Several people report seeing warnings about inconsistent radix tree
nodes followed by crashes in the workingset code, which all looked like
use-after-free access from the shadow node shrinker.
Dave Jones managed to reproduce the issue with a debug patch applied,
which confirmed that the radix tree shrinking indeed frees shadow nodes
while they are still linked to the shadow LRU:
WARNING: CPU: 2 PID: 53 at lib/radix-tree.c:643 delete_node+0x1e4/0x200
CPU: 2 PID: 53 Comm: kswapd0 Not tainted 4.10.0-rc2-think+ #3
Call Trace:
delete_node+0x1e4/0x200
__radix_tree_delete_node+0xd/0x10
shadow_lru_isolate+0xe6/0x220
__list_lru_walk_one.isra.4+0x9b/0x190
list_lru_walk_one+0x23/0x30
scan_shadow_nodes+0x2e/0x40
shrink_slab.part.44+0x23d/0x5d0
shrink_node+0x22c/0x330
kswapd+0x392/0x8f0
This is the WARN_ON_ONCE(!list_empty(&node->private_list)) placed in the
inlined radix_tree_shrink().
The problem is with 14b468791f ("mm: workingset: move shadow entry
tracking to radix tree exceptional tracking"), which passes an update
callback into the radix tree to link and unlink shadow leaf nodes when
tree entries change, but forgot to pass the callback when reclaiming a
shadow node.
While the reclaimed shadow node itself is unlinked by the shrinker, its
deletion from the tree can cause the left-most leaf node in the tree to
be shrunk. If that happens to be a shadow node as well, we don't unlink
it from the LRU as we should.
Consider this tree, where the s are shadow entries:
root->rnode
|
[0 n]
| |
[s ] [sssss]
Now the shadow node shrinker reclaims the rightmost leaf node through
the shadow node LRU:
root->rnode
|
[0 ]
|
[s ]
Because the parent of the deleted node is the first level below the
root and has only one child in the left-most slot, the intermediate
level is shrunk and the node containing the single shadow is put in
its place:
root->rnode
|
[s ]
The shrinker again sees a single left-most slot in a first level node
and thus decides to store the shadow in root->rnode directly and free
the node - which is a leaf node on the shadow node LRU.
root->rnode
|
s
Without the update callback, the freed node remains on the shadow LRU,
where it causes later shrinker runs to crash.
Pass the node updater callback into __radix_tree_delete_node() in case
the deletion causes the left-most branch in the tree to collapse too.
Also add warnings when linked nodes are freed right away, rather than
wait for the use-after-free when the list is scanned much later.
Fixes: 14b468791f ("mm: workingset: move shadow entry tracking to radix tree exceptional tracking")
Reported-by: Dave Chinner <david@fromorbit.com>
Reported-by: Hugh Dickins <hughd@google.com>
Reported-by: Andrea Arcangeli <aarcange@redhat.com>
Reported-and-tested-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Chris Leech <cleech@redhat.com>
Cc: Lee Duncan <lduncan@suse.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is an exceptionally complicated function with just one caller
(tag_pages_for_writeback). We devote a large portion of the runtime of
the test suite to testing this one function which has one caller. By
introducing the new function radix_tree_iter_tag_set(), we can eliminate
all of the complexity while keeping the performance. The caller can now
use a fairly standard radix_tree_for_each() loop, and it doesn't need to
worry about tricksy things like 'start' wrapping.
The test suite continues to spend a large amount of time investigating
this function, but now it's testing the underlying primitives such as
radix_tree_iter_resume() and the radix_tree_for_each_tagged() iterator
which are also used by other parts of the kernel.
Link: http://lkml.kernel.org/r/1480369871-5271-57-git-send-email-mawilcox@linuxonhyperv.com
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Tested-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This fixes several interlinked problems with the iterators in the
presence of multiorder entries.
1. radix_tree_iter_next() would only advance by one slot, which would
result in the iterators returning the same entry more than once if
there were sibling entries.
2. radix_tree_next_slot() could return an internal pointer instead of
a user pointer if a tagged multiorder entry was immediately followed by
an entry of lower order.
3. radix_tree_next_slot() expanded to a lot more code than it used to
when multiorder support was compiled in. And I wasn't comfortable with
entry_to_node() being in a header file.
