13 Jan, 2012
1 commit
-
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
Cc: Jan Kara
Cc: Dave Chinner
Cc: Mel Gorman
Cc: Nai Xia
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
01 Nov, 2011
1 commit
-
radix_tree_tag_get()'s BUG (when it sees a tag after saw_unset_tag) was
unsafe and removed in 2.6.34, but the pointless saw_unset_tag left behind.Remove it now, and return 0 as soon as we see unset tag - we already rely
upon the root tag to be correct, returning 0 immediately if it's not set.Signed-off-by: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
04 Aug, 2011
2 commits
-
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
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
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
Acked-by: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
26 Jan, 2011
1 commit
-
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
#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 ffffffff8100c0b2I 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
Acked-by: Jan Kara
Cc: Dave Chinner
Cc: Nick Piggin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
12 Nov, 2010
1 commit
-
Salman Qazi describes the following radix-tree bug:
In the following case, we get can get a deadlock:
0. The radix tree contains two items, one has the index 0.
1. The reader (in this case find_get_pages) takes the rcu_read_lock.
2. The reader acquires slot(s) for item(s) including the index 0 item.
3. The non-zero index item is deleted, and as a consequence the other item is
moved to the root of the tree. The place where it used to be is queued for
deletion after the readers finish.
3b. The zero item is deleted, removing it from the direct slot, it remains in
the rcu-delayed indirect node.
4. The reader looks at the index 0 slot, and finds that the page has 0 ref
count
5. The reader looks at it again, hoping that the item will either be freed or
the ref count will increase. This never happens, as the slot it is looking
at will never be updated. Also, this slot can never be reclaimed because
the reader is holding rcu_read_lock and is in an infinite loop.The fix is to re-use the same "indirect" pointer case that requires a slot
lookup retry into a general "retry the lookup" bit.Signed-off-by: Nick Piggin
Reported-by: Salman Qazi
Cc:
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
07 Oct, 2010
1 commit
-
…ck/linux-2.6-rcu into core/rcu
23 Aug, 2010
4 commits
-
…/linux-2.6-rcu into core/rcu
-
* 'radix-tree' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/xfsdev:
radix-tree: radix_tree_range_tag_if_tagged() can set incorrect tags
radix-tree: clear all tags in radix_tree_node_rcu_free -
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
Acked-by: Jan Kara -
Commit f446daaea9d4a420d16c606f755f3689dcb2d0ce ("mm: implement
writeback livelock avoidance using page tagging") introduced a new
radix tree tag, increasing the number of tags in each node from 2 to
3. It did not, however, fix up the code in
radix_tree_node_rcu_free() that cleans up after radix_tree_shrink()
and hence could leave stray tags set in the new tag array.The result is that the livelock avoidance code added in the the
above commit would hit stale tags when doing tag based lookups,
resulting in livelocks when trying to traverse the tree.Fix this problem in radix_tree_node_rcu_free() so it doesn't happen
again in the future by using a loop to walk all the tags up to
RADIX_TREE_MAX_TAGS to clear the stray tags radix_tree_shrink()
leaves behind.Signed-off-by: Dave Chinner
Acked-by: Nick Piggin
Acked-by: Jan Kara
21 Aug, 2010
1 commit
-
When radix_tree_maxindex() is ~0UL, it can happen that scanning overflows
index and tree traversal code goes astray reading memory until it hits
unreadable memory. Check for overflow and exit in that case.Signed-off-by: Jan Kara
Cc: Christoph Hellwig
Cc: Nick Piggin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
20 Aug, 2010
1 commit
-
Signed-off-by: Arnd Bergmann
Signed-off-by: Paul E. McKenney
Cc: Nick Piggin
Reviewed-by: Josh Triplett
10 Aug, 2010
1 commit
-
Implement function for setting one tag if another tag is set for each item
in given range.Signed-off-by: Jan Kara
Cc: Dave Chinner
Cc: Nick Piggin
Cc: Chris Mason
Cc: Theodore Ts'o
Cc: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
