge the hole
761 * (distance between the end of our previous node and the start of
762 * or next), without poisoning the link so that we can restore it
763 * later in drm_mm_scan_remove_block().
764 */
765 hole = list_prev_entry(node, node_list);
766 DRM_MM_BUG_ON(list_next_entry(hole, node_list) != node);
767 __list_del_entry(&node->node_list);
768
769 hole_start = __drm_mm_hole_node_start(hole);
770 hole_end = __drm_mm_hole_node_end(hole);
771
772 col_start = hole_start;
773 col_end = hole_end;
774 if (mm->color_adjust)
775 mm->color_adjust(hole, scan->color, &col_start, &col_end);
776
777 adj_start = max(col_start, scan->range_start);
778 adj_end = min(col_end, scan->range_end);
779 if (adj_end <= adj_start || adj_end - adj_start < scan->size)
780 return false;
781
782 if (scan->mode == DRM_MM_INSERT_HIGH)
783 adj_start = adj_end - scan->size;
784
785 if (scan->alignment) {
786 u64 rem;
787
788 if (likely(scan->remainder_mask))
789 rem = adj_start & scan->remainder_mask;
790 else
791 div64_u64_rem(adj_start, scan->alignment, &rem);
792 if (rem) {
793 adj_start -= rem;
794 if (scan->mode != DRM_MM_INSERT_HIGH)
795 adj_start += scan->alignment;
796 if (adj_start < max(col_start, scan->range_start) ||
797 min(col_end, scan->range_end) - adj_start < scan->size)
798 return false;
799
800 if (adj_end <= adj_start ||
801 adj_end - adj_start < scan->size)
802 return false;
803 }
804 }
805
806 scan->hit_start = adj_start;
807 scan->hit_end = adj_start + scan->size;
808
809 DRM_MM_BUG_ON(scan->hit_start >= scan->hit_end);
810 DRM_MM_BUG_ON(scan->hit_start < hole_start);
811 DRM_MM_BUG_ON(scan->hit_end > hole_end);
812
813 return true;
814}
815EXPORT_SYMBOL(drm_mm_scan_add_block);
816
817/**
818 * drm_mm_scan_remove_block - remove a node from the scan list
819 * @scan: the active drm_mm scanner
820 * @node: drm_mm_node to remove
821 *
822 * Nodes **must** be removed in exactly the reverse order from the scan list as
823 * they have been added (e.g. using list_add() as they are added and then
824 * list_for_each() over that eviction list to remove), otherwise the internal
825 * state of the memory manager will be corrupted.
826 *
827 * When the scan list is empty, the selected memory nodes can be freed. An
828 * immediately following drm_mm_insert_node_in_range_generic() or one of the
829 * simpler versions of that function with !DRM_MM_SEARCH_BEST will then return
830 * the just freed block (because it's at the top of the free_stack list).
831 *
832 * Returns:
833 * True if this block should be evicted, false otherwise. Will always
834 * return false when no hole has been found.
835 */
836bool drm_mm_scan_remove_block(struct drm_mm_scan *scan,
837 struct drm_mm_node *node)
838{
839 struct drm_mm_node *prev_node;
840
841 DRM_MM_BUG_ON(node->mm != scan->mm);
842 DRM_MM_BUG_ON(!drm_mm_node_scanned_block(node));
843 __clear_bit(DRM_MM_NODE_SCANNED_BIT, &node->flags);
844
845 DRM_MM_BUG_ON(!node->mm->scan_active);
846 node->mm->scan_active--;
847
848 /* During drm_mm_scan_add_block() we decoupled this node leaving
849 * its pointers intact. Now that the caller is walking back along
850 * the eviction list we can restore this block into its rightful
851 * place on the full node_list. To confirm that the caller is walking
852 * backwards correctly we check that prev_node->next == node->next,
853 * i.e. both believe the same node should be on the other side of the
854 * hole.
855 */
856 prev_node = list_prev_entry(node, node_list);
857 DRM_MM_BUG_ON(list_next_entry(prev_node, node_list) !=
858 list_next_entry(node, node_list));
859 list_add(&node->node_list, &prev_node->node_list);
860
861 return (node->start + node->size > scan->hit_start &&
862 node->start < scan->hit_end);
863}
864EXPORT_SYMBOL(drm_mm_scan_remove_block);
865
866/**
867 * drm_mm_scan_color_evict - evict overlapping nodes on either side of hole
868 * @scan: drm_mm scan with target hole
869 *
870 * After completing an eviction scan and removing the selected nodes, we may
871 * need to remove a few more nodes from either side of the target hole if
872 * mm.color_adjust is being used.
873 *
874 * Returns:
875 * A node to evict, or NULL if there are no overlapping nodes.
876 */
877struct drm_mm_node *drm_mm_scan_color_evict(struct drm_mm_scan *scan)
878{
879 struct drm_mm *mm = scan->mm;
880 struct drm_mm_node *hole;
881 u64 hole_start, hole_end;
882
883 DRM_MM_BUG_ON(list_empty(&mm->hole_stack));
884
885 if (!mm->color_adjust)
886 return NULL;
887
888 /*
889 * The hole found during scanning should ideally be the first element
890 * in the hole_stack list, but due to side-effects in the driver it
891 * may not be.