Fixing radix_tree_iter_next() for the presence of sibling entries
necessarily involves examining the contents of the radix tree, so we now
need to pass 'slot' to radix_tree_iter_next(), and we need to change the
calling convention so it is called *before* dropping the lock which
protects the tree. Also rename it to radix_tree_iter_resume(), as some
people thought it was necessary to call radix_tree_iter_next() each time
around the loop.
radix_tree_next_slot() becomes closer to how it looked before multiorder
support was introduced. It only checks to see if the next entry in the
chunk is a sibling entry or a pointer to a node; this should be rare
enough that handling this case out of line is not a performance impact
(and such impact is amortised by the fact that the entry we just
processed was a multiorder entry). Also, radix_tree_next_slot() used to
force a new chunk lookup for untagged entries, which is more expensive
than the out of line sibling entry skipping.
Link: http://lkml.kernel.org/r/1480369871-5271-55-git-send-email-mawilcox@linuxonhyperv.com
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
Tested-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, we track the shadow entries in the page cache in the upper
bits of the radix_tree_node->count, behind the back of the radix tree
implementation. Because the radix tree code has no awareness of them,
we rely on random subtleties throughout the implementation (such as the
node->count != 1 check in the shrinking code, which is meant to exclude
multi-entry nodes but also happens to skip nodes with only one shadow
entry, as that's accounted in the upper bits). This is error prone and
has, in fact, caused the bug fixed in d3798ae8c6 ("mm: filemap: don't
plant shadow entries without radix tree node").
To remove these subtleties, this patch moves shadow entry tracking from
the upper bits of node->count to the existing counter for exceptional
entries. node->count goes back to being a simple counter of valid
entries in the tree node and can be shrunk to a single byte.
This vastly simplifies the page cache code. All accounting happens
natively inside the radix tree implementation, and maintaining the LRU
linkage of shadow nodes is consolidated into a single function in the
workingset code that is called for leaf nodes affected by a change in
the page cache tree.
This also removes the last user of the __radix_delete_node() return
value. Eliminate it.
Link: http://lkml.kernel.org/r/20161117193211.GE23430@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The bug in khugepaged fixed earlier in this series shows that radix tree
slot replacement is fragile; and it will become more so when not only
NULL<->!NULL transitions need to be caught but transitions from and to
exceptional entries as well. We need checks.
Re-implement radix_tree_replace_slot() on top of the sanity-checked
__radix_tree_replace(). This requires existing callers to also pass the
radix tree root, but it'll warn us when somebody replaces slots with
contents that need proper accounting (transitions between NULL entries,
real entries, exceptional entries) and where a replacement through the
slot pointer would corrupt the radix tree node counts.
Link: http://lkml.kernel.org/r/20161117193021.GB23430@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Suggested-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The way the page cache is sneaking shadow entries of evicted pages into
the radix tree past the node entry accounting and tracking them manually
in the upper bits of node->count is fraught with problems.
These shadow entries are marked in the tree as exceptional entries,
which are a native concept to the radix tree. Maintain an explicit
counter of exceptional entries in the radix tree node. Subsequent
patches will switch shadow entry tracking over to that counter.
DAX and shmem are the other users of exceptional entries. Since slot
replacements that change the entry type from regular to exceptional must
now be accounted, introduce a __radix_tree_replace() function that does
replacement and accounting, and switch DAX and shmem over.
The increase in radix tree node size is temporary. A followup patch
switches the shadow tracking to this new scheme and we'll no longer need
the upper bits in node->count and shrink that back to one byte.
Link: http://lkml.kernel.org/r/20161117192945.GA23430@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are four cases I can see where we could end up with a NULL 'slot' in
radix_tree_next_slot(). Yet radix_tree_next_slot() never actually checks
whether 'slot' is NULL. It just happens that for the cases where 'slot'
is NULL, some other combination of factors prevents us from dereferencing
it.
It would be very easy for someone to unwittingly change one of these
factors without realizing that we are implicitly depending on it to save
us from a NULL pointer dereference.
Add a comment documenting the things that allow 'slot' to be safely passed
as NULL to radix_tree_next_slot().
Here are details on the four cases:
1) radix_tree_iter_retry() via a non-tagged iteration like
radix_tree_for_each_slot(). In this case we currently aren't seeing a bug
because radix_tree_iter_retry() sets
iter->next_index = iter->index;
which means that in in the else case in radix_tree_next_slot(), 'count' is
zero, so we skip over the while() loop and effectively just return NULL
without ever dereferencing 'slot'.