28 May, 2010
1 commit
-
radix_tree_prev_hole() used LONG_MAX to detect underflow; however,
ULONG_MAX is clearly what was intended, both here and by its only user
(count_history_pages at mm/readahead.c).Reviewed-by: Wu Fengguang
Signed-off-by: Cesar Eduardo Barros
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
10 Apr, 2010
1 commit
-
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
Signed-off-by: David Howells
Acked-by: Nick Piggin
Signed-off-by: Linus Torvalds
30 Mar, 2010
1 commit
-
…it slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
25 Feb, 2010
1 commit
-
Because the radix tree is used with many different locking
designs, we cannot do any effective checking without changing
the radix-tree APIs. It might make sense to do this later, but
only if the RCU lockdep checking proves itself sufficiently
valuable.Signed-off-by: Paul E. McKenney
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: mathieu.desnoyers@polymtl.ca
Cc: josh@joshtriplett.org
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
Cc: Valdis.Kletnieks@vt.edu
Cc: dhowells@redhat.com
LKML-Reference:
Signed-off-by: Ingo Molnar
20 Nov, 2009
2 commits
-
Don't delete pending pages from the page-store tracking tree, but rather send
them for another write as they've presumably been updated.Signed-off-by: David Howells
-
__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: [] __fscache_uncache_page+0xdb/0x160 [fscache]but task is already holding lock:
(&cookie->lock){+.+.-.}, at: [] __fscache_write_page+0x15c/0x3f3 [fscache]other info that might help us debug this:
5 locks held by nfsiod/7916:
#0: (nfsiod){+.+.+.}, at: [] worker_thread+0x19a/0x2e2
#1: (&task->u.tk_work#2){+.+.+.}, at: [] worker_thread+0x19a/0x2e2
#2: (&cookie->lock){+.+.-.}, at: [] __fscache_write_page+0x15c/0x3f3 [fscache]
#3: (&object->lock#2){+.+.-.}, at: [] __fscache_write_page+0x197/0x3f3 [fscache]
#4: (&cookie->stores_lock){+.+...}, at: [] __fscache_write_page+0x19f/0x3f3 [fscache]stack backtrace:
Pid: 7916, comm: nfsiod Not tainted 2.6.32-rc6-cachefs #24
Call Trace:
[] __lock_acquire+0x1649/0x16e3
[] ? __lock_acquire+0x7b7/0x16e3
[] ? dump_trace+0x248/0x257
[] lock_acquire+0x57/0x6d
[] ? __fscache_uncache_page+0xdb/0x160 [fscache]
[] _spin_lock+0x2c/0x3b
[] ? __fscache_uncache_page+0xdb/0x160 [fscache]
[] __fscache_uncache_page+0xdb/0x160 [fscache]
[] ? __fscache_check_page_write+0x0/0x71 [fscache]
[] nfs_fscache_release_page+0x86/0xc4 [nfs]
[] nfs_release_page+0x3c/0x41 [nfs]
[] try_to_release_page+0x32/0x3b
[] shrink_page_list+0x316/0x4ac
[] ? mark_held_locks+0x52/0x70
[] ? _spin_unlock_irq+0x2b/0x31
[] shrink_inactive_list+0x392/0x67c
[] ? mark_held_locks+0x52/0x70
[] shrink_list+0x8d/0x8f
[] shrink_zone+0x278/0x33c
[] ? ktime_get_ts+0xad/0xba
[] try_to_free_pages+0x22e/0x392
[] ? isolate_pages_global+0x0/0x212
[] __alloc_pages_nodemask+0x3dc/0x5cf
[] cache_alloc_refill+0x34d/0x6c1
[] ? radix_tree_node_alloc+0x52/0x5c
[] kmem_cache_alloc+0xb2/0x118
[] radix_tree_node_alloc+0x52/0x5c
[] radix_tree_insert+0x57/0x19c
[] __fscache_write_page+0x1e3/0x3f3 [fscache]
[] __nfs_readpage_to_fscache+0x58/0x11e [nfs]
[] nfs_readpage_release+0x34/0x9b [nfs]
[] nfs_readpage_release_full+0x32/0x4b [nfs]
[] rpc_release_calldata+0x12/0x14 [sunrpc]
[] rpc_free_task+0x59/0x61 [sunrpc]
[] rpc_async_release+0x10/0x12 [sunrpc]
[] worker_thread+0x1ef/0x2e2
[] ? worker_thread+0x19a/0x2e2
[] ? thread_return+0x3e/0x101
[] ? rpc_async_release+0x0/0x12 [sunrpc]
[] ? autoremove_wake_function+0x0/0x34
[] ? trace_hardirqs_on+0xd/0xf
[] ? worker_thread+0x0/0x2e2
[] kthread+0x7a/0x82
[] child_rip+0xa/0x20
[] ? restore_args+0x0/0x30
[] ? add_wait_queue+0x15/0x44
[] ? kthread+0x0/0x82
[] ? child_rip+0x0/0x20Signed-off-by: David Howells
17 Jun, 2009
2 commits
-
radix_tree_lookup() and radix_tree_lookup_slot() have much the
same code except for the return value.Introduce radix_tree_lookup_element() to do the real work.
/*
* is_slot == 1 : search for the slot.
* is_slot == 0 : search for the node.
*/
static void * radix_tree_lookup_element(struct radix_tree_root *root,
unsigned long index, int is_slot);Signed-off-by: Huang Shijie
Cc: Nick Piggin
Cc: Christoph Lameter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The counterpart of radix_tree_next_hole(). To be used by context readahead.