892 */
893 list_for_each_entry(hole, &mm->hole_stack, hole_stack) {
894 hole_start = __drm_mm_hole_node_start(hole);
895 hole_end = hole_start + hole->hole_size;
896
897 if (hole_start <= scan->hit_start &&
898 hole_end >= scan->hit_end)
899 break;
900 }
901
902 /* We should only be called after we found the hole previously */
903 DRM_MM_BUG_ON(&hole->hole_stack == &mm->hole_stack);
904 if (unlikely(&hole->hole_stack == &mm->hole_stack))
905 return NULL;
906
907 DRM_MM_BUG_ON(hole_start > scan->hit_start);
908 DRM_MM_BUG_ON(hole_end < scan->hit_end);
909
910 mm->color_adjust(hole, scan->color, &hole_start, &hole_end);
911 if (hole_start > scan->hit_start)
912 return hole;
913 if (hole_end < scan->hit_end)
914 return list_next_entry(hole, node_list);
915
916 return NULL;
917}
918EXPORT_SYMBOL(drm_mm_scan_color_evict);
919
920/**
921 * drm_mm_init - initialize a drm-mm allocator
922 * @mm: the drm_mm structure to initialize
923 * @start: start of the range managed by @mm
924 * @size: end of the range managed by @mm
925 *
926 * Note that @mm must be cleared to 0 before calling this function.
927 */
928void drm_mm_init(struct drm_mm *mm, u64 start, u64 size)
929{
930 DRM_MM_BUG_ON(start + size <= start);
931
932 mm->color_adjust = NULL;
933
934 INIT_LIST_HEAD(&mm->hole_stack);
935 mm->interval_tree = RB_ROOT_CACHED;
936 mm->holes_size = RB_ROOT_CACHED;
937 mm->holes_addr = RB_ROOT;
938
939 /* Clever trick to avoid a special case in the free hole tracking. */
940 INIT_LIST_HEAD(&mm->head_node.node_list);
941 mm->head_node.flags = 0;
942 mm->head_node.mm = mm;
943 mm->head_node.start = start + size;
944 mm->head_node.size = -size;
945 add_hole(&mm->head_node);
946
947 mm->scan_active = 0;
948
949#ifdef CONFIG_DRM_DEBUG_MM
950 stack_depot_init();
951#endif
952}
953EXPORT_SYMBOL(drm_mm_init);
954
955/**
956 * drm_mm_takedown - clean up a drm_mm allocator
957 * @mm: drm_mm allocator to clean up
958 *
959 * Note that it is a bug to call this function on an allocator which is not
960 * clean.
961 */
962void drm_mm_takedown(struct drm_mm *mm)
963{
964 if (WARN(!drm_mm_clean(mm),
965 "Memory manager not clean during takedown.\n"))
966 show_leaks(mm);
967}
968EXPORT_SYMBOL(drm_mm_takedown);
969
970static u64 drm_mm_dump_hole(struct drm_printer *p, const struct drm_mm_node *entry)
971{
972 u64 start, size;
973
974 size = entry->hole_size;
975 if (size) {
976 start = drm_mm_hole_node_start(entry);
977 drm_printf(p, "%#018llx-%#018llx: %llu: free\n",
978 start, start + size, size);
979 }
980
981 return size;
982}
983/**
984 * drm_mm_print - print allocator state
985 * @mm: drm_mm allocator to print
986 * @p: DRM printer to use
987 */
988void drm_mm_print(const struct drm_mm *mm, struct drm_printer *p)
989{
990 const struct drm_mm_node *entry;
991 u64 total_used = 0, total_free = 0, total = 0;
992
993 total_free += drm_mm_dump_hole(p, &mm->head_node);
994
995 drm_mm_for_each_node(entry, mm) {
996 drm_printf(p, "%#018llx-%#018llx: %llu: used\n", entry->start,
997 entry->start + entry->size, entry->size);
998 total_used += entry->size;
999 total_free += drm_mm_dump_hole(p, entry);
1000 }
1001 total = total_free + total_used;
1002
1003 drm_printf(p, "total: %llu, used %llu free %llu\n", total,
1004 total_used, total_free);
1005}
1006EXPORT_SYMBOL(drm_mm_print);
1/**************************************************************************
2 *
3 * Copyright 2006 Tungsten Graphics, Inc., Bismarck, ND., USA.
4 * Copyright 2016 Intel Corporation
5 * All Rights Reserved.
6 *
7 * Permission is hereby granted, free of charge, to any person obtaining a
8 * copy of this software and associated documentation files (the
9 * "Software"), to deal in the Software without restriction, including
10 * without limitation the rights to use, copy, modify, merge, publish,
11 * distribute, sub license, and/or sell copies of the Software, and to
12 * permit persons to whom the Software is furnished to do so, subject to
13 * the following conditions:
14 *
15 * The above copyright notice and this permission notice (including the
16 * next paragraph) shall be included in all copies or substantial portions
17 * of the Software.
18 *
19 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
20 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
21 * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
22 * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
23 * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
24 * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
25 * USE OR OTHER DEALINGS IN THE SOFTWARE.
26 *
27 *
28 **************************************************************************/
29
30/*
31 * Generic simple memory manager implementation. Intended to be used as a base
32 * class implementation for more advanced memory managers.
33 *
34 * Note that the algorithm used is quite simple and there might be substantial
35 * performance gains if a smarter free list is implemented. Currently it is
36 * just an unordered stack of free regions. This could easily be improved if
37 * an RB-tree is used instead. At least if we expect heavy fragmentation.
38 *
39 * Aligned allocations can also see improvement.
