1 // SPDX-License-Identifier: MIT 2 /* 3 * Copyright © 2021 Intel Corporation 4 */ 5 6 #include <linux/kmemleak.h> 7 #include <linux/module.h> 8 #include <linux/sizes.h> 9 10 #include <drm/drm_buddy.h> 11 12 static struct kmem_cache *slab_blocks; 13 14 static struct drm_buddy_block *drm_block_alloc(struct drm_buddy *mm, 15 struct drm_buddy_block *parent, 16 unsigned int order, 17 u64 offset) 18 { 19 struct drm_buddy_block *block; 20 21 BUG_ON(order > DRM_BUDDY_MAX_ORDER); 22 23 block = kmem_cache_zalloc(slab_blocks, GFP_KERNEL); 24 if (!block) 25 return NULL; 26 27 block->header = offset; 28 block->header |= order; 29 block->parent = parent; 30 31 BUG_ON(block->header & DRM_BUDDY_HEADER_UNUSED); 32 return block; 33 } 34 35 static void drm_block_free(struct drm_buddy *mm, 36 struct drm_buddy_block *block) 37 { 38 kmem_cache_free(slab_blocks, block); 39 } 40 41 static void list_insert_sorted(struct drm_buddy *mm, 42 struct drm_buddy_block *block) 43 { 44 struct drm_buddy_block *node; 45 struct list_head *head; 46 47 head = &mm->free_list[drm_buddy_block_order(block)]; 48 if (list_empty(head)) { 49 list_add(&block->link, head); 50 return; 51 } 52 53 list_for_each_entry(node, head, link) 54 if (drm_buddy_block_offset(block) < drm_buddy_block_offset(node)) 55 break; 56 57 __list_add(&block->link, node->link.prev, &node->link); 58 } 59 60 static void mark_allocated(struct drm_buddy_block *block) 61 { 62 block->header &= ~DRM_BUDDY_HEADER_STATE; 63 block->header |= DRM_BUDDY_ALLOCATED; 64 65 list_del(&block->link); 66 } 67 68 static void mark_free(struct drm_buddy *mm, 69 struct drm_buddy_block *block) 70 { 71 block->header &= ~DRM_BUDDY_HEADER_STATE; 72 block->header |= DRM_BUDDY_FREE; 73 74 list_insert_sorted(mm, block); 75 } 76 77 static void mark_split(struct drm_buddy_block *block) 78 { 79 block->header &= ~DRM_BUDDY_HEADER_STATE; 80 block->header |= DRM_BUDDY_SPLIT; 81 82 list_del(&block->link); 83 } 84 85 /** 86 * drm_buddy_init - init memory manager 87 * 88 * @mm: DRM buddy manager to initialize 89 * @size: size in bytes to manage 90 * @chunk_size: minimum page size in bytes for our allocations 91 * 92 * Initializes the memory manager and its resources. 93 * 94 * Returns: 95 * 0 on success, error code on failure. 96 */ 97 int drm_buddy_init(struct drm_buddy *mm, u64 size, u64 chunk_size) 98 { 99 unsigned int i; 100 u64 offset; 101 102 if (size < chunk_size) 103 return -EINVAL; 104 105 if (chunk_size < PAGE_SIZE) 106 return -EINVAL; 107 108 if (!is_power_of_2(chunk_size)) 109 return -EINVAL; 110 111 size = round_down(size, chunk_size); 112 113 mm->size = size; 114 mm->avail = size; 115 mm->chunk_size = chunk_size; 116 mm->max_order = ilog2(size) - ilog2(chunk_size); 117 118 BUG_ON(mm->max_order > DRM_BUDDY_MAX_ORDER); 119 120 mm->free_list = kmalloc_array(mm->max_order + 1, 121 sizeof(struct list_head), 122 GFP_KERNEL); 123 if (!mm->free_list) 124 return -ENOMEM; 125 126 for (i = 0; i <= mm->max_order; ++i) 127 INIT_LIST_HEAD(&mm->free_list[i]); 128 129 mm->n_roots = hweight64(size); 130 131 mm->roots = kmalloc_array(mm->n_roots, 132 sizeof(struct drm_buddy_block *), 133 GFP_KERNEL); 134 if (!mm->roots) 135 goto out_free_list; 136 137 offset = 0; 138 i = 0; 139 140 /* 141 * Split into power-of-two blocks, in