We shrink a radix tree when its root node has only one child, in the left
most slot. The child becomes the new root node. To perform this
operation in a manner compatible with concurrent lockless lookups, we
atomically switch the root pointer from the parent to its child.
However a concurrent lockless lookup may now have loaded a pointer to the
parent (and is presently deciding what to do next). For this reason, we
also have to keep the parent node in a valid state after shrinking the
tree, until the next RCU grace period -- otherwise this lookup with the
parent pointer may not do the right thing. Notably, we need to keep the
child in the left most slot there in case that is requested by the lookup.
This is all pretty standard RCU stuff. It is worth repeating because in
my eagerness to obey the radix tree node constructor scheme, I had broken
it by zeroing the radix tree node before the grace period.
What could happen is that a lookup can load the parent pointer, then
decide it wants to follow the left most child slot, only to find the slot
contained NULL due to the concurrent shrinker having zeroed the parent
node before waiting for a grace period. The lookup would return a false
negative as a result.
Fix it by doing that clearing in the RCU callback. I would normally want
to rip out the constructor entirely, but radix tree nodes are one of those
places where they make sense (only few cachelines will be touched soon
after allocation).
This was never actually found in any lockless pagecache testing or by the
test harness, but by seeing the odd problem with my scalable vmap rewrite.
I have not tickled the test harness into reproducing it yet, but I'll
keep working at it.
Fortunately, it is not a problem anywhere lockless pagecache is used in
mainline kernels (pagecache probe is not a guarantee, and brd does not
have concurrent lookups and deletes).
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <clameter@sgi.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Most pagecache (and some other) radix tree insertions have the great
opportunity to preallocate a few nodes with relaxed gfp flags. But the
preallocation is squandered when it comes time to allocate a node, we
default to first attempting a GFP_ATOMIC allocation -- that doesn't
normally fail, but it can eat into atomic memory reserves that we don't
need to be using.
Another upshot of this is that it removes the sometimes highly contended
zone->lock from underneath tree_lock. Pagecache insertions are always
performed with a radix tree preload, and after this change, such a
situation will never fall back to kmem_cache_alloc within
radix_tree_node_alloc.
David Miller reports seeing this allocation fail on a highly threaded
sparc64 system:
[527319.459981] dd: page allocation failure. order:0, mode:0x20
[527319.460403] Call Trace:
[527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8
[527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8
[527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90
[527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260
[527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0
[527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134
[527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280
[527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214
[527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428
[527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170
[527319.461217] [00000000004badac] do_sync_read+0x88/0xd0
[527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c
[527319.461361] [00000000004bb920] sys_read+0x34/0x60
[527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40
The calltrace is significant: __do_page_cache_readahead allocates a number
of pages with GFP_KERNEL, and hence it should have reclaimed sufficient
memory to satisfy GFP_ATOMIC allocations. However after the list of pages
goes to mpage_readpages, there can be significant intervals (including disk
IO) before all the pages are inserted into the radix-tree. So the reserves
can easily be depleted at that point. The patch is confirmed to fix the
problem.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Negative shifts are not allowed in C (the result is undefined). Same thing
with full-width shifts.
It works on most platforms but not on the VAX with gcc 4.0.1 (it results in an
"operand reserved" fault).
Shifting by more than the width of the value on the left is also not
allowed. I think the extra '>> 1' tacked on at the end in the original
code was an attempt to work around that. Getting rid of that is an extra
feature of this patch.
Here's the chapter and verse, taken from the final draft of the C99
standard ("6.5.7 Bitwise shift operators", paragraph 3):
"The integer promotions are performed on each of the operands. The
type of the result is that of the promoted left operand. If the
value of the right operand is negative or is greater than or equal
to the width of the promoted left operand, the behavior is
undefined."
Thank you to Jan-Benedict Glaw, Christoph Hellwig, Maciej Rozycki, Pekka
Enberg, Andreas Schwab, and Christoph Lameter for review. Special thanks
to Andreas for spotting that my fix only removed half the undefined
behaviour.