40 *
41 * Authors:
42 * Thomas Hellström <thomas-at-tungstengraphics-dot-com>
43 */
44
45#include <drm/drmP.h>
46#include <drm/drm_mm.h>
47#include <linux/slab.h>
48#include <linux/seq_file.h>
49#include <linux/export.h>
50#include <linux/interval_tree_generic.h>
51
52/**
53 * DOC: Overview
54 *
55 * drm_mm provides a simple range allocator. The drivers are free to use the
56 * resource allocator from the linux core if it suits them, the upside of drm_mm
57 * is that it's in the DRM core. Which means that it's easier to extend for
58 * some of the crazier special purpose needs of gpus.
59 *
60 * The main data struct is &drm_mm, allocations are tracked in &drm_mm_node.
61 * Drivers are free to embed either of them into their own suitable
62 * datastructures. drm_mm itself will not do any memory allocations of its own,
63 * so if drivers choose not to embed nodes they need to still allocate them
64 * themselves.
65 *
66 * The range allocator also supports reservation of preallocated blocks. This is
67 * useful for taking over initial mode setting configurations from the firmware,
68 * where an object needs to be created which exactly matches the firmware's
69 * scanout target. As long as the range is still free it can be inserted anytime
70 * after the allocator is initialized, which helps with avoiding looped
71 * dependencies in the driver load sequence.
72 *
73 * drm_mm maintains a stack of most recently freed holes, which of all
74 * simplistic datastructures seems to be a fairly decent approach to clustering
75 * allocations and avoiding too much fragmentation. This means free space
76 * searches are O(num_holes). Given that all the fancy features drm_mm supports
77 * something better would be fairly complex and since gfx thrashing is a fairly
78 * steep cliff not a real concern. Removing a node again is O(1).
79 *
80 * drm_mm supports a few features: Alignment and range restrictions can be
81 * supplied. Furthermore every &drm_mm_node has a color value (which is just an
82 * opaque unsigned long) which in conjunction with a driver callback can be used
83 * to implement sophisticated placement restrictions. The i915 DRM driver uses
84 * this to implement guard pages between incompatible caching domains in the
85 * graphics TT.
86 *
87 * Two behaviors are supported for searching and allocating: bottom-up and
88 * top-down. The default is bottom-up. Top-down allocation can be used if the
89 * memory area has different restrictions, or just to reduce fragmentation.
90 *
91 * Finally iteration helpers to walk all nodes and all holes are provided as are
92 * some basic allocator dumpers for debugging.
93 *
94 * Note that this range allocator is not thread-safe, drivers need to protect
95 * modifications with their own locking. The idea behind this is that for a full
96 * memory manager additional data needs to be protected anyway, hence internal
97 * locking would be fully redundant.
98 */
99
100#ifdef CONFIG_DRM_DEBUG_MM
101#include <linux/stackdepot.h>
102
103#define STACKDEPTH 32
104#define BUFSZ 4096
105
106static noinline void save_stack(struct drm_mm_node *node)
107{
108 unsigned long entries[STACKDEPTH];
109 struct stack_trace trace = {
110 .entries = entries,
111 .max_entries = STACKDEPTH,
112 .skip = 1
113 };
114
115 save_stack_trace(&trace);
116 if (trace.nr_entries != 0 &&
117 trace.entries[trace.nr_entries-1] == ULONG_MAX)
118 trace.nr_entries--;
119
120 /* May be called under spinlock, so avoid sleeping */
121 node->stack = depot_save_stack(&trace, GFP_NOWAIT);
122}
123
124static void show_leaks(struct drm_mm *mm)
125{
126 struct drm_mm_node *node;
127 unsigned long entries[STACKDEPTH];
128 char *buf;
129
130 buf = kmalloc(BUFSZ, GFP_KERNEL);
131 if (!buf)
132 return;
133
134 list_for_each_entry(node, drm_mm_nodes(mm), node_list) {
135 struct stack_trace trace = {
136 .entries = entries,
137 .max_entries = STACKDEPTH
138 };
139
140 if (!node->stack) {
141 DRM_ERROR("node [%08llx + %08llx]: unknown owner\n",
142 node->start, node->size);
143 continue;
144 }
145
146 depot_fetch_stack(node->stack, &trace);
147 snprint_stack_trace(buf, BUFSZ, &trace, 0);
148 DRM_ERROR("node [%08llx + %08llx]: inserted at\n%s",
149 node->start, node->size, buf);
150 }
151
152 kfree(buf);
153}
154
155#undef STACKDEPTH
156#undef BUFSZ
157#else
158static void save_stack(struct drm_mm_node *node) { }
159static void show_leaks(struct drm_mm *mm) { }
160#endif
161
162#define START(node) ((node)->start)
163#define LAST(node) ((node)->start + (node)->size - 1)
164
165INTERVAL_TREE_DEFINE(struct drm_mm_node, rb,
166 u64, __subtree_last,
167 START, LAST, static inline, drm_mm_interval_tree)
168
169struct drm_mm_node *
170__drm_mm_interval_first(const struct drm_mm *mm, u64 start, u64 last)
171{
172 return drm_mm_interval_tree_iter_first((struct rb_root_cached *)&mm->interval_tree,
173 start, last) ?: (struct drm_mm_node *)&mm->head_node;
174}
175EXPORT_SYMBOL(__drm_mm_interval_first);
176
177static void drm_mm_interval_tree_add_node(struct drm_mm_node *hole_node,
178 struct drm_mm_node *node)
179{
180 struct drm_mm *mm = hole_node->mm;
181 struct rb_node **link, *rb;
182 struct drm_mm_node *parent;
183 bool leftmost;
184
185 node->__subtree_last = LAST(node);
186
187 if (hole_node->allocated) {
