03 Apr, 2009
2 commits
-
Impact: also output kfree(NULL) entries
This patch moves the trace_kfree() calls before the ZERO_OR_NULL_PTR
check so that we can trace call-sites that call kfree() with NULL many
times which might be an indication of a bug.Signed-off-by: Pekka Enberg
Cc: Eduard - Gabriel Munteanu
LKML-Reference:
Signed-off-by: Ingo Molnar -
kmemtrace now uses tracepoints instead of markers. We no longer need to
use format specifiers to pass arguments.Signed-off-by: Eduard - Gabriel Munteanu
[ folded: Use the new TP_PROTO and TP_ARGS to fix the build. ]
[ folded: fix build when CONFIG_KMEMTRACE is disabled. ]
[ folded: define tracepoints when CONFIG_TRACEPOINTS is enabled. ]
Signed-off-by: Pekka Enberg
LKML-Reference:
Signed-off-by: Ingo Molnar
02 Apr, 2009
1 commit
-
Conflicts:
include/linux/slub_def.h
lib/Kconfig.debug
mm/slob.c
mm/slub.c
31 Mar, 2009
1 commit
-
* 'locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (33 commits)
lockdep: fix deadlock in lockdep_trace_alloc
lockdep: annotate reclaim context (__GFP_NOFS), fix SLOB
lockdep: annotate reclaim context (__GFP_NOFS), fix
lockdep: build fix for !PROVE_LOCKING
lockstat: warn about disabled lock debugging
lockdep: use stringify.h
lockdep: simplify check_prev_add_irq()
lockdep: get_user_chars() redo
lockdep: simplify get_user_chars()
lockdep: add comments to mark_lock_irq()
lockdep: remove macro usage from mark_held_locks()
lockdep: fully reduce mark_lock_irq()
lockdep: merge the !_READ mark_lock_irq() helpers
lockdep: merge the _READ mark_lock_irq() helpers
lockdep: simplify mark_lock_irq() helpers #3
lockdep: further simplify mark_lock_irq() helpers
lockdep: simplify the mark_lock_irq() helpers
lockdep: split up mark_lock_irq()
lockdep: generate usage strings
lockdep: generate the state bit definitions
...
24 Mar, 2009
1 commit
-
…pic/slub/cleanups' and 'topic/slub/perf' into for-linus
23 Mar, 2009
1 commit
-
Use get_track() in set_track()
Signed-off-by: Akinobu Mita
Cc: Christoph Lameter
Cc: Pekka Enberg
Signed-off-by: Pekka Enberg
05 Mar, 2009
1 commit
25 Feb, 2009
1 commit
-
As suggested by Christoph Lameter, rename calculate_min_partial() to
set_min_partial() as the function doesn't really do any calculations.Cc: Christoph Lameter
Signed-off-by: David Rientjes
Signed-off-by: Pekka Enberg
23 Feb, 2009
2 commits
-
Now that a cache's min_partial has been moved to struct kmem_cache, it's
possible to easily tune it from userspace by adding a sysfs attribute.It may not be desirable to keep a large number of partial slabs around
if a cache is used infrequently and memory, especially when constrained
by a cgroup, is scarce. It's better to allow userspace to set the
minimum policy per cache instead of relying explicitly on
kmem_cache_shrink().The memory savings from simply moving min_partial from struct
kmem_cache_node to struct kmem_cache is obviously not significant
(unless maybe you're from SGI or something), at the largest it's# allocated caches * (MAX_NUMNODES - 1) * sizeof(unsigned long)
The true savings occurs when userspace reduces the number of partial
slabs that would otherwise be wasted, especially on machines with a
large number of nodes (ia64 with CONFIG_NODES_SHIFT at 10 for default?).
As well as the kernel estimates ideal values for n->min_partial and
ensures it's within a sane range, userspace has no other input other
than writing to /sys/kernel/slab/cache/shrink.There simply isn't any better heuristic to add when calculating the
partial values for a better estimate that works for all possible caches.
