linux_dsm_epyc7002/lib/radix-tree.c

1440 lines
38 KiB
C
Raw Normal View History

/*
* Copyright (C) 2001 Momchil Velikov
* Portions Copyright (C) 2001 Christoph Hellwig
* Copyright (C) 2005 SGI, Christoph Lameter
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
* Copyright (C) 2006 Nick Piggin
radix-tree: introduce bit-optimized iterator A series of radix tree cleanups, and usage of them in the core pagecache code. Micro-benchmark: lookup 14 slots (typical page-vector size) in radix-tree there earch <step> slot filled and tagged before/after - nsec per full scan through tree * Intel Sandy Bridge i7-2620M 4Mb L3 New code always faster * AMD Athlon 6000+ 2x1Mb L2, without L3 New code generally faster, Minor degradation (marked with "*") for huge sparse trees * i386 on Sandy Bridge New code faster for common cases: tagged and dense trees. Some degradations for non-tagged lookup on sparse trees. Ideally, there might help __ffs() analog for searching first non-zero long element in array, gcc sometimes cannot optimize this loop corretly. Numbers: CPU: Intel Sandy Bridge i7-2620M 4Mb L3 radix-tree with 1024 slots: tagged lookup step 1 before 7156 after 3613 step 2 before 5399 after 2696 step 3 before 4779 after 1928 step 4 before 4456 after 1429 step 5 before 4292 after 1213 step 6 before 4183 after 1052 step 7 before 4157 after 951 step 8 before 4016 after 812 step 9 before 3952 after 851 step 10 before 3937 after 732 step 11 before 4023 after 709 step 12 before 3872 after 657 step 13 before 3892 after 633 step 14 before 3720 after 591 step 15 before 3879 after 578 step 16 before 3561 after 513 normal lookup step 1 before 4266 after 3301 step 2 before 2695 after 2129 step 3 before 2083 after 1712 step 4 before 1801 after 1534 step 5 before 1628 after 1313 step 6 before 1551 after 1263 step 7 before 1475 after 1185 step 8 before 1432 after 1167 step 9 before 1373 after 1092 step 10 before 1339 after 1134 step 11 before 1292 after 1056 step 12 before 1319 after 1030 step 13 before 1276 after 1004 step 14 before 1256 after 987 step 15 before 1228 after 992 step 16 before 1247 after 999 radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1086102841 after 674196409 step 2 before 816839155 after 498138306 step 7 before 599728907 after 240676762 step 15 before 555729253 after 185219677 step 63 before 606637748 after 128585664 step 64 before 608384432 after 102945089 step 65 before 596987114 after 123996019 step 128 before 304459225 after 56783056 step 256 before 158846855 after 31232481 step 512 before 86085652 after 18950595 step 12345 before 6517189 after 1674057 normal lookup step 1 before 626064869 after 544418266 step 2 before 418809975 after 336321473 step 7 before 242303598 after 207755560 step 15 before 208380563 after 176496355 step 63 before 186854206 after 167283638 step 64 before 176188060 after 170143976 step 65 before 185139608 after 167487116 step 128 before 88181865 after 86913490 step 256 before 45733628 after 45143534 step 512 before 24506038 after 23859036 step 12345 before 2177425 after 2018662 * AMD Athlon 6000+ 2x1Mb L2, without L3 radix-tree with 1024 slots: tag-lookup step 1 before 8164 after 5379 step 2 before 5818 after 5581 step 3 before 4959 after 4213 step 4 before 4371 after 3386 step 5 before 4204 after 2997 step 6 before 4950 after 2744 step 7 before 4598 after 2480 step 8 before 4251 after 2288 step 9 before 4262 after 2243 step 10 before 4175 after 2131 step 11 before 3999 after 2024 step 12 before 3979 after 1994 step 13 before 3842 after 1929 step 14 before 3750 after 1810 step 15 before 3735 after 1810 step 16 before 3532 after 1660 normal-lookup step 1 before 7875 after 5847 step 2 before 4808 after 4071 step 3 before 4073 after 3462 step 4 before 3677 after 3074 step 5 before 4308 after 2978 step 6 before 3911 after 3807 step 7 before 3635 after 3522 step 8 before 3313 after 3202 step 9 before 3280 after 3257 step 10 before 3166 after 3083 step 11 before 3066 after 3026 step 12 before 2985 after 2982 step 13 before 2925 after 2924 step 14 before 2834 after 2808 step 15 before 2805 after 2803 step 16 before 2647 after 2622 radix-tree with 1024*1024*128 slots: tag-lookup step 1 before 1288059720 after 951736580 step 2 before 961292300 after 884212140 step 7 before 768905140 after 547267580 step 15 before 771319480 after 456550640 step 63 before 504847640 after 242704304 step 64 before 392484800 after 177920786 step 65 before 491162160 after 246895264 step 128 before 208084064 after 97348392 step 256 before 112401035 after 51408126 step 512 before 75825834 after 29145070 step 12345 before 5603166 after 2847330 normal-lookup step 1 before 1025677120 after 861375100 step 2 before 647220080 after 572258540 step 7 before 505518960 after 484041813 step 15 before 430483053 after 444815320 * step 63 before 388113453 after 404250546 * step 64 before 374154666 after 396027440 * step 65 before 381423973 after 396704853 * step 128 before 190078700 after 202619384 * step 256 before 100886756 after 102829108 * step 512 before 64074505 after 56158720 step 12345 before 4237289 after 4422299 * * i686 on Sandy bridge radix-tree with 1024 slots: tagged lookup step 1 before 7990 after 4019 step 2 before 5698 after 2897 step 3 before 5013 after 2475 step 4 before 4630 after 1721 step 5 before 4346 after 1759 step 6 before 4299 after 1556 step 7 before 4098 after 1513 step 8 before 4115 after 1222 step 9 before 3983 after 1390 step 10 before 4077 after 1207 step 11 before 3921 after 1231 step 12 before 3894 after 1116 step 13 before 3840 after 1147 step 14 before 3799 after 1090 step 15 before 3797 after 1059 step 16 before 3783 after 745 normal lookup step 1 before 5103 after 3499 step 2 before 3299 after 2550 step 3 before 2489 after 2370 step 4 before 2034 after 2302 * step 5 before 1846 after 2268 * step 6 before 1752 after 2249 * step 7 before 1679 after 2164 * step 8 before 1627 after 2153 * step 9 before 1542 after 2095 * step 10 before 1479 after 2109 * step 11 before 1469 after 2009 * step 12 before 1445 after 2039 * step 13 before 1411 after 2013 * step 14 before 1374 after 2046 * step 15 before 1340 after 1975 * step 16 before 1331 after 2000 * radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1225865377 after 667153553 step 2 before 842427423 after 471533007 step 7 before 609296153 after 276260116 step 15 before 544232060 after 226859105 step 63 before 519209199 after 141343043 step 64 before 588980279 after 141951339 step 65 before 521099710 after 138282060 step 128 before 298476778 after 83390628 step 256 before 149358342 after 43602609 step 512 before 76994713 after 22911077 step 12345 before 5328666 after 1472111 normal lookup step 1 before 819284564 after 533635310 step 2 before 512421605 after 364956155 step 7 before 271443305 after 305721345 * step 15 before 223591630 after 273960216 * step 63 before 190320247 after 217770207 * step 64 before 178538168 after 267411372 * step 65 before 186400423 after 215347937 * step 128 before 88106045 after 140540612 * step 256 before 44812420 after 70660377 * step 512 before 24435438 after 36328275 * step 12345 before 2123924 after 2148062 * bloat-o-meter delta for this patchset + patchset with related shmem cleanups bloat-o-meter: x86_64 add/remove: 4/3 grow/shrink: 5/6 up/down: 928/-939 (-11) function old new delta radix_tree_next_chunk - 499 +499 shmem_unuse 428 554 +126 shmem_radix_tree_replace 131 227 +96 find_get_pages_tag 354 419 +65 find_get_pages_contig 345 407 +62 find_get_pages 362 396 +34 __kstrtab_radix_tree_next_chunk - 22 +22 __ksymtab_radix_tree_next_chunk - 16 +16 __kcrctab_radix_tree_next_chunk - 8 +8 radix_tree_gang_lookup_slot 204 203 -1 static.shmem_xattr_set 384 381 -3 radix_tree_gang_lookup_tag_slot 208 191 -17 radix_tree_gang_lookup 231 187 -44 radix_tree_gang_lookup_tag 247 199 -48 shmem_unlock_mapping 278 190 -88 __lookup 217 - -217 __lookup_tag 242 - -242 radix_tree_locate_item 279 - -279 bloat-o-meter: i386 add/remove: 3/3 grow/shrink: 8/9 up/down: 1075/-1275 (-200) function old new delta radix_tree_next_chunk - 757 +757 shmem_unuse 352 449 +97 find_get_pages_contig 269 322 +53 shmem_radix_tree_replace 113 154 +41 find_get_pages_tag 277 318 +41 dcache_dir_lseek 426 458 +32 __kstrtab_radix_tree_next_chunk - 22 +22 vc_do_resize 968 977 +9 snd_pcm_lib_read1 725 733 +8 __ksymtab_radix_tree_next_chunk - 8 +8 netlbl_cipsov4_list 1120 1127 +7 find_get_pages 293 291 -2 new_slab 467 459 -8 bitfill_unaligned_rev 425 417 -8 radix_tree_gang_lookup_tag_slot 177 146 -31 blk_dump_cmd 267 229 -38 radix_tree_gang_lookup_slot 212 134 -78 shmem_unlock_mapping 221 128 -93 radix_tree_gang_lookup_tag 275 162 -113 radix_tree_gang_lookup 255 126 -129 __lookup 227 - -227 __lookup_tag 271 - -271 radix_tree_locate_item 277 - -277 This patch: Implement a clean, simple and effective radix-tree iteration routine. Iterating divided into two phases: * lookup next chunk in radix-tree leaf node * iterating through slots in this chunk Main iterator function radix_tree_next_chunk() returns pointer to first slot, and stores in the struct radix_tree_iter index of next-to-last slot. For tagged-iterating it also constuct bitmask of tags for retunted chunk. All additional logic implemented as static-inline functions and macroses. Also adds radix_tree_find_next_bit() static-inline variant of find_next_bit() optimized for small constant size arrays, because find_next_bit() too heavy for searching in an array with one/two long elements. [akpm@linux-foundation.org: rework comments a bit] Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org> Tested-by: Hugh Dickins <hughd@google.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-03-29 04:42:53 +07:00
* Copyright (C) 2012 Konstantin Khlebnikov
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation; either version 2, or (at
* your option) any later version.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include <linux/errno.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/radix-tree.h>
#include <linux/percpu.h>
#include <linux/slab.h>
#include <linux/notifier.h>
#include <linux/cpu.h>
#include <linux/string.h>
#include <linux/bitops.h>
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
#include <linux/rcupdate.h>
#ifdef __KERNEL__
#define RADIX_TREE_MAP_SHIFT (CONFIG_BASE_SMALL ? 4 : 6)
#else
#define RADIX_TREE_MAP_SHIFT 3 /* For more stressful testing */
#endif
#define RADIX_TREE_MAP_SIZE (1UL << RADIX_TREE_MAP_SHIFT)
#define RADIX_TREE_MAP_MASK (RADIX_TREE_MAP_SIZE-1)
#define RADIX_TREE_TAG_LONGS \
((RADIX_TREE_MAP_SIZE + BITS_PER_LONG - 1) / BITS_PER_LONG)
struct radix_tree_node {
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
unsigned int height; /* Height from the bottom */
unsigned int count;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
union {
struct radix_tree_node *parent; /* Used when ascending tree */
struct rcu_head rcu_head; /* Used when freeing node */
};
void __rcu *slots[RADIX_TREE_MAP_SIZE];
unsigned long tags[RADIX_TREE_MAX_TAGS][RADIX_TREE_TAG_LONGS];
};
#define RADIX_TREE_INDEX_BITS (8 /* CHAR_BIT */ * sizeof(unsigned long))
#define RADIX_TREE_MAX_PATH (DIV_ROUND_UP(RADIX_TREE_INDEX_BITS, \
RADIX_TREE_MAP_SHIFT))
/*
* The height_to_maxindex array needs to be one deeper than the maximum
* path as height 0 holds only 1 entry.
*/
static unsigned long height_to_maxindex[RADIX_TREE_MAX_PATH + 1] __read_mostly;
/*
* Radix tree node cache.
*/
static struct kmem_cache *radix_tree_node_cachep;
/*
* Per-cpu pool of preloaded nodes
*/
struct radix_tree_preload {
int nr;
struct radix_tree_node *nodes[RADIX_TREE_MAX_PATH];
};
static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads) = { 0, };
static inline void *ptr_to_indirect(void *ptr)
{
return (void *)((unsigned long)ptr | RADIX_TREE_INDIRECT_PTR);
}
static inline void *indirect_to_ptr(void *ptr)
{
return (void *)((unsigned long)ptr & ~RADIX_TREE_INDIRECT_PTR);
}
static inline gfp_t root_gfp_mask(struct radix_tree_root *root)
{
return root->gfp_mask & __GFP_BITS_MASK;
}
radix-tree: fix small lockless radix-tree bug We shrink a radix tree when its root node has only one child, in the left most slot. The child becomes the new root node. To perform this operation in a manner compatible with concurrent lockless lookups, we atomically switch the root pointer from the parent to its child. However a concurrent lockless lookup may now have loaded a pointer to the parent (and is presently deciding what to do next). For this reason, we also have to keep the parent node in a valid state after shrinking the tree, until the next RCU grace period -- otherwise this lookup with the parent pointer may not do the right thing. Notably, we need to keep the child in the left most slot there in case that is requested by the lookup. This is all pretty standard RCU stuff. It is worth repeating because in my eagerness to obey the radix tree node constructor scheme, I had broken it by zeroing the radix tree node before the grace period. What could happen is that a lookup can load the parent pointer, then decide it wants to follow the left most child slot, only to find the slot contained NULL due to the concurrent shrinker having zeroed the parent node before waiting for a grace period. The lookup would return a false negative as a result. Fix it by doing that clearing in the RCU callback. I would normally want to rip out the constructor entirely, but radix tree nodes are one of those places where they make sense (only few cachelines will be touched soon after allocation). This was never actually found in any lockless pagecache testing or by the test harness, but by seeing the odd problem with my scalable vmap rewrite. I have not tickled the test harness into reproducing it yet, but I'll keep working at it. Fortunately, it is not a problem anywhere lockless pagecache is used in mainline kernels (pagecache probe is not a guarantee, and brd does not have concurrent lookups and deletes). Signed-off-by: Nick Piggin <npiggin@suse.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-06-13 05:21:52 +07:00
static inline void tag_set(struct radix_tree_node *node, unsigned int tag,
int offset)
{
__set_bit(offset, node->tags[tag]);
}
static inline void tag_clear(struct radix_tree_node *node, unsigned int tag,
int offset)
{
__clear_bit(offset, node->tags[tag]);
}
static inline int tag_get(struct radix_tree_node *node, unsigned int tag,
int offset)
{
return test_bit(offset, node->tags[tag]);
}
static inline void root_tag_set(struct radix_tree_root *root, unsigned int tag)
{
root->gfp_mask |= (__force gfp_t)(1 << (tag + __GFP_BITS_SHIFT));
}
static inline void root_tag_clear(struct radix_tree_root *root, unsigned int tag)
{
root->gfp_mask &= (__force gfp_t)~(1 << (tag + __GFP_BITS_SHIFT));
}
static inline void root_tag_clear_all(struct radix_tree_root *root)
{
root->gfp_mask &= __GFP_BITS_MASK;
}
static inline int root_tag_get(struct radix_tree_root *root, unsigned int tag)
{
return (__force unsigned)root->gfp_mask & (1 << (tag + __GFP_BITS_SHIFT));
}
/*
* Returns 1 if any slot in the node has this tag set.
