If "cpuset.memory_migrate" is set, when a process is moved from one
cpuset to another with a different memory node mask, pages in used by
the process are migrated to the new set of nodes. This was performed
synchronously in the ->attach() callback, which is synchronized
against process management. Recently, the synchronization was changed
from per-process rwsem to global percpu rwsem for simplicity and
optimization.
Combined with the synchronous mm migration, this led to deadlocks
because mm migration could schedule a work item which may in turn try
to create a new worker blocking on the process management lock held
from cgroup process migration path.
This heavy an operation shouldn't be performed synchronously from that
deep inside cgroup migration in the first place. This patch punts the
actual migration to an ordered workqueue and updates cgroup process
migration and cpuset config update paths to flush the workqueue after
all locks are released. This way, the operations still seem
synchronous to userland without entangling mm migration with process
management synchronization. CPU hotplug can also invoke mm migration
but there's no reason for it to wait for mm migrations and thus
doesn't synchronize against their completions.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: stable@vger.kernel.org # v4.4+
There is a seqcounter that protects against spurious allocation failures
when a task is changing the allowed nodes in a cpuset. There is no need
to check the seqcounter until a cpuset exists.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Christoph Lameter <cl@linux.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vitaly Wool <vitalywool@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The oom killer takes task_lock() in a couple of places solely to protect
printing the task's comm.
A process's comm, including current's comm, may change due to
/proc/pid/comm or PR_SET_NAME.
The comm will always be NULL-terminated, so the worst race scenario would
only be during update. We can tolerate a comm being printed that is in
the middle of an update to avoid taking the lock.
Other locations in the kernel have already dropped task_lock() when
printing comm, so this is consistent.
Signed-off-by: David Rientjes <rientjes@google.com>
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Current cpuset API for checking if a zone/node is allowed to allocate
from looks rather awkward. We have hardwall and softwall versions of
cpuset_node_allowed with the softwall version doing literally the same
as the hardwall version if __GFP_HARDWALL is passed to it in gfp flags.
If it isn't, the softwall version may check the given node against the
enclosing hardwall cpuset, which it needs to take the callback lock to
do.
Such a distinction was introduced by commit 02a0e53d82 ("cpuset:
rework cpuset_zone_allowed api"). Before, we had the only version with
the __GFP_HARDWALL flag determining its behavior. The purpose of the
commit was to avoid sleep-in-atomic bugs when someone would mistakenly
call the function without the __GFP_HARDWALL flag for an atomic
allocation. The suffixes introduced were intended to make the callers
think before using the function.
However, since the callback lock was converted from mutex to spinlock by
the previous patch, the softwall check function cannot sleep, and these
precautions are no longer necessary.
So let's simplify the API back to the single check.
Suggested-by: David Rientjes <rientjes@google.com>
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: Christoph Lameter <cl@linux.com>
Acked-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull cgroup updates from Tejun Heo:
"Nothing too interesting. Just a handful of cleanup patches"
* 'for-3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
Revert "cgroup: remove redundant variable in cgroup_mount()"
cgroup: remove redundant variable in cgroup_mount()
cgroup: fix missing unlock in cgroup_release_agent()
cgroup: remove CGRP_RELEASABLE flag
perf/cgroup: Remove perf_put_cgroup()
cgroup: remove redundant check in cgroup_ino()
cpuset: simplify proc_cpuset_show()
cgroup: simplify proc_cgroup_show()
cgroup: use a per-cgroup work for release agent
cgroup: remove bogus comments
cgroup: remove redundant code in cgroup_rmdir()
cgroup: remove some useless forward declarations
cgroup: fix a typo in comment.
When we change cpuset.memory_spread_{page,slab}, cpuset will flip
PF_SPREAD_{PAGE,SLAB} bit of tsk->flags for each task in that cpuset.
This should be done using atomic bitops, but currently we don't,
which is broken.
Tetsuo reported a hard-to-reproduce kernel crash on RHEL6, which happened
when one thread tried to clear PF_USED_MATH while at the same time another
thread tried to flip PF_SPREAD_PAGE/PF_SPREAD_SLAB. They both operate on
the same task.
Here's the full report:
https://lkml.org/lkml/2014/9/19/230
To fix this, we make PF_SPREAD_PAGE and PF_SPREAD_SLAB atomic flags.
v4:
- updated mm/slab.c. (Fengguang Wu)
- updated Documentation.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Miao Xie <miaox@cn.fujitsu.com>
Cc: Kees Cook <keescook@chromium.org>
Fixes: 950592f7b9 ("cpusets: update tasks' page/slab spread flags in time")
Cc: <stable@vger.kernel.org> # 2.6.31+
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Use the ONE macro instead of REG, and we can simplify proc_cpuset_show().
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
If cpusets are not in use then we still check a global variable on every
page allocation. Use jump labels to avoid the overhead.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jan Kara <jack@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since put_mems_allowed() is strictly optional, its a seqcount retry, we
don't need to evaluate the function if the allocation was in fact
successful, saving a smp_rmb some loads and comparisons on some relative
fast-paths.
Since the naming, get/put_mems_allowed() does suggest a mandatory
pairing, rename the interface, as suggested by Mel, to resemble the
seqcount interface.
This gives us: read_mems_allowed_begin() and read_mems_allowed_retry(),
where it is important to note that the return value of the latter call
is inverted from its previous incarnation.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull VFS updates from Al Viro,
Misc cleanups all over the place, mainly wrt /proc interfaces (switch
create_proc_entry to proc_create(), get rid of the deprecated
create_proc_read_entry() in favor of using proc_create_data() and
seq_file etc).