Signed-off-by: Wu Fengguang
Cc: Vladislav Bolkhovitin
Cc: Jens Axboe
Cc: Jeff Moyer
Cc: Nick Piggin
Cc: Ying Han
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
08 Jan, 2009
1 commit
-
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (24 commits)
trivial: chack -> check typo fix in main Makefile
trivial: Add a space (and a comma) to a printk in 8250 driver
trivial: Fix misspelling of "firmware" in docs for ncr53c8xx/sym53c8xx
trivial: Fix misspelling of "firmware" in powerpc Makefile
trivial: Fix misspelling of "firmware" in usb.c
trivial: Fix misspelling of "firmware" in qla1280.c
trivial: Fix misspelling of "firmware" in a100u2w.c
trivial: Fix misspelling of "firmware" in megaraid.c
trivial: Fix misspelling of "firmware" in ql4_mbx.c
trivial: Fix misspelling of "firmware" in acpi_memhotplug.c
trivial: Fix misspelling of "firmware" in ipw2100.c
trivial: Fix misspelling of "firmware" in atmel.c
trivial: Fix misspelled firmware in Kconfig
trivial: fix an -> a typos in documentation and comments
trivial: fix then -> than typos in comments and documentation
trivial: update Jesper Juhl CREDITS entry with new email
trivial: fix singal -> signal typo
trivial: Fix incorrect use of "loose" in event.c
trivial: printk: fix indentation of new_text_line declaration
trivial: rtc-stk17ta8: fix sparse warning
...
07 Jan, 2009
1 commit
-
radix_tree_preloads is unused outside of this file, make it static.
Noticed by sparse:
lib/radix-tree.c:84:1: warning: symbol 'per_cpu__radix_tree_preloads' was not declared. Should it be static?Signed-off-by: Harvey Harrison
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
06 Jan, 2009
1 commit
-
And some 80-line cleanups.
Signed-off-by: Wu Fengguang
Signed-off-by: Jiri Kosina
27 Jul, 2008
2 commits
-
Kmem cache passed to constructor is only needed for constructors that are
themselves multiplexeres. Nobody uses this "feature", nor does anybody uses
passed kmem cache in non-trivial way, so pass only pointer to object.Non-trivial places are:
arch/powerpc/mm/init_64.c
arch/powerpc/mm/hugetlbpage.cThis is flag day, yes.
Signed-off-by: Alexey Dobriyan
Acked-by: Pekka Enberg
Acked-by: Christoph Lameter
Cc: Jon Tollefson
Cc: Nick Piggin
Cc: Matt Mackall
[akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c]
[akpm@linux-foundation.org: fix mm/slab.c]
[akpm@linux-foundation.org: fix ubifs]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Introduce gang_lookup_slot() and gang_lookup_slot_tag() functions, which
are used by lockless pagecache.Signed-off-by: Nick Piggin
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Hugh Dickins
Cc: "Paul E. McKenney"
Reviewed-by: Peter Zijlstra
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
05 Jul, 2008
1 commit
-
Remove all clameter@sgi.com addresses from the kernel tree since they will
become invalid on June 27th. Change my maintainer email address for the
slab allocators to cl@linux-foundation.org (which will be the new email
address for the future).Signed-off-by: Christoph Lameter
Signed-off-by: Christoph Lameter
Cc: Pekka Enberg
Cc: Stephen Rothwell
Cc: Matt Mackall
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
13 Jun, 2008
1 commit
-
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
Acked-by: Peter Zijlstra
Cc: "Paul E. McKenney"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
28 Apr, 2008
1 commit
-
Migrate flags must be set on slab creation as agreed upon when the antifrag
logic was reviewed. Otherwise some slabs of a slabcache will end up in the
unmovable and others in the reclaimable section depending on which flag was
active when a new slab page was allocated.This likely slid in somehow when antifrag was merged. Remove it.
The buffer_heads are always allocated with __GFP_RECLAIMABLE because the
SLAB_RECLAIM_ACCOUNT option is set. The set_migrateflags() never had any
effect there.Radix tree allocations are not directly reclaimable but they are allocated
with __GFP_RECLAIMABLE set on each allocation. We now set
SLAB_RECLAIM_ACCOUNT on radix tree slab creation making sure that radix
tree slabs are consistently placed in the reclaimable section. Radix tree
slabs will also be accounted as such.There is then no user left of set_migratepages. So remove it.