And since it's currently a static value, the user really has no way of
reclaiming that wasted space, which can be significant when constrained
by a cgroup (either cpusets or, later, memory controller slab limits)
without shrinking it entirely.This also allows the user to specify that increased fragmentation and
more partial slabs are actually desired to avoid the cost of allocating
new slabs at runtime for specific caches.There's also no reason why this should be a per-struct kmem_cache_node
value in the first place. You could argue that a machine would have
such node size asymmetries that it should be specified on a per-node
basis, but we know nobody is doing that right now since it's a purely
static value at the moment and there's no convenient way to tune that
via slub's sysfs interface.Cc: Christoph Lameter
Signed-off-by: David Rientjes
Signed-off-by: Pekka Enberg -
Although it allows for better cacheline use, it is unnecessary to save a
copy of the cache's min_partial value in each kmem_cache_node.Cc: Christoph Lameter
Signed-off-by: David Rientjes
Signed-off-by: Pekka Enberg
20 Feb, 2009
4 commits
-
…/slab-2.6 into tracing/kmemtrace
Conflicts:
mm/slub.c -
As a preparational patch to bump up page allocator pass-through threshold,
introduce two new constants SLUB_MAX_SIZE and SLUB_PAGE_SHIFT and convert
mm/slub.c to use them.Reported-by: "Zhang, Yanmin"
Tested-by: "Zhang, Yanmin"
Signed-off-by: Christoph Lameter
Signed-off-by: Pekka Enberg -
The default order of kmalloc-8192 on 2*4 stoakley is an issue of
calculate_order.slab_size order name
-------------------------------------------------
4096 3 sgpool-128
8192 2 kmalloc-8192
16384 3 kmalloc-16384kmalloc-8192's default order is smaller than sgpool-128's.
On 4*4 tigerton machine, a similiar issue appears on another kmem_cache.
Function calculate_order uses 'min_objects /= 2;' to shrink. Plus size
calculation/checking in slab_order, sometimes above issue appear.Below patch against 2.6.29-rc2 fixes it.
I checked the default orders of all kmem_cache and they don't become
smaller than before. So the patch wouldn't hurt performance.Signed-off-by Zhang Yanmin
Signed-off-by: Pekka Enberg -
As a preparational patch to bump up page allocator pass-through threshold,
introduce two new constants SLUB_MAX_SIZE and SLUB_PAGE_SHIFT and convert
mm/slub.c to use them.Reported-by: "Zhang, Yanmin"
Tested-by: "Zhang, Yanmin"
Signed-off-by: Christoph Lameter
Signed-off-by: Pekka Enberg
15 Feb, 2009
1 commit
-
Here is another version, with the incremental patch rolled up, and
added reclaim context annotation to kswapd, and allocation tracing
to slab allocators (which may only ever reach the page allocator
in rare cases, so it is good to put annotations here too).Haven't tested this version as such, but it should be getting closer
to merge worthy ;)--
After noticing some code in mm/filemap.c accidentally perform a __GFP_FS
allocation when it should not have been, I thought it might be a good idea to
try to catch this kind of thing with lockdep.I coded up a little idea that seems to work. Unfortunately the system has to
actually be in __GFP_FS page reclaim, then take the lock, before it will mark
it. But at least that might still be some orders of magnitude more common
(and more debuggable) than an actual deadlock condition, so we have some
improvement I hope (the concept is no less complete than discovery of a lock's
interrupt contexts).I guess we could even do the same thing with __GFP_IO (normal reclaim), and
even GFP_NOIO locks too... but filesystems will have the most locks and fiddly
code paths, so let's start there and see how it goes.It *seems* to work. I did a quick test.