7kloc removed.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (204 commits)
don't bother with deferred freeing of fdtables
proc: Move non-public stuff from linux/proc_fs.h to fs/proc/internal.h
proc: Make the PROC_I() and PDE() macros internal to procfs
proc: Supply a function to remove a proc entry by PDE
take cgroup_open() and cpuset_open() to fs/proc/base.c
ppc: Clean up scanlog
ppc: Clean up rtas_flash driver somewhat
hostap: proc: Use remove_proc_subtree()
drm: proc: Use remove_proc_subtree()
drm: proc: Use minor->index to label things, not PDE->name
drm: Constify drm_proc_list[]
zoran: Don't print proc_dir_entry data in debug
reiserfs: Don't access the proc_dir_entry in r_open(), r_start() r_show()
proc: Supply an accessor for getting the data from a PDE's parent
airo: Use remove_proc_subtree()
rtl8192u: Don't need to save device proc dir PDE
rtl8187se: Use a dir under /proc/net/r8180/
proc: Add proc_mkdir_data()
proc: Move some bits from linux/proc_fs.h to linux/{of.h,signal.h,tty.h}
proc: Move PDE_NET() to fs/proc/proc_net.c
...
N_HIGH_MEMORY stands for the nodes that has normal or high memory.
N_MEMORY stands for the nodes that has any memory.
The code here need to handle with the nodes which have memory, we should
use N_MEMORY instead.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Lin Feng <linfeng@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Separate out the cpuset related handling for CPU/Memory online/offline.
This also helps us exploit the most obvious and basic level of optimization
that any notification mechanism (CPU/Mem online/offline) has to offer us:
"We *know* why we have been invoked. So stop pretending that we are lost,
and do only the necessary amount of processing!".
And while at it, rename scan_for_empty_cpusets() to
scan_cpusets_upon_hotplug(), which is more appropriate considering how
it is restructured.
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20120524141650.3692.48637.stgit@srivatsabhat.in.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Ingo Molnar.
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
cpusets: Remove an unused variable
sched/rt: Improve pick_next_highest_task_rt()
sched: Fix select_fallback_rq() vs cpu_active/cpu_online
sched/x86/smp: Do not enable IRQs over calibrate_delay()
sched: Fix compiler warning about declared inline after use
MAINTAINERS: Update email address for SCHEDULER and PERF EVENTS
Commit 5fbd036b55 ("sched: Cleanup cpu_active madness"), which was
supposed to finally sort the cpu_active mess, instead uncovered more.
Since CPU_STARTING is ran before setting the cpu online, there's a
(small) window where the cpu has active,!online.
If during this time there's a wakeup of a task that used to reside on
that cpu select_task_rq() will use select_fallback_rq() to compute an
alternative cpu to run on since we find !online.
select_fallback_rq() however will compute the new cpu against
cpu_active, this means that it can return the same cpu it started out
with, the !online one, since that cpu is in fact marked active.
This results in us trying to scheduling a task on an offline cpu and
triggering a WARN in the IPI code.
The solution proposed by Chuansheng Liu of setting cpu_active in
set_cpu_online() is buggy, firstly not all archs actually use
set_cpu_online(), secondly, not all archs call set_cpu_online() with
IRQs disabled, this means we would introduce either the same race or
the race from fd8a7de17 ("x86: cpu-hotplug: Prevent softirq wakeup on
wrong CPU") -- albeit much narrower.
[ By setting online first and active later we have a window of
online,!active, fresh and bound kthreads have task_cpu() of 0 and
since cpu0 isn't in tsk_cpus_allowed() we end up in
select_fallback_rq() which excludes !active, resulting in a reset
of ->cpus_allowed and the thread running all over the place. ]
The solution is to re-work select_fallback_rq() to require active
_and_ online. This makes the active,!online case work as expected,
OTOH archs running CPU_STARTING after setting online are now
vulnerable to the issue from fd8a7de17 -- these are alpha and
blackfin.
Reported-by: Chuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: linux-alpha@vger.kernel.org
Link: http://lkml.kernel.org/n/tip-hubqk1i10o4dpvlm06gq7v6j@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit c0ff7453bb ("cpuset,mm: fix no node to alloc memory when
changing cpuset's mems") wins a super prize for the largest number of
memory barriers entered into fast paths for one commit.
[get|put]_mems_allowed is incredibly heavy with pairs of full memory
barriers inserted into a number of hot paths. This was detected while
investigating at large page allocator slowdown introduced some time
after 2.6.32. The largest portion of this overhead was shown by
oprofile to be at an mfence introduced by this commit into the page
allocator hot path.
For extra style points, the commit introduced the use of yield() in an
implementation of what looks like a spinning mutex.
This patch replaces the full memory barriers on both read and write
sides with a sequence counter with just read barriers on the fast path
side. This is much cheaper on some architectures, including x86. The
main bulk of the patch is the retry logic if the nodemask changes in a
manner that can cause a false failure.
While updating the nodemask, a check is made to see if a false failure
is a risk. If it is, the sequence number gets bumped and parallel
allocators will briefly stall while the nodemask update takes place.
In a page fault test microbenchmark, oprofile samples from
__alloc_pages_nodemask went from 4.53% of all samples to 1.15%. The
actual results were
3.3.0-rc3 3.3.0-rc3
rc3-vanilla nobarrier-v2r1
Clients 1 UserTime 0.07 ( 0.00%) 0.08 (-14.19%)
Clients 2 UserTime 0.07 ( 0.00%) 0.07 ( 2.72%)
Clients 4 UserTime 0.08 ( 0.00%) 0.07 ( 3.29%)
Clients 1 SysTime 0.70 ( 0.00%) 0.65 ( 6.65%)
Clients 2 SysTime 0.85 ( 0.00%) 0.82 ( 3.65%)
Clients 4 SysTime 1.41 ( 0.00%) 1.41 ( 0.32%)
Clients 1 WallTime 0.77 ( 0.00%) 0.74 ( 4.19%)
Clients 2 WallTime 0.47 ( 0.00%) 0.45 ( 3.73%)
Clients 4 WallTime 0.38 ( 0.00%) 0.37 ( 1.58%)
Clients 1 Flt/sec/cpu 497620.28 ( 0.00%) 520294.53 ( 4.56%)
Clients 2 Flt/sec/cpu 414639.05 ( 0.00%) 429882.01 ( 3.68%)
Clients 4 Flt/sec/cpu 257959.16 ( 0.00%) 258761.48 ( 0.31%)
Clients 1 Flt/sec 495161.39 ( 0.00%) 517292.87 ( 4.47%)
Clients 2 Flt/sec 820325.95 ( 0.00%) 850289.77 ( 3.65%)
Clients 4 Flt/sec 1020068.93 ( 0.00%) 1022674.06 ( 0.26%)
MMTests Statistics: duration
Sys Time Running Test (seconds) 135.68 132.17
User+Sys Time Running Test (seconds) 164.2 160.13
Total Elapsed Time (seconds) 123.46 120.87
The overall improvement is small but the System CPU time is much
improved and roughly in correlation to what oprofile reported (these
performance figures are without profiling so skew is expected). The
actual number of page faults is noticeably improved.