Signed-off-by: Peter Lund <firefly@vax64.dk>
Christoph Lameter <clameter@sgi.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Andreas Schwab <schwab@suse.de>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: WU Fengguang <wfg@mail.ustc.edu.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
A while back, Nick Piggin introduced a patch to reduce the node memory
usage for small files (commit cfd9b7df4a):
-#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 <jmoyer@redhat.com>
Acked-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <npiggin@suse.de>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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 <nickpiggin@yahoo.com.au>
Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Slab destructors were no longer supported after Christoph's
c59def9f22 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 <lethal@linux-sh.org>
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 <dgc@sgi.com>
Signed-Off-By: Tim Shimmin <tes@sgi.com>
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 <rjw@sisk.pl>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There was lots of #ifdef noise in the kernel due to hotcpu_notifier(fn,
prio) not correctly marking 'fn' as used in the !HOTPLUG_CPU case, and thus
generating compiler warnings of unused symbols, hence forcing people to add
#ifdefs.
the compiler can skip truly unused functions just fine:
text data bss dec hex filename
1624412 728710 3674856 6027978 5bfaca vmlinux.before
1624412 728710 3674856 6027978 5bfaca vmlinux.after
[akpm@osdl.org: topology.c fix]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make radix tree lookups safe to be performed without locks. Readers are
protected against nodes being deleted by using RCU based freeing. Readers
are protected against new node insertion by using memory barriers to ensure
the node itself will be properly written before it is visible in the radix
tree.
Each radix tree node keeps a record of their height (above leaf nodes).
This height does not change after insertion -- when the radix tree is
extended, higher nodes are only inserted in the top. So a lookup can take
the pointer to what is *now* the root node, and traverse down it even if
the tree is concurrently extended and this node becomes a subtree of a new
root.
"Direct" pointers (tree height of 0, where root->rnode points directly to
the data item) are handled by using the low bit of the pointer to signal
whether rnode is a direct pointer or a pointer to a radix tree node.
When a reader wants to traverse the next branch, they will take a copy of
the pointer. This pointer will be either NULL (and the branch is empty) or
non-NULL (and will point to a valid node).
[akpm@osdl.org: cleanups]
[Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications]
[clameter@sgi.com: build fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Replace all uses of kmem_cache_t with struct kmem_cache.
The patch was generated using the following script:
#!/bin/sh
#
# Replace one string by another in all the kernel sources.
#
set -e
for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do
quilt add $file
sed -e "1,\$s/$1/$2/g" $file >/tmp/$$
mv /tmp/$$ $file
quilt refresh
done
The script was run like this
sh replace kmem_cache_t "struct kmem_cache"
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In radix_tree_tag_get(), return normalized value of 0/1, as indicated
by its comment.
Signed-off-by: Wu Fengguang <wfg@mail.ustc.edu.cn>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The comment states: 'Setting a tag on a not-present item is a BUG.' Hence
if 'index' is larger than the maxindex; the item _cannot_ be presen; it
should also be a BUG.
Also, this allows the following statement (assume a fresh tree):
radix_tree_tag_set(root, 16, 1);
to fail silently, but when preceded by:
radix_tree_insert(root, 32, item);
it would BUG, because the height has been extended by the insert.
In neither case was 16 present.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reduce radix tree node memory usage by about a factor of 4 for small files
(< 64K). There are pointer traversal and memory usage costs for large
files with dense pagecache.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The ability to have height 0 radix trees (a direct pointer to the data item
rather than going through a full node->slot) quietly disappeared with
old-2.6-bkcvs commit ffee171812d51652f9ba284302d9e5c5cc14bdfd. On 64-bit
machines this causes nearly 600 bytes to be used for every <= 4K file in
pagecache.
Re-introduce this feature, root tags stored in spare ->gfp_mask bits.