188 rb = &hole_node->rb;
189 while (rb) {
190 parent = rb_entry(rb, struct drm_mm_node, rb);
191 if (parent->__subtree_last >= node->__subtree_last)
192 break;
193
194 parent->__subtree_last = node->__subtree_last;
195 rb = rb_parent(rb);
196 }
197
198 rb = &hole_node->rb;
199 link = &hole_node->rb.rb_right;
200 leftmost = false;
201 } else {
202 rb = NULL;
203 link = &mm->interval_tree.rb_root.rb_node;
204 leftmost = true;
205 }
206
207 while (*link) {
208 rb = *link;
209 parent = rb_entry(rb, struct drm_mm_node, rb);
210 if (parent->__subtree_last < node->__subtree_last)
211 parent->__subtree_last = node->__subtree_last;
212 if (node->start < parent->start) {
213 link = &parent->rb.rb_left;
214 } else {
215 link = &parent->rb.rb_right;
216 leftmost = false;
217 }
218 }
219
220 rb_link_node(&node->rb, rb, link);
221 rb_insert_augmented_cached(&node->rb, &mm->interval_tree, leftmost,
222 &drm_mm_interval_tree_augment);
223}
224
225#define RB_INSERT(root, member, expr) do { \
226 struct rb_node **link = &root.rb_node, *rb = NULL; \
227 u64 x = expr(node); \
228 while (*link) { \
229 rb = *link; \
230 if (x < expr(rb_entry(rb, struct drm_mm_node, member))) \
231 link = &rb->rb_left; \
232 else \
233 link = &rb->rb_right; \
234 } \
235 rb_link_node(&node->member, rb, link); \
236 rb_insert_color(&node->member, &root); \
237} while (0)
238
239#define HOLE_SIZE(NODE) ((NODE)->hole_size)
240#define HOLE_ADDR(NODE) (__drm_mm_hole_node_start(NODE))
241
242static void add_hole(struct drm_mm_node *node)
243{
244 struct drm_mm *mm = node->mm;
245
246 node->hole_size =
247 __drm_mm_hole_node_end(node) - __drm_mm_hole_node_start(node);
248 DRM_MM_BUG_ON(!drm_mm_hole_follows(node));
249
250 RB_INSERT(mm->holes_size, rb_hole_size, HOLE_SIZE);
251 RB_INSERT(mm->holes_addr, rb_hole_addr, HOLE_ADDR);
252
253 list_add(&node->hole_stack, &mm->hole_stack);
254}
255
256static void rm_hole(struct drm_mm_node *node)
257{
258 DRM_MM_BUG_ON(!drm_mm_hole_follows(node));
259
260 list_del(&node->hole_stack);
261 rb_erase(&node->rb_hole_size, &node->mm->holes_size);
262 rb_erase(&node->rb_hole_addr, &node->mm->holes_addr);
263 node->hole_size = 0;
264
265 DRM_MM_BUG_ON(drm_mm_hole_follows(node));
266}
267
268static inline struct drm_mm_node *rb_hole_size_to_node(struct rb_node *rb)
269{
270 return rb_entry_safe(rb, struct drm_mm_node, rb_hole_size);
271}
272
273static inline struct drm_mm_node *rb_hole_addr_to_node(struct rb_node *rb)
274{
275 return rb_entry_safe(rb, struct drm_mm_node, rb_hole_addr);
276}
277
278static inline u64 rb_hole_size(struct rb_node *rb)
279{
280 return rb_entry(rb, struct drm_mm_node, rb_hole_size)->hole_size;
281}
282
283static struct drm_mm_node *best_hole(struct drm_mm *mm, u64 size)
284{
285 struct rb_node *best = NULL;
286 struct rb_node **link = &mm->holes_size.rb_node;
287
288 while (*link) {
289 struct rb_node *rb = *link;
290
291 if (size <= rb_hole_size(rb)) {
292 link = &rb->rb_left;
293 best = rb;
294 } else {
295 link = &rb->rb_right;
296 }
297 }
298
299 return rb_hole_size_to_node(best);
300}
301
302static struct drm_mm_node *find_hole(struct drm_mm *mm, u64 addr)
303{
304 struct drm_mm_node *node = NULL;
305 struct rb_node **link = &mm->holes_addr.rb_node;
306
307 while (*link) {
308 u64 hole_start;
309
310 node = rb_hole_addr_to_node(*link);
311 hole_start = __drm_mm_hole_node_start(node);
312
313 if (addr < hole_start)
314 link = &node->rb_hole_addr.rb_left;
315 else if (addr > hole_start + node->hole_size)
316 link = &node->rb_hole_addr.rb_right;
317 else
318 break;
319 }
320
321 return node;
322}
323
324static struct drm_mm_node *
325first_hole(struct drm_mm *mm,
326 u64 start, u64 end, u64 size,
327 enum drm_mm_insert_mode mode)
328{
329 if (RB_EMPTY_ROOT(&mm->holes_size))
330 return NULL;
331
332 switch (mode) {
333 default:
334 case DRM_MM_INSERT_BEST:
335 return best_hole(mm, size);
336
337 case DRM_MM_INSERT_LOW:
338 return find_hole(mm, start);
339
340 case DRM_MM_INSERT_HIGH:
341 return find_hole(mm, end);
342
343 case DRM_MM_INSERT_EVICT:
344 return list_first_entry_or_null(&mm->hole_stack,
345 struct drm_mm_node,
346 hole_stack);
347 }
348}
349
350static struct drm_mm_node *
351next_hole(struct drm_mm *mm,
352 struct drm_mm_node *node,
353 enum drm_mm_insert_mode mode)
354{
355 switch (mode) {
356 default:
357 case DRM_MM_INSERT_BEST:
358 return rb_hole_size_to_node(rb_next(&node->rb_hole_size));
359
360 case DRM_MM_INSERT_LOW:
361 return rb_hole_addr_to_node(rb_next(&node->rb_hole_addr));
362
363 case DRM_MM_INSERT_HIGH:
364 return rb_hole_addr_to_node(rb_prev(&node->rb_hole_addr));
365
366 case DRM_MM_INSERT_EVICT:
367 node = list_next_entry(node, hole_stack);
368 return &node->hole_stack == &mm->hole_stack ? NULL : node;
369 }
370}
371
372/**
373 * drm_mm_reserve_node - insert an pre-initialized node
374 * @mm: drm_mm allocator to insert @node into
375 * @node: drm_mm_node to insert
376 *
377 * This functions inserts an already set-up &drm_mm_node into the allocator,
378 * meaning that start, size and color must be set by the caller. All other
379 * fields must be cleared to 0. This is useful to initialize the allocator with
380 * preallocated objects which must be set-up before the range allocator can be
381 * set-up, e.g. when taking over a firmware framebuffer.