Signed-off-by: Christoph Lameter
Cc: Mel Gorman
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
06 Feb, 2008
1 commit
-
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/0x40The 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
Cc: "David S. Miller"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
17 Oct, 2007
6 commits
-
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
Christoph Lameter
Cc: Christoph Hellwig
Cc: "Maciej W. Rozycki"
Cc: Pekka Enberg
Cc: Andreas Schwab
Cc: Nick Piggin
Cc: WU Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Slab constructors currently have a flags parameter that is never used. And
the order of the arguments is opposite to other slab functions. The object
pointer is placed before the kmem_cache pointer.Convert
ctor(void *object, struct kmem_cache *s, unsigned long flags)
to
ctor(struct kmem_cache *s, void *object)
throughout the kernel
[akpm@linux-foundation.org: coupla fixes]
Signed-off-by: Christoph Lameter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch marks a number of allocations that are either short-lived such as
network buffers or are reclaimable such as inode allocations. When something
like updatedb is called, long-lived and unmovable kernel allocations tend to
be spread throughout the address space which increases fragmentation.This patch groups these allocations together as much as possible by adding a
new MIGRATE_TYPE. The MIGRATE_RECLAIMABLE type is for allocations that can be
reclaimed on demand, but not moved. i.e. they can be migrated by deleting
them and re-reading the information from elsewhere.Signed-off-by: Mel Gorman
Cc: Andy Whitcroft
Cc: Christoph Lameter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
A while back, Nick Piggin introduced a patch to reduce the node memory
usage for small files (commit cfd9b7df4abd3257c9e381b0e445817b26a51c0c):-#define RADIX_TREE_MAP_SHIFT 6
+#define RADIX_TREE_MAP_SHIFT (CONFIG_BASE_SMALL ? 4 : 6)Unfortunately, he didn't take into account the fact that the
calculation of the maximum path was based on an assumption of having
to round up:#define RADIX_TREE_MAX_PATH (RADIX_TREE_INDEX_BITS/RADIX_TREE_MAP_SHIFT + 2)
So, if CONFIG_BASE_SMALL is set, you will end up with a
RADIX_TREE_MAX_PATH that is one greater than necessary. The practical
upshot of this is just a bit of wasted memory (one long in the
height_to_maxindex array, an extra pre-allocated radix tree node per
cpu, and extra stack usage in a couple of functions), but it seems
worth getting right.It's also worth noting that I never build with CONFIG_BASE_SMALL.
What I did to test this was duplicate the code in a small user-space
program and check the results of the calculations for max path and the
contents of the height_to_maxindex array.Signed-off-by: Jeff Moyer
Acked-by: Nick Piggin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Rather than sign direct radix-tree pointers with a special bit, sign the
indirect one that hangs off the root. This means that, given a lookup_slot
operation, the invalid result will be differentiated from the valid
(previously, valid results could have the bit either set or clear).This does not affect slot lookups which occur under lock -- they can never
return an invalid result. Is needed in future for lockless pagecache.Signed-off-by: Nick Piggin
Acked-by: Peter Zijlstra
Cc: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Introduce radix_tree_next_hole(root, index, max_scan) to scan radix tree for
the first hole. It will be used in interleaved readahead.The implementation is dumb and obviously correct. It can help debug(and
document) the possible smart one in future.Cc: Nick Piggin
Signed-off-by: Fengguang Wu
Cc: Rusty Russell
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
20 Jul, 2007
1 commit
-
Slab destructors were no longer supported after Christoph's
c59def9f222d44bb7e2f0a559f2906191a0862d7 change. They've been
BUGs for both slab and slub, and slob never supported them
either.This rips out support for the dtor pointer from kmem_cache_create()
completely and fixes up every single callsite in the kernel (there were
about 224, not including the slab allocator definitions themselves,
or the documentation references).Signed-off-by: Paul Mundt
14 Jul, 2007
1 commit
-
XFS filestreams functionality uses radix trees and the preload
functions. XFS can be built as a module and hence we need
radix_tree_preload() exported. radix_tree_preload_end() is a
static inline, so it doesn't need exporting.Signed-Off-By: Dave Chinner
Signed-Off-By: Tim Shimmin
10 May, 2007
1 commit
-
Since nonboot CPUs are now disabled after tasks and devices have been
frozen and the CPU hotplug infrastructure is used for this purpose, we need
special CPU hotplug notifications that will help the CPU-hotplug-aware
subsystems distinguish normal CPU hotplug events from CPU hotplug events
related to a system-wide suspend or resume operation in progress. This
patch introduces such notifications and causes them to be used during
suspend and resume transitions. It also changes all of the
CPU-hotplug-aware subsystems to take these notifications into consideration
(for now they are handled in the same way as the corresponding "normal"
ones).[oleg@tv-sign.ru: cleanups]
Signed-off-by: Rafael J. Wysocki
Cc: Gautham R Shenoy
Cc: Pavel Machek
Signed-off-by: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