Simplify radix_tree_delete's complex tag clearing arrangement (which would
become even more complex) by just falling back to tag clearing functions
(the pagecache radix-tree never uses this path anyway, so the icache
savings will mean it's actually a speedup).
On my 4GB G5, this saves 8MB RAM per kernel kernel source+object tree in
pagecache.
Pagecache lookup, insertion, and removal speed for small files will also be
improved.
This makes RCU radix tree harder, but it's worth it.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Documentation changes to help radix tree users avoid overrunning the tags
array. RADIX_TREE_TAGS moves to linux/radix-tree.h and is now known as
RADIX_TREE_MAX_TAGS (Nick Piggin's idea). Tag parameters are changed to
unsigned, and some comments are updated.
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If a tag is set for a node being deleted from a radix_tree, then that
tag gets cleared from the parent of the node, even if it is set for some
siblings of the node begin deleted.
This patch changes the logic to include a test for any_tag_set similar
to the logic a little futher down. Care is taken to ensure that
'nr_cleared_tags' remains equals to the number of entries in the 'tags'
array which are set to '0' (which means that this tag is not set in the
tree below pathp->node, and should be cleared at pathp->node and
possibly above.
[ Nick says: "Linus FYI, I was able to modify the radix tree test
harness to catch the bug and can no longer trigger it after the fix.
Resulting code passes all other harness tests as well of course." ]
Signed-off-by: Neil Brown <neilb@suse.de>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Shrink the height of a radix tree when it is partially truncated - we only do
shrinkage of full truncation at present.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Correctly determine the tags to be cleared in radix_tree_delete() so we
don't keep moving up the tree clearing tags that we don't need to. For
example, if a tag is simply not set in the deleted item, nor anywhere up
the tree, radix_tree_delete() would attempt to clear it up the entire
height of the tree.
Also, tag_set() was made conditional so as not to dirty too many cachelines
high up in the radix tree. Instead, put this logic into
radix_tree_tag_set().
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Introduce helper any_tag_set() rather than repeat the same code sequence 4
times.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reiser4 uses radix trees to solve a trouble reiser4_readdir has serving nfs
requests.
Unfortunately, radix tree api lacks an operation suitable for modifying
existing entry. This patch adds radix_tree_lookup_slot which returns pointer
to found item within the tree. That location can be then updated.
Both Nick and Christoph Lameter have patches which need this as well.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- added typedef unsigned int __nocast gfp_t;
- replaced __nocast uses for gfp flags with gfp_t - it gives exactly
the same warnings as far as sparse is concerned, doesn't change
generated code (from gcc point of view we replaced unsigned int with
typedef) and documents what's going on far better.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix the sparse warning "implicit cast to nocast type"
Signed-off-by: Victor Fusco <victor@cetuc.puc-rio.br>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Simple patch to radix_tree_tag_get() to return different values for non
present node and tag unset.
The function is not used by any in-kernel callers (yet), but this
information is definitely useful.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- There is frequent use of indirections in the radix code. This patch
removes those indirections, makes the code more readable and allows
the compilers to generate better code.
- Removing indirections allows the removal of several casts.
- Removing indirections allows the reduction of the radix_tree_path
size from 3 to 2 words.
- Use pathp-> consistently.
- Remove unnecessary tmp variable in radix_tree_insert
- Separate the upper layer processing from the lowest layer in __lookup()
in order to make it easier to understand what is going on and allow
compilers to generate better code for the loop.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a new section called ".data.read_mostly" for data items that are read
frequently and rarely written to like cpumaps etc.
If these maps are placed in the .data section then these frequenly read
items may end up in cachelines with data is is frequently updated. In that
case all processors in an SMP system must needlessly reload the cachelines
again and again containing elements of those frequently used variables.
The ability to share these cachelines will allow each cpu in an SMP system
to keep local copies of those shared cachelines thereby optimizing
performance.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Shobhit Dayal <shobhit@calsoftinc.com>
Signed-off-by: Christoph Lameter <christoph@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!