For benchmarks like kernel builds, the overall benefit is marginal but
the system CPU time is slightly reduced.
To test the actual bug the commit fixed I opened two terminals. The
first ran within a cpuset and continually ran a small program that
faulted 100M of anonymous data. In a second window, the nodemask of the
cpuset was continually randomised in a loop.
Without the commit, the program would fail every so often (usually
within 10 seconds) and obviously with the commit everything worked fine.
With this patch applied, it also worked fine so the fix should be
functionally equivalent.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Miao Xie <miaox@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The rule is, we have to update tsk->rt.nr_cpus_allowed if we change
tsk->cpus_allowed. Otherwise RT scheduler may confuse.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/4DD4B3FA.5060901@jp.fujitsu.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently, when a cpu goes down, cpu_active is cleared before
CPU_DOWN_PREPARE starts and cpuset configuration is updated from a
default priority cpu notifier. When a cpu is coming up, it's set
before CPU_ONLINE but cpuset configuration again is updated from the
same cpu notifier.
For cpu notifiers, this presents an inconsistent state. Threads which
a CPU_DOWN_PREPARE notifier expects to be bound to the CPU can be
migrated to other cpus because the cpu is no more inactive.
Fix it by updating cpu_active in the highest priority cpu notifier and
cpuset configuration in the second highest when a cpu is coming up.
Down path is updated similarly. This guarantees that all other cpu
notifiers see consistent cpu_active and cpuset configuration.
cpuset_track_online_cpus() notifier is converted to
cpuset_update_active_cpus() which just updates the configuration and
now called from cpuset_cpu_[in]active() notifiers registered from
sched_init_smp(). If cpuset is disabled, cpuset_update_active_cpus()
degenerates into partition_sched_domains() making separate notifier
for !CONFIG_CPUSETS unnecessary.
This problem is triggered by cmwq. During CPU_DOWN_PREPARE, hotplug
callback creates a kthread and kthread_bind()s it to the target cpu,
and the thread is expected to run on that cpu.
* Ingo's test discovered __cpuinit/exit markups were incorrect.
Fixed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Menage <menage@google.com>
We have observed several workloads running on multi-node systems where
memory is assigned unevenly across the nodes in the system. There are
numerous reasons for this but one is the round-robin rotor in
cpuset_mem_spread_node().
For example, a simple test that writes a multi-page file will allocate
pages on nodes 0 2 4 6 ... Odd nodes are skipped. (Sometimes it
allocates on odd nodes & skips even nodes).
An example is shown below. The program "lfile" writes a file consisting
of 10 pages. The program then mmaps the file & uses get_mempolicy(...,
MPOL_F_NODE) to determine the nodes where the file pages were allocated.
The output is shown below:
# ./lfile
allocated on nodes: 2 4 6 0 1 2 6 0 2
There is a single rotor that is used for allocating both file pages & slab
pages. Writing the file allocates both a data page & a slab page
(buffer_head). This advances the RR rotor 2 nodes for each page
allocated.
A quick confirmation seems to confirm this is the cause of the uneven
allocation:
# echo 0 >/dev/cpuset/memory_spread_slab
# ./lfile
allocated on nodes: 6 7 8 9 0 1 2 3 4 5
This patch introduces a second rotor that is used for slab allocations.
Signed-off-by: Jack Steiner <steiner@sgi.com>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Paul Menage <menage@google.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Before applying this patch, cpuset updates task->mems_allowed and
mempolicy by setting all new bits in the nodemask first, and clearing all
old unallowed bits later. But in the way, the allocator may find that
there is no node to alloc memory.
The reason is that cpuset rebinds the task's mempolicy, it cleans the
nodes which the allocater can alloc pages on, for example:
(mpol: mempolicy)
task1 task1's mpol task2
alloc page 1
alloc on node0? NO 1
1 change mems from 1 to 0
1 rebind task1's mpol
0-1 set new bits
0 clear disallowed bits
alloc on node1? NO 0
...
can't alloc page
goto oom
This patch fixes this problem by expanding the nodes range first(set newly
allowed bits) and shrink it lazily(clear newly disallowed bits). So we
use a variable to tell the write-side task that read-side task is reading
nodemask, and the write-side task clears newly disallowed nodes after
read-side task ends the current memory allocation.
[akpm@linux-foundation.org: fix spello]
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Paul Menage <menage@google.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Ravikiran Thirumalai <kiran@scalex86.org>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce cpuset_cpus_allowed_fallback() helper to fix the cpuset problems
with select_fallback_rq(). It can be called from any context and can't use
any cpuset locks including task_lock(). It is called when the task doesn't
have online cpus in ->cpus_allowed but ttwu/etc must be able to find a
suitable cpu.
I am not proud of this patch. Everything which needs such a fat comment
can't be good even if correct. But I'd prefer to not change the locking
rules in the code I hardly understand, and in any case I believe this
simple change make the code much more correct compared to deadlocks we
currently have.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100315091027.GA9155@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch just states the fact the cpusets/cpuhotplug interaction is
broken and removes the deadlockable code which only pretends to work.
- cpuset_lock() doesn't really work. It is needed for
cpuset_cpus_allowed_locked() but we can't take this lock in
try_to_wake_up()->select_fallback_rq() path.