* Otherwise returns 0.
*/
static inline int any_tag_set(struct radix_tree_node *node, unsigned int tag)
{
int idx;
for (idx = 0; idx < RADIX_TREE_TAG_LONGS; idx++) {
if (node->tags[tag][idx])
return 1;
}
return 0;
}
radix-tree: introduce bit-optimized iterator A series of radix tree cleanups, and usage of them in the core pagecache code. Micro-benchmark: lookup 14 slots (typical page-vector size) in radix-tree there earch <step> slot filled and tagged before/after - nsec per full scan through tree * Intel Sandy Bridge i7-2620M 4Mb L3 New code always faster * AMD Athlon 6000+ 2x1Mb L2, without L3 New code generally faster, Minor degradation (marked with "*") for huge sparse trees * i386 on Sandy Bridge New code faster for common cases: tagged and dense trees. Some degradations for non-tagged lookup on sparse trees. Ideally, there might help __ffs() analog for searching first non-zero long element in array, gcc sometimes cannot optimize this loop corretly. Numbers: CPU: Intel Sandy Bridge i7-2620M 4Mb L3 radix-tree with 1024 slots: tagged lookup step 1 before 7156 after 3613 step 2 before 5399 after 2696 step 3 before 4779 after 1928 step 4 before 4456 after 1429 step 5 before 4292 after 1213 step 6 before 4183 after 1052 step 7 before 4157 after 951 step 8 before 4016 after 812 step 9 before 3952 after 851 step 10 before 3937 after 732 step 11 before 4023 after 709 step 12 before 3872 after 657 step 13 before 3892 after 633 step 14 before 3720 after 591 step 15 before 3879 after 578 step 16 before 3561 after 513 normal lookup step 1 before 4266 after 3301 step 2 before 2695 after 2129 step 3 before 2083 after 1712 step 4 before 1801 after 1534 step 5 before 1628 after 1313 step 6 before 1551 after 1263 step 7 before 1475 after 1185 step 8 before 1432 after 1167 step 9 before 1373 after 1092 step 10 before 1339 after 1134 step 11 before 1292 after 1056 step 12 before 1319 after 1030 step 13 before 1276 after 1004 step 14 before 1256 after 987 step 15 before 1228 after 992 step 16 before 1247 after 999 radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1086102841 after 674196409 step 2 before 816839155 after 498138306 step 7 before 599728907 after 240676762 step 15 before 555729253 after 185219677 step 63 before 606637748 after 128585664 step 64 before 608384432 after 102945089 step 65 before 596987114 after 123996019 step 128 before 304459225 after 56783056 step 256 before 158846855 after 31232481 step 512 before 86085652 after 18950595 step 12345 before 6517189 after 1674057 normal lookup step 1 before 626064869 after 544418266 step 2 before 418809975 after 336321473 step 7 before 242303598 after 207755560 step 15 before 208380563 after 176496355 step 63 before 186854206 after 167283638 step 64 before 176188060 after 170143976 step 65 before 185139608 after 167487116 step 128 before 88181865 after 86913490 step 256 before 45733628 after 45143534 step 512 before 24506038 after 23859036 step 12345 before 2177425 after 2018662 * AMD Athlon 6000+ 2x1Mb L2, without L3 radix-tree with 1024 slots: tag-lookup step 1 before 8164 after 5379 step 2 before 5818 after 5581 step 3 before 4959 after 4213 step 4 before 4371 after 3386 step 5 before 4204 after 2997 step 6 before 4950 after 2744 step 7 before 4598 after 2480 step 8 before 4251 after 2288 step 9 before 4262 after 2243 step 10 before 4175 after 2131 step 11 before 3999 after 2024 step 12 before 3979 after 1994 step 13 before 3842 after 1929 step 14 before 3750 after 1810 step 15 before 3735 after 1810 step 16 before 3532 after 1660 normal-lookup step 1 before 7875 after 5847 step 2 before 4808 after 4071 step 3 before 4073 after 3462 step 4 before 3677 after 3074 step 5 before 4308 after 2978 step 6 before 3911 after 3807 step 7 before 3635 after 3522 step 8 before 3313 after 3202 step 9 before 3280 after 3257 step 10 before 3166 after 3083 step 11 before 3066 after 3026 step 12 before 2985 after 2982 step 13 before 2925 after 2924 step 14 before 2834 after 2808 step 15 before 2805 after 2803 step 16 before 2647 after 2622 radix-tree with 1024*1024*128 slots: tag-lookup step 1 before 1288059720 after 951736580 step 2 before 961292300 after 884212140 step 7 before 768905140 after 547267580 step 15 before 771319480 after 456550640 step 63 before 504847640 after 242704304 step 64 before 392484800 after 177920786 step 65 before 491162160 after 246895264 step 128 before 208084064 after 97348392 step 256 before 112401035 after 51408126 step 512 before 75825834 after 29145070 step 12345 before 5603166 after 2847330 normal-lookup step 1 before 1025677120 after 861375100 step 2 before 647220080 after 572258540 step 7 before 505518960 after 484041813 step 15 before 430483053 after 444815320 * step 63 before 388113453 after 404250546 * step 64 before 374154666 after 396027440 * step 65 before 381423973 after 396704853 * step 128 before 190078700 after 202619384 * step 256 before 100886756 after 102829108 * step 512 before 64074505 after 56158720 step 12345 before 4237289 after 4422299 * * i686 on Sandy bridge radix-tree with 1024 slots: tagged lookup step 1 before 7990 after 4019 step 2 before 5698 after 2897 step 3 before 5013 after 2475 step 4 before 4630 after 1721 step 5 before 4346 after 1759 step 6 before 4299 after 1556 step 7 before 4098 after 1513 step 8 before 4115 after 1222 step 9 before 3983 after 1390 step 10 before 4077 after 1207 step 11 before 3921 after 1231 step 12 before 3894 after 1116 step 13 before 3840 after 1147 step 14 before 3799 after 1090 step 15 before 3797 after 1059 step 16 before 3783 after 745 normal lookup step 1 before 5103 after 3499 step 2 before 3299 after 2550 step 3 before 2489 after 2370 step 4 before 2034 after 2302 * step 5 before 1846 after 2268 * step 6 before 1752 after 2249 * step 7 before 1679 after 2164 * step 8 before 1627 after 2153 * step 9 before 1542 after 2095 * step 10 before 1479 after 2109 * step 11 before 1469 after 2009 * step 12 before 1445 after 2039 * step 13 before 1411 after 2013 * step 14 before 1374 after 2046 * step 15 before 1340 after 1975 * step 16 before 1331 after 2000 * radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1225865377 after 667153553 step 2 before 842427423 after 471533007 step 7 before 609296153 after 276260116 step 15 before 544232060 after 226859105 step 63 before 519209199 after 141343043 step 64 before 588980279 after 141951339 step 65 before 521099710 after 138282060 step 128 before 298476778 after 83390628 step 256 before 149358342 after 43602609 step 512 before 76994713 after 22911077 step 12345 before 5328666 after 1472111 normal lookup step 1 before 819284564 after 533635310 step 2 before 512421605 after 364956155 step 7 before 271443305 after 305721345 * step 15 before 223591630 after 273960216 * step 63 before 190320247 after 217770207 * step 64 before 178538168 after 267411372 * step 65 before 186400423 after 215347937 * step 128 before 88106045 after 140540612 * step 256 before 44812420 after 70660377 * step 512 before 24435438 after 36328275 * step 12345 before 2123924 after 2148062 * bloat-o-meter delta for this patchset + patchset with related shmem cleanups bloat-o-meter: x86_64 add/remove: 4/3 grow/shrink: 5/6 up/down: 928/-939 (-11) function old new delta radix_tree_next_chunk - 499 +499 shmem_unuse 428 554 +126 shmem_radix_tree_replace 131 227 +96 find_get_pages_tag 354 419 +65 find_get_pages_contig 345 407 +62 find_get_pages 362 396 +34 __kstrtab_radix_tree_next_chunk - 22 +22 __ksymtab_radix_tree_next_chunk - 16 +16 __kcrctab_radix_tree_next_chunk - 8 +8 radix_tree_gang_lookup_slot 204 203 -1 static.shmem_xattr_set 384 381 -3 radix_tree_gang_lookup_tag_slot 208 191 -17 radix_tree_gang_lookup 231 187 -44 radix_tree_gang_lookup_tag 247 199 -48 shmem_unlock_mapping 278 190 -88 __lookup 217 - -217 __lookup_tag 242 - -242 radix_tree_locate_item 279 - -279 bloat-o-meter: i386 add/remove: 3/3 grow/shrink: 8/9 up/down: 1075/-1275 (-200) function old new delta radix_tree_next_chunk - 757 +757 shmem_unuse 352 449 +97 find_get_pages_contig 269 322 +53 shmem_radix_tree_replace 113 154 +41 find_get_pages_tag 277 318 +41 dcache_dir_lseek 426 458 +32 __kstrtab_radix_tree_next_chunk - 22 +22 vc_do_resize 968 977 +9 snd_pcm_lib_read1 725 733 +8 __ksymtab_radix_tree_next_chunk - 8 +8 netlbl_cipsov4_list 1120 1127 +7 find_get_pages 293 291 -2 new_slab 467 459 -8 bitfill_unaligned_rev 425 417 -8 radix_tree_gang_lookup_tag_slot 177 146 -31 blk_dump_cmd 267 229 -38 radix_tree_gang_lookup_slot 212 134 -78 shmem_unlock_mapping 221 128 -93 radix_tree_gang_lookup_tag 275 162 -113 radix_tree_gang_lookup 255 126 -129 __lookup 227 - -227 __lookup_tag 271 - -271 radix_tree_locate_item 277 - -277 This patch: Implement a clean, simple and effective radix-tree iteration routine. Iterating divided into two phases: * lookup next chunk in radix-tree leaf node * iterating through slots in this chunk Main iterator function radix_tree_next_chunk() returns pointer to first slot, and stores in the struct radix_tree_iter index of next-to-last slot. For tagged-iterating it also constuct bitmask of tags for retunted chunk. All additional logic implemented as static-inline functions and macroses. Also adds radix_tree_find_next_bit() static-inline variant of find_next_bit() optimized for small constant size arrays, because find_next_bit() too heavy for searching in an array with one/two long elements. [akpm@linux-foundation.org: rework comments a bit] Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org> Tested-by: Hugh Dickins <hughd@google.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-03-29 04:42:53 +07:00
/**
* radix_tree_find_next_bit - find the next set bit in a memory region
*
* @addr: The address to base the search on
* @size: The bitmap size in bits
* @offset: The bitnumber to start searching at
*
* Unrollable variant of find_next_bit() for constant size arrays.
* Tail bits starting from size to roundup(size, BITS_PER_LONG) must be zero.
* Returns next bit offset, or size if nothing found.
*/
static __always_inline unsigned long
radix_tree_find_next_bit(const unsigned long *addr,
unsigned long size, unsigned long offset)
{
if (!__builtin_constant_p(size))
return find_next_bit(addr, size, offset);
if (offset < size) {
unsigned long tmp;
addr += offset / BITS_PER_LONG;
tmp = *addr >> (offset % BITS_PER_LONG);
if (tmp)
return __ffs(tmp) + offset;
offset = (offset + BITS_PER_LONG) & ~(BITS_PER_LONG - 1);
while (offset < size) {
tmp = *++addr;
if (tmp)
return __ffs(tmp) + offset;
offset += BITS_PER_LONG;
}
}
return size;
}
/*
* This assumes that the caller has performed appropriate preallocation, and
* that the caller has pinned this thread of control to the current CPU.