=================================
[ INFO: inconsistent lock state ]
2.6.28-rc6-00007-ged31348-dirty #26
---------------------------------
inconsistent {in-reclaim-W} -> {ov-reclaim-W} usage.
modprobe/8526 [HC0[0]:SC0[0]:HE1:SE1] takes:
(testlock){--..}, at: [] brd_init+0x55/0x216 [brd]
{in-reclaim-W} state was registered at:
[] __lock_acquire+0x75b/0x1a60
[] lock_acquire+0x91/0xc0
[] mutex_lock_nested+0xb1/0x310
[] brd_init+0x2b/0x216 [brd]
[] _stext+0x3b/0x170
[] sys_init_module+0xaf/0x1e0
[] system_call_fastpath+0x16/0x1b
[] 0xffffffffffffffff
irq event stamp: 3929
hardirqs last enabled at (3929): [] mutex_lock_nested+0x285/0x310
hardirqs last disabled at (3928): [] mutex_lock_nested+0x59/0x310
softirqs last enabled at (3732): [] sk_filter+0x83/0xe0
softirqs last disabled at (3730): [] sk_filter+0x16/0xe0other info that might help us debug this:
1 lock held by modprobe/8526:
#0: (testlock){--..}, at: [] brd_init+0x55/0x216 [brd]stack backtrace:
Pid: 8526, comm: modprobe Not tainted 2.6.28-rc6-00007-ged31348-dirty #26
Call Trace:
[] print_usage_bug+0x193/0x1d0
[] mark_lock+0xaf0/0xca0
[] mark_held_locks+0x55/0xc0
[] ? brd_init+0x0/0x216 [brd]
[] trace_reclaim_fs+0x2a/0x60
[] __alloc_pages_internal+0x475/0x580
[] ? mutex_lock_nested+0x26e/0x310
[] ? brd_init+0x0/0x216 [brd]
[] brd_init+0x6a/0x216 [brd]
[] ? brd_init+0x0/0x216 [brd]
[] _stext+0x3b/0x170
[] ? mutex_unlock+0x9/0x10
[] ? __mutex_unlock_slowpath+0x10d/0x180
[] ? trace_hardirqs_on_caller+0x12c/0x190
[] sys_init_module+0xaf/0x1e0
[] system_call_fastpath+0x16/0x1bSigned-off-by: Nick Piggin
Signed-off-by: Peter Zijlstra
Signed-off-by: Ingo Molnar
13 Feb, 2009
1 commit
-
…ing/urgent' and 'linus' into tracing/core
12 Feb, 2009
1 commit
-
Commit 7b2cd92adc5430b0c1adeb120971852b4ea1ab08 ("crypto: api - Fix
zeroing on free") added modular user of ksize(). Export that to fix
crypto.ko compilation.Cc: Herbert Xu
Signed-off-by: Kirill A. Shutemov
Signed-off-by: Pekka Enberg
03 Feb, 2009
1 commit
28 Jan, 2009
1 commit
-
The per cpu array of kmem_cache_cpu structures accomodates
NR_KMEM_CACHE_CPU such structs.When this array overflows and a struct is allocated by kmalloc(), it may
have an address at the upper bound of this array. If this happens, it
does not get freed and the per cpu kmem_cache_cpu_free pointer will be out
of bounds after kmem_cache_destroy() or cpu offlining.Cc: Christoph Lameter
Signed-off-by: David Rientjes
Signed-off-by: Pekka Enberg
14 Jan, 2009
1 commit
-
There's no reason to use ->objsize from struct kmem_cache in slab_free() for
the SLAB_DEBUG_OBJECTS case. All it does is generate extra cache pressure as we
try very hard not to touch struct kmem_cache in the fast-path.Signed-off-by: Pekka Enberg
11 Jan, 2009
1 commit
06 Jan, 2009
2 commits
-
It is always "an" if there is a vowel _spoken_ (not written).
So it is:
"an hour" (spoken vowel)
but
"a uniform" (spoken 'j')Signed-off-by: Frederik Schwarzer
Signed-off-by: Jiri Kosina
01 Jan, 2009
1 commit
-
Impact: Use new API
Convert kernel mm functions to use struct cpumask.
We skip include/linux/percpu.h and mm/allocpercpu.c, which are in flux.
Signed-off-by: Rusty Russell
Signed-off-by: Mike Travis
Reviewed-by: Christoph Lameter
31 Dec, 2008
3 commits
-
Conflicts:
arch/x86/kernel/io_apic.c
-
Conflicts:
mm/slub.c
30 Dec, 2008
2 commits
-
Impact: new tracer plugin
This patch adapts kmemtrace raw events tracing to the unified tracing API.