case we are given a size that is 142 * not itself a power-of-two. 143 */ 144 do { 145 struct drm_buddy_block *root; 146 unsigned int order; 147 u64 root_size; 148 149 order = ilog2(size) - ilog2(chunk_size); 150 root_size = chunk_size << order; 151 152 root = drm_block_alloc(mm, NULL, order, offset); 153 if (!root) 154 goto out_free_roots; 155 156 mark_free(mm, root); 157 158 BUG_ON(i > mm->max_order); 159 BUG_ON(drm_buddy_block_size(mm, root) < chunk_size); 160 161 mm->roots[i] = root; 162 163 offset += root_size; 164 size -= root_size; 165 i++; 166 } while (size); 167 168 return 0; 169 170 out_free_roots: 171 while (i--) 172 drm_block_free(mm, mm->roots[i]); 173 kfree(mm->roots); 174 out_free_list: 175 kfree(mm->free_list); 176 return -ENOMEM; 177 } 178 EXPORT_SYMBOL(drm_buddy_init); 179 180 /** 181 * drm_buddy_fini - tear down the memory manager 182 * 183 * @mm: DRM buddy manager to free 184 * 185 * Cleanup memory manager resources and the freelist 186 */ 187 void drm_buddy_fini(struct drm_buddy *mm) 188 { 189 int i; 190 191 for (i = 0; i < mm->n_roots; ++i) { 192 WARN_ON(!drm_buddy_block_is_free(mm->roots[i])); 193 drm_block_free(mm, mm->roots[i]); 194 } 195 196 WARN_ON(mm->avail != mm->size); 197 198 kfree(mm->roots); 199 kfree(mm->free_list); 200 } 201 EXPORT_SYMBOL(drm_buddy_fini); 202 203 static int split_block(struct drm_buddy *mm, 204 struct drm_buddy_block *block) 205 { 206 unsigned int block_order = drm_buddy_block_order(block) - 1; 207 u64 offset = drm_buddy_block_offset(block); 208 209 BUG_ON(!drm_buddy_block_is_free(block)); 210 BUG_ON(!drm_buddy_block_order(block)); 211 212 block->left = drm_block_alloc(mm, block, block_order, offset); 213 if (!block->left) 214 return -ENOMEM; 215 216 block->right = drm_block_alloc(mm, block, block_order, 217 offset + (mm->chunk_size << block_order)); 218 if (!block->right) { 219 drm_block_free(mm, block->left); 220 return -ENOMEM; 221 } 222 223 mark_free(mm, block->left); 224 mark_free(mm, block->right); 225 226 mark_split(block); 227 228 return 0; 229 } 230 231 static struct drm_buddy_block * 232 __get_buddy(struct drm_buddy_block *block) 233 { 234 struct drm_buddy_block *parent; 235 236 parent = block->parent; 237 if (!parent) 238 return NULL; 239 240 if (parent->left == block) 241 return parent->right; 242 243 return parent->left; 244 } 245 246 /** 247 * drm_get_buddy - get buddy address 248 * 249 * @block: DRM buddy block 250 * 251 * Returns the corresponding buddy block for @block, or NULL 252 * if this is a root block and can't be merged further. 253 * Requires some kind of locking to protect against 254 * any concurrent allocate and free operations. 255 */ 256 struct drm_buddy_block * 257 drm_get_buddy(struct drm_buddy_block *block) 258 { 259 return __get_buddy(block); 260 } 261 EXPORT_SYMBOL(drm_get_buddy); 262 263 static void __drm_buddy_free(struct drm_buddy *mm, 264 struct drm_buddy_block *block) 265 { 266 struct drm_buddy_block *parent; 267 268 while ((parent = block->parent)) { 269 struct drm_buddy_block *buddy; 270 271 buddy = __get_buddy(block); 272 273 if (!drm_buddy_block_is_free(buddy)) 274 break; 275 276 list_del(&buddy->link); 277 278 drm_block_free(mm, block); 279 drm_block_free(mm, buddy); 280 281 block = parent; 282 } 283 284 