*/
static struct radix_tree_node *
radix_tree_node_alloc(struct radix_tree_root *root)
{
radix-tree: avoid atomic allocations for preloaded insertions Most pagecache (and some other) radix tree insertions have the great opportunity to preallocate a few nodes with relaxed gfp flags. But the preallocation is squandered when it comes time to allocate a node, we default to first attempting a GFP_ATOMIC allocation -- that doesn't normally fail, but it can eat into atomic memory reserves that we don't need to be using. Another upshot of this is that it removes the sometimes highly contended zone->lock from underneath tree_lock. Pagecache insertions are always performed with a radix tree preload, and after this change, such a situation will never fall back to kmem_cache_alloc within radix_tree_node_alloc. David Miller reports seeing this allocation fail on a highly threaded sparc64 system: [527319.459981] dd: page allocation failure. order:0, mode:0x20 [527319.460403] Call Trace: [527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8 [527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8 [527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90 [527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260 [527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0 [527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134 [527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280 [527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214 [527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428 [527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170 [527319.461217] [00000000004badac] do_sync_read+0x88/0xd0 [527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c [527319.461361] [00000000004bb920] sys_read+0x34/0x60 [527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40 The calltrace is significant: __do_page_cache_readahead allocates a number of pages with GFP_KERNEL, and hence it should have reclaimed sufficient memory to satisfy GFP_ATOMIC allocations. However after the list of pages goes to mpage_readpages, there can be significant intervals (including disk IO) before all the pages are inserted into the radix-tree. So the reserves can easily be depleted at that point. The patch is confirmed to fix the problem. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 13:29:10 +07:00
struct radix_tree_node *ret = NULL;
gfp_t gfp_mask = root_gfp_mask(root);
radix-tree: avoid atomic allocations for preloaded insertions Most pagecache (and some other) radix tree insertions have the great opportunity to preallocate a few nodes with relaxed gfp flags. But the preallocation is squandered when it comes time to allocate a node, we default to first attempting a GFP_ATOMIC allocation -- that doesn't normally fail, but it can eat into atomic memory reserves that we don't need to be using. Another upshot of this is that it removes the sometimes highly contended zone->lock from underneath tree_lock. Pagecache insertions are always performed with a radix tree preload, and after this change, such a situation will never fall back to kmem_cache_alloc within radix_tree_node_alloc. David Miller reports seeing this allocation fail on a highly threaded sparc64 system: [527319.459981] dd: page allocation failure. order:0, mode:0x20 [527319.460403] Call Trace: [527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8 [527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8 [527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90 [527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260 [527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0 [527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134 [527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280 [527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214 [527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428 [527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170 [527319.461217] [00000000004badac] do_sync_read+0x88/0xd0 [527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c [527319.461361] [00000000004bb920] sys_read+0x34/0x60 [527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40 The calltrace is significant: __do_page_cache_readahead allocates a number of pages with GFP_KERNEL, and hence it should have reclaimed sufficient memory to satisfy GFP_ATOMIC allocations. However after the list of pages goes to mpage_readpages, there can be significant intervals (including disk IO) before all the pages are inserted into the radix-tree. So the reserves can easily be depleted at that point. The patch is confirmed to fix the problem. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 13:29:10 +07:00
if (!(gfp_mask & __GFP_WAIT)) {
struct radix_tree_preload *rtp;
radix-tree: avoid atomic allocations for preloaded insertions Most pagecache (and some other) radix tree insertions have the great opportunity to preallocate a few nodes with relaxed gfp flags. But the preallocation is squandered when it comes time to allocate a node, we default to first attempting a GFP_ATOMIC allocation -- that doesn't normally fail, but it can eat into atomic memory reserves that we don't need to be using. Another upshot of this is that it removes the sometimes highly contended zone->lock from underneath tree_lock. Pagecache insertions are always performed with a radix tree preload, and after this change, such a situation will never fall back to kmem_cache_alloc within radix_tree_node_alloc. David Miller reports seeing this allocation fail on a highly threaded sparc64 system: [527319.459981] dd: page allocation failure. order:0, mode:0x20 [527319.460403] Call Trace: [527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8 [527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8 [527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90 [527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260 [527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0 [527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134 [527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280 [527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214 [527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428 [527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170 [527319.461217] [00000000004badac] do_sync_read+0x88/0xd0 [527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c [527319.461361] [00000000004bb920] sys_read+0x34/0x60 [527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40 The calltrace is significant: __do_page_cache_readahead allocates a number of pages with GFP_KERNEL, and hence it should have reclaimed sufficient memory to satisfy GFP_ATOMIC allocations. However after the list of pages goes to mpage_readpages, there can be significant intervals (including disk IO) before all the pages are inserted into the radix-tree. So the reserves can easily be depleted at that point. The patch is confirmed to fix the problem. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 13:29:10 +07:00
/*
* Provided the caller has preloaded here, we will always
* succeed in getting a node here (and never reach
* kmem_cache_alloc)
*/
rtp = &__get_cpu_var(radix_tree_preloads);
if (rtp->nr) {
ret = rtp->nodes[rtp->nr - 1];
rtp->nodes[rtp->nr - 1] = NULL;
rtp->nr--;
}
}
radix-tree: avoid atomic allocations for preloaded insertions Most pagecache (and some other) radix tree insertions have the great opportunity to preallocate a few nodes with relaxed gfp flags. But the preallocation is squandered when it comes time to allocate a node, we default to first attempting a GFP_ATOMIC allocation -- that doesn't normally fail, but it can eat into atomic memory reserves that we don't need to be using. Another upshot of this is that it removes the sometimes highly contended zone->lock from underneath tree_lock. Pagecache insertions are always performed with a radix tree preload, and after this change, such a situation will never fall back to kmem_cache_alloc within radix_tree_node_alloc. David Miller reports seeing this allocation fail on a highly threaded sparc64 system: [527319.459981] dd: page allocation failure. order:0, mode:0x20 [527319.460403] Call Trace: [527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8 [527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8 [527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90 [527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260 [527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0 [527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134 [527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280 [527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214 [527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428 [527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170 [527319.461217] [00000000004badac] do_sync_read+0x88/0xd0 [527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c [527319.461361] [00000000004bb920] sys_read+0x34/0x60 [527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40 The calltrace is significant: __do_page_cache_readahead allocates a number of pages with GFP_KERNEL, and hence it should have reclaimed sufficient memory to satisfy GFP_ATOMIC allocations. However after the list of pages goes to mpage_readpages, there can be significant intervals (including disk IO) before all the pages are inserted into the radix-tree. So the reserves can easily be depleted at that point. The patch is confirmed to fix the problem. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 13:29:10 +07:00
if (ret == NULL)
ret = kmem_cache_alloc(radix_tree_node_cachep, gfp_mask);
radix-tree: avoid atomic allocations for preloaded insertions Most pagecache (and some other) radix tree insertions have the great opportunity to preallocate a few nodes with relaxed gfp flags. But the preallocation is squandered when it comes time to allocate a node, we default to first attempting a GFP_ATOMIC allocation -- that doesn't normally fail, but it can eat into atomic memory reserves that we don't need to be using. Another upshot of this is that it removes the sometimes highly contended zone->lock from underneath tree_lock. Pagecache insertions are always performed with a radix tree preload, and after this change, such a situation will never fall back to kmem_cache_alloc within radix_tree_node_alloc. David Miller reports seeing this allocation fail on a highly threaded sparc64 system: [527319.459981] dd: page allocation failure. order:0, mode:0x20 [527319.460403] Call Trace: [527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8 [527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8 [527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90 [527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260 [527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0 [527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134 [527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280 [527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214 [527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428 [527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170 [527319.461217] [00000000004badac] do_sync_read+0x88/0xd0 [527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c [527319.461361] [00000000004bb920] sys_read+0x34/0x60 [527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40 The calltrace is significant: __do_page_cache_readahead allocates a number of pages with GFP_KERNEL, and hence it should have reclaimed sufficient memory to satisfy GFP_ATOMIC allocations. However after the list of pages goes to mpage_readpages, there can be significant intervals (including disk IO) before all the pages are inserted into the radix-tree. So the reserves can easily be depleted at that point. The patch is confirmed to fix the problem. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-05 13:29:10 +07:00
BUG_ON(radix_tree_is_indirect_ptr(ret));
return ret;
}
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
static void radix_tree_node_rcu_free(struct rcu_head *head)
{
struct radix_tree_node *node =
container_of(head, struct radix_tree_node, rcu_head);
int i;
radix-tree: fix small lockless radix-tree bug We shrink a radix tree when its root node has only one child, in the left most slot. The child becomes the new root node. To perform this operation in a manner compatible with concurrent lockless lookups, we atomically switch the root pointer from the parent to its child. However a concurrent lockless lookup may now have loaded a pointer to the parent (and is presently deciding what to do next). For this reason, we also have to keep the parent node in a valid state after shrinking the tree, until the next RCU grace period -- otherwise this lookup with the parent pointer may not do the right thing. Notably, we need to keep the child in the left most slot there in case that is requested by the lookup. This is all pretty standard RCU stuff. It is worth repeating because in my eagerness to obey the radix tree node constructor scheme, I had broken it by zeroing the radix tree node before the grace period. What could happen is that a lookup can load the parent pointer, then decide it wants to follow the left most child slot, only to find the slot contained NULL due to the concurrent shrinker having zeroed the parent node before waiting for a grace period. The lookup would return a false negative as a result. Fix it by doing that clearing in the RCU callback. I would normally want to rip out the constructor entirely, but radix tree nodes are one of those places where they make sense (only few cachelines will be touched soon after allocation). This was never actually found in any lockless pagecache testing or by the test harness, but by seeing the odd problem with my scalable vmap rewrite. I have not tickled the test harness into reproducing it yet, but I'll keep working at it. Fortunately, it is not a problem anywhere lockless pagecache is used in mainline kernels (pagecache probe is not a guarantee, and brd does not have concurrent lookups and deletes). Signed-off-by: Nick Piggin <npiggin@suse.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-06-13 05:21:52 +07:00
/*
* must only free zeroed nodes into the slab. radix_tree_shrink
* can leave us with a non-NULL entry in the first slot, so clear
* that here to make sure.
*/
for (i = 0; i < RADIX_TREE_MAX_TAGS; i++)
tag_clear(node, i, 0);
radix-tree: fix small lockless radix-tree bug We shrink a radix tree when its root node has only one child, in the left most slot. The child becomes the new root node. To perform this operation in a manner compatible with concurrent lockless lookups, we atomically switch the root pointer from the parent to its child. However a concurrent lockless lookup may now have loaded a pointer to the parent (and is presently deciding what to do next). For this reason, we also have to keep the parent node in a valid state after shrinking the tree, until the next RCU grace period -- otherwise this lookup with the parent pointer may not do the right thing. Notably, we need to keep the child in the left most slot there in case that is requested by the lookup. This is all pretty standard RCU stuff. It is worth repeating because in my eagerness to obey the radix tree node constructor scheme, I had broken it by zeroing the radix tree node before the grace period. What could happen is that a lookup can load the parent pointer, then decide it wants to follow the left most child slot, only to find the slot contained NULL due to the concurrent shrinker having zeroed the parent node before waiting for a grace period. The lookup would return a false negative as a result. Fix it by doing that clearing in the RCU callback. I would normally want to rip out the constructor entirely, but radix tree nodes are one of those places where they make sense (only few cachelines will be touched soon after allocation). This was never actually found in any lockless pagecache testing or by the test harness, but by seeing the odd problem with my scalable vmap rewrite. I have not tickled the test harness into reproducing it yet, but I'll keep working at it. Fortunately, it is not a problem anywhere lockless pagecache is used in mainline kernels (pagecache probe is not a guarantee, and brd does not have concurrent lookups and deletes). Signed-off-by: Nick Piggin <npiggin@suse.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-06-13 05:21:52 +07:00
node->slots[0] = NULL;
node->count = 0;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
kmem_cache_free(radix_tree_node_cachep, node);
}
static inline void
radix_tree_node_free(struct radix_tree_node *node)
{
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
call_rcu(&node->rcu_head, radix_tree_node_rcu_free);
}
/*
* Load up this CPU's radix_tree_node buffer with sufficient objects to
* ensure that the addition of a single element in the tree cannot fail. On
* success, return zero, with preemption disabled. On error, return -ENOMEM
* with preemption not disabled.
FS-Cache: Use radix tree preload correctly in tracking of pages to be stored __fscache_write_page() attempts to load the radix tree preallocation pool for the CPU it is on before calling radix_tree_insert(), as the insertion must be done inside a pair of spinlocks. Use of the preallocation pool, however, is contingent on the radix tree being initialised without __GFP_WAIT specified. __fscache_acquire_cookie() was passing GFP_NOFS to INIT_RADIX_TREE() - but that includes __GFP_WAIT. The solution is to AND out __GFP_WAIT. Additionally, the banner comment to radix_tree_preload() is altered to make note of this prerequisite. Possibly there should be a WARN_ON() too. Without this fix, I have seen the following recursive deadlock caused by radix_tree_insert() attempting to allocate memory inside the spinlocked region, which resulted in FS-Cache being called back into to release memory - which required the spinlock already held. ============================================= [ INFO: possible recursive locking detected ] 2.6.32-rc6-cachefs #24 --------------------------------------------- nfsiod/7916 is trying to acquire lock: (&cookie->lock){+.+.-.}, at: [<ffffffffa0076872>] __fscache_uncache_page+0xdb/0x160 [fscache] but task is already holding lock: (&cookie->lock){+.+.-.}, at: [<ffffffffa0076acc>] __fscache_write_page+0x15c/0x3f3 [fscache] other info that might help us debug this: 5 locks held by nfsiod/7916: #0: (nfsiod){+.+.+.}, at: [<ffffffff81048290>] worker_thread+0x19a/0x2e2 #1: (&task->u.tk_work#2){+.+.+.}, at: [<ffffffff81048290>] worker_thread+0x19a/0x2e2 #2: (&cookie->lock){+.+.-.}, at: [<ffffffffa0076acc>] __fscache_write_page+0x15c/0x3f3 [fscache] #3: (&object->lock#2){+.+.-.}, at: [<ffffffffa0076b07>] __fscache_write_page+0x197/0x3f3 [fscache] #4: (&cookie->stores_lock){+.+...}, at: [<ffffffffa0076b0f>] __fscache_write_page+0x19f/0x3f3 [fscache] stack backtrace: Pid: 7916, comm: nfsiod Not tainted 2.6.32-rc6-cachefs #24 Call Trace: [<ffffffff8105ac7f>] __lock_acquire+0x1649/0x16e3 [<ffffffff81059ded>] ? __lock_acquire+0x7b7/0x16e3 [<ffffffff8100e27d>] ? dump_trace+0x248/0x257 [<ffffffff8105ad70>] lock_acquire+0x57/0x6d [<ffffffffa0076872>] ? __fscache_uncache_page+0xdb/0x160 [fscache] [<ffffffff8135467c>] _spin_lock+0x2c/0x3b [<ffffffffa0076872>] ? __fscache_uncache_page+0xdb/0x160 [fscache] [<ffffffffa0076872>] __fscache_uncache_page+0xdb/0x160 [fscache] [<ffffffffa0077eb7>] ? __fscache_check_page_write+0x0/0x71 [fscache] [<ffffffffa00b4755>] nfs_fscache_release_page+0x86/0xc4 [nfs] [<ffffffffa00907f0>] nfs_release_page+0x3c/0x41 [nfs] [<ffffffff81087ffb>] try_to_release_page+0x32/0x3b [<ffffffff81092c2b>] shrink_page_list+0x316/0x4ac [<ffffffff81058a9b>] ? mark_held_locks+0x52/0x70 [<ffffffff8135451b>] ? _spin_unlock_irq+0x2b/0x31 [<ffffffff81093153>] shrink_inactive_list+0x392/0x67c [<ffffffff81058a9b>] ? mark_held_locks+0x52/0x70 [<ffffffff810934ca>] shrink_list+0x8d/0x8f [<ffffffff81093744>] shrink_zone+0x278/0x33c [<ffffffff81052c70>] ? ktime_get_ts+0xad/0xba [<ffffffff8109453b>] try_to_free_pages+0x22e/0x392 [<ffffffff8109184c>] ? isolate_pages_global+0x0/0x212 [<ffffffff8108e16b>] __alloc_pages_nodemask+0x3dc/0x5cf [<ffffffff810ae24a>] cache_alloc_refill+0x34d/0x6c1 [<ffffffff811bcf74>] ? radix_tree_node_alloc+0x52/0x5c [<ffffffff810ae929>] kmem_cache_alloc+0xb2/0x118 [<ffffffff811bcf74>] radix_tree_node_alloc+0x52/0x5c [<ffffffff811bcfd5>] radix_tree_insert+0x57/0x19c [<ffffffffa0076b53>] __fscache_write_page+0x1e3/0x3f3 [fscache] [<ffffffffa00b4248>] __nfs_readpage_to_fscache+0x58/0x11e [nfs] [<ffffffffa009bb77>] nfs_readpage_release+0x34/0x9b [nfs] [<ffffffffa009c0d9>] nfs_readpage_release_full+0x32/0x4b [nfs] [<ffffffffa0006cff>] rpc_release_calldata+0x12/0x14 [sunrpc] [<ffffffffa0006e2d>] rpc_free_task+0x59/0x61 [sunrpc] [<ffffffffa0006f03>] rpc_async_release+0x10/0x12 [sunrpc] [<ffffffff810482e5>] worker_thread+0x1ef/0x2e2 [<ffffffff81048290>] ? worker_thread+0x19a/0x2e2 [<ffffffff81352433>] ? thread_return+0x3e/0x101 [<ffffffffa0006ef3>] ? rpc_async_release+0x0/0x12 [sunrpc] [<ffffffff8104bff5>] ? autoremove_wake_function+0x0/0x34 [<ffffffff81058d25>] ? trace_hardirqs_on+0xd/0xf [<ffffffff810480f6>] ? worker_thread+0x0/0x2e2 [<ffffffff8104bd21>] kthread+0x7a/0x82 [<ffffffff8100beda>] child_rip+0xa/0x20 [<ffffffff8100b87c>] ? restore_args+0x0/0x30 [<ffffffff8104c2b9>] ? add_wait_queue+0x15/0x44 [<ffffffff8104bca7>] ? kthread+0x0/0x82 [<ffffffff8100bed0>] ? child_rip+0x0/0x20 Signed-off-by: David Howells <dhowells@redhat.com>
2009-11-20 01:11:14 +07:00
*
* To make use of this facility, the radix tree must be initialised without
* __GFP_WAIT being passed to INIT_RADIX_TREE().