2) radix_tree_iter_retry() via tagged iteration like
radix_tree_for_each_tagged(). This case was giving us NULL pointer
dereferences in testing, and was fixed with this commit:
commit 3cb9185c67 ("radix-tree: fix radix_tree_iter_retry() for tagged
iterators.")
This fix doesn't explicitly check for 'slot' being NULL, though, it works
around the NULL pointer dereference by instead zeroing iter->tags in
radix_tree_iter_retry(), which makes us bail out of the if() case in
radix_tree_next_slot() before we dereference 'slot'.
3) radix_tree_iter_next() via via a non-tagged iteration like
radix_tree_for_each_slot(). This currently happens in shmem_tag_pins()
and shmem_partial_swap_usage().
As with non-tagged iteration, 'count' in the else case of
radix_tree_next_slot() is zero, so we skip over the while() loop and
effectively just return NULL without ever dereferencing 'slot'.
4) radix_tree_iter_next() via tagged iteration like
radix_tree_for_each_tagged(). This happens in shmem_wait_for_pins().
radix_tree_iter_next() zeros out iter->tags, so we end up exiting
radix_tree_next_slot() here:
if (flags & RADIX_TREE_ITER_TAGGED) {
void *canon = slot;
iter->tags >>= 1;
if (unlikely(!iter->tags))
return NULL;
Link: http://lkml.kernel.org/r/20160815194237.25967-2-ross.zwisler@linux.intel.com
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Shuah Khan <shuahkh@osg.samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When the underflow checks were added to workingset_node_shadow_dec(),
they triggered immediately:
kernel BUG at ./include/linux/swap.h:276!
invalid opcode: 0000 [#1] SMP
Modules linked in: isofs usb_storage fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_REJECT nf_reject_ipv6
soundcore wmi acpi_als pinctrl_sunrisepoint kfifo_buf tpm_tis industrialio acpi_pad pinctrl_intel tpm_tis_core tpm nfsd auth_rpcgss nfs_acl lockd grace sunrpc dm_crypt
CPU: 0 PID: 20929 Comm: blkid Not tainted 4.8.0-rc8-00087-gbe67d60ba944 #1
Hardware name: System manufacturer System Product Name/Z170-K, BIOS 1803 05/06/2016
task: ffff8faa93ecd940 task.stack: ffff8faa7f478000
RIP: page_cache_tree_insert+0xf1/0x100
Call Trace:
__add_to_page_cache_locked+0x12e/0x270
add_to_page_cache_lru+0x4e/0xe0
mpage_readpages+0x112/0x1d0
blkdev_readpages+0x1d/0x20
__do_page_cache_readahead+0x1ad/0x290
force_page_cache_readahead+0xaa/0x100
page_cache_sync_readahead+0x3f/0x50
generic_file_read_iter+0x5af/0x740
blkdev_read_iter+0x35/0x40
__vfs_read+0xe1/0x130
vfs_read+0x96/0x130
SyS_read+0x55/0xc0
entry_SYSCALL_64_fastpath+0x13/0x8f
Code: 03 00 48 8b 5d d8 65 48 33 1c 25 28 00 00 00 44 89 e8 75 19 48 83 c4 18 5b 41 5c 41 5d 41 5e 5d c3 0f 0b 41 bd ef ff ff ff eb d7 <0f> 0b e8 88 68 ef ff 0f 1f 84 00
RIP page_cache_tree_insert+0xf1/0x100
This is a long-standing bug in the way shadow entries are accounted in
the radix tree nodes. The shrinker needs to know when radix tree nodes
contain only shadow entries, no pages, so node->count is split in half
to count shadows in the upper bits and pages in the lower bits.
Unfortunately, the radix tree implementation doesn't know of this and
assumes all entries are in node->count. When there is a shadow entry
directly in root->rnode and the tree is later extended, the radix tree
implementation will copy that entry into the new node and and bump its
node->count, i.e. increases the page count bits. Once the shadow gets
removed and we subtract from the upper counter, node->count underflows
and triggers the warning. Afterwards, without node->count reaching 0
again, the radix tree node is leaked.
Limit shadow entries to when we have actual radix tree nodes and can
count them properly. That means we lose the ability to detect refaults
from files that had only the first page faulted in at eviction time.