mark_free(mm, block); 285 } 286 287 /** 288 * drm_buddy_free_block - free a block 289 * 290 * @mm: DRM buddy manager 291 * @block: block to be freed 292 */ 293 void drm_buddy_free_block(struct drm_buddy *mm, 294 struct drm_buddy_block *block) 295 { 296 BUG_ON(!drm_buddy_block_is_allocated(block)); 297 mm->avail += drm_buddy_block_size(mm, block); 298 __drm_buddy_free(mm, block); 299 } 300 EXPORT_SYMBOL(drm_buddy_free_block); 301 302 /** 303 * drm_buddy_free_list - free blocks 304 * 305 * @mm: DRM buddy manager 306 * @objects: input list head to free blocks 307 */ 308 void drm_buddy_free_list(struct drm_buddy *mm, struct list_head *objects) 309 { 310 struct drm_buddy_block *block, *on; 311 312 list_for_each_entry_safe(block, on, objects, link) { 313 drm_buddy_free_block(mm, block); 314 cond_resched(); 315 } 316 INIT_LIST_HEAD(objects); 317 } 318 EXPORT_SYMBOL(drm_buddy_free_list); 319 320 static inline bool overlaps(u64 s1, u64 e1, u64 s2, u64 e2) 321 { 322 return s1 <= e2 && e1 >= s2; 323 } 324 325 static inline bool contains(u64 s1, u64 e1, u64 s2, u64 e2) 326 { 327 return s1 <= s2 && e1 >= e2; 328 } 329 330 static struct drm_buddy_block * 331 alloc_range_bias(struct drm_buddy *mm, 332 u64 start, u64 end, 333 unsigned int order) 334 { 335 struct drm_buddy_block *block; 336 struct drm_buddy_block *buddy; 337 LIST_HEAD(dfs); 338 int err; 339 int i; 340 341 end = end - 1; 342 343 for (i = 0; i < mm->n_roots; ++i) 344 list_add_tail(&mm->roots[i]->tmp_link, &dfs); 345 346 do { 347 u64 block_start; 348 u64 block_end; 349 350 block = list_first_entry_or_null(&dfs, 351 struct drm_buddy_block, 352 tmp_link); 353 if (!block) 354 break; 355 356 list_del(&block->tmp_link); 357 358 if (drm_buddy_block_order(block) < order) 359 continue; 360 361 block_start = drm_buddy_block_offset(block); 362 block_end = block_start + drm_buddy_block_size(mm, block) - 1; 363 364 if (!overlaps(start, end, block_start, block_end)) 365 continue; 366 367 if (drm_buddy_block_is_allocated(block)) 368 continue; 369 370 if (contains(start, end, block_start, block_end) && 371 order == drm_buddy_block_order(block)) { 372 /* 373 * Find the free block within the range. 374 */ 375 if (drm_buddy_block_is_free(block)) 376 return block; 377 378 continue; 379 } 380 381 if (!drm_buddy_block_is_split(block)) { 382 err = split_block(mm, block); 383 if (unlikely(err)) 384 goto err_undo; 385 } 386 387 list_add(&block->right->tmp_link, &dfs); 388 list_add(&block->left->tmp_link, &dfs); 389 } while (1); 390 391 return ERR_PTR(-ENOSPC); 392 393 err_undo: 394 /* 395 * We really don't want to leave around a bunch of split blocks, since 396 * bigger is better, so make sure we merge everything back before we 397 * free the allocated blocks. 398 */ 399 buddy = __get_buddy(block); 400 if (buddy && 401 (drm_buddy_block_is_free(block) && 402 drm_buddy_block_is_free(buddy))) 403 __drm_buddy_free(mm, block); 404 return ERR_PTR(err); 405 } 406 407 static struct drm_buddy_block * 408 get_maxblock(struct drm_buddy *mm, unsigned int order) 409 { 410 struct drm_buddy_block *max_block = NULL, *node; 411 unsigned int i; 412 413 for (i = order; i <= mm->max_order; ++i) { 414 if (!list_empty(&mm->free_list[i])) { 415 node = list_last_entry(&mm->free_list[i], 416 struct drm_buddy_block, 