*/
int radix_tree_preload(gfp_t gfp_mask)
{
struct radix_tree_preload *rtp;
struct radix_tree_node *node;
int ret = -ENOMEM;
preempt_disable();
rtp = &__get_cpu_var(radix_tree_preloads);
while (rtp->nr < ARRAY_SIZE(rtp->nodes)) {
preempt_enable();
node = kmem_cache_alloc(radix_tree_node_cachep, gfp_mask);
if (node == NULL)
goto out;
preempt_disable();
rtp = &__get_cpu_var(radix_tree_preloads);
if (rtp->nr < ARRAY_SIZE(rtp->nodes))
rtp->nodes[rtp->nr++] = node;
else
kmem_cache_free(radix_tree_node_cachep, node);
}
ret = 0;
out:
return ret;
}
EXPORT_SYMBOL(radix_tree_preload);
/*
* Return the maximum key which can be store into a
* radix tree with height HEIGHT.
*/
static inline unsigned long radix_tree_maxindex(unsigned int height)
{
return height_to_maxindex[height];
}
/*
* Extend a radix tree so it can store key @index.
*/
static int radix_tree_extend(struct radix_tree_root *root, unsigned long index)
{
struct radix_tree_node *node;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
struct radix_tree_node *slot;
unsigned int height;
int tag;
/* Figure out what the height should be. */
height = root->height + 1;
while (index > radix_tree_maxindex(height))
height++;
if (root->rnode == NULL) {
root->height = height;
goto out;
}
do {
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
unsigned int newheight;
if (!(node = radix_tree_node_alloc(root)))
return -ENOMEM;
/* Propagate the aggregated tag info into the new root */
for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) {
if (root_tag_get(root, tag))
tag_set(node, tag, 0);
}
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
/* Increase the height. */
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
newheight = root->height+1;
node->height = newheight;
node->count = 1;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
node->parent = NULL;
slot = root->rnode;
if (newheight > 1) {
slot = indirect_to_ptr(slot);
slot->parent = node;
}
node->slots[0] = slot;
node = ptr_to_indirect(node);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
rcu_assign_pointer(root->rnode, node);
root->height = newheight;
} while (height > root->height);
out:
return 0;
}
/**
* radix_tree_insert - insert into a radix tree
* @root: radix tree root
* @index: index key
* @item: item to insert
*
* Insert an item into the radix tree at position @index.
*/
int radix_tree_insert(struct radix_tree_root *root,
unsigned long index, void *item)
{
struct radix_tree_node *node = NULL, *slot;
unsigned int height, shift;
int offset;
int error;
BUG_ON(radix_tree_is_indirect_ptr(item));
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
/* Make sure the tree is high enough. */
if (index > radix_tree_maxindex(root->height)) {
error = radix_tree_extend(root, index);
if (error)
return error;
}
slot = indirect_to_ptr(root->rnode);
height = root->height;
shift = (height-1) * RADIX_TREE_MAP_SHIFT;
offset = 0; /* uninitialised var warning */
while (height > 0) {
if (slot == NULL) {
/* Have to add a child node. */
if (!(slot = radix_tree_node_alloc(root)))
return -ENOMEM;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
slot->height = height;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
slot->parent = node;
if (node) {
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
rcu_assign_pointer(node->slots[offset], slot);
node->count++;
} else
rcu_assign_pointer(root->rnode, ptr_to_indirect(slot));
}
/* Go a level down */
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
node = slot;
slot = node->slots[offset];
shift -= RADIX_TREE_MAP_SHIFT;
height--;
}
if (slot != NULL)
return -EEXIST;
if (node) {
node->count++;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
rcu_assign_pointer(node->slots[offset], item);
BUG_ON(tag_get(node, 0, offset));
BUG_ON(tag_get(node, 1, offset));
} else {
rcu_assign_pointer(root->rnode, item);
BUG_ON(root_tag_get(root, 0));
BUG_ON(root_tag_get(root, 1));
}
return 0;
}
EXPORT_SYMBOL(radix_tree_insert);
/*
* is_slot == 1 : search for the slot.
* is_slot == 0 : search for the node.
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
*/
static void *radix_tree_lookup_element(struct radix_tree_root *root,
unsigned long index, int is_slot)
{
unsigned int height, shift;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
struct radix_tree_node *node, **slot;
node = rcu_dereference_raw(root->rnode);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (node == NULL)
return NULL;
if (!radix_tree_is_indirect_ptr(node)) {
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (index > 0)
return NULL;
return is_slot ? (void *)&root->rnode : node;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
}
node = indirect_to_ptr(node);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
height = node->height;
if (index > radix_tree_maxindex(height))
return NULL;
shift = (height-1) * RADIX_TREE_MAP_SHIFT;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
do {
slot = (struct radix_tree_node **)
(node->slots + ((index>>shift) & RADIX_TREE_MAP_MASK));
node = rcu_dereference_raw(*slot);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (node == NULL)
return NULL;
shift -= RADIX_TREE_MAP_SHIFT;
height--;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
} while (height > 0);
return is_slot ? (void *)slot : indirect_to_ptr(node);
}
/**
* radix_tree_lookup_slot - lookup a slot in a radix tree
* @root: radix tree root
* @index: index key
*
* Returns: the slot corresponding to the position @index in the
* radix tree @root. This is useful for update-if-exists operations.
*
* This function can be called under rcu_read_lock iff the slot is not
* modified by radix_tree_replace_slot, otherwise it must be called
* exclusive from other writers. Any dereference of the slot must be done
* using radix_tree_deref_slot.
*/
void **radix_tree_lookup_slot(struct radix_tree_root *root, unsigned long index)
{
return (void **)radix_tree_lookup_element(root, index, 1);
}
EXPORT_SYMBOL(radix_tree_lookup_slot);
/**
* radix_tree_lookup - perform lookup operation on a radix tree
* @root: radix tree root
* @index: index key
*
* Lookup the item at the position @index in the radix tree @root.
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
*
* This function can be called under rcu_read_lock, however the caller
* must manage lifetimes of leaf nodes (eg. RCU may also be used to free
* them safely). No RCU barriers are required to access or modify the
* returned item, however.
*/
void *radix_tree_lookup(struct radix_tree_root *root, unsigned long index)
{
return radix_tree_lookup_element(root, index, 0);
}
EXPORT_SYMBOL(radix_tree_lookup);
/**
* radix_tree_tag_set - set a tag on a radix tree node
* @root: radix tree root
* @index: index key
* @tag: tag index
*
* Set the search tag (which must be < RADIX_TREE_MAX_TAGS)
* corresponding to @index in the radix tree. From
* the root all the way down to the leaf node.
*
* Returns the address of the tagged item. Setting a tag on a not-present
* item is a bug.
*/
void *radix_tree_tag_set(struct radix_tree_root *root,
unsigned long index, unsigned int tag)
{
unsigned int height, shift;
struct radix_tree_node *slot;
height = root->height;
BUG_ON(index > radix_tree_maxindex(height));
slot = indirect_to_ptr(root->rnode);
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
while (height > 0) {
int offset;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
if (!tag_get(slot, tag, offset))
tag_set(slot, tag, offset);
slot = slot->slots[offset];
BUG_ON(slot == NULL);
shift -= RADIX_TREE_MAP_SHIFT;
height--;
}
/* set the root's tag bit */
if (slot && !root_tag_get(root, tag))
root_tag_set(root, tag);
return slot;
}
EXPORT_SYMBOL(radix_tree_tag_set);
/**
* radix_tree_tag_clear - clear a tag on a radix tree node
* @root: radix tree root
* @index: index key
* @tag: tag index
*
* Clear the search tag (which must be < RADIX_TREE_MAX_TAGS)
* corresponding to @index in the radix tree. If
* this causes the leaf node to have no tags set then clear the tag in the
* next-to-leaf node, etc.
*
* Returns the address of the tagged item on success, else NULL. ie:
* has the same return value and semantics as radix_tree_lookup().
*/
void *radix_tree_tag_clear(struct radix_tree_root *root,
unsigned long index, unsigned int tag)
{
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
struct radix_tree_node *node = NULL;
struct radix_tree_node *slot = NULL;
unsigned int height, shift;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
int uninitialized_var(offset);
height = root->height;
if (index > radix_tree_maxindex(height))
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
shift = height * RADIX_TREE_MAP_SHIFT;
slot = indirect_to_ptr(root->rnode);
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
while (shift) {
if (slot == NULL)
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
shift -= RADIX_TREE_MAP_SHIFT;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
node = slot;
slot = slot->slots[offset];
}
if (slot == NULL)
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
while (node) {
if (!tag_get(node, tag, offset))
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
tag_clear(node, tag, offset);
if (any_tag_set(node, tag))
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
index >>= RADIX_TREE_MAP_SHIFT;
offset = index & RADIX_TREE_MAP_MASK;
node = node->parent;
}
/* clear the root's tag bit */
if (root_tag_get(root, tag))
root_tag_clear(root, tag);
out:
return slot;
}
EXPORT_SYMBOL(radix_tree_tag_clear);
/**
* radix_tree_tag_get - get a tag on a radix tree node
* @root: radix tree root
* @index: index key
* @tag: tag index (< RADIX_TREE_MAX_TAGS)
*
* Return values:
*
* 0: tag not present or not set
* 1: tag set
radix_tree_tag_get() is not as safe as the docs make out [ver #2] radix_tree_tag_get() is not safe to use concurrently with radix_tree_tag_set() or radix_tree_tag_clear(). The problem is that the double tag_get() in radix_tree_tag_get(): if (!tag_get(node, tag, offset)) saw_unset_tag = 1; if (height == 1) { int ret = tag_get(node, tag, offset); may see the value change due to the action of set/clear. RCU is no protection against this as no pointers are being changed, no nodes are being replaced according to a COW protocol - set/clear alter the node directly. The documentation in linux/radix-tree.h, however, says that radix_tree_tag_get() is an exception to the rule that "any function modifying the tree or tags (...) must exclude other modifications, and exclude any functions reading the tree". The problem is that the next statement in radix_tree_tag_get() checks that the tag doesn't vary over time: BUG_ON(ret && saw_unset_tag); This has been seen happening in FS-Cache: https://www.redhat.com/archives/linux-cachefs/2010-April/msg00013.html To this end, remove the BUG_ON() from radix_tree_tag_get() and note in various comments that the value of the tag may change whilst the RCU read lock is held, and thus that the return value of radix_tree_tag_get() may not be relied upon unless radix_tree_tag_set/clear() and radix_tree_delete() are excluded from running concurrently with it. Reported-by: Romain DEGEZ <romain.degez@smartjog.com> Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-04-07 04:36:20 +07:00
*
* Note that the return value of this function may not be relied on, even if
* the RCU lock is held, unless tag modification and node deletion are excluded
* from concurrency.