To enable and use this tracer, just do the following:
echo kmemtrace > /debugfs/tracing/current_tracer
cat /debugfs/tracing/traceYou will have the following output:
# tracer: kmemtrace
#
#
# ALLOC TYPE REQ GIVEN FLAGS POINTER NODE CALLER
# FREE | | | | | | | |
# |type_id 1 call_site 18446744071565527833 ptr 18446612134395152256
type_id 0 call_site 18446744071565585597 ptr 18446612134405955584 bytes_req 4096 bytes_alloc 4096 gfp_flags 208 node -1
type_id 1 call_site 18446744071565585534 ptr 18446612134405955584
type_id 0 call_site 18446744071565585597 ptr 18446612134405955584 bytes_req 4096 bytes_alloc 4096 gfp_flags 208 node -1
type_id 0 call_site 18446744071565636711 ptr 18446612134345164672 bytes_req 240 bytes_alloc 240 gfp_flags 208 node -1
type_id 1 call_site 18446744071565585534 ptr 18446612134405955584
type_id 0 call_site 18446744071565585597 ptr 18446612134405955584 bytes_req 4096 bytes_alloc 4096 gfp_flags 208 node -1
type_id 0 call_site 18446744071565636711 ptr 18446612134345164912 bytes_req 240 bytes_alloc 240 gfp_flags 208 node -1
type_id 1 call_site 18446744071565585534 ptr 18446612134405955584
type_id 0 call_site 18446744071565585597 ptr 18446612134405955584 bytes_req 4096 bytes_alloc 4096 gfp_flags 208 node -1
type_id 0 call_site 18446744071565636711 ptr 18446612134345165152 bytes_req 240 bytes_alloc 240 gfp_flags 208 node -1
type_id 0 call_site 18446744071566144042 ptr 18446612134346191680 bytes_req 1304 bytes_alloc 1312 gfp_flags 208 node -1
type_id 1 call_site 18446744071565585534 ptr 18446612134405955584
type_id 0 call_site 18446744071565585597 ptr 18446612134405955584 bytes_req 4096 bytes_alloc 4096 gfp_flags 208 node -1
type_id 1 call_site 18446744071565585534 ptr 18446612134405955584That was to stay backward compatible with the format output produced in
inux/tracepoint.h.This is the default ouput, but note that I tried something else.
If you change an option:
echo kmem_minimalistic > /debugfs/trace_options
and then cat /debugfs/trace, you will have the following output:
# tracer: kmemtrace
#
#
# ALLOC TYPE REQ GIVEN FLAGS POINTER NODE CALLER
# FREE | | | | | | | |
# |- C 0xffff88007c088780 file_free_rcu
+ K 4096 4096 000000d0 0xffff88007cad6000 -1 getname
- C 0xffff88007cad6000 putname
+ K 4096 4096 000000d0 0xffff88007cad6000 -1 getname
+ K 240 240 000000d0 0xffff8800790dc780 -1 d_alloc
- C 0xffff88007cad6000 putname
+ K 4096 4096 000000d0 0xffff88007cad6000 -1 getname
+ K 240 240 000000d0 0xffff8800790dc870 -1 d_alloc
- C 0xffff88007cad6000 putname
+ K 4096 4096 000000d0 0xffff88007cad6000 -1 getname
+ K 240 240 000000d0 0xffff8800790dc960 -1 d_alloc
+ K 1304 1312 000000d0 0xffff8800791d7340 -1 reiserfs_alloc_inode
- C 0xffff88007cad6000 putname
+ K 4096 4096 000000d0 0xffff88007cad6000 -1 getname
- C 0xffff88007cad6000 putname
+ K 992 1000 000000d0 0xffff880079045b58 -1 alloc_inode
+ K 768 1024 000080d0 0xffff88007c096400 -1 alloc_pipe_info
+ K 240 240 000000d0 0xffff8800790dca50 -1 d_alloc
+ K 272 320 000080d0 0xffff88007c088780 -1 get_empty_filp
+ K 272 320 000080d0 0xffff88007c088000 -1 get_empty_filpYeah I shall confess kmem_minimalistic should be: kmem_alternative.