417 link); 418 if (!max_block) { 419 max_block = node; 420 continue; 421 } 422 423 if (drm_buddy_block_offset(node) > 424 drm_buddy_block_offset(max_block)) { 425 max_block = node; 426 } 427 } 428 } 429 430 return max_block; 431 } 432 433 static struct drm_buddy_block * 434 alloc_from_freelist(struct drm_buddy *mm, 435 unsigned int order, 436 unsigned long flags) 437 { 438 struct drm_buddy_block *block = NULL; 439 unsigned int tmp; 440 int err; 441 442 if (flags & DRM_BUDDY_TOPDOWN_ALLOCATION) { 443 block = get_maxblock(mm, order); 444 if (block) 445 /* Store the obtained block order */ 446 tmp = drm_buddy_block_order(block); 447 } else { 448 for (tmp = order; tmp <= mm->max_order; ++tmp) { 449 if (!list_empty(&mm->free_list[tmp])) { 450 block = list_last_entry(&mm->free_list[tmp], 451 struct drm_buddy_block, 452 link); 453 if (block) 454 break; 455 } 456 } 457 } 458 459 if (!block) 460 return ERR_PTR(-ENOSPC); 461 462 BUG_ON(!drm_buddy_block_is_free(block)); 463 464 while (tmp != order) { 465 err = split_block(mm, block); 466 if (unlikely(err)) 467 goto err_undo; 468 469 block = block->right; 470 tmp--; 471 } 472 return block; 473 474 err_undo: 475 if (tmp != order) 476 __drm_buddy_free(mm, block); 477 return ERR_PTR(err); 478 } 479 480 static int __alloc_range(struct drm_buddy *mm, 481 struct list_head *dfs, 482 u64 start, u64 size, 483 struct list_head *blocks, 484 u64 *total_allocated_on_err) 485 { 486 struct drm_buddy_block *block; 487 struct drm_buddy_block *buddy; 488 u64 total_allocated = 0; 489 LIST_HEAD(allocated); 490 u64 end; 491 int err; 492 493 end = start + size - 1; 494 495 do { 496 u64 block_start; 497 u64 block_end; 498 499 block = list_first_entry_or_null(dfs, 500 struct drm_buddy_block, 501 tmp_link); 502 if (!block) 503 break; 504 505 list_del(&block->tmp_link); 506 507 block_start = drm_buddy_block_offset(block); 508 block_end = block_start + drm_buddy_block_size(mm, block) - 1; 509 510 if (!overlaps(start, end, block_start, block_end)) 511 continue; 512 513 if (drm_buddy_block_is_allocated(block)) { 514 err = -ENOSPC; 515 goto err_free; 516 } 517 518 if (contains(start, end, block_start, block_end)) { 519 if (!drm_buddy_block_is_free(block)) { 520 err = -ENOSPC; 521 goto err_free; 522 } 523 524 mark_allocated(block); 525 total_allocated += drm_buddy_block_size(mm, block); 526 mm->avail -= drm_buddy_block_size(mm, block); 527 list_add_tail(&block->link, &allocated); 528 continue; 529 } 530 531 if (!drm_buddy_block_is_split(block)) { 532 err = split_block(mm, block); 533 if (unlikely(err)) 534 goto err_undo; 535 } 536 537 list_add(&block->right->tmp_link, dfs); 538 list_add(&block->left->tmp_link, dfs); 539 } while (1); 540 541 list_splice_tail(&allocated, blocks); 542 return 0; 543 544 err_undo: 545 /* 546 * We really don't want to leave around a bunch of split blocks, since 547 * bigger is better, so make sure we merge everything back before we 548 * free the allocated blocks. 