Fixes: 449dd6984d ("mm: keep page cache radix tree nodes in check")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-and-tested-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The bottom two bits of radix tree entries are reserved for special use
by the radix tree code itself. A comment detailing their usage was
added by commit 3bcadd6fa6 ("radix-tree: free up the bottom bit of
exceptional entries for reuse")
This comment states that if the bottom two bits are '11', this means
that this is a locked exceptional entry.
It turns out that this bit combination was never actually used. Radix
tree locking for DAX was indeed implemented, but it actually used the
third LSB:
/* We use lowest available exceptional entry bit for locking */
#define RADIX_DAX_ENTRY_LOCK (1 << RADIX_TREE_EXCEPTIONAL_SHIFT)
This locking code was also made specific to the DAX code instead of
being generally implemented in radix-tree.h.
So, fix the comment.
Link: http://lkml.kernel.org/r/1468997731-2155-1-git-send-email-ross.zwisler@linux.intel.com
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Jan Kara <jack@suse.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
radix_tree_iter_retry() resets slot to NULL, but it doesn't reset tags.
Then NULL slot and non-zero iter.tags passed to radix_tree_next_slot()
leading to crash:
RIP: radix_tree_next_slot include/linux/radix-tree.h:473
find_get_pages_tag+0x334/0x930 mm/filemap.c:1452
....
Call Trace:
pagevec_lookup_tag+0x3a/0x80 mm/swap.c:960
mpage_prepare_extent_to_map+0x321/0xa90 fs/ext4/inode.c:2516
ext4_writepages+0x10be/0x2b20 fs/ext4/inode.c:2736
do_writepages+0x97/0x100 mm/page-writeback.c:2364
__filemap_fdatawrite_range+0x248/0x2e0 mm/filemap.c:300
filemap_write_and_wait_range+0x121/0x1b0 mm/filemap.c:490
ext4_sync_file+0x34d/0xdb0 fs/ext4/fsync.c:115
vfs_fsync_range+0x10a/0x250 fs/sync.c:195
vfs_fsync fs/sync.c:209
do_fsync+0x42/0x70 fs/sync.c:219
SYSC_fdatasync fs/sync.c:232
SyS_fdatasync+0x19/0x20 fs/sync.c:230
entry_SYSCALL_64_fastpath+0x23/0xc1 arch/x86/entry/entry_64.S:207
We must reset iterator's tags to bail out from radix_tree_next_slot()
and go to the slow-path in radix_tree_next_chunk().
Fixes: 46437f9a55 ("radix-tree: fix race in gang lookup")
Link: http://lkml.kernel.org/r/1468495196-10604-1-git-send-email-aryabinin@virtuozzo.com
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We are guaranteed that pointers to radix_tree_nodes always have the
bottom two bits clear (because they come from a slab cache, and slab
caches have a minimum alignment of sizeof(void *)), so we can redefine
'radix_tree_is_internal_node' to only return true if the bottom two bits
have value '01'. This frees up one quarter of the potential values for
use by the user.
Idea from Neil Brown.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Suggested-by: Neil Brown <neilb@suse.de>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Jan Kara <jack@suse.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In addition to replacing the entry, we also clear all associated tags.
This is really a one-off special for page_cache_tree_delete() which had
far too much detailed knowledge about how the radix tree works.
For efficiency, factor node_tag_clear() out of radix_tree_tag_clear() It
can be used by radix_tree_delete_item() as well as
radix_tree_replace_clear_tags().
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Jan Kara <jack@suse.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This enables the macros radix_tree_for_each_slot() and friends to be
used with multi-order entries.
The way that this works is that we treat all entries in a given slots[]
array as a single chunk. If the index given to radix_tree_next_chunk()
happens to point us to a sibling entry, we will back up iter->index so
that it points to the canonical entry, and that will be the place where
we start our iteration.
As we're processing a chunk in radix_tree_next_slot(), we process
canonical entries, skip over sibling entries, and restart the chunk
lookup if we find a non-sibling indirect pointer. This drops back to
the radix_tree_next_chunk() code, which will re-walk the tree and look
for another chunk.
This allows us to properly handle multi-order entries mixed with other
entries that are at various heights in the radix tree.