- cpuset_lock() is deadlockable. Suppose that a task T bound to CPU takes
callback_mutex. If cpu_down(CPU) happens before T drops callback_mutex
stop_machine() preempts T, then migration_call(CPU_DEAD) tries to take
cpuset_lock() and hangs forever because CPU is already dead and thus
T can't be scheduled.
- cpuset_cpus_allowed_locked() is deadlockable too. It takes task_lock()
which is not irq-safe, but try_to_wake_up() can be called from irq.
Kill them, and change select_fallback_rq() to use cpu_possible_mask, like
we currently do without CONFIG_CPUSETS.
Also, with or without this patch, with or without CONFIG_CPUSETS, the
callers of select_fallback_rq() can race with each other or with
set_cpus_allowed() pathes.
The subsequent patches try to to fix these problems.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100315091003.GA9123@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix allocating page cache/slab object on the unallowed node when memory
spread is set by updating tasks' mems_allowed after its cpuset's mems is
changed.
In order to update tasks' mems_allowed in time, we must modify the code of
memory policy. Because the memory policy is applied in the process's
context originally. After applying this patch, one task directly
manipulates anothers mems_allowed, and we use alloc_lock in the
task_struct to protect mems_allowed and memory policy of the task.
But in the fast path, we didn't use lock to protect them, because adding a
lock may lead to performance regression. But if we don't add a lock,the
task might see no nodes when changing cpuset's mems_allowed to some
non-overlapping set. In order to avoid it, we set all new allowed nodes,
then clear newly disallowed ones.
[lee.schermerhorn@hp.com:
The rework of mpol_new() to extract the adjusting of the node mask to
apply cpuset and mpol flags "context" breaks set_mempolicy() and mbind()
with MPOL_PREFERRED and a NULL nodemask--i.e., explicit local
allocation. Fix this by adding the check for MPOL_PREFERRED and empty
node mask to mpol_new_mpolicy().
Remove the now unneeded 'nodes = NULL' from mpol_new().
Note that mpol_new_mempolicy() is always called with a non-NULL
'nodes' parameter now that it has been removed from mpol_new().
Therefore, we don't need to test nodes for NULL before testing it for
'empty'. However, just to be extra paranoid, add a VM_BUG_ON() to
verify this assumption.]
[lee.schermerhorn@hp.com:
I don't think the function name 'mpol_new_mempolicy' is descriptive
enough to differentiate it from mpol_new().
This function applies cpuset set context, usually constraining nodes
to those allowed by the cpuset. However, when the 'RELATIVE_NODES flag
is set, it also translates the nodes. So I settled on
'mpol_set_nodemask()', because the comment block for mpol_new() mentions
that we need to call this function to "set nodes".
Some additional minor line length, whitespace and typo cleanup.]
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Paul Menage <menage@google.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The cpuset_zone_allowed() variants are actually only a function of the
zone's node.
Cc: Paul Menage <menage@google.com>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: cleanup
Time to clean up remaining laggards using the old cpu_ functions.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Trond.Myklebust@netapp.com
Impact: cleanups, use new cpumask API
Final trivial cleanups: mainly s/cpumask_t/struct cpumask
Note there is a FIXME in generate_sched_domains(). A future patch will
change struct cpumask *doms to struct cpumask *doms[].
(I suppose Rusty will do this.)
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Mike Travis <travis@sgi.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When cpusets are enabled, it's necessary to print the triggering task's
set of allowable nodes so the subsequently printed meminfo can be
interpreted correctly.
We also print the task's cpuset name for informational purposes.
[rientjes@google.com: task lock current before dereferencing cpuset]
Cc: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After adding a node into the machine, top cpuset's mems isn't updated.
By reviewing the code, we found that the update function
cpuset_track_online_nodes()
was invoked after node_states[N_ONLINE] changes. It is wrong because
N_ONLINE just means node has pgdat, and if node has/added memory, we use
N_HIGH_MEMORY. So, We should invoke the update function after
node_states[N_HIGH_MEMORY] changes, just like its commit says.
This patch fixes it. And we use notifier of memory hotplug instead of
direct calling of cpuset_track_online_nodes().
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Acked-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Paul Menage <menage@google.com
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
What I realized recently is that calling rebuild_sched_domains() in
arch_reinit_sched_domains() by itself is not enough when cpusets are enabled.
partition_sched_domains() code is trying to avoid unnecessary domain rebuilds
and will not actually rebuild anything if new domain masks match the old ones.
What this means is that doing
echo 1 > /sys/devices/system/cpu/sched_mc_power_savings
on a system with cpusets enabled will not take affect untill something changes
in the cpuset setup (ie new sets created or deleted).
This patch fixes restore correct behaviour where domains must be rebuilt in
order to enable MC powersaving flags.
Test on quad-core Core2 box with both CONFIG_CPUSETS and !CONFIG_CPUSETS.
Also tested on dual-core Core2 laptop. Lockdep is happy and things are working
as expected.
Signed-off-by: Max Krasnyansky <maxk@qualcomm.com>
Tested-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This is based on Linus' idea of creating cpu_active_map that prevents
scheduler load balancer from migrating tasks to the cpu that is going
down.
It allows us to simplify domain management code and avoid unecessary
domain rebuilds during cpu hotplug event handling.
Please ignore the cpusets part for now. It needs some more work in order
to avoid crazy lock nesting. Although I did simplfy and unify domain
reinitialization logic. We now simply call partition_sched_domains() in
all the cases. This means that we're using exact same code paths as in
cpusets case and hence the test below cover cpusets too.
Cpuset changes to make rebuild_sched_domains() callable from various
contexts are in the separate patch (right next after this one).
This not only boots but also easily handles
while true; do make clean; make -j 8; done
and
while true; do on-off-cpu 1; done
at the same time.
(on-off-cpu 1 simple does echo 0/1 > /sys/.../cpu1/online thing).
Suprisingly the box (dual-core Core2) is quite usable. In fact I'm typing
this on right now in gnome-terminal and things are moving just fine.
Also this is running with most of the debug features enabled (lockdep,
mutex, etc) no BUG_ONs or lockdep complaints so far.
I believe I addressed all of the Dmitry's comments for original Linus'
version. I changed both fair and rt balancer to mask out non-active cpus.