549 */ 550 buddy = __get_buddy(block); 551 if (buddy && 552 (drm_buddy_block_is_free(block) && 553 drm_buddy_block_is_free(buddy))) 554 __drm_buddy_free(mm, block); 555 556 err_free: 557 if (err == -ENOSPC && total_allocated_on_err) { 558 list_splice_tail(&allocated, blocks); 559 *total_allocated_on_err = total_allocated; 560 } else { 561 drm_buddy_free_list(mm, &allocated); 562 } 563 564 return err; 565 } 566 567 static int __drm_buddy_alloc_range(struct drm_buddy *mm, 568 u64 start, 569 u64 size, 570 u64 *total_allocated_on_err, 571 struct list_head *blocks) 572 { 573 LIST_HEAD(dfs); 574 int i; 575 576 for (i = 0; i < mm->n_roots; ++i) 577 list_add_tail(&mm->roots[i]->tmp_link, &dfs); 578 579 return __alloc_range(mm, &dfs, start, size, 580 blocks, total_allocated_on_err); 581 } 582 583 static int __alloc_contig_try_harder(struct drm_buddy *mm, 584 u64 size, 585 u64 min_block_size, 586 struct list_head *blocks) 587 { 588 u64 rhs_offset, lhs_offset, lhs_size, filled; 589 struct drm_buddy_block *block; 590 struct list_head *list; 591 LIST_HEAD(blocks_lhs); 592 unsigned long pages; 593 unsigned int order; 594 u64 modify_size; 595 int err; 596 597 modify_size = rounddown_pow_of_two(size); 598 pages = modify_size >> ilog2(mm->chunk_size); 599 order = fls(pages) - 1; 600 if (order == 0) 601 return -ENOSPC; 602 603 list = &mm->free_list[order]; 604 if (list_empty(list)) 605 return -ENOSPC; 606 607 list_for_each_entry_reverse(block, list, link) { 608 /* Allocate blocks traversing RHS */ 609 rhs_offset = drm_buddy_block_offset(block); 610 err = __drm_buddy_alloc_range(mm, rhs_offset, size, 611 &filled, blocks); 612 if (!err || err != -ENOSPC) 613 return err; 614 615 lhs_size = max((size - filled), min_block_size); 616 if (!IS_ALIGNED(lhs_size, min_block_size)) 617 lhs_size = round_up(lhs_size, min_block_size); 618 619 /* Allocate blocks traversing LHS */ 620 lhs_offset = drm_buddy_block_offset(block) - lhs_size; 621 err = __drm_buddy_alloc_range(mm, lhs_offset, lhs_size, 622 NULL, &blocks_lhs); 623 if (!err) { 624 list_splice(&blocks_lhs, blocks); 625 return 0; 626 } else if (err != -ENOSPC) { 627 drm_buddy_free_list(mm, blocks); 628 return err; 629 } 630 /* Free blocks for the next iteration */ 631 drm_buddy_free_list(mm, blocks); 632 } 633 634 return -ENOSPC; 635 } 636 637 /** 638 * drm_buddy_block_trim - free unused pages 639 * 640 * @mm: DRM buddy manager 641 * @new_size: original size requested 642 * @blocks: Input and output list of allocated blocks. 643 * MUST contain single block as input to be trimmed. 644 * On success will contain the newly allocated blocks 645 * making up the @new_size. Blocks always appear in 646 * ascending order 647 * 648 * For contiguous allocation, we round up the size to the nearest 649 * power of two value, drivers consume *actual* size, so remaining 650 * portions are unused and can be optionally freed with this function 651 * 652 * Returns: 653 * 0 on success, error code on failure. 654 */ 655 int drm_buddy_block_trim(struct drm_buddy *mm, 656 u64 new_size, 657 struct list_head *blocks) 658 { 659 struct drm_buddy_block *parent; 660 struct drm_buddy_block *block; 661 LIST_HEAD(dfs); 662 u64 new_start; 663 int err; 664 665 if (!list_is_singular(blocks)) 666 return -EINVAL; 667 668 block = list_first_entry(blocks, 669 struct drm_buddy_block, 670 link); 671 672 if (WARN_ON(!drm_buddy_block_is_allocated(block))) 673 return -EINVAL; 674 675 if (new_size > drm_buddy_block_size(mm, block)) 676 return -EINVAL; 677 678 if (!new_size || !IS_ALIGNED(new_size, mm->chunk_size)) 679 return -EINVAL; 680 