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Jan Kara <jack@suse.com>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit e614523653 ("radix_tree: add support for multi-order entries")
left the impression that the support for multiorder radix tree entries
was functional. As soon as Ross tried to use it, it became apparent
that my testing was completely inadequate, and it didn't even work a
little bit for orders that were not a multiple of shift.
This series of patches is the result of about 6 weeks of redesign,
reimplementation, testing, arguing and hair-pulling. The great news is
that the test-suite is now far better than it was. That's reflected in
the diffstat for the test-suite alone:
12 files changed, 436 insertions(+), 28 deletions(-)
The highlight for users of the tree is that the restriction on the order
of inserted entries being >= RADIX_TREE_MAP_SHIFT is now gone; the radix
tree now supports any order between 0 and 64.
For those who are interested in how the tree works, patch 9 is probably
the most interesting one as it introduces the new machinery for handling
sibling entries.
I've tried to be fair in attributing authorship to the person who
contributed the majority of the code in each patch; Ross has been an
invaluable partner in the development of this support and it's fair to
say that each of us has code in every commit.
I should also express my appreciation of the 0day testing. It prompted
me that I was bloating the tinyconfig in an unacceptable way, and it
bisected to a commit which contained a rather nasty memory-corruption
bug.
This patch (of 29):
The irqdomain code was checking for 0 or 1 entries, not 0 entries like
the comment said they were. Introduce a new helper that will actually
check for an empty tree.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
shmem likes to occasionally drop the lock, schedule, then reacqire the
lock and continue with the iteration from the last place it left off.
This is currently done with a pretty ugly goto. Introduce
radix_tree_iter_next() and use it throughout shmem.c.
[koct9i@gmail.com: fix bug in radix_tree_iter_next() for tagged iteration]
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With huge pages, it is convenient to have the radix tree be able to
return an entry that covers multiple indices. Previous attempts to deal
with the problem have involved inserting N duplicate entries, which is a
waste of memory and leads to problems trying to handle aliased tags, or
probing the tree multiple times to find alternative entries which might
cover the requested index.
This approach inserts one canonical entry into the tree for a given
range of indices, and may also insert other entries in order to ensure
that lookups find the canonical entry.
This solution only tolerates inserting powers of two that are greater
than the fanout of the tree. If we wish to expand the radix tree's
abilities to support large-ish pages that is less than the fanout at the
penultimate level of the tree, then we would need to add one more step
in lookup to ensure that any sibling nodes in the final level of the
tree are dereferenced and we return the canonical entry that they
reference.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Helper radix_tree_iter_retry() resets next_index to the current index.
In following radix_tree_next_slot current chunk size becomes zero. This
isn't checked and it tries to dereference null pointer in slot.
Tagged iterator is fine because retry happens only at slot 0 where tag
bitmask in iter->tags is filled with single bit.
Fixes: 46437f9a55 ("radix-tree: fix race in gang lookup")
Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Jeremiah Mahler <jmmahler@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the indirect_ptr bit is set on a slot, that indicates we need to redo
the lookup. Introduce a new function radix_tree_iter_retry() which
forces the loop to retry the lookup by setting 'slot' to NULL and
turning the iterator back to point at the problematic entry.
This is a pretty rare problem to hit at the moment; the lookup has to
race with a grow of the radix tree from a height of 0. The consequences
of hitting this race are that gang lookup could return a pointer to a
radix_tree_node instead of a pointer to whatever the user had inserted
in the tree.
Fixes: cebbd29e1c ("radix-tree: rewrite gang lookup using iterator")
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add support for tracking dirty DAX entries in the struct address_space
radix tree. This tree is already used for dirty page writeback, and it
already supports the use of exceptional (non struct page*) entries.
In order to properly track dirty DAX pages we will insert new
exceptional entries into the radix tree that represent dirty DAX PTE or
PMD pages. These exceptional entries will also contain the writeback
addresses for the PTE or PMD faults that we can use at fsync/msync time.
There are currently two types of exceptional entries (shmem and shadow)
that can be placed into the radix tree, and this adds a third. We rely
on the fact that only one type of exceptional entry can be found in a
given radix tree based on its usage. This happens for free with DAX vs
shmem but we explicitly prevent shadow entries from being added to radix
trees for DAX mappings.