And replaced cpu_is_offline() with !cpu_active() in the main scheduler
code where it made sense (to me).
Signed-off-by: Max Krasnyanskiy <maxk@qualcomm.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Gregory Haskins <ghaskins@novell.com>
Cc: dmitry.adamushko@gmail.com
Cc: pj@sgi.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The MPOL_BIND policy creates a zonelist that is used for allocations
controlled by that mempolicy. As the per-node zonelist is already being
filtered based on a zone id, this patch adds a version of __alloc_pages() that
takes a nodemask for further filtering. This eliminates the need for
MPOL_BIND to create a custom zonelist.
A positive benefit of this is that allocations using MPOL_BIND now use the
local node's distance-ordered zonelist instead of a custom node-id-ordered
zonelist. I.e., pages will be allocated from the closest allowed node with
available memory.
[Lee.Schermerhorn@hp.com: Mempolicy: update stale documentation and comments]
[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask]
[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask rework]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Modify cpuset_cpus_allowed to return the currently allowed cpuset
via a pointer argument instead of as the function return value.
* Use new set_cpus_allowed_ptr function.
* Cleanup CPU_MASK_ALL and NODE_MASK_ALL uses.
Depends on:
[sched-devel]: sched: add new set_cpus_allowed_ptr function
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Kosaki Motohito noted that "numactl --interleave=all ..." failed in the
presence of memoryless nodes. This patch attempts to fix that problem.
Some background:
numactl --interleave=all calls set_mempolicy(2) with a fully populated
[out to MAXNUMNODES] nodemask. set_mempolicy() [in do_set_mempolicy()]
calls contextualize_policy() which requires that the nodemask be a
subset of the current task's mems_allowed; else EINVAL will be returned.
A task's mems_allowed will always be a subset of node_states[N_HIGH_MEMORY]
i.e., nodes with memory. So, a fully populated nodemask will be
declared invalid if it includes memoryless nodes.
NOTE: the same thing will occur when running in a cpuset
with restricted mem_allowed--for the same reason:
node mask contains dis-allowed nodes.
mbind(2), on the other hand, just masks off any nodes in the nodemask
that are not included in the caller's mems_allowed.
In each case [mbind() and set_mempolicy()], mpol_check_policy() will
complain [again, resulting in EINVAL] if the nodemask contains any
memoryless nodes. This is somewhat redundant as mpol_new() will remove
memoryless nodes for interleave policy, as will bind_zonelist()--called
by mpol_new() for BIND policy.
Proposed fix:
1) modify contextualize_policy logic to:
a) remember whether the incoming node mask is empty.
b) if not, restrict the nodemask to allowed nodes, as is
currently done in-line for mbind(). This guarantees
that the resulting mask includes only nodes with memory.
NOTE: this is a [benign, IMO] change in behavior for
set_mempolicy(). Dis-allowed nodes will be
silently ignored, rather than returning an error.
c) fold this code into mpol_check_policy(), replace 2 calls to
contextualize_policy() to call mpol_check_policy() directly
and remove contextualize_policy().
2) In existing mpol_check_policy() logic, after "contextualization":
a) MPOL_DEFAULT: require that in coming mask "was_empty"
b) MPOL_{BIND|INTERLEAVE}: require that contextualized nodemask
contains at least one node.
c) add a case for MPOL_PREFERRED: if in coming was not empty
and resulting mask IS empty, user specified invalid nodes.
Return EINVAL.
c) remove the now redundant check for memoryless nodes
3) remove the now redundant masking of policy nodes for interleave
policy from mpol_new().
4) Now that mpol_check_policy() contextualizes the nodemask, remove
the in-line nodes_and() from sys_mbind(). I believe that this
restores mbind() to the behavior before the memoryless-nodes
patch series. E.g., we'll no longer treat an invalid nodemask
with MPOL_PREFERRED as local allocation.
[ Patch history:
v1 -> v2:
- Communicate whether or not incoming node mask was empty to
mpol_check_policy() for better error checking.
- As suggested by David Rientjes, remove the now unused
cpuset_nodes_subset_current_mems_allowed() from cpuset.h
v2 -> v3:
- As suggested by Kosaki Motohito, fold the "contextualization"
of policy nodemask into mpol_check_policy(). Looks a little
cleaner. ]
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Tested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently we possibly lookup the pid in the wrong pid namespace. So
seq_file convert proc_pid_status which ensures the proper pid namespaces is
passed in.
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: another build fix]
[akpm@linux-foundation.org: s390 build fix]
[akpm@linux-foundation.org: fix task_name() output]
[akpm@linux-foundation.org: fix nommu build]
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Andrew Morgan <morgan@kernel.org>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When a cpu is disabled, move_task_off_dead_cpu() is called for tasks that have
been running on that cpu.
Currently, such a task is migrated:
1) to any cpu on the same node as the disabled cpu, which is both online
and among that task's cpus_allowed
2) to any cpu which is both online and among that task's cpus_allowed
It is typical of a multithreaded application running on a large NUMA system to
have its tasks confined to a cpuset so as to cluster them near the memory that
they share. Furthermore, it is typical to explicitly place such a task on a
specific cpu in that cpuset. And in that case the task's cpus_allowed
includes only a single cpu.
This patch would insert a preference to migrate such a task to some cpu within
its cpuset (and set its cpus_allowed to its entire cpuset).
With this patch, migrate the task to:
1) to any cpu on the same node as the disabled cpu, which is both online
and among that task's cpus_allowed
2) to any online cpu within the task's cpuset
3) to any cpu which is both online and among that task's cpus_allowed
In order to do this, move_task_off_dead_cpu() must make a call to
cpuset_cpus_allowed_locked(), a new subset of cpuset_cpus_allowed(), that will
not block. (name change - per Oleg's suggestion)
Calls are made to cpuset_lock() and cpuset_unlock() in migration_call() to set
the cpuset mutex during the whole migrate_live_tasks() and
migrate_dead_tasks() procedure.