681 if (new_size == drm_buddy_block_size(mm, block)) 682 return 0; 683 684 list_del(&block->link); 685 mark_free(mm, block); 686 mm->avail += drm_buddy_block_size(mm, block); 687 688 /* Prevent recursively freeing this node */ 689 parent = block->parent; 690 block->parent = NULL; 691 692 new_start = drm_buddy_block_offset(block); 693 list_add(&block->tmp_link, &dfs); 694 err = __alloc_range(mm, &dfs, new_start, new_size, blocks, NULL); 695 if (err) { 696 mark_allocated(block); 697 mm->avail -= drm_buddy_block_size(mm, block); 698 list_add(&block->link, blocks); 699 } 700 701 block->parent = parent; 702 return err; 703 } 704 EXPORT_SYMBOL(drm_buddy_block_trim); 705 706 /** 707 * drm_buddy_alloc_blocks - allocate power-of-two blocks 708 * 709 * @mm: DRM buddy manager to allocate from 710 * @start: start of the allowed range for this block 711 * @end: end of the allowed range for this block 712 * @size: size of the allocation 713 * @min_block_size: alignment of the allocation 714 * @blocks: output list head to add allocated blocks 715 * @flags: DRM_BUDDY_*_ALLOCATION flags 716 * 717 * alloc_range_bias() called on range limitations, which traverses 718 * the tree and returns the desired block. 719 * 720 * alloc_from_freelist() called when *no* range restrictions 721 * are enforced, which picks the block from the freelist. 722 * 723 * Returns: 724 * 0 on success, error code on failure. 725 */ 726 int drm_buddy_alloc_blocks(struct drm_buddy *mm, 727 u64 start, u64 end, u64 size, 728 u64 min_block_size, 729 struct list_head *blocks, 730 unsigned long flags) 731 { 732 struct drm_buddy_block *block = NULL; 733 u64 original_size, original_min_size; 734 unsigned int min_order, order; 735 LIST_HEAD(allocated); 736 unsigned long pages; 737 int err; 738 739 if (size < mm->chunk_size) 740 return -EINVAL; 741 742 if (min_block_size < mm->chunk_size) 743 return -EINVAL; 744 745 if (!is_power_of_2(min_block_size)) 746 return -EINVAL; 747 748 if (!IS_ALIGNED(start | end | size, mm->chunk_size)) 749 return -EINVAL; 750 751 if (end > mm->size) 752 return -EINVAL; 753 754 if (range_overflows(start, size, mm->size)) 755 return -EINVAL; 756 757 /* Actual range allocation */ 758 if (start + size == end) 759 return __drm_buddy_alloc_range(mm, start, size, NULL, blocks); 760 761 original_size = size; 762 original_min_size = min_block_size; 763 764 /* Roundup the size to power of 2 */ 765 if (flags & DRM_BUDDY_CONTIGUOUS_ALLOCATION) { 766 size = roundup_pow_of_two(size); 767 min_block_size = size; 768 /* Align size value to min_block_size */ 769 } else if (!IS_ALIGNED(size, min_block_size)) { 770 size = round_up(size, min_block_size); 771 } 772 773 pages = size >> ilog2(mm->chunk_size); 774 order = fls(pages) - 1; 775 min_order = ilog2(min_block_size) - ilog2(mm->chunk_size); 776 777 do { 778 order = min(order, (unsigned int)fls(pages) - 1); 779 BUG_ON(order > mm->max_order); 780 BUG_ON(order < min_order); 781 782 do { 783 if (flags & DRM_BUDDY_RANGE_ALLOCATION) 784 /* Allocate traversing within the range */ 785 block = alloc_range_bias(mm, start, end, order); 786 else 787 /* Allocate from freelist */ 788 block = alloc_from_freelist(mm, order, flags); 789 790 if (!IS_ERR(block)) 791 break; 792 793 if (order-- == min_order) { 794 if (flags & DRM_BUDDY_CONTIGUOUS_ALLOCATION && 795 !