The only shadow entries that would be generated for DAX radix trees
would be to track zero page mappings that were created for holes. These
pages would receive minimal benefit from having shadow entries, and the
choice to have only one type of exceptional entry in a given radix tree
makes the logic simpler both in clear_exceptional_entry() and in the
rest of DAX.
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jan Kara <jack@suse.com>
Cc: Jeff Layton <jlayton@poochiereds.net>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Matthew Wilcox <matthew.r.wilcox@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This text refers to the "first 7 functions", which was correct when
written but became incorrect when Johannes Weiner added another function
to the list in 139e561660 ("lib: radix_tree: tree node interface").
Change the text to correctly refer to the first 8 functions.
Signed-off-by: Adam Barth <aurorean@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Previously, page cache radix tree nodes were freed after reclaim emptied
out their page pointers. But now reclaim stores shadow entries in their
place, which are only reclaimed when the inodes themselves are
reclaimed. This is problematic for bigger files that are still in use
after they have a significant amount of their cache reclaimed, without
any of those pages actually refaulting. The shadow entries will just
sit there and waste memory. In the worst case, the shadow entries will
accumulate until the machine runs out of memory.
To get this under control, the VM will track radix tree nodes
exclusively containing shadow entries on a per-NUMA node list. Per-NUMA
rather than global because we expect the radix tree nodes themselves to
be allocated node-locally and we want to reduce cross-node references of
otherwise independent cache workloads. A simple shrinker will then
reclaim these nodes on memory pressure.
A few things need to be stored in the radix tree node to implement the
shadow node LRU and allow tree deletions coming from the list:
1. There is no index available that would describe the reverse path
from the node up to the tree root, which is needed to perform a
deletion. To solve this, encode in each node its offset inside the
parent. This can be stored in the unused upper bits of the same
member that stores the node's height at no extra space cost.
2. The number of shadow entries needs to be counted in addition to the
regular entries, to quickly detect when the node is ready to go to
the shadow node LRU list. The current entry count is an unsigned
int but the maximum number of entries is 64, so a shadow counter
can easily be stored in the unused upper bits.
3. Tree modification needs tree lock and tree root, which are located
in the address space, so store an address_space backpointer in the
node. The parent pointer of the node is in a union with the 2-word
rcu_head, so the backpointer comes at no extra cost as well.
4. The node needs to be linked to an LRU list, which requires a list
head inside the node. This does increase the size of the node, but
it does not change the number of objects that fit into a slab page.
[akpm@linux-foundation.org: export the right function]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Minchan Kim <minchan@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Bob Liu <bob.liu@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Metin Doslu <metin@citusdata.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Ozgun Erdogan <ozgun@citusdata.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Roman Gushchin <klamm@yandex-team.ru>
Cc: Ryan Mallon <rmallon@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make struct radix_tree_node part of the public interface and provide API
functions to create, look up, and delete whole nodes. Refactor the
existing insert, look up, delete functions on top of these new node
primitives.
This will allow the VM to track and garbage collect page cache radix
tree nodes.
[sasha.levin@oracle.com: return correct error code on insertion failure]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Bob Liu <bob.liu@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Metin Doslu <metin@citusdata.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Ozgun Erdogan <ozgun@citusdata.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Roman Gushchin <klamm@yandex-team.ru>
Cc: Ryan Mallon <rmallon@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With users of radix_tree_preload() run from interrupt (block/blk-ioc.c is
one such possible user), the following race can happen:
radix_tree_preload()
...
radix_tree_insert()
radix_tree_node_alloc()
if (rtp->nr) {
ret = rtp->nodes[rtp->nr - 1];
<interrupt>
...
radix_tree_preload()
...
radix_tree_insert()
radix_tree_node_alloc()
if (rtp->nr) {
ret = rtp->nodes[rtp->nr - 1];
And we give out one radix tree node twice. That clearly results in radix
tree corruption with different results (usually OOPS) depending on which
two users of radix tree race.
We fix the problem by making radix_tree_node_alloc() always allocate fresh
radix tree nodes when in interrupt. Using preloading when in interrupt
doesn't make sense since all the allocations have to be atomic anyway and
we cannot steal nodes from process-context users because some users rely
on radix_tree_insert() succeeding after radix_tree_preload().
in_interrupt() check is somewhat ugly but we cannot simply key off passed
gfp_mask as that is acquired from root_gfp_mask() and thus the same for
all preload users.