[akpm@linux-foundation.org: build fix]
[pj@sgi.com: Fix indentation and spacing]
Signed-off-by: Cliff Wickman <cpw@sgi.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove the filesystem support logic from the cpusets system and makes cpusets
a cgroup subsystem
The "cpuset" filesystem becomes a dummy filesystem; attempts to mount it get
passed through to the cgroup filesystem with the appropriate options to
emulate the old cpuset filesystem behaviour.
Signed-off-by: Paul Menage <menage@google.com>
Cc: Serge E. Hallyn <serue@us.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Kirill Korotaev <dev@openvz.org>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Instead of testing for overlap in the memory nodes of the the nearest
exclusive ancestor of both current and the candidate task, it is better to
simply test for intersection between the task's mems_allowed in their task
descriptors. This does not require taking callback_mutex since it is only
used as a hint in the badness scoring.
Tasks that do not have an intersection in their mems_allowed with the current
task are not explicitly restricted from being OOM killed because it is quite
possible that the candidate task has allocated memory there before and has
since changed its mems_allowed.
Cc: Andrea Arcangeli <andrea@suse.de>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
cpusets try to ensure that any node added to a cpuset's mems_allowed is
on-line and contains memory. The assumption was that online nodes contained
memory. Thus, it is possible to add memoryless nodes to a cpuset and then add
tasks to this cpuset. This results in continuous series of oom-kill and
apparent system hang.
Change cpusets to use node_states[N_HIGH_MEMORY] [a.k.a. node_memory_map] in
place of node_online_map when vetting memories. Return error if admin
attempts to write a non-empty mems_allowed node mask containing only
memoryless-nodes.
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@skynet.ie>
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>
Many struct file_operations in the kernel can be "const". Marking them const
moves these to the .rodata section, which avoids false sharing with potential
dirty data. In addition it'll catch accidental writes at compile time to
these shared resources.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/proc/base.c:1869: warning: initialization discards qualifiers from pointer target type
fs/proc/base.c:2150: warning: initialization discards qualifiers from pointer target type
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Elaborate the API for calling cpuset_zone_allowed(), so that users have to
explicitly choose between the two variants:
cpuset_zone_allowed_hardwall()
cpuset_zone_allowed_softwall()
Until now, whether or not you got the hardwall flavor depended solely on
whether or not you or'd in the __GFP_HARDWALL gfp flag to the gfp_mask
argument.
If you didn't specify __GFP_HARDWALL, you implicitly got the softwall
version.
Unfortunately, this meant that users would end up with the softwall version
without thinking about it. Since only the softwall version might sleep,
this led to bugs with possible sleeping in interrupt context on more than
one occassion.
The hardwall version requires that the current tasks mems_allowed allows
the node of the specified zone (or that you're in interrupt or that
__GFP_THISNODE is set or that you're on a one cpuset system.)
The softwall version, depending on the gfp_mask, might allow a node if it
was allowed in the nearest enclusing cpuset marked mem_exclusive (which
requires taking the cpuset lock 'callback_mutex' to evaluate.)
This patch removes the cpuset_zone_allowed() call, and forces the caller to
explicitly choose between the hardwall and the softwall case.
If the caller wants the gfp_mask to determine this choice, they should (1)
be sure they can sleep or that __GFP_HARDWALL is set, and (2) invoke the
cpuset_zone_allowed_softwall() routine.
This adds another 100 or 200 bytes to the kernel text space, due to the few
lines of nearly duplicate code at the top of both cpuset_zone_allowed_*
routines. It should save a few instructions executed for the calls that
turned into calls of cpuset_zone_allowed_hardwall, thanks to not having to
set (before the call) then check (within the call) the __GFP_HARDWALL flag.
For the most critical call, from get_page_from_freelist(), the same
instructions are executed as before -- the old cpuset_zone_allowed()
routine it used to call is the same code as the
cpuset_zone_allowed_softwall() routine that it calls now.
Not a perfect win, but seems worth it, to reduce this chance of hitting a
sleeping with irq off complaint again.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- move some file_operations structs into the .rodata section
- move static strings from policy_types[] array into the .rodata section
- fix generic seq_operations usages, so that those structs may be defined
as "const" as well
[akpm@osdl.org: couple of fixes]
Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Optimize the critical zonelist scanning for free pages in the kernel memory
allocator by caching the zones that were found to be full recently, and
skipping them.
Remembers the zones in a zonelist that were short of free memory in the
last second. And it stashes a zone-to-node table in the zonelist struct,
to optimize that conversion (minimize its cache footprint.)
Recent changes:
This differs in a significant way from a similar patch that I
posted a week ago. Now, instead of having a nodemask_t of
recently full nodes, I have a bitmask of recently full zones.
This solves a problem that last weeks patch had, which on
systems with multiple zones per node (such as DMA zone) would
take seeing any of these zones full as meaning that all zones
on that node were full.
Also I changed names - from "zonelist faster" to "zonelist cache",
as that seemed to better convey what we're doing here - caching
some of the key zonelist state (for faster access.)
See below for some performance benchmark results. After all that
discussion with David on why I didn't need them, I went and got
some ;). I wanted to verify that I had not hurt the normal case
of memory allocation noticeably. At least for my one little
microbenchmark, I found (1) the normal case wasn't affected, and
(2) workloads that forced scanning across multiple nodes for
memory improved up to 10% fewer System CPU cycles and lower
elapsed clock time ('sys' and 'real'). Good. See details, below.
I didn't have the logic in get_page_from_freelist() for various
full nodes and zone reclaim failures correct. That should be
fixed up now - notice the new goto labels zonelist_scan,
this_zone_full, and try_next_zone, in get_page_from_freelist().
There are two reasons I persued this alternative, over some earlier
proposals that would have focused on optimizing the fake numa
emulation case by caching the last useful zone:
1) Contrary to what I said before, we (SGI, on large ia64 sn2 systems)
have seen real customer loads where the cost to scan the zonelist
was a problem, due to many nodes being full of memory before
we got to a node we could use. Or at least, I think we have.
This was related to me by another engineer, based on experiences
from some time past. So this is not guaranteed. Most likely, though.