*/
int radix_tree_tag_get(struct radix_tree_root *root,
unsigned long index, unsigned int tag)
{
unsigned int height, shift;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
struct radix_tree_node *node;
/* check the root's tag bit */
if (!root_tag_get(root, tag))
return 0;
node = rcu_dereference_raw(root->rnode);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (node == NULL)
return 0;
if (!radix_tree_is_indirect_ptr(node))
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
return (index == 0);
node = indirect_to_ptr(node);
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
height = node->height;
if (index > radix_tree_maxindex(height))
return 0;
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
for ( ; ; ) {
int offset;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (node == NULL)
return 0;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (!tag_get(node, tag, offset))
return 0;
radix_tree_tag_get() is not as safe as the docs make out [ver #2] radix_tree_tag_get() is not safe to use concurrently with radix_tree_tag_set() or radix_tree_tag_clear(). The problem is that the double tag_get() in radix_tree_tag_get(): if (!tag_get(node, tag, offset)) saw_unset_tag = 1; if (height == 1) { int ret = tag_get(node, tag, offset); may see the value change due to the action of set/clear. RCU is no protection against this as no pointers are being changed, no nodes are being replaced according to a COW protocol - set/clear alter the node directly. The documentation in linux/radix-tree.h, however, says that radix_tree_tag_get() is an exception to the rule that "any function modifying the tree or tags (...) must exclude other modifications, and exclude any functions reading the tree". The problem is that the next statement in radix_tree_tag_get() checks that the tag doesn't vary over time: BUG_ON(ret && saw_unset_tag); This has been seen happening in FS-Cache: https://www.redhat.com/archives/linux-cachefs/2010-April/msg00013.html To this end, remove the BUG_ON() from radix_tree_tag_get() and note in various comments that the value of the tag may change whilst the RCU read lock is held, and thus that the return value of radix_tree_tag_get() may not be relied upon unless radix_tree_tag_set/clear() and radix_tree_delete() are excluded from running concurrently with it. Reported-by: Romain DEGEZ <romain.degez@smartjog.com> Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-04-07 04:36:20 +07:00
if (height == 1)
return 1;
node = rcu_dereference_raw(node->slots[offset]);
shift -= RADIX_TREE_MAP_SHIFT;
height--;
}
}
EXPORT_SYMBOL(radix_tree_tag_get);
radix-tree: introduce bit-optimized iterator A series of radix tree cleanups, and usage of them in the core pagecache code. Micro-benchmark: lookup 14 slots (typical page-vector size) in radix-tree there earch <step> slot filled and tagged before/after - nsec per full scan through tree * Intel Sandy Bridge i7-2620M 4Mb L3 New code always faster * AMD Athlon 6000+ 2x1Mb L2, without L3 New code generally faster, Minor degradation (marked with "*") for huge sparse trees * i386 on Sandy Bridge New code faster for common cases: tagged and dense trees. Some degradations for non-tagged lookup on sparse trees. Ideally, there might help __ffs() analog for searching first non-zero long element in array, gcc sometimes cannot optimize this loop corretly. Numbers: CPU: Intel Sandy Bridge i7-2620M 4Mb L3 radix-tree with 1024 slots: tagged lookup step 1 before 7156 after 3613 step 2 before 5399 after 2696 step 3 before 4779 after 1928 step 4 before 4456 after 1429 step 5 before 4292 after 1213 step 6 before 4183 after 1052 step 7 before 4157 after 951 step 8 before 4016 after 812 step 9 before 3952 after 851 step 10 before 3937 after 732 step 11 before 4023 after 709 step 12 before 3872 after 657 step 13 before 3892 after 633 step 14 before 3720 after 591 step 15 before 3879 after 578 step 16 before 3561 after 513 normal lookup step 1 before 4266 after 3301 step 2 before 2695 after 2129 step 3 before 2083 after 1712 step 4 before 1801 after 1534 step 5 before 1628 after 1313 step 6 before 1551 after 1263 step 7 before 1475 after 1185 step 8 before 1432 after 1167 step 9 before 1373 after 1092 step 10 before 1339 after 1134 step 11 before 1292 after 1056 step 12 before 1319 after 1030 step 13 before 1276 after 1004 step 14 before 1256 after 987 step 15 before 1228 after 992 step 16 before 1247 after 999 radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1086102841 after 674196409 step 2 before 816839155 after 498138306 step 7 before 599728907 after 240676762 step 15 before 555729253 after 185219677 step 63 before 606637748 after 128585664 step 64 before 608384432 after 102945089 step 65 before 596987114 after 123996019 step 128 before 304459225 after 56783056 step 256 before 158846855 after 31232481 step 512 before 86085652 after 18950595 step 12345 before 6517189 after 1674057 normal lookup step 1 before 626064869 after 544418266 step 2 before 418809975 after 336321473 step 7 before 242303598 after 207755560 step 15 before 208380563 after 176496355 step 63 before 186854206 after 167283638 step 64 before 176188060 after 170143976 step 65 before 185139608 after 167487116 step 128 before 88181865 after 86913490 step 256 before 45733628 after 45143534 step 512 before 24506038 after 23859036 step 12345 before 2177425 after 2018662 * AMD Athlon 6000+ 2x1Mb L2, without L3 radix-tree with 1024 slots: tag-lookup step 1 before 8164 after 5379 step 2 before 5818 after 5581 step 3 before 4959 after 4213 step 4 before 4371 after 3386 step 5 before 4204 after 2997 step 6 before 4950 after 2744 step 7 before 4598 after 2480 step 8 before 4251 after 2288 step 9 before 4262 after 2243 step 10 before 4175 after 2131 step 11 before 3999 after 2024 step 12 before 3979 after 1994 step 13 before 3842 after 1929 step 14 before 3750 after 1810 step 15 before 3735 after 1810 step 16 before 3532 after 1660 normal-lookup step 1 before 7875 after 5847 step 2 before 4808 after 4071 step 3 before 4073 after 3462 step 4 before 3677 after 3074 step 5 before 4308 after 2978 step 6 before 3911 after 3807 step 7 before 3635 after 3522 step 8 before 3313 after 3202 step 9 before 3280 after 3257 step 10 before 3166 after 3083 step 11 before 3066 after 3026 step 12 before 2985 after 2982 step 13 before 2925 after 2924 step 14 before 2834 after 2808 step 15 before 2805 after 2803 step 16 before 2647 after 2622 radix-tree with 1024*1024*128 slots: tag-lookup step 1 before 1288059720 after 951736580 step 2 before 961292300 after 884212140 step 7 before 768905140 after 547267580 step 15 before 771319480 after 456550640 step 63 before 504847640 after 242704304 step 64 before 392484800 after 177920786 step 65 before 491162160 after 246895264 step 128 before 208084064 after 97348392 step 256 before 112401035 after 51408126 step 512 before 75825834 after 29145070 step 12345 before 5603166 after 2847330 normal-lookup step 1 before 1025677120 after 861375100 step 2 before 647220080 after 572258540 step 7 before 505518960 after 484041813 step 15 before 430483053 after 444815320 * step 63 before 388113453 after 404250546 * step 64 before 374154666 after 396027440 * step 65 before 381423973 after 396704853 * step 128 before 190078700 after 202619384 * step 256 before 100886756 after 102829108 * step 512 before 64074505 after 56158720 step 12345 before 4237289 after 4422299 * * i686 on Sandy bridge radix-tree with 1024 slots: tagged lookup step 1 before 7990 after 4019 step 2 before 5698 after 2897 step 3 before 5013 after 2475 step 4 before 4630 after 1721 step 5 before 4346 after 1759 step 6 before 4299 after 1556 step 7 before 4098 after 1513 step 8 before 4115 after 1222 step 9 before 3983 after 1390 step 10 before 4077 after 1207 step 11 before 3921 after 1231 step 12 before 3894 after 1116 step 13 before 3840 after 1147 step 14 before 3799 after 1090 step 15 before 3797 after 1059 step 16 before 3783 after 745 normal lookup step 1 before 5103 after 3499 step 2 before 3299 after 2550 step 3 before 2489 after 2370 step 4 before 2034 after 2302 * step 5 before 1846 after 2268 * step 6 before 1752 after 2249 * step 7 before 1679 after 2164 * step 8 before 1627 after 2153 * step 9 before 1542 after 2095 * step 10 before 1479 after 2109 * step 11 before 1469 after 2009 * step 12 before 1445 after 2039 * step 13 before 1411 after 2013 * step 14 before 1374 after 2046 * step 15 before 1340 after 1975 * step 16 before 1331 after 2000 * radix-tree with 1024*1024*128 slots: tagged lookup step 1 before 1225865377 after 667153553 step 2 before 842427423 after 471533007 step 7 before 609296153 after 276260116 step 15 before 544232060 after 226859105 step 63 before 519209199 after 141343043 step 64 before 588980279 after 141951339 step 65 before 521099710 after 138282060 step 128 before 298476778 after 83390628 step 256 before 149358342 after 43602609 step 512 before 76994713 after 22911077 step 12345 before 5328666 after 1472111 normal lookup step 1 before 819284564 after 533635310 step 2 before 512421605 after 364956155 step 7 before 271443305 after 305721345 * step 15 before 223591630 after 273960216 * step 63 before 190320247 after 217770207 * step 64 before 178538168 after 267411372 * step 65 before 186400423 after 215347937 * step 128 before 88106045 after 140540612 * step 256 before 44812420 after 70660377 * step 512 before 24435438 after 36328275 * step 12345 before 2123924 after 2148062 * bloat-o-meter delta for this patchset + patchset with related shmem cleanups bloat-o-meter: x86_64 add/remove: 4/3 grow/shrink: 5/6 up/down: 928/-939 (-11) function old new delta radix_tree_next_chunk - 499 +499 shmem_unuse 428 554 +126 shmem_radix_tree_replace 131 227 +96 find_get_pages_tag 354 419 +65 find_get_pages_contig 345 407 +62 find_get_pages 362 396 +34 __kstrtab_radix_tree_next_chunk - 22 +22 __ksymtab_radix_tree_next_chunk - 16 +16 __kcrctab_radix_tree_next_chunk - 8 +8 radix_tree_gang_lookup_slot 204 203 -1 static.shmem_xattr_set 384 381 -3 radix_tree_gang_lookup_tag_slot 208 191 -17 radix_tree_gang_lookup 231 187 -44 radix_tree_gang_lookup_tag 247 199 -48 shmem_unlock_mapping 278 190 -88 __lookup 217 - -217 __lookup_tag 242 - -242 radix_tree_locate_item 279 - -279 bloat-o-meter: i386 add/remove: 3/3 grow/shrink: 8/9 up/down: 1075/-1275 (-200) function old new delta radix_tree_next_chunk - 757 +757 shmem_unuse 352 449 +97 find_get_pages_contig 269 322 +53 shmem_radix_tree_replace 113 154 +41 find_get_pages_tag 277 318 +41 dcache_dir_lseek 426 458 +32 __kstrtab_radix_tree_next_chunk - 22 +22 vc_do_resize 968 977 +9 snd_pcm_lib_read1 725 733 +8 __ksymtab_radix_tree_next_chunk - 8 +8 netlbl_cipsov4_list 1120 1127 +7 find_get_pages 293 291 -2 new_slab 467 459 -8 bitfill_unaligned_rev 425 417 -8 radix_tree_gang_lookup_tag_slot 177 146 -31 blk_dump_cmd 267 229 -38 radix_tree_gang_lookup_slot 212 134 -78 shmem_unlock_mapping 221 128 -93 radix_tree_gang_lookup_tag 275 162 -113 radix_tree_gang_lookup 255 126 -129 __lookup 227 - -227 __lookup_tag 271 - -271 radix_tree_locate_item 277 - -277 This patch: Implement a clean, simple and effective radix-tree iteration routine. Iterating divided into two phases: * lookup next chunk in radix-tree leaf node * iterating through slots in this chunk Main iterator function radix_tree_next_chunk() returns pointer to first slot, and stores in the struct radix_tree_iter index of next-to-last slot. For tagged-iterating it also constuct bitmask of tags for retunted chunk. All additional logic implemented as static-inline functions and macroses. Also adds radix_tree_find_next_bit() static-inline variant of find_next_bit() optimized for small constant size arrays, because find_next_bit() too heavy for searching in an array with one/two long elements. [akpm@linux-foundation.org: rework comments a bit] Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org> Tested-by: Hugh Dickins <hughd@google.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-03-29 04:42:53 +07:00
/**
* radix_tree_next_chunk - find next chunk of slots for iteration
*
* @root: radix tree root
* @iter: iterator state
* @flags: RADIX_TREE_ITER_* flags and tag index
* Returns: pointer to chunk first slot, or NULL if iteration is over
*/
void **radix_tree_next_chunk(struct radix_tree_root *root,
struct radix_tree_iter *iter, unsigned flags)
{
unsigned shift, tag = flags & RADIX_TREE_ITER_TAG_MASK;
struct radix_tree_node *rnode, *node;
unsigned long index, offset;
if ((flags & RADIX_TREE_ITER_TAGGED) && !root_tag_get(root, tag))
return NULL;
/*
* Catch next_index overflow after ~0UL. iter->index never overflows
* during iterating; it can be zero only at the beginning.
* And we cannot overflow iter->next_index in a single step,
* because RADIX_TREE_MAP_SHIFT < BITS_PER_LONG.
*/
index = iter->next_index;
if (!index && iter->index)
return NULL;
rnode = rcu_dereference_raw(root->rnode);
if (radix_tree_is_indirect_ptr(rnode)) {
rnode = indirect_to_ptr(rnode);
} else if (rnode && !index) {
/* Single-slot tree */
iter->index = 0;
iter->next_index = 1;
iter->tags = 1;
return (void **)&root->rnode;
} else
return NULL;
restart:
shift = (rnode->height - 1) * RADIX_TREE_MAP_SHIFT;
offset = index >> shift;
/* Index outside of the tree */
if (offset >= RADIX_TREE_MAP_SIZE)
return NULL;
node = rnode;
while (1) {
if ((flags & RADIX_TREE_ITER_TAGGED) ?
!test_bit(offset, node->tags[tag]) :
!node->slots[offset]) {
/* Hole detected */
if (flags & RADIX_TREE_ITER_CONTIG)
return NULL;
if (flags & RADIX_TREE_ITER_TAGGED)
offset = radix_tree_find_next_bit(
node->tags[tag],
RADIX_TREE_MAP_SIZE,
offset + 1);
else
while (++offset < RADIX_TREE_MAP_SIZE) {
if (node->slots[offset])
break;
}
index &= ~((RADIX_TREE_MAP_SIZE << shift) - 1);
index += offset << shift;
/* Overflow after ~0UL */
if (!index)
return NULL;
if (offset == RADIX_TREE_MAP_SIZE)
goto restart;
}
/* This is leaf-node */
if (!shift)
break;
node = rcu_dereference_raw(node->slots[offset]);
if (node == NULL)
goto restart;
shift -= RADIX_TREE_MAP_SHIFT;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
}
/* Update the iterator state */
iter->index = index;
iter->next_index = (index | RADIX_TREE_MAP_MASK) + 1;
/* Construct iter->tags bit-mask from node->tags[tag] array */
if (flags & RADIX_TREE_ITER_TAGGED) {
unsigned tag_long, tag_bit;
tag_long = offset / BITS_PER_LONG;
tag_bit = offset % BITS_PER_LONG;
iter->tags = node->tags[tag][tag_long] >> tag_bit;
/* This never happens if RADIX_TREE_TAG_LONGS == 1 */
if (tag_long < RADIX_TREE_TAG_LONGS - 1) {
/* Pick tags from next element */
if (tag_bit)
iter->tags |= node->tags[tag][tag_long + 1] <<
(BITS_PER_LONG - tag_bit);
/* Clip chunk size, here only BITS_PER_LONG tags */
iter->next_index = index + BITS_PER_LONG;
}
}
return node->slots + offset;
}
EXPORT_SYMBOL(radix_tree_next_chunk);
/**
* radix_tree_range_tag_if_tagged - for each item in given range set given
* tag if item has another tag set
* @root: radix tree root
* @first_indexp: pointer to a starting index of a range to scan
* @last_index: last index of a range to scan
* @nr_to_tag: maximum number items to tag
* @iftag: tag index to test
* @settag: tag index to set if tested tag is set
*
* This function scans range of radix tree from first_index to last_index
* (inclusive). For each item in the range if iftag is set, the function sets
* also settag. The function stops either after tagging nr_to_tag items or
* after reaching last_index.
*
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
* The tags must be set from the leaf level only and propagated back up the
* path to the root. We must do this so that we resolve the full path before
* setting any tags on intermediate nodes. If we set tags as we descend, then
* we can get to the leaf node and find that the index that has the iftag
* set is outside the range we are scanning. This reults in dangling tags and
* can lead to problems with later tag operations (e.g. livelocks on lookups).
*
* The function returns number of leaves where the tag was set and sets
* *first_indexp to the first unscanned index.
* WARNING! *first_indexp can wrap if last_index is ULONG_MAX. Caller must
* be prepared to handle that.
*/
unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root,
unsigned long *first_indexp, unsigned long last_index,
unsigned long nr_to_tag,
unsigned int iftag, unsigned int settag)
{
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
unsigned int height = root->height;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
struct radix_tree_node *node = NULL;
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
struct radix_tree_node *slot;
unsigned int shift;
unsigned long tagged = 0;
unsigned long index = *first_indexp;
last_index = min(last_index, radix_tree_maxindex(height));
if (index > last_index)
return 0;
if (!nr_to_tag)
return 0;
if (!root_tag_get(root, iftag)) {
*first_indexp = last_index + 1;
return 0;
}
if (height == 0) {
*first_indexp = last_index + 1;
root_tag_set(root, settag);
return 1;
}
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
slot = indirect_to_ptr(root->rnode);
for (;;) {
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
unsigned long upindex;
int offset;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
if (!slot->slots[offset])
goto next;
if (!tag_get(slot, iftag, offset))
goto next;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
if (shift) {
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
/* Go down one level */
shift -= RADIX_TREE_MAP_SHIFT;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
node = slot;
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
slot = slot->slots[offset];
continue;
}
/* tag the leaf */
tagged++;
tag_set(slot, settag, offset);
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
/* walk back up the path tagging interior nodes */
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
upindex = index;
while (node) {
upindex >>= RADIX_TREE_MAP_SHIFT;
offset = upindex & RADIX_TREE_MAP_MASK;
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
/* stop if we find a node with the tag already set */
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
if (tag_get(node, settag, offset))
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
break;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
tag_set(node, settag, offset);
node = node->parent;
}
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags Commit ebf8aa44beed48cd17893a83d92a4403e5f9d9e2 ("radix-tree: omplement function radix_tree_range_tag_if_tagged") does not safely set tags on on intermediate tree nodes. The code walks down the tree setting tags before it has fully resolved the path to the leaf under the assumption there will be a leaf slot with the tag set in the range it is searching. Unfortunately, this is not a valid assumption - we can abort after setting a tag on an intermediate node if we overrun the number of tags we are allowed to set in a batch, or stop scanning because we we have passed the last scan index before we reach a leaf slot with the tag we are searching for set. As a result, we can leave the function with tags set on intemediate nodes which can be tripped over later by tag-based lookups. The result of these stale tags is that lookup may end prematurely or livelock because the lookup cannot make progress. The fix for the problem involves reocrding the traversal path we take to the leaf nodes, and only propagating the tags back up the tree once the tag is set in the leaf node slot. We are already recording the path for efficient traversal, so there is no additional overhead to do the intermediately node tag setting in this manner. This fixes a radix tree lookup livelock triggered by the new writeback sync livelock avoidance code introduced in commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement writeback livelock avoidance using page tagging"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Acked-by: Jan Kara <jack@suse.cz>
2010-08-23 07:33:53 +07:00
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
/*
* Small optimization: now clear that node pointer.