382 *
383 * Returns:
384 * 0 on success, -ENOSPC if there's no hole where @node is.
385 */
386int drm_mm_reserve_node(struct drm_mm *mm, struct drm_mm_node *node)
387{
388 u64 end = node->start + node->size;
389 struct drm_mm_node *hole;
390 u64 hole_start, hole_end;
391 u64 adj_start, adj_end;
392
393 end = node->start + node->size;
394 if (unlikely(end <= node->start))
395 return -ENOSPC;
396
397 /* Find the relevant hole to add our node to */
398 hole = find_hole(mm, node->start);
399 if (!hole)
400 return -ENOSPC;
401
402 adj_start = hole_start = __drm_mm_hole_node_start(hole);
403 adj_end = hole_end = hole_start + hole->hole_size;
404
405 if (mm->color_adjust)
406 mm->color_adjust(hole, node->color, &adj_start, &adj_end);
407
408 if (adj_start > node->start || adj_end < end)
409 return -ENOSPC;
410
411 node->mm = mm;
412
413 list_add(&node->node_list, &hole->node_list);
414 drm_mm_interval_tree_add_node(hole, node);
415 node->allocated = true;
416 node->hole_size = 0;
417
418 rm_hole(hole);
419 if (node->start > hole_start)
420 add_hole(hole);
421 if (end < hole_end)
422 add_hole(node);
423
424 save_stack(node);
425 return 0;
426}
427EXPORT_SYMBOL(drm_mm_reserve_node);
428
429/**
430 * drm_mm_insert_node_in_range - ranged search for space and insert @node
431 * @mm: drm_mm to allocate from
432 * @node: preallocate node to insert
433 * @size: size of the allocation
434 * @alignment: alignment of the allocation
435 * @color: opaque tag value to use for this node
436 * @range_start: start of the allowed range for this node
437 * @range_end: end of the allowed range for this node
438 * @mode: fine-tune the allocation search and placement
439 *
440 * The preallocated @node must be cleared to 0.
441 *
442 * Returns:
443 * 0 on success, -ENOSPC if there's no suitable hole.
444 */
445int drm_mm_insert_node_in_range(struct drm_mm * const mm,
446 struct drm_mm_node * const node,
447 u64 size, u64 alignment,
448 unsigned long color,
449 u64 range_start, u64 range_end,
450 enum drm_mm_insert_mode mode)
451{
452 struct drm_mm_node *hole;
453 u64 remainder_mask;
454
455 DRM_MM_BUG_ON(range_start >= range_end);
456
457 if (unlikely(size == 0 || range_end - range_start < size))
458 return -ENOSPC;
459
460 if (alignment <= 1)
461 alignment = 0;
462
463 remainder_mask = is_power_of_2(alignment) ? alignment - 1 : 0;
464 for (hole = first_hole(mm, range_start, range_end, size, mode); hole;
465 hole = next_hole(mm, hole, mode)) {
466 u64 hole_start = __drm_mm_hole_node_start(hole);
467 u64 hole_end = hole_start + hole->hole_size;
468 u64 adj_start, adj_end;
469 u64 col_start, col_end;
470
471 if (mode == DRM_MM_INSERT_LOW && hole_start >= range_end)
472 break;
473
474 if (mode == DRM_MM_INSERT_HIGH && hole_end <= range_start)
475 break;
476
477 col_start = hole_start;
478 col_end = hole_end;
479 if (mm->color_adjust)
480 mm->color_adjust(hole, color, &col_start, &col_end);
481
482 adj_start = max(col_start, range_start);
483 adj_end = min(col_end, range_end);
484
485 if (adj_end <= adj_start || adj_end - adj_start < size)
486 continue;
487
488 if (mode == DRM_MM_INSERT_HIGH)
489 adj_start = adj_end - size;
490
491 if (alignment) {
492 u64 rem;
493
494 if (likely(remainder_mask))
495 rem = adj_start & remainder_mask;
496 else
497 div64_u64_rem(adj_start, alignment, &rem);
498 if (rem) {
499 adj_start -= rem;
500 if (mode != DRM_MM_INSERT_HIGH)
501 adj_start += alignment;
502
503 if (adj_start < max(col_start, range_start) ||
504 min(col_end, range_end) - adj_start < size)
505 continue;
506
507 if (adj_end <= adj_start ||
508 adj_end - adj_start < size)
509 continue;
510 }
511 }
512
513 node->mm = mm;
514 node->size = size;
515 node->start = adj_start;
516 node->color = color;
517 node->hole_size = 0;
518
519 list_add(&node->node_list, &hole->node_list);
520 drm_mm_interval_tree_add_node(hole, node);
521 node->allocated = true;
522
523 rm_hole(hole);
524 if (adj_start > hole_start)
525 add_hole(hole);
526 if (adj_start + size < hole_end)
527 add_hole(node);
528
529 save_stack(node);
530 return 0;
531 }
532
533 return -ENOSPC;
534}
535EXPORT_SYMBOL(drm_mm_insert_node_in_range);
536
537/**
538 * drm_mm_remove_node - Remove a memory node from the allocator.