The following approach should help such real numa systems just as
much as it helps fake numa systems, or any combination thereof.
2) The effort to distinguish fake from real numa, using node_distance,
so that we could cache a fake numa node and optimize choosing
it over equivalent distance fake nodes, while continuing to
properly scan all real nodes in distance order, was going to
require a nasty blob of zonelist and node distance munging.
The following approach has no new dependency on node distances or
zone sorting.
See comment in the patch below for a description of what it actually does.
Technical details of note (or controversy):
- See the use of "zlc_active" and "did_zlc_setup" below, to delay
adding any work for this new mechanism until we've looked at the
first zone in zonelist. I figured the odds of the first zone
having the memory we needed were high enough that we should just
look there, first, then get fancy only if we need to keep looking.
- Some odd hackery was needed to add items to struct zonelist, while
not tripping up the custom zonelists built by the mm/mempolicy.c
code for MPOL_BIND. My usual wordy comments below explain this.
Search for "MPOL_BIND".
- Some per-node data in the struct zonelist is now modified frequently,
with no locking. Multiple CPU cores on a node could hit and mangle
this data. The theory is that this is just performance hint data,
and the memory allocator will work just fine despite any such mangling.
The fields at risk are the struct 'zonelist_cache' fields 'fullzones'
(a bitmask) and 'last_full_zap' (unsigned long jiffies). It should
all be self correcting after at most a one second delay.
- This still does a linear scan of the same lengths as before. All
I've optimized is making the scan faster, not algorithmically
shorter. It is now able to scan a compact array of 'unsigned
short' in the case of many full nodes, so one cache line should
cover quite a few nodes, rather than each node hitting another
one or two new and distinct cache lines.
- If both Andi and Nick don't find this too complicated, I will be
(pleasantly) flabbergasted.
- I removed the comment claiming we only use one cachline's worth of
zonelist. We seem, at least in the fake numa case, to have put the
lie to that claim.
- I pay no attention to the various watermarks and such in this performance
hint. A node could be marked full for one watermark, and then skipped
over when searching for a page using a different watermark. I think
that's actually quite ok, as it will tend to slightly increase the
spreading of memory over other nodes, away from a memory stressed node.
===============
Performance - some benchmark results and analysis:
This benchmark runs a memory hog program that uses multiple
threads to touch alot of memory as quickly as it can.
Multiple runs were made, touching 12, 38, 64 or 90 GBytes out of
the total 96 GBytes on the system, and using 1, 19, 37, or 55
threads (on a 56 CPU system.) System, user and real (elapsed)
timings were recorded for each run, shown in units of seconds,
in the table below.
Two kernels were tested - 2.6.18-mm3 and the same kernel with
this zonelist caching patch added. The table also shows the
percentage improvement the zonelist caching sys time is over
(lower than) the stock *-mm kernel.
number 2.6.18-mm3 zonelist-cache delta (< 0 good) percent
GBs N ------------ -------------- ---------------- systime
mem threads sys user real sys user real sys user real better
12 1 153 24 177 151 24 176 -2 0 -1 1%
12 19 99 22 8 99 22 8 0 0 0 0%
12 37 111 25 6 112 25 6 1 0 0 -0%
12 55 115 25 5 110 23 5 -5 -2 0 4%
38 1 502 74 576 497 73 570 -5 -1 -6 0%
38 19 426 78 48 373 76 39 -53 -2 -9 12%
38 37 544 83 36 547 82 36 3 -1 0 -0%
38 55 501 77 23 511 80 24 10 3 1 -1%
64 1 917 125 1042 890 124 1014 -27 -1 -28 2%
64 19 1118 138 119 965 141 103 -153 3 -16 13%
64 37 1202 151 94 1136 150 81 -66 -1 -13 5%
64 55 1118 141 61 1072 140 58 -46 -1 -3 4%
90 1 1342 177 1519 1275 174 1450 -67 -3 -69 4%
90 19 2392 199 192 2116 189 176 -276 -10 -16 11%
90 37 3313 238 175 2972 225 145 -341 -13 -30 10%
90 55 1948 210 104 1843 213 100 -105 3 -4 5%
Notes:
1) This test ran a memory hog program that started a specified number N of
threads, and had each thread allocate and touch 1/N'th of
the total memory to be used in the test run in a single loop,
writing a constant word to memory, one store every 4096 bytes.
Watching this test during some earlier trial runs, I would see
each of these threads sit down on one CPU and stay there, for
the remainder of the pass, a different CPU for each thread.
2) The 'real' column is not comparable to the 'sys' or 'user' columns.
The 'real' column is seconds wall clock time elapsed, from beginning
to end of that test pass. The 'sys' and 'user' columns are total
CPU seconds spent on that test pass. For a 19 thread test run,
for example, the sum of 'sys' and 'user' could be up to 19 times the
number of 'real' elapsed wall clock seconds.
3) Tests were run on a fresh, single-user boot, to minimize the amount
of memory already in use at the start of the test, and to minimize
the amount of background activity that might interfere.
4) Tests were done on a 56 CPU, 28 Node system with 96 GBytes of RAM.
5) Notice that the 'real' time gets large for the single thread runs, even
though the measured 'sys' and 'user' times are modest. I'm not sure what
that means - probably something to do with it being slow for one thread to
be accessing memory along ways away. Perhaps the fake numa system, running
ostensibly the same workload, would not show this substantial degradation
of 'real' time for one thread on many nodes -- lets hope not.
6) The high thread count passes (one thread per CPU - on 55 of 56 CPUs)
ran quite efficiently, as one might expect. Each pair of threads needed
to allocate and touch the memory on the node the two threads shared, a
pleasantly parallizable workload.
7) The intermediate thread count passes, when asking for alot of memory forcing
them to go to a few neighboring nodes, improved the most with this zonelist
caching patch.
Conclusions:
* This zonelist cache patch probably makes little difference one way or the
other for most workloads on real numa hardware, if those workloads avoid
heavy off node allocations.
* For memory intensive workloads requiring substantial off-node allocations
on real numa hardware, this patch improves both kernel and elapsed timings
up to ten per-cent.