* Since all of this slot's ancestors now have the tag set
* from setting it above, we have no further need to walk
* back up the tree setting tags, until we update slot to
* point to another radix_tree_node.
*/
node = NULL;
next:
/* Go to next item at level determined by 'shift' */
index = ((index >> shift) + 1) << shift;
/* Overflow can happen when last_index is ~0UL... */
if (index > last_index || !index)
break;
if (tagged >= nr_to_tag)
break;
while (((index >> shift) & RADIX_TREE_MAP_MASK) == 0) {
/*
* We've fully scanned this node. Go up. Because
* last_index is guaranteed to be in the tree, what
* we do below cannot wander astray.
*/
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
slot = slot->parent;
shift += RADIX_TREE_MAP_SHIFT;
}
}
/*
radix_tree: radix_tree_gang_lookup_tag_slot() may never return Executed command: fsstress -d /mnt -n 600 -p 850 crash> bt PID: 7947 TASK: ffff880160546a70 CPU: 0 COMMAND: "fsstress" #0 [ffff8800dfc07d00] machine_kexec at ffffffff81030db9 #1 [ffff8800dfc07d70] crash_kexec at ffffffff810a7952 #2 [ffff8800dfc07e40] oops_end at ffffffff814aa7c8 #3 [ffff8800dfc07e70] die_nmi at ffffffff814aa969 #4 [ffff8800dfc07ea0] do_nmi_callback at ffffffff8102b07b #5 [ffff8800dfc07f10] do_nmi at ffffffff814aa514 #6 [ffff8800dfc07f50] nmi at ffffffff814a9d60 [exception RIP: __lookup_tag+100] RIP: ffffffff812274b4 RSP: ffff88016056b998 RFLAGS: 00000287 RAX: 0000000000000000 RBX: 0000000000000002 RCX: 0000000000000006 RDX: 000000000000001d RSI: ffff88016056bb18 RDI: ffff8800c85366e0 RBP: ffff88016056b9c8 R8: ffff88016056b9e8 R9: 0000000000000000 R10: 000000000000000e R11: ffff8800c8536908 R12: 0000000000000010 R13: 0000000000000040 R14: ffffffffffffffc0 R15: ffff8800c85366e0 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 <NMI exception stack> #7 [ffff88016056b998] __lookup_tag at ffffffff812274b4 #8 [ffff88016056b9d0] radix_tree_gang_lookup_tag_slot at ffffffff81227605 #9 [ffff88016056ba20] find_get_pages_tag at ffffffff810fc110 #10 [ffff88016056ba80] pagevec_lookup_tag at ffffffff81105e85 #11 [ffff88016056baa0] write_cache_pages at ffffffff81104c47 #12 [ffff88016056bbd0] generic_writepages at ffffffff81105014 #13 [ffff88016056bbe0] do_writepages at ffffffff81105055 #14 [ffff88016056bbf0] __filemap_fdatawrite_range at ffffffff810fb2cb #15 [ffff88016056bc40] filemap_write_and_wait_range at ffffffff810fb32a #16 [ffff88016056bc70] generic_file_direct_write at ffffffff810fb3dc #17 [ffff88016056bce0] __generic_file_aio_write at ffffffff810fcee5 #18 [ffff88016056bda0] generic_file_aio_write at ffffffff810fd085 #19 [ffff88016056bdf0] do_sync_write at ffffffff8114f9ea #20 [ffff88016056bf00] vfs_write at ffffffff8114fcf8 #21 [ffff88016056bf30] sys_write at ffffffff81150691 #22 [ffff88016056bf80] system_call_fastpath at ffffffff8100c0b2 I think this root cause is the following: radix_tree_range_tag_if_tagged() always tags the root tag with settag if the root tag is set with iftag even if there are no iftag tags in the specified range (Of course, there are some iftag tags outside the specified range). =============================================================================== [[[Detailed description]]] (1) Why cannot radix_tree_gang_lookup_tag_slot() return forever? __lookup_tag(): - Return with 0. - Return with the index which is not bigger than the old one as the input parameter. Therefore the following "while" repeats forever because the above conditions cause "ret" not to be updated and the cur_index cannot be changed into the bigger one. (So, radix_tree_gang_lookup_tag_slot() cannot return forever.) radix_tree_gang_lookup_tag_slot(): 1178 while (ret < max_items) { 1179 unsigned int slots_found; 1180 unsigned long next_index; /* Index of next search */ 1181 1182 if (cur_index > max_index) 1183 break; 1184 slots_found = __lookup_tag(node, results + ret, 1185 cur_index, max_items - ret, &next_index, tag); 1186 ret += slots_found; // cannot update ret because slots_found == 0. // so, this while loops forever. 1187 if (next_index == 0) 1188 break; 1189 cur_index = next_index; 1190 } (2) Why does __lookup_tag() return with 0 and doesn't update the index? Assuming the following: - the one of the slot in radix_tree_node is NULL. - the one of the tag which corresponds to the slot sets with PAGECACHE_TAG_TOWRITE or other. - In a certain height(!=0), the corresponding index is 0. a) __lookup_tag() notices that the tag is set. 1005 static unsigned int 1006 __lookup_tag(struct radix_tree_node *slot, void ***results, unsigned long index, 1007 unsigned int max_items, unsigned long *next_index, unsigned int tag) 1008 { 1009 unsigned int nr_found = 0; 1010 unsigned int shift, height; 1011 1012 height = slot->height; 1013 if (height == 0) 1014 goto out; 1015 shift = (height-1) * RADIX_TREE_MAP_SHIFT; 1016 1017 while (height > 0) { 1018 unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK ; 1019 1020 for (;;) { 1021 if (tag_get(slot, tag, i)) 1022 break; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ * the index is not updated yet. b) __lookup_tag() notices that the slot is NULL. 1023 index &= ~((1UL << shift) - 1); 1024 index += 1UL << shift; 1025 if (index == 0) 1026 goto out; /* 32-bit wraparound */ 1027 i++; 1028 if (i == RADIX_TREE_MAP_SIZE) 1029 goto out; 1030 } 1031 height--; 1032 if (height == 0) { /* Bottom level: grab some items */ ... 1055 } 1056 shift -= RADIX_TREE_MAP_SHIFT; 1057 slot = rcu_dereference_raw(slot->slots[i]); 1058 if (slot == NULL) 1059 break; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ c) __lookup_tag() doesn't update the index and return with 0. 1060 } 1061 out: 1062 *next_index = index; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 1063 return nr_found; 1064 } (3) Why is the slot NULL even if the tag is set? Because radix_tree_range_tag_if_tagged() always sets the root tag with PAGECACHE_TAG_TOWRITE if the root tag is set with PAGECACHE_TAG_DIRTY, even if there is no tag which can be set with PAGECACHE_TAG_TOWRITE in the specified range (from *first_indexp to last_index). Of course, some PAGECACHE_TAG_DIRTY nodes must exist outside the specified range. (radix_tree_range_tag_if_tagged() is called only from tag_pages_for_writeback()) 640 unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root, 641 unsigned long *first_indexp, unsigned long last_index, 642 unsigned long nr_to_tag, 643 unsigned int iftag, unsigned int settag) 644 { 645 unsigned int height = root->height; 646 struct radix_tree_path path[height]; 647 struct radix_tree_path *pathp = path; 648 struct radix_tree_node *slot; 649 unsigned int shift; 650 unsigned long tagged = 0; 651 unsigned long index = *first_indexp; 652 653 last_index = min(last_index, radix_tree_maxindex(height)); 654 if (index > last_index) 655 return 0; 656 if (!nr_to_tag) 657 return 0; 658 if (!root_tag_get(root, iftag)) { 659 *first_indexp = last_index + 1; 660 return 0; 661 } 662 if (height == 0) { 663 *first_indexp = last_index + 1; 664 root_tag_set(root, settag); 665 return 1; 666 } ... 733 root_tag_set(root, settag); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 734 *first_indexp = index; 735 736 return tagged; 737 } As the result, there is no radix_tree_node which is set with PAGECACHE_TAG_TOWRITE but the root tag(radix_tree_root) is set with PAGECACHE_TAG_TOWRITE. [figure: inside radix_tree] (Please see the figure with typewriter font) =========================================== [roottag = DIRTY] | tag=0:NOTHING tag[0 0 0 1] 1:DIRTY [x x x +] 2:WRITEBACK | 3:DIRTY,WRITEBACK p 4:TOWRITE <---> 5:DIRTY,TOWRITE ... specified range (index: 0 to 2) * There is no DIRTY tag within the specified range. (But there is a DIRTY tag outside that range.) | | | | | | | | | after calling tag_pages_for_writeback() | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | p is "page". tag[0 0 0 1] x is NULL. [x x x +] +- is a pointer to "page". | p * But TOWRITE tag is set on the root tag. ============================================ After that, radix_tree_extend() via radix_tree_insert() is called when the page is added. This function sets the new radix_tree_node with PAGECACHE_TAG_TOWRITE to succeed the status of the root tag. 246 static int radix_tree_extend(struct radix_tree_root *root, unsigned long index) 247 { 248 struct radix_tree_node *node; 249 unsigned int height; 250 int tag; 251 252 /* Figure out what the height should be. */ 253 height = root->height + 1; 254 while (index > radix_tree_maxindex(height)) 255 height++; 256 257 if (root->rnode == NULL) { 258 root->height = height; 259 goto out; 260 } 261 262 do { 263 unsigned int newheight; 264 if (!(node = radix_tree_node_alloc(root))) 265 return -ENOMEM; 266 267 /* Increase the height. */ 268 node->slots[0] = radix_tree_indirect_to_ptr(root->rnode); 269 270 /* Propagate the aggregated tag info into the new root */ 271 for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) { 272 if (root_tag_get(root, tag)) 273 tag_set(node, tag, 0); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 274 } =========================================== [roottag = DIRTY,TOWRITE] | : tag[0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p (new page) | | | | | | | | | after calling radix_tree_insert | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | tag [5 0 0 0] * DIRTY and TOWRITE tags are [+ + x x] succeeded to the new node. | | tag [0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p ============================================ After that, the index 3 page is released by remove_from_page_cache(). Then we can make the situation that the tag is set with PAGECACHE_TAG_TOWRITE and that the slot which corresponds to the tag is NULL. =========================================== [roottag = DIRTY,TOWRITE] | tag [5 0 0 0] [+ + x x] | | tag [0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p (remove) | | | | | | | | | after calling remove_page_cache | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | tag [4 0 0 0] * Only DIRTY tag is cleared [x + x x] because no TOWRITE tag is existed | in the bottom node. [0 0 0 0] [+ x x x] | p ============================================ To solve this problem Change to that radix_tree_tag_if_tagged() doesn't tag the root tag if it doesn't set any tags within the specified range. Like this. ============================================ 640 unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root, 641 unsigned long *first_indexp, unsigned long last_index, 642 unsigned long nr_to_tag, 643 unsigned int iftag, unsigned int settag) 644 { 650 unsigned long tagged = 0; ... 733 if (tagged) ^^^^^^^^^^^^^^^^^^^^^^^^ 734 root_tag_set(root, settag); 735 *first_indexp = index; 736 737 return tagged; 738 } ============================================ Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-26 06:07:32 +07:00
* We need not to tag the root tag if there is no tag which is set with
* settag within the range from *first_indexp to last_index.