539 * @node: drm_mm_node to remove
540 *
541 * This just removes a node from its drm_mm allocator. The node does not need to
542 * be cleared again before it can be re-inserted into this or any other drm_mm
543 * allocator. It is a bug to call this function on a unallocated node.
544 */
545void drm_mm_remove_node(struct drm_mm_node *node)
546{
547 struct drm_mm *mm = node->mm;
548 struct drm_mm_node *prev_node;
549
550 DRM_MM_BUG_ON(!node->allocated);
551 DRM_MM_BUG_ON(node->scanned_block);
552
553 prev_node = list_prev_entry(node, node_list);
554
555 if (drm_mm_hole_follows(node))
556 rm_hole(node);
557
558 drm_mm_interval_tree_remove(node, &mm->interval_tree);
559 list_del(&node->node_list);
560 node->allocated = false;
561
562 if (drm_mm_hole_follows(prev_node))
563 rm_hole(prev_node);
564 add_hole(prev_node);
565}
566EXPORT_SYMBOL(drm_mm_remove_node);
567
568/**
569 * drm_mm_replace_node - move an allocation from @old to @new
570 * @old: drm_mm_node to remove from the allocator
571 * @new: drm_mm_node which should inherit @old's allocation
572 *
573 * This is useful for when drivers embed the drm_mm_node structure and hence
574 * can't move allocations by reassigning pointers. It's a combination of remove
575 * and insert with the guarantee that the allocation start will match.
576 */
577void drm_mm_replace_node(struct drm_mm_node *old, struct drm_mm_node *new)
578{
579 struct drm_mm *mm = old->mm;
580
581 DRM_MM_BUG_ON(!old->allocated);
582
583 *new = *old;
584
585 list_replace(&old->node_list, &new->node_list);
586 rb_replace_node_cached(&old->rb, &new->rb, &mm->interval_tree);
587
588 if (drm_mm_hole_follows(old)) {
589 list_replace(&old->hole_stack, &new->hole_stack);
590 rb_replace_node(&old->rb_hole_size,
591 &new->rb_hole_size,
592 &mm->holes_size);
593 rb_replace_node(&old->rb_hole_addr,
594 &new->rb_hole_addr,
595 &mm->holes_addr);
596 }
597
598 old->allocated = false;
599 new->allocated = true;
600}
601EXPORT_SYMBOL(drm_mm_replace_node);
602
603/**
604 * DOC: lru scan roster
605 *
606 * Very often GPUs need to have continuous allocations for a given object. When
607 * evicting objects to make space for a new one it is therefore not most
608 * efficient when we simply start to select all objects from the tail of an LRU
609 * until there's a suitable hole: Especially for big objects or nodes that
610 * otherwise have special allocation constraints there's a good chance we evict
611 * lots of (smaller) objects unnecessarily.
612 *
613 * The DRM range allocator supports this use-case through the scanning
614 * interfaces. First a scan operation needs to be initialized with
615 * drm_mm_scan_init() or drm_mm_scan_init_with_range(). The driver adds
616 * objects to the roster, probably by walking an LRU list, but this can be
617 * freely implemented. Eviction candiates are added using
618 * drm_mm_scan_add_block() until a suitable hole is found or there are no
619 * further evictable objects. Eviction roster metadata is tracked in &struct
620 * drm_mm_scan.
621 *
622 * The driver must walk through all objects again in exactly the reverse
623 * order to restore the allocator state. Note that while the allocator is used
624 * in the scan mode no other operation is allowed.
625 *
626 * Finally the driver evicts all objects selected (drm_mm_scan_remove_block()
627 * reported true) in the scan, and any overlapping nodes after color adjustment
628 * (drm_mm_scan_color_evict()). Adding and removing an object is O(1), and
629 * since freeing a node is also O(1) the overall complexity is
630 * O(scanned_objects). So like the free stack which needs to be walked before a
631 * scan operation even begins this is linear in the number of objects. It
632 * doesn't seem to hurt too badly.
633 */
634
635/**
636 * drm_mm_scan_init_with_range - initialize range-restricted lru scanning
637 * @scan: scan state
638 * @mm: drm_mm to scan
639 * @size: size of the allocation
640 * @alignment: alignment of the allocation
641 * @color: opaque tag value to use for the allocation
642 * @start: start of the allowed range for the allocation
643 * @end: end of the allowed range for the allocation
644 * @mode: fine-tune the allocation search and placement
645 *
646 * This simply sets up the scanning routines with the parameters for the desired
647 * hole.
648 *
649 * Warning:
650 * As long as the scan list is non-empty, no other operations than
651 * adding/removing nodes to/from the scan list are allowed.