* For fake numa systems, I'm optimistic, but will have to leave that up to
Rohit Seth to actually test (once I get him a 2.6.18 backport.)
Signed-off-by: Paul Jackson <pj@sgi.com>
Cc: Rohit Seth <rohitseth@google.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: David Rientjes <rientjes@cs.washington.edu>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Change the list of memory nodes allowed to tasks in the top (root) nodeset
to dynamically track what cpus are online, using a call to a cpuset hook
from the memory hotplug code. Make this top cpus file read-only.
On systems that have cpusets configured in their kernel, but that aren't
actively using cpusets (for some distros, this covers the majority of
systems) all tasks end up in the top cpuset.
If that system does support memory hotplug, then these tasks cannot make
use of memory nodes that are added after system boot, because the memory
nodes are not allowed in the top cpuset. This is a surprising regression
over earlier kernels that didn't have cpusets enabled.
One key motivation for this change is to remain consistent with the
behaviour for the top_cpuset's 'cpus', which is also read-only, and which
automatically tracks the cpu_online_map.
This change also has the minor benefit that it fixes a long standing,
little noticed, minor bug in cpusets. The cpuset performance tweak to
short circuit the cpuset_zone_allowed() check on systems with just a single
cpuset (see 'number_of_cpusets', in linux/cpuset.h) meant that simply
changing the 'mems' of the top_cpuset had no affect, even though the change
(the write system call) appeared to succeed. With the following change,
that write to the 'mems' file fails -EACCES, and the 'mems' file stubbornly
refuses to be changed via user space writes. Thus no one should be mislead
into thinking they've changed the top_cpusets's 'mems' when in affect they
haven't.
In order to keep the behaviour of cpusets consistent between systems
actively making use of them and systems not using them, this patch changes
the behaviour of the 'mems' file in the top (root) cpuset, making it read
only, and making it automatically track the value of node_online_map. Thus
tasks in the top cpuset will have automatic use of hot plugged memory nodes
allowed by their cpuset.
[akpm@osdl.org: build fix]
[bunk@stusta.de: build fix]
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch provides the implementation and cpuset interface for an alternative
memory allocation policy that can be applied to certain kinds of memory
allocations, such as the page cache (file system buffers) and some slab caches
(such as inode caches).
The policy is called "memory spreading." If enabled, it spreads out these
kinds of memory allocations over all the nodes allowed to a task, instead of
preferring to place them on the node where the task is executing.
All other kinds of allocations, including anonymous pages for a tasks stack
and data regions, are not affected by this policy choice, and continue to be
allocated preferring the node local to execution, as modified by the NUMA
mempolicy.
There are two boolean flag files per cpuset that control where the kernel
allocates pages for the file system buffers and related in kernel data
structures. They are called 'memory_spread_page' and 'memory_spread_slab'.
If the per-cpuset boolean flag file 'memory_spread_page' is set, then the
kernel will spread the file system buffers (page cache) evenly over all the
nodes that the faulting task is allowed to use, instead of preferring to put
those pages on the node where the task is running.
If the per-cpuset boolean flag file 'memory_spread_slab' is set, then the
kernel will spread some file system related slab caches, such as for inodes
and dentries evenly over all the nodes that the faulting task is allowed to
use, instead of preferring to put those pages on the node where the task is
running.
The implementation is simple. Setting the cpuset flags 'memory_spread_page'
or 'memory_spread_cache' turns on the per-process flags PF_SPREAD_PAGE or
PF_SPREAD_SLAB, respectively, for each task that is in the cpuset or
subsequently joins that cpuset. In subsequent patches, the page allocation
calls for the affected page cache and slab caches are modified to perform an
inline check for these flags, and if set, a call to a new routine
cpuset_mem_spread_node() returns the node to prefer for the allocation.
The cpuset_mem_spread_node() routine is also simple. It uses the value of a
per-task rotor cpuset_mem_spread_rotor to select the next node in the current
tasks mems_allowed to prefer for the allocation.
This policy can provide substantial improvements for jobs that need to place
thread local data on the corresponding node, but that need to access large
file system data sets that need to be spread across the several nodes in the
jobs cpuset in order to fit. Without this patch, especially for jobs that
might have one thread reading in the data set, the memory allocation across
the nodes in the jobs cpuset can become very uneven.
A couple of Copyright year ranges are updated as well. And a couple of email
addresses that can be found in the MAINTAINERS file are removed.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The problem, reported in:
http://bugzilla.kernel.org/show_bug.cgi?id=5859
and by various other email messages and lkml posts is that the cpuset hook
in the oom (out of memory) code can try to take a cpuset semaphore while
holding the tasklist_lock (a spinlock).
One must not sleep while holding a spinlock.
The fix seems easy enough - move the cpuset semaphore region outside the
tasklist_lock region.
This required a few lines of mechanism to implement. The oom code where
the locking needs to be changed does not have access to the cpuset locks,
which are internal to kernel/cpuset.c only. So I provided a couple more
cpuset interface routines, available to the rest of the kernel, which
simple take and drop the lock needed here (cpusets callback_sem).
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove a couple of more lines of code from the cpuset hooks in the page
allocation code path.
There was a check for a NULL cpuset pointer in the routine
cpuset_update_task_memory_state() that was only needed during system boot,
after the memory subsystem was initialized, before the cpuset subsystem was
initialized, to catch a NULL task->cpuset pointer.
Add a cpuset_init_early() routine, just before the mem_init() call in
init/main.c, that sets up just enough of the init tasks cpuset structure to
render cpuset_update_task_memory_state() calls harmless.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Easy little optimization hack to avoid actually having to call
cpuset_zone_allowed() and check mems_allowed, in the main page allocation
routine, __alloc_pages(). This saves several CPU cycles per page allocation
on systems not using cpusets.
A counter is updated each time a cpuset is created or removed, and whenever
there is only one cpuset in the system, it must be the root cpuset, which
contains all CPUs and all Memory Nodes. In that case, when the counter is
one, all allocations are allowed.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Tejun Heo