*/
radix_tree: radix_tree_gang_lookup_tag_slot() may never return Executed command: fsstress -d /mnt -n 600 -p 850 crash> bt PID: 7947 TASK: ffff880160546a70 CPU: 0 COMMAND: "fsstress" #0 [ffff8800dfc07d00] machine_kexec at ffffffff81030db9 #1 [ffff8800dfc07d70] crash_kexec at ffffffff810a7952 #2 [ffff8800dfc07e40] oops_end at ffffffff814aa7c8 #3 [ffff8800dfc07e70] die_nmi at ffffffff814aa969 #4 [ffff8800dfc07ea0] do_nmi_callback at ffffffff8102b07b #5 [ffff8800dfc07f10] do_nmi at ffffffff814aa514 #6 [ffff8800dfc07f50] nmi at ffffffff814a9d60 [exception RIP: __lookup_tag+100] RIP: ffffffff812274b4 RSP: ffff88016056b998 RFLAGS: 00000287 RAX: 0000000000000000 RBX: 0000000000000002 RCX: 0000000000000006 RDX: 000000000000001d RSI: ffff88016056bb18 RDI: ffff8800c85366e0 RBP: ffff88016056b9c8 R8: ffff88016056b9e8 R9: 0000000000000000 R10: 000000000000000e R11: ffff8800c8536908 R12: 0000000000000010 R13: 0000000000000040 R14: ffffffffffffffc0 R15: ffff8800c85366e0 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 <NMI exception stack> #7 [ffff88016056b998] __lookup_tag at ffffffff812274b4 #8 [ffff88016056b9d0] radix_tree_gang_lookup_tag_slot at ffffffff81227605 #9 [ffff88016056ba20] find_get_pages_tag at ffffffff810fc110 #10 [ffff88016056ba80] pagevec_lookup_tag at ffffffff81105e85 #11 [ffff88016056baa0] write_cache_pages at ffffffff81104c47 #12 [ffff88016056bbd0] generic_writepages at ffffffff81105014 #13 [ffff88016056bbe0] do_writepages at ffffffff81105055 #14 [ffff88016056bbf0] __filemap_fdatawrite_range at ffffffff810fb2cb #15 [ffff88016056bc40] filemap_write_and_wait_range at ffffffff810fb32a #16 [ffff88016056bc70] generic_file_direct_write at ffffffff810fb3dc #17 [ffff88016056bce0] __generic_file_aio_write at ffffffff810fcee5 #18 [ffff88016056bda0] generic_file_aio_write at ffffffff810fd085 #19 [ffff88016056bdf0] do_sync_write at ffffffff8114f9ea #20 [ffff88016056bf00] vfs_write at ffffffff8114fcf8 #21 [ffff88016056bf30] sys_write at ffffffff81150691 #22 [ffff88016056bf80] system_call_fastpath at ffffffff8100c0b2 I think this root cause is the following: radix_tree_range_tag_if_tagged() always tags the root tag with settag if the root tag is set with iftag even if there are no iftag tags in the specified range (Of course, there are some iftag tags outside the specified range). =============================================================================== [[[Detailed description]]] (1) Why cannot radix_tree_gang_lookup_tag_slot() return forever? __lookup_tag(): - Return with 0. - Return with the index which is not bigger than the old one as the input parameter. Therefore the following "while" repeats forever because the above conditions cause "ret" not to be updated and the cur_index cannot be changed into the bigger one. (So, radix_tree_gang_lookup_tag_slot() cannot return forever.) radix_tree_gang_lookup_tag_slot(): 1178 while (ret < max_items) { 1179 unsigned int slots_found; 1180 unsigned long next_index; /* Index of next search */ 1181 1182 if (cur_index > max_index) 1183 break; 1184 slots_found = __lookup_tag(node, results + ret, 1185 cur_index, max_items - ret, &next_index, tag); 1186 ret += slots_found; // cannot update ret because slots_found == 0. // so, this while loops forever. 1187 if (next_index == 0) 1188 break; 1189 cur_index = next_index; 1190 } (2) Why does __lookup_tag() return with 0 and doesn't update the index? Assuming the following: - the one of the slot in radix_tree_node is NULL. - the one of the tag which corresponds to the slot sets with PAGECACHE_TAG_TOWRITE or other. - In a certain height(!=0), the corresponding index is 0. a) __lookup_tag() notices that the tag is set. 1005 static unsigned int 1006 __lookup_tag(struct radix_tree_node *slot, void ***results, unsigned long index, 1007 unsigned int max_items, unsigned long *next_index, unsigned int tag) 1008 { 1009 unsigned int nr_found = 0; 1010 unsigned int shift, height; 1011 1012 height = slot->height; 1013 if (height == 0) 1014 goto out; 1015 shift = (height-1) * RADIX_TREE_MAP_SHIFT; 1016 1017 while (height > 0) { 1018 unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK ; 1019 1020 for (;;) { 1021 if (tag_get(slot, tag, i)) 1022 break; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ * the index is not updated yet. b) __lookup_tag() notices that the slot is NULL. 1023 index &= ~((1UL << shift) - 1); 1024 index += 1UL << shift; 1025 if (index == 0) 1026 goto out; /* 32-bit wraparound */ 1027 i++; 1028 if (i == RADIX_TREE_MAP_SIZE) 1029 goto out; 1030 } 1031 height--; 1032 if (height == 0) { /* Bottom level: grab some items */ ... 1055 } 1056 shift -= RADIX_TREE_MAP_SHIFT; 1057 slot = rcu_dereference_raw(slot->slots[i]); 1058 if (slot == NULL) 1059 break; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ c) __lookup_tag() doesn't update the index and return with 0. 1060 } 1061 out: 1062 *next_index = index; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 1063 return nr_found; 1064 } (3) Why is the slot NULL even if the tag is set? Because radix_tree_range_tag_if_tagged() always sets the root tag with PAGECACHE_TAG_TOWRITE if the root tag is set with PAGECACHE_TAG_DIRTY, even if there is no tag which can be set with PAGECACHE_TAG_TOWRITE in the specified range (from *first_indexp to last_index). Of course, some PAGECACHE_TAG_DIRTY nodes must exist outside the specified range. (radix_tree_range_tag_if_tagged() is called only from tag_pages_for_writeback()) 640 unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root, 641 unsigned long *first_indexp, unsigned long last_index, 642 unsigned long nr_to_tag, 643 unsigned int iftag, unsigned int settag) 644 { 645 unsigned int height = root->height; 646 struct radix_tree_path path[height]; 647 struct radix_tree_path *pathp = path; 648 struct radix_tree_node *slot; 649 unsigned int shift; 650 unsigned long tagged = 0; 651 unsigned long index = *first_indexp; 652 653 last_index = min(last_index, radix_tree_maxindex(height)); 654 if (index > last_index) 655 return 0; 656 if (!nr_to_tag) 657 return 0; 658 if (!root_tag_get(root, iftag)) { 659 *first_indexp = last_index + 1; 660 return 0; 661 } 662 if (height == 0) { 663 *first_indexp = last_index + 1; 664 root_tag_set(root, settag); 665 return 1; 666 } ... 733 root_tag_set(root, settag); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 734 *first_indexp = index; 735 736 return tagged; 737 } As the result, there is no radix_tree_node which is set with PAGECACHE_TAG_TOWRITE but the root tag(radix_tree_root) is set with PAGECACHE_TAG_TOWRITE. [figure: inside radix_tree] (Please see the figure with typewriter font) =========================================== [roottag = DIRTY] | tag=0:NOTHING tag[0 0 0 1] 1:DIRTY [x x x +] 2:WRITEBACK | 3:DIRTY,WRITEBACK p 4:TOWRITE <---> 5:DIRTY,TOWRITE ... specified range (index: 0 to 2) * There is no DIRTY tag within the specified range. (But there is a DIRTY tag outside that range.) | | | | | | | | | after calling tag_pages_for_writeback() | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | p is "page". tag[0 0 0 1] x is NULL. [x x x +] +- is a pointer to "page". | p * But TOWRITE tag is set on the root tag. ============================================ After that, radix_tree_extend() via radix_tree_insert() is called when the page is added. This function sets the new radix_tree_node with PAGECACHE_TAG_TOWRITE to succeed the status of the root tag. 246 static int radix_tree_extend(struct radix_tree_root *root, unsigned long index) 247 { 248 struct radix_tree_node *node; 249 unsigned int height; 250 int tag; 251 252 /* Figure out what the height should be. */ 253 height = root->height + 1; 254 while (index > radix_tree_maxindex(height)) 255 height++; 256 257 if (root->rnode == NULL) { 258 root->height = height; 259 goto out; 260 } 261 262 do { 263 unsigned int newheight; 264 if (!(node = radix_tree_node_alloc(root))) 265 return -ENOMEM; 266 267 /* Increase the height. */ 268 node->slots[0] = radix_tree_indirect_to_ptr(root->rnode); 269 270 /* Propagate the aggregated tag info into the new root */ 271 for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) { 272 if (root_tag_get(root, tag)) 273 tag_set(node, tag, 0); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 274 } =========================================== [roottag = DIRTY,TOWRITE] | : tag[0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p (new page) | | | | | | | | | after calling radix_tree_insert | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | tag [5 0 0 0] * DIRTY and TOWRITE tags are [+ + x x] succeeded to the new node. | | tag [0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p ============================================ After that, the index 3 page is released by remove_from_page_cache(). Then we can make the situation that the tag is set with PAGECACHE_TAG_TOWRITE and that the slot which corresponds to the tag is NULL. =========================================== [roottag = DIRTY,TOWRITE] | tag [5 0 0 0] [+ + x x] | | tag [0 0 0 1] [0 0 0 0] [x x x +] [+ x x x] | | p p (remove) | | | | | | | | | after calling remove_page_cache | | | | | | | | | v v v v v v v v v [roottag = DIRTY,TOWRITE] | tag [4 0 0 0] * Only DIRTY tag is cleared [x + x x] because no TOWRITE tag is existed | in the bottom node. [0 0 0 0] [+ x x x] | p ============================================ To solve this problem Change to that radix_tree_tag_if_tagged() doesn't tag the root tag if it doesn't set any tags within the specified range. Like this. ============================================ 640 unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root, 641 unsigned long *first_indexp, unsigned long last_index, 642 unsigned long nr_to_tag, 643 unsigned int iftag, unsigned int settag) 644 { 650 unsigned long tagged = 0; ... 733 if (tagged) ^^^^^^^^^^^^^^^^^^^^^^^^ 734 root_tag_set(root, settag); 735 *first_indexp = index; 736 737 return tagged; 738 } ============================================ Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-26 06:07:32 +07:00
if (tagged > 0)
root_tag_set(root, settag);
*first_indexp = index;
return tagged;
}
EXPORT_SYMBOL(radix_tree_range_tag_if_tagged);
/**
* radix_tree_next_hole - find the next hole (not-present entry)
* @root: tree root
* @index: index key
* @max_scan: maximum range to search
*
* Search the set [index, min(index+max_scan-1, MAX_INDEX)] for the lowest
* indexed hole.
*
* Returns: the index of the hole if found, otherwise returns an index
* outside of the set specified (in which case 'return - index >= max_scan'
* will be true). In rare cases of index wrap-around, 0 will be returned.
*
* radix_tree_next_hole may be called under rcu_read_lock. However, like
* radix_tree_gang_lookup, this will not atomically search a snapshot of
* the tree at a single point in time. For example, if a hole is created
* at index 5, then subsequently a hole is created at index 10,
* radix_tree_next_hole covering both indexes may return 10 if called
* under rcu_read_lock.
*/
unsigned long radix_tree_next_hole(struct radix_tree_root *root,
unsigned long index, unsigned long max_scan)
{
unsigned long i;
for (i = 0; i < max_scan; i++) {
if (!radix_tree_lookup(root, index))
break;
index++;
if (index == 0)
break;
}
return index;
}
EXPORT_SYMBOL(radix_tree_next_hole);
/**
* radix_tree_prev_hole - find the prev hole (not-present entry)
* @root: tree root
* @index: index key
* @max_scan: maximum range to search
*
* Search backwards in the range [max(index-max_scan+1, 0), index]
* for the first hole.
*
* Returns: the index of the hole if found, otherwise returns an index
* outside of the set specified (in which case 'index - return >= max_scan'
* will be true). In rare cases of wrap-around, ULONG_MAX will be returned.
*
* radix_tree_next_hole may be called under rcu_read_lock. However, like
* radix_tree_gang_lookup, this will not atomically search a snapshot of
* the tree at a single point in time. For example, if a hole is created
* at index 10, then subsequently a hole is created at index 5,
* radix_tree_prev_hole covering both indexes may return 5 if called under
* rcu_read_lock.
*/
unsigned long radix_tree_prev_hole(struct radix_tree_root *root,
unsigned long index, unsigned long max_scan)
{
unsigned long i;
for (i = 0; i < max_scan; i++) {
if (!radix_tree_lookup(root, index))
break;
index--;
if (index == ULONG_MAX)
break;
}
return index;
}
EXPORT_SYMBOL(radix_tree_prev_hole);
/**
* radix_tree_gang_lookup - perform multiple lookup on a radix tree
* @root: radix tree root
* @results: where the results of the lookup are placed
* @first_index: start the lookup from this key
* @max_items: place up to this many items at *results
*
* Performs an index-ascending scan of the tree for present items. Places
* them at *@results and returns the number of items which were placed at
* *@results.
*
* The implementation is naive.
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
*
* Like radix_tree_lookup, radix_tree_gang_lookup may be called under
* rcu_read_lock. In this case, rather than the returned results being
* an atomic snapshot of the tree at a single point in time, the semantics
* of an RCU protected gang lookup are as though multiple radix_tree_lookups
* have been issued in individual locks, and results stored in 'results'.
*/
unsigned int
radix_tree_gang_lookup(struct radix_tree_root *root, void **results,
unsigned long first_index, unsigned int max_items)
{
struct radix_tree_iter iter;
void **slot;
unsigned int ret = 0;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (unlikely(!max_items))
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
return 0;
radix_tree_for_each_slot(slot, root, &iter, first_index) {
results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot));
if (!results[ret])
continue;
if (++ret == max_items)
break;
}
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
return ret;
}
EXPORT_SYMBOL(radix_tree_gang_lookup);
/**
* radix_tree_gang_lookup_slot - perform multiple slot lookup on radix tree
* @root: radix tree root
* @results: where the results of the lookup are placed
radix_tree: exceptional entries and indices A patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its peculiar swap vector, instead keeping a file's swap entries in the same radix tree as its struct page pointers: thus saving memory, and simplifying its code and locking. This patch: The radix_tree is used by several subsystems for different purposes. A major use is to store the struct page pointers of a file's pagecache for memory management. But what if mm wanted to store something other than page pointers there too? The low bit of a radix_tree entry is already used to denote an indirect pointer, for internal use, and the unlikely radix_tree_deref_retry() case. Define the next bit as denoting an exceptional entry, and supply inline functions radix_tree_exception() to return non-0 in either unlikely case, and radix_tree_exceptional_entry() to return non-0 in the second case. If a subsystem already uses radix_tree with that bit set, no problem: it does not affect internal workings at all, but is defined for the convenience of those storing well-aligned pointers in the radix_tree. The radix_tree_gang_lookups have an implicit assumption that the caller can deduce the offset of each entry returned e.g. by the page->index of a struct page. But that may not be feasible for some kinds of item to be stored there. radix_tree_gang_lookup_slot() allow for an optional indices argument, output array in which to return those offsets. The same could be added to other radix_tree_gang_lookups, but for now keep it to the only one for which we need it. Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-08-04 06:21:18 +07:00
* @indices: where their indices should be placed (but usually NULL)
* @first_index: start the lookup from this key
* @max_items: place up to this many items at *results
*
* Performs an index-ascending scan of the tree for present items. Places
* their slots at *@results and returns the number of items which were
* placed at *@results.
*
* The implementation is naive.
*
* Like radix_tree_gang_lookup as far as RCU and locking goes. Slots must
* be dereferenced with radix_tree_deref_slot, and if using only RCU
* protection, radix_tree_deref_slot may fail requiring a retry.
*/
unsigned int
radix_tree: exceptional entries and indices A patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its peculiar swap vector, instead keeping a file's swap entries in the same radix tree as its struct page pointers: thus saving memory, and simplifying its code and locking. This patch: The radix_tree is used by several subsystems for different purposes. A major use is to store the struct page pointers of a file's pagecache for memory management. But what if mm wanted to store something other than page pointers there too? The low bit of a radix_tree entry is already used to denote an indirect pointer, for internal use, and the unlikely radix_tree_deref_retry() case. Define the next bit as denoting an exceptional entry, and supply inline functions radix_tree_exception() to return non-0 in either unlikely case, and radix_tree_exceptional_entry() to return non-0 in the second case. If a subsystem already uses radix_tree with that bit set, no problem: it does not affect internal workings at all, but is defined for the convenience of those storing well-aligned pointers in the radix_tree. The radix_tree_gang_lookups have an implicit assumption that the caller can deduce the offset of each entry returned e.g. by the page->index of a struct page. But that may not be feasible for some kinds of item to be stored there. radix_tree_gang_lookup_slot() allow for an optional indices argument, output array in which to return those offsets. The same could be added to other radix_tree_gang_lookups, but for now keep it to the only one for which we need it. Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-08-04 06:21:18 +07:00
radix_tree_gang_lookup_slot(struct radix_tree_root *root,
void ***results, unsigned long *indices,
unsigned long first_index, unsigned int max_items)
{
struct radix_tree_iter iter;
void **slot;
unsigned int ret = 0;
if (unlikely(!max_items))
return 0;
radix_tree_for_each_slot(slot, root, &iter, first_index) {
results[ret] = slot;
radix_tree: exceptional entries and indices A patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its peculiar swap vector, instead keeping a file's swap entries in the same radix tree as its struct page pointers: thus saving memory, and simplifying its code and locking. This patch: The radix_tree is used by several subsystems for different purposes. A major use is to store the struct page pointers of a file's pagecache for memory management. But what if mm wanted to store something other than page pointers there too? The low bit of a radix_tree entry is already used to denote an indirect pointer, for internal use, and the unlikely radix_tree_deref_retry() case. Define the next bit as denoting an exceptional entry, and supply inline functions radix_tree_exception() to return non-0 in either unlikely case, and radix_tree_exceptional_entry() to return non-0 in the second case. If a subsystem already uses radix_tree with that bit set, no problem: it does not affect internal workings at all, but is defined for the convenience of those storing well-aligned pointers in the radix_tree. The radix_tree_gang_lookups have an implicit assumption that the caller can deduce the offset of each entry returned e.g. by the page->index of a struct page. But that may not be feasible for some kinds of item to be stored there. radix_tree_gang_lookup_slot() allow for an optional indices argument, output array in which to return those offsets. The same could be added to other radix_tree_gang_lookups, but for now keep it to the only one for which we need it. Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-08-04 06:21:18 +07:00
if (indices)
indices[ret] = iter.index;
if (++ret == max_items)
break;
}
return ret;
}
EXPORT_SYMBOL(radix_tree_gang_lookup_slot);
/**
* radix_tree_gang_lookup_tag - perform multiple lookup on a radix tree
* based on a tag
* @root: radix tree root
* @results: where the results of the lookup are placed
* @first_index: start the lookup from this key
* @max_items: place up to this many items at *results
* @tag: the tag index (< RADIX_TREE_MAX_TAGS)
*
* Performs an index-ascending scan of the tree for present items which
* have the tag indexed by @tag set. Places the items at *@results and
* returns the number of items which were placed at *@results.