652 */
653void drm_mm_scan_init_with_range(struct drm_mm_scan *scan,
654 struct drm_mm *mm,
655 u64 size,
656 u64 alignment,
657 unsigned long color,
658 u64 start,
659 u64 end,
660 enum drm_mm_insert_mode mode)
661{
662 DRM_MM_BUG_ON(start >= end);
663 DRM_MM_BUG_ON(!size || size > end - start);
664 DRM_MM_BUG_ON(mm->scan_active);
665
666 scan->mm = mm;
667
668 if (alignment <= 1)
669 alignment = 0;
670
671 scan->color = color;
672 scan->alignment = alignment;
673 scan->remainder_mask = is_power_of_2(alignment) ? alignment - 1 : 0;
674 scan->size = size;
675 scan->mode = mode;
676
677 DRM_MM_BUG_ON(end <= start);
678 scan->range_start = start;
679 scan->range_end = end;
680
681 scan->hit_start = U64_MAX;
682 scan->hit_end = 0;
683}
684EXPORT_SYMBOL(drm_mm_scan_init_with_range);
685
686/**
687 * drm_mm_scan_add_block - add a node to the scan list
688 * @scan: the active drm_mm scanner
689 * @node: drm_mm_node to add
690 *
691 * Add a node to the scan list that might be freed to make space for the desired
692 * hole.
693 *
694 * Returns:
695 * True if a hole has been found, false otherwise.
696 */
697bool drm_mm_scan_add_block(struct drm_mm_scan *scan,
698 struct drm_mm_node *node)
699{
700 struct drm_mm *mm = scan->mm;
701 struct drm_mm_node *hole;
702 u64 hole_start, hole_end;
703 u64 col_start, col_end;
704 u64 adj_start, adj_end;
705
706 DRM_MM_BUG_ON(node->mm != mm);
707 DRM_MM_BUG_ON(!node->allocated);
708 DRM_MM_BUG_ON(node->scanned_block);
709 node->scanned_block = true;
710 mm->scan_active++;
711
712 /* Remove this block from the node_list so that we enlarge the hole
713 * (distance between the end of our previous node and the start of
714 * or next), without poisoning the link so that we can restore it
715 * later in drm_mm_scan_remove_block().
716 */
717 hole = list_prev_entry(node, node_list);
718 DRM_MM_BUG_ON(list_next_entry(hole, node_list) != node);
719 __list_del_entry(&node->node_list);
720
721 hole_start = __drm_mm_hole_node_start(hole);
722 hole_end = __drm_mm_hole_node_end(hole);
723
724 col_start = hole_start;
725 col_end = hole_end;
726 if (mm->color_adjust)
727 mm->color_adjust(hole, scan->color, &col_start, &col_end);
728
729 adj_start = max(col_start, scan->range_start);
730 adj_end = min(col_end, scan->range_end);
731 if (adj_end <= adj_start || adj_end - adj_start < scan->size)
732 return false;
733
734 if (scan->mode == DRM_MM_INSERT_HIGH)
735 adj_start = adj_end - scan->size;
736
737 if (scan->alignment) {
738 u64 rem;
739
740 if (likely(scan->remainder_mask))
741 rem = adj_start & scan->remainder_mask;
742 else
743 div64_u64_rem(adj_start, scan->alignment, &rem);
744 if (rem) {
745 adj_start -= rem;
746 if (scan->mode != DRM_MM_INSERT_HIGH)
747 adj_start += scan->alignment;
748 if (adj_start < max(col_start, scan->range_start) ||
749 min(col_end, scan->range_end) - adj_start < scan->size)
750 return false;
751
752 if (adj_end <= adj_start ||
753 adj_end - adj_start < scan->size)
754 return false;
755 }
756 }
757
758 scan->hit_start = adj_start;
759 scan->hit_end = adj_start + scan->size;
760
761 DRM_MM_BUG_ON(scan->hit_start >= scan->hit_end);
762 DRM_MM_BUG_ON(scan->hit_start < hole_start);
763 DRM_MM_BUG_ON(scan->hit_end > hole_end);
764
765 return true;
766}
767EXPORT_SYMBOL(drm_mm_scan_add_block);
768
769/**
770 * drm_mm_scan_remove_block - remove a node from the scan list
771 * @scan: the active drm_mm scanner
772 * @node: drm_mm_node to remove
773 *
774 * Nodes **must** be removed in exactly the reverse order from the scan list as
775 * they have been added (e.g. using list_add() as they are added and then
776 * list_for_each() over that eviction list to remove), otherwise the internal
777 * state of the memory manager will be corrupted.
778 *
779 * When the scan list is empty, the selected memory nodes can be freed. An
780 * immediately following drm_mm_insert_node_in_range_generic() or one of the
781 * simpler versions of that function with !DRM_MM_SEARCH_BEST will then return
782 * the just freed block (because its at the top of the free_stack list).
783 *
784 * Returns:
785 * True if this block should be evicted, false otherwise. Will always
786 * return false when no hole has been found.
787 */
788bool drm_mm_scan_remove_block(struct drm_mm_scan *scan,
789 struct drm_mm_node *node)
790{
791 struct drm_mm_node *prev_node;
792
793 DRM_MM_BUG_ON(node->mm != scan->mm);
794 DRM_MM_BUG_ON(!node->scanned_block);
795 node->scanned_block = false;
796
797 DRM_MM_BUG_ON(!node->mm->scan_active);
798 node->mm->scan_active--;
799
800 /* During drm_mm_scan_add_block() we decoupled this node leaving
801 * its pointers intact. Now that the caller is walking back along
802 * the eviction list we can restore this block into its rightful
803 * place on the full node_list. To confirm that the caller is walking
804 * backwards correctly we check that prev_node->next == node->next,
805 * i.e. both believe the same node should be on the other side of the
806 * hole.