Another part of the fix is to avoid node preallocation in
radix_tree_preload() when passed gfp_mask doesn't allow waiting. Again,
preallocation in such case doesn't make sense and when preallocation would
happen in interrupt we could possibly leak some allocated nodes. However,
some users of radix_tree_preload() require following radix_tree_insert()
to succeed. To avoid unexpected effects for these users,
radix_tree_preload() only warns if passed gfp mask doesn't allow waiting
and we provide a new function radix_tree_maybe_preload() for those users
which get different gfp mask from different call sites and which are
prepared to handle radix_tree_insert() failure.
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch fixes bug in macro radix_tree_for_each_contig().
If radix_tree_next_slot() sees NULL in next slot it returns NULL, but following
radix_tree_next_chunk() switches iterating into next chunk. As result iterating
becomes non-contiguous and breaks vfs "splice" and all its users.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Reported-and-bisected-by: Hans de Bruin <jmdebruin@xmsnet.nl>
Reported-and-bisected-by: Ondrej Zary <linux@rainbow-software.org>
Reported-bisected-and-tested-by: Toralf Förster <toralf.foerster@gmx.de>
Link: https://lkml.org/lkml/2012/6/5/64
Cc: stable <stable@vger.kernel.org> # 3.4.x
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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>
If a header file is making use of BUG, BUG_ON, BUILD_BUG_ON, or any
other BUG variant in a static inline (i.e. not in a #define) then
that header really should be including <linux/bug.h> and not just
expecting it to be implicitly present.
We can make this change risk-free, since if the files using these
headers didn't have exposure to linux/bug.h already, they would have
been causing compile failures/warnings.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
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>
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>
migrate_pages() -> unmap_and_move() only calls rcu_read_lock() for
anonymous pages, as introduced by git commit
989f89c57e ("fix rcu_read_lock() in page
migraton"). The point of the RCU protection there is part of getting a
stable reference to anon_vma and is only held for anon pages as file pages
are locked which is sufficient protection against freeing.
However, while a file page's mapping is being migrated, the radix tree is
double checked to ensure it is the expected page. This uses
radix_tree_deref_slot() -> rcu_dereference() without the RCU lock held
triggering the following warning.
[ 173.674290] ===================================================
[ 173.676016] [ INFO: suspicious rcu_dereference_check() usage. ]
[ 173.676016] ---------------------------------------------------
[ 173.676016] include/linux/radix-tree.h:145 invoked rcu_dereference_check() without protection!
[ 173.676016]
[ 173.676016] other info that might help us debug this:
[ 173.676016]
[ 173.676016]
[ 173.676016] rcu_scheduler_active = 1, debug_locks = 0
[ 173.676016] 1 lock held by hugeadm/2899:
[ 173.676016] #0: (&(&inode->i_data.tree_lock)->rlock){..-.-.}, at: [<c10e3d2b>] migrate_page_move_mapping+0x40/0x1ab
[ 173.676016]
[ 173.676016] stack backtrace:
[ 173.676016] Pid: 2899, comm: hugeadm Not tainted 2.6.37-rc5-autobuild
[ 173.676016] Call Trace:
[ 173.676016] [<c128cc01>] ? printk+0x14/0x1b
[ 173.676016] [<c1063502>] lockdep_rcu_dereference+0x7d/0x86
[ 173.676016] [<c10e3db5>] migrate_page_move_mapping+0xca/0x1ab
[ 173.676016] [<c10e41ad>] migrate_page+0x23/0x39
[ 173.676016] [<c10e491b>] buffer_migrate_page+0x22/0x107
[ 173.676016] [<c10e48f9>] ? buffer_migrate_page+0x0/0x107
[ 173.676016] [<c10e425d>] move_to_new_page+0x9a/0x1ae
[ 173.676016] [<c10e47e6>] migrate_pages+0x1e7/0x2fa
This patch introduces radix_tree_deref_slot_protected() which calls
rcu_dereference_protected(). Users of it must pass in the
mapping->tree_lock that is protecting this dereference. Holding the tree
lock protects against parallel updaters of the radix tree meaning that
rcu_dereference_protected is allowable.
[akpm@linux-foundation.org: remove unneeded casts]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Milton Miller <miltonm@bga.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: <stable@kernel.org> [2.6.37.early]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <npiggin@kernel.dk>
Reported-by: Salman Qazi <sqazi@google.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Weiner