*/
unsigned int
radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
unsigned long first_index, unsigned int max_items,
unsigned int tag)
{
struct radix_tree_iter iter;
void **slot;
unsigned int ret = 0;
if (unlikely(!max_items))
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
return 0;
radix_tree_for_each_tagged(slot, root, &iter, first_index, tag) {
results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot));
if (!results[ret])
continue;
if (++ret == max_items)
break;
}
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
return ret;
}
EXPORT_SYMBOL(radix_tree_gang_lookup_tag);
/**
* radix_tree_gang_lookup_tag_slot - perform multiple slot lookup on a
* radix tree based on a tag
* @root: radix tree root
* @results: where the results of the lookup are placed
* @first_index: start the lookup from this key
* @max_items: place up to this many items at *results
* @tag: the tag index (< RADIX_TREE_MAX_TAGS)
*
* Performs an index-ascending scan of the tree for present items which
* have the tag indexed by @tag set. Places the slots at *@results and
* returns the number of slots which were placed at *@results.
*/
unsigned int
radix_tree_gang_lookup_tag_slot(struct radix_tree_root *root, void ***results,
unsigned long first_index, unsigned int max_items,
unsigned int tag)
{
struct radix_tree_iter iter;
void **slot;
unsigned int ret = 0;
if (unlikely(!max_items))
return 0;
radix_tree_for_each_tagged(slot, root, &iter, first_index, tag) {
results[ret] = slot;
if (++ret == max_items)
break;
}
return ret;
}
EXPORT_SYMBOL(radix_tree_gang_lookup_tag_slot);
tmpfs radix_tree: locate_item to speed up swapoff We have already acknowledged that swapoff of a tmpfs file is slower than it was before conversion to the generic radix_tree: a little slower there will be acceptable, if the hotter paths are faster. But it was a shock to find swapoff of a 500MB file 20 times slower on my laptop, taking 10 minutes; and at that rate it significantly slows down my testing. Now, most of that turned out to be overhead from PROVE_LOCKING and PROVE_RCU: without those it was only 4 times slower than before; and more realistic tests on other machines don't fare as badly. I've tried a number of things to improve it, including tagging the swap entries, then doing lookup by tag: I'd expected that to halve the time, but in practice it's erratic, and often counter-productive. The only change I've so far found to make a consistent improvement, is to short-circuit the way we go back and forth, gang lookup packing entries into the array supplied, then shmem scanning that array for the target entry. Scanning in place doubles the speed, so it's now only twice as slow as before (or three times slower when the PROVEs are on). So, add radix_tree_locate_item() as an expedient, once-off, single-caller hack to do the lookup directly in place. #ifdef it on CONFIG_SHMEM and CONFIG_SWAP, as much to document its limited applicability as save space in other configurations. And, sadly, #include sched.h for cond_resched(). Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-08-04 06:21:27 +07:00
#if defined(CONFIG_SHMEM) && defined(CONFIG_SWAP)
#include <linux/sched.h> /* for cond_resched() */
/*
* This linear search is at present only useful to shmem_unuse_inode().
*/
static unsigned long __locate(struct radix_tree_node *slot, void *item,
unsigned long index, unsigned long *found_index)
{
unsigned int shift, height;
unsigned long i;
height = slot->height;
shift = (height-1) * RADIX_TREE_MAP_SHIFT;
for ( ; height > 1; height--) {
i = (index >> shift) & RADIX_TREE_MAP_MASK;
for (;;) {
if (slot->slots[i] != NULL)
break;
index &= ~((1UL << shift) - 1);
index += 1UL << shift;
if (index == 0)
goto out; /* 32-bit wraparound */
i++;
if (i == RADIX_TREE_MAP_SIZE)
goto out;
}
shift -= RADIX_TREE_MAP_SHIFT;
slot = rcu_dereference_raw(slot->slots[i]);
if (slot == NULL)
goto out;
}
/* Bottom level: check items */
for (i = 0; i < RADIX_TREE_MAP_SIZE; i++) {
if (slot->slots[i] == item) {
*found_index = index + i;
index = 0;
goto out;
}
}
index += RADIX_TREE_MAP_SIZE;
out:
return index;
}
/**
* radix_tree_locate_item - search through radix tree for item
* @root: radix tree root
* @item: item to be found
*
* Returns index where item was found, or -1 if not found.
* Caller must hold no lock (since this time-consuming function needs
* to be preemptible), and must check afterwards if item is still there.
*/
unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item)
{
struct radix_tree_node *node;
unsigned long max_index;
unsigned long cur_index = 0;
unsigned long found_index = -1;
do {
rcu_read_lock();
node = rcu_dereference_raw(root->rnode);
if (!radix_tree_is_indirect_ptr(node)) {
rcu_read_unlock();
if (node == item)
found_index = 0;
break;
}
node = indirect_to_ptr(node);
max_index = radix_tree_maxindex(node->height);
if (cur_index > max_index)
break;
cur_index = __locate(node, item, cur_index, &found_index);
rcu_read_unlock();
cond_resched();
} while (cur_index != 0 && cur_index <= max_index);
return found_index;
}
#else
unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item)
{
return -1;
}
#endif /* CONFIG_SHMEM && CONFIG_SWAP */
/**
* radix_tree_shrink - shrink height of a radix tree to minimal
* @root radix tree root
*/
static inline void radix_tree_shrink(struct radix_tree_root *root)
{
/* try to shrink tree height */
while (root->height > 0) {
struct radix_tree_node *to_free = root->rnode;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
struct radix_tree_node *slot;
BUG_ON(!radix_tree_is_indirect_ptr(to_free));
to_free = indirect_to_ptr(to_free);
/*
* The candidate node has more than one child, or its child
* is not at the leftmost slot, we cannot shrink.
*/
if (to_free->count != 1)
break;
if (!to_free->slots[0])
break;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
/*
* We don't need rcu_assign_pointer(), since we are simply
* moving the node from one part of the tree to another: if it
* was safe to dereference the old pointer to it
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
* (to_free->slots[0]), it will be safe to dereference the new
* one (root->rnode) as far as dependent read barriers go.
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
*/
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
slot = to_free->slots[0];
if (root->height > 1) {
slot->parent = NULL;
slot = ptr_to_indirect(slot);
}
root->rnode = slot;
root->height--;
/*
* We have a dilemma here. The node's slot[0] must not be
* NULLed in case there are concurrent lookups expecting to
* find the item. However if this was a bottom-level node,
* then it may be subject to the slot pointer being visible
* to callers dereferencing it. If item corresponding to
* slot[0] is subsequently deleted, these callers would expect
* their slot to become empty sooner or later.
*
* For example, lockless pagecache will look up a slot, deref
* the page pointer, and if the page is 0 refcount it means it
* was concurrently deleted from pagecache so try the deref
* again. Fortunately there is already a requirement for logic
* to retry the entire slot lookup -- the indirect pointer
* problem (replacing direct root node with an indirect pointer
* also results in a stale slot). So tag the slot as indirect
* to force callers to retry.
*/
if (root->height == 0)
*((unsigned long *)&to_free->slots[0]) |=
RADIX_TREE_INDIRECT_PTR;
radix_tree_node_free(to_free);
}
}
/**
* radix_tree_delete - delete an item from a radix tree
* @root: radix tree root
* @index: index key
*
* Remove the item at @index from the radix tree rooted at @root.
*
* Returns the address of the deleted item, or NULL if it was not present.
*/
void *radix_tree_delete(struct radix_tree_root *root, unsigned long index)
{
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
struct radix_tree_node *node = NULL;
struct radix_tree_node *slot = NULL;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
struct radix_tree_node *to_free;
unsigned int height, shift;
int tag;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
int uninitialized_var(offset);
height = root->height;
if (index > radix_tree_maxindex(height))
goto out;
slot = root->rnode;
if (height == 0) {
root_tag_clear_all(root);
root->rnode = NULL;
goto out;
}
slot = indirect_to_ptr(slot);
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
shift = height * RADIX_TREE_MAP_SHIFT;
do {
if (slot == NULL)
goto out;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
shift -= RADIX_TREE_MAP_SHIFT;
offset = (index >> shift) & RADIX_TREE_MAP_MASK;
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
node = slot;
slot = slot->slots[offset];
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
} while (shift);
if (slot == NULL)
goto out;
/*
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
* Clear all tags associated with the item to be deleted.
* This way of doing it would be inefficient, but seldom is any set.
*/
for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) {
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
if (tag_get(node, tag, offset))
radix_tree_tag_clear(root, index, tag);
}
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
to_free = NULL;
/* Now free the nodes we do not need anymore */
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
while (node) {
node->slots[offset] = NULL;
node->count--;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
/*
* Queue the node for deferred freeing after the
* last reference to it disappears (set NULL, above).
*/
if (to_free)
radix_tree_node_free(to_free);
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
if (node->count) {
if (node == indirect_to_ptr(root->rnode))
radix_tree_shrink(root);
goto out;
}
/* Node with zero slots in use so free it */
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
to_free = node;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
index >>= RADIX_TREE_MAP_SHIFT;
offset = index & RADIX_TREE_MAP_MASK;
node = node->parent;
}
radix_tree: take radix_tree_path off stack Down, down in the deepest depths of GFP_NOIO page reclaim, we have shrink_page_list() calling __remove_mapping() calling __delete_from_ swap_cache() or __delete_from_page_cache(). You would not expect those to need much stack, but in fact they call radix_tree_delete(): which declares a 192-byte radix_tree_path array on its stack (to record the node,offsets it visits when descending, in case it needs to ascend to update them). And if any tag is still set [1], that calls radix_tree_tag_clear(), which declares a further such 192-byte radix_tree_path array on the stack. (At least we have interrupts disabled here, so won't then be pushing registers too.) That was probably a good choice when most users were 32-bit (array of half the size), and adding fields to radix_tree_node would have bloated it unnecessarily. But nowadays many are 64-bit, and each radix_tree_node contains a struct rcu_head, which is only used when freeing; whereas the radix_tree_path info is only used for updating the tree (deleting, clearing tags or setting tags if tagged) when a lock must be held, of no interest when accessing the tree locklessly. So add a parent pointer to the radix_tree_node, in union with the rcu_head, and remove all uses of the radix_tree_path. There would be space in that union to save the offset when descending as before (we can argue that a lock must already be held to exclude other users), but recalculating it when ascending is both easy (a constant shift and a constant mask) and uncommon, so it seems better just to do that. Two little optimizations: no need to decrement height when descending, adjusting shift is enough; and once radix_tree_tag_if_tagged() has set tag on a node and its ancestors, it need not ascend from that node again. perf on the radix tree test harness reports radix_tree_insert() as 2% slower (now having to set parent), but radix_tree_delete() 24% faster. Surely that's an exaggeration from rtth's artificially low map shift 3, but forcing it back to 6 still rates radix_tree_delete() 8% faster. [1] Can a pagecache tag (dirty, writeback or towrite) actually still be set at the time of radix_tree_delete()? Perhaps not if the filesystem is well-behaved. But although I've not tracked any stack overflow down to this cause, I have observed a curious case in which a dirty tag is set and left set on tmpfs: page migration's migrate_page_copy() happens to use __set_page_dirty_nobuffers() to set PageDirty on the newpage, and that sets PAGECACHE_TAG_DIRTY as a side-effect - harmless to a filesystem which doesn't use tags, except for this stack depth issue. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Dave Chinner <david@fromorbit.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nai Xia <nai.xia@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-13 08:20:41 +07:00
root_tag_clear_all(root);
root->height = 0;
root->rnode = NULL;
[PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:44 +07:00
if (to_free)
radix_tree_node_free(to_free);
out:
return slot;
}
EXPORT_SYMBOL(radix_tree_delete);
/**
* radix_tree_tagged - test whether any items in the tree are tagged
* @root: radix tree root
* @tag: tag to test
*/
int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag)
{
return root_tag_get(root, tag);
}
EXPORT_SYMBOL(radix_tree_tagged);
static void
radix_tree_node_ctor(void *node)
{
memset(node, 0, sizeof(struct radix_tree_node));
}
static __init unsigned long __maxindex(unsigned int height)
{
avoid negative (and full-width) shifts in radix-tree.c Negative shifts are not allowed in C (the result is undefined). Same thing with full-width shifts. It works on most platforms but not on the VAX with gcc 4.0.1 (it results in an "operand reserved" fault). Shifting by more than the width of the value on the left is also not allowed. I think the extra '>> 1' tacked on at the end in the original code was an attempt to work around that. Getting rid of that is an extra feature of this patch. Here's the chapter and verse, taken from the final draft of the C99 standard ("6.5.7 Bitwise shift operators", paragraph 3): "The integer promotions are performed on each of the operands. The type of the result is that of the promoted left operand. If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined." Thank you to Jan-Benedict Glaw, Christoph Hellwig, Maciej Rozycki, Pekka Enberg, Andreas Schwab, and Christoph Lameter for review. Special thanks to Andreas for spotting that my fix only removed half the undefined behaviour. Signed-off-by: Peter Lund <firefly@vax64.dk> Christoph Lameter <clameter@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: "Maciej W. Rozycki" <macro@linux-mips.org> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Cc: Andreas Schwab <schwab@suse.de> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: WU Fengguang <wfg@mail.ustc.edu.cn> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 13:29:35 +07:00
unsigned int width = height * RADIX_TREE_MAP_SHIFT;
int shift = RADIX_TREE_INDEX_BITS - width;
if (shift < 0)
return ~0UL;
if (shift >= BITS_PER_LONG)
return 0UL;
return ~0UL >> shift;
}
static __init void radix_tree_init_maxindex(void)
{
unsigned int i;
for (i = 0; i < ARRAY_SIZE(height_to_maxindex); i++)
height_to_maxindex[i] = __maxindex(i);
}
static int radix_tree_callback(struct notifier_block *nfb,
unsigned long action,
void *hcpu)
{
int cpu = (long)hcpu;
struct radix_tree_preload *rtp;
/* Free per-cpu pool of perloaded nodes */
if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
rtp = &per_cpu(radix_tree_preloads, cpu);
while (rtp->nr) {
kmem_cache_free(radix_tree_node_cachep,
rtp->nodes[rtp->nr-1]);
rtp->nodes[rtp->nr-1] = NULL;
rtp->nr--;
}
}
return NOTIFY_OK;
}
void __init radix_tree_init(void)
{
radix_tree_node_cachep = kmem_cache_create("radix_tree_node",
sizeof(struct radix_tree_node), 0,
SLAB_PANIC | SLAB_RECLAIM_ACCOUNT,
radix_tree_node_ctor);
radix_tree_init_maxindex();
hotcpu_notifier(radix_tree_callback, 0);
}