807 */
808 prev_node = list_prev_entry(node, node_list);
809 DRM_MM_BUG_ON(list_next_entry(prev_node, node_list) !=
810 list_next_entry(node, node_list));
811 list_add(&node->node_list, &prev_node->node_list);
812
813 return (node->start + node->size > scan->hit_start &&
814 node->start < scan->hit_end);
815}
816EXPORT_SYMBOL(drm_mm_scan_remove_block);
817
818/**
819 * drm_mm_scan_color_evict - evict overlapping nodes on either side of hole
820 * @scan: drm_mm scan with target hole
821 *
822 * After completing an eviction scan and removing the selected nodes, we may
823 * need to remove a few more nodes from either side of the target hole if
824 * mm.color_adjust is being used.
825 *
826 * Returns:
827 * A node to evict, or NULL if there are no overlapping nodes.
828 */
829struct drm_mm_node *drm_mm_scan_color_evict(struct drm_mm_scan *scan)
830{
831 struct drm_mm *mm = scan->mm;
832 struct drm_mm_node *hole;
833 u64 hole_start, hole_end;
834
835 DRM_MM_BUG_ON(list_empty(&mm->hole_stack));
836
837 if (!mm->color_adjust)
838 return NULL;
839
840 /*
841 * The hole found during scanning should ideally be the first element
842 * in the hole_stack list, but due to side-effects in the driver it
843 * may not be.
844 */
845 list_for_each_entry(hole, &mm->hole_stack, hole_stack) {
846 hole_start = __drm_mm_hole_node_start(hole);
847 hole_end = hole_start + hole->hole_size;
848
849 if (hole_start <= scan->hit_start &&
850 hole_end >= scan->hit_end)
851 break;
852 }
853
854 /* We should only be called after we found the hole previously */
855 DRM_MM_BUG_ON(&hole->hole_stack == &mm->hole_stack);
856 if (unlikely(&hole->hole_stack == &mm->hole_stack))
857 return NULL;
858
859 DRM_MM_BUG_ON(hole_start > scan->hit_start);
860 DRM_MM_BUG_ON(hole_end < scan->hit_end);
861
862 mm->color_adjust(hole, scan->color, &hole_start, &hole_end);
863 if (hole_start > scan->hit_start)
864 return hole;
865 if (hole_end < scan->hit_end)
866 return list_next_entry(hole, node_list);
867
868 return NULL;
869}
870EXPORT_SYMBOL(drm_mm_scan_color_evict);
871
872/**
873 * drm_mm_init - initialize a drm-mm allocator
874 * @mm: the drm_mm structure to initialize
875 * @start: start of the range managed by @mm
876 * @size: end of the range managed by @mm
877 *
878 * Note that @mm must be cleared to 0 before calling this function.
879 */
880void drm_mm_init(struct drm_mm *mm, u64 start, u64 size)
881{
882 DRM_MM_BUG_ON(start + size <= start);
883
884 mm->color_adjust = NULL;
885
886 INIT_LIST_HEAD(&mm->hole_stack);
887 mm->interval_tree = RB_ROOT_CACHED;
888 mm->holes_size = RB_ROOT;
889 mm->holes_addr = RB_ROOT;
890
891 /* Clever trick to avoid a special case in the free hole tracking. */
892 INIT_LIST_HEAD(&mm->head_node.node_list);
893 mm->head_node.allocated = false;
894 mm->head_node.mm = mm;
895 mm->head_node.start = start + size;
896 mm->head_node.size = -size;
897 add_hole(&mm->head_node);
898
899 mm->scan_active = 0;
900}
901EXPORT_SYMBOL(drm_mm_init);
902
903/**
904 * drm_mm_takedown - clean up a drm_mm allocator
905 * @mm: drm_mm allocator to clean up
906 *
907 * Note that it is a bug to call this function on an allocator which is not
908 * clean.
909 */
910void drm_mm_takedown(struct drm_mm *mm)
911{
912 if (WARN(!drm_mm_clean(mm),
913 "Memory manager not clean during takedown.\n"))
914 show_leaks(mm);
915}
916EXPORT_SYMBOL(drm_mm_takedown);
917
918static u64 drm_mm_dump_hole(struct drm_printer *p, const struct drm_mm_node *entry)
919{
920 u64 start, size;
921
922 size = entry->hole_size;
923 if (size) {
924 start = drm_mm_hole_node_start(entry);
925 drm_printf(p, "%#018llx-%#018llx: %llu: free\n",
926 start, start + size, size);
927 }
928
929 return size;
930}
931/**
932 * drm_mm_print - print allocator state
933 * @mm: drm_mm allocator to print
934 * @p: DRM printer to use
935 */
936void drm_mm_print(const struct drm_mm *mm, struct drm_printer *p)
937{
938 const struct drm_mm_node *entry;
939 u64 total_used = 0, total_free = 0, total = 0;
940
941 total_free += drm_mm_dump_hole(p, &mm->head_node);
942
943 drm_mm_for_each_node(entry, mm) {
944 drm_printf(p, "%#018llx-%#018llx: %llu: used\n", entry->start,
945 entry->start + entry->size, entry->size);
946 total_used += entry->size;
947 total_free += drm_mm_dump_hole(p, entry);
948 }
949 total = total_free + total_used;
950
951 drm_printf(p, "total: %llu, used %llu free %llu\n", total,
952 total_used, total_free);
953}
954EXPORT_SYMBOL(drm_mm_print);