(flags & DRM_BUDDY_RANGE_ALLOCATION)) 796 /* 797 * Try contiguous block allocation through 798 * try harder method 799 */ 800 return __alloc_contig_try_harder(mm, 801 original_size, 802 original_min_size, 803 blocks); 804 err = -ENOSPC; 805 goto err_free; 806 } 807 } while (1); 808 809 mark_allocated(block); 810 mm->avail -= drm_buddy_block_size(mm, block); 811 kmemleak_update_trace(block); 812 list_add_tail(&block->link, &allocated); 813 814 pages -= BIT(order); 815 816 if (!pages) 817 break; 818 } while (1); 819 820 /* Trim the allocated block to the required size */ 821 if (original_size != size) { 822 struct list_head *trim_list; 823 LIST_HEAD(temp); 824 u64 trim_size; 825 826 trim_list = &allocated; 827 trim_size = original_size; 828 829 if (!list_is_singular(&allocated)) { 830 block = list_last_entry(&allocated, typeof(*block), link); 831 list_move(&block->link, &temp); 832 trim_list = &temp; 833 trim_size = drm_buddy_block_size(mm, block) - 834 (size - original_size); 835 } 836 837 drm_buddy_block_trim(mm, 838 trim_size, 839 trim_list); 840 841 if (!list_empty(&temp)) 842 list_splice_tail(trim_list, &allocated); 843 } 844 845 list_splice_tail(&allocated, blocks); 846 return 0; 847 848 err_free: 849 drm_buddy_free_list(mm, &allocated); 850 return err; 851 } 852 EXPORT_SYMBOL(drm_buddy_alloc_blocks); 853 854 /** 855 * drm_buddy_block_print - print block information 856 * 857 * @mm: DRM buddy manager 858 * @block: DRM buddy block 859 * @p: DRM printer to use 860 */ 861 void drm_buddy_block_print(struct drm_buddy *mm, 862 struct drm_buddy_block *block, 863 struct drm_printer *p) 864 { 865 u64 start = drm_buddy_block_offset(block); 866 u64 size = drm_buddy_block_size(mm, block); 867 868 drm_printf(p, "%#018llx-%#018llx: %llu\n", start, start + size, size); 869 } 870 EXPORT_SYMBOL(drm_buddy_block_print); 871 872 /** 873 * drm_buddy_print - print allocator state 874 * 875 * @mm: DRM buddy manager 876 * @p: DRM printer to use 877 */ 878 void drm_buddy_print(struct drm_buddy *mm, struct drm_printer *p) 879 { 880 int order; 881 882 drm_printf(p, "chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB\n", 883 mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20); 884 885 for (order = mm->max_order; order >= 0; order--) { 886 struct drm_buddy_block *block; 887 u64 count = 0, free; 888 889 list_for_each_entry(block, &mm->free_list[order], link) { 890 BUG_ON(!drm_buddy_block_is_free(block)); 891 count++; 892 } 893 894 drm_printf(p, "order-%2d ", order); 895 896 free = count * (mm->chunk_size << order); 897 if (free < SZ_1M) 898 drm_printf(p, "free: %8llu KiB", free >> 10); 899 else 900 drm_printf(p, "free: %8llu MiB", free >> 20); 901 902 drm_printf(p, ", blocks: %llu\n", count); 903 } 904 } 905 EXPORT_SYMBOL(drm_buddy_print); 906 907 static void drm_buddy_module_exit(void) 908 { 909 kmem_cache_destroy(slab_blocks); 910 } 911 912 static int __init drm_buddy_module_init(void) 913 { 914 slab_blocks = KMEM_CACHE(drm_buddy_block, 0); 915 if (!slab_blocks) 916 return -ENOMEM; 917 918 return 0; 919 } 920 921 module_init(drm_buddy_module_init); 922 module_exit(drm_buddy_module_exit); 923 924 MODULE_DESCRIPTION("DRM Buddy Allocator"); 925 MODULE_LICENSE("Dual MIT/GPL"); 926