Whatever, I find it more readable but this a personal opinion of course.
We can drop it if you want.On the ALLOC/FREE column, + means an allocation and - a free.
On the type column, you have K = kmalloc, C = cache, P = page
I would like the flags to be GFP_* strings but that would not be easy to not
break the column with strings....About the node...it seems to always be -1. I don't know why but that shouldn't
be difficult to find.I moved linux/tracepoint.h to trace/tracepoint.h as well. I think that would
be more easy to find the tracer headers if they are all in their common
directory.Signed-off-by: Frederic Weisbecker
Signed-off-by: Ingo Molnar -
Impact: avoid conflicts with kmemcheck
kmemcheck modifies the same area of slab.c and slub.c - move the
include lines up a bit.Signed-off-by: Ingo Molnar
29 Dec, 2008
10 commits
-
…penberg/slab-2.6 into tracing/kmemtrace
-
Now that we use _RET_IP_ there's no need to cast 'caller' to unsigned long.
Signed-off-by: Pekka Enberg
-
This patch adds kmemtrace hooks for __kmalloc_track_caller() and
__kmalloc_node_track_caller(). Currently, they set the call site pointer
to the value recieved as a parameter. (This could change if we implement
stack trace exporting in kmemtrace.)Signed-off-by: Eduard - Gabriel Munteanu
Signed-off-by: Pekka Enberg -
This adds hooks for the SLUB allocator, to allow tracing with kmemtrace.
Signed-off-by: Eduard - Gabriel Munteanu
Signed-off-by: Pekka Enberg -
This patch replaces __builtin_return_address(0) with _RET_IP_, since a
previous patch moved _RET_IP_ and _THIS_IP_ to include/linux/kernel.h and
they're widely available now. This makes for shorter and easier to read
code.[penberg@cs.helsinki.fi: remove _RET_IP_ casts to void pointer]
Signed-off-by: Eduard - Gabriel Munteanu
Signed-off-by: Pekka Enberg -
Conflicts:
mm/slub.c
Signed-off-by: Pekka Enberg
-
If a slab cache is mergeable and the sysfs alias cannot be added, the
target cache shall have its refcount decremented. kmem_cache_create()
will return NULL, so if kmem_cache_destroy() is ever called on the target
cache, it will never be freed if the refcount has been leaked.Likewise, if a slab cache is not mergeable and the sysfs link cannot be
added, the new cache shall be removed from the slab_caches list.
kmem_cache_create() will return NULL, so it will be impossible to call
kmem_cache_destroy() on it.Both of these operations require slub_lock since refcount of all slab
caches and slab_caches are protected by the lock.In the mergeable case, it would be better to restore objsize and offset
back to their original values, but this could race with another merge
since slub_lock was dropped.Cc: Christoph Lameter
Signed-off-by: David Rientjes
Signed-off-by: Pekka Enberg -
Currently SLUB doesn't warn about __GFP_WAIT. Add it into slab_alloc().
Acked-by: Christoph Lameter
Signed-off-by: OGAWA Hirofumi
Signed-off-by: Pekka Enberg -
Currently fault-injection capability for SLAB allocator is only
available to SLAB. This patch makes it available to SLUB, too.[penberg@cs.helsinki.fi: unify slab and slub implementations]
Cc: Christoph Lameter
Cc: Matt Mackall
Signed-off-by: Akinobu Mita
Signed-off-by: Pekka Enberg
13 Dec, 2008
1 commit
-
…t_scnprintf to take pointers.
Impact: change calling convention of existing cpumask APIs
Most cpumask functions started with cpus_: these have been replaced by
cpumask_ ones which take struct cpumask pointers as expected.These four functions don't have good replacement names; fortunately
they're rarely used, so we just change them over.Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: paulus@samba.org
Cc: mingo@redhat.com
Cc: tony.luck@intel.com
Cc: ralf@linux-mips.org
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: cl@linux-foundation.org
Cc: srostedt@redhat.com