commit 7cd2b0a34a upstream.
Oleg reports a division by zero error on zero-length write() to the
percpu_pagelist_fraction sysctl:
divide error: 0000 [#1] SMP DEBUG_PAGEALLOC
CPU: 1 PID: 9142 Comm: badarea_io Not tainted 3.15.0-rc2-vm-nfs+ #19
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
task: ffff8800d5aeb6e0 ti: ffff8800d87a2000 task.ti: ffff8800d87a2000
RIP: 0010: percpu_pagelist_fraction_sysctl_handler+0x84/0x120
RSP: 0018:ffff8800d87a3e78 EFLAGS: 00010246
RAX: 0000000000000f89 RBX: ffff88011f7fd000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000010
RBP: ffff8800d87a3e98 R08: ffffffff81d002c8 R09: ffff8800d87a3f50
R10: 000000000000000b R11: 0000000000000246 R12: 0000000000000060
R13: ffffffff81c3c3e0 R14: ffffffff81cfddf8 R15: ffff8801193b0800
FS: 00007f614f1e9740(0000) GS:ffff88011f440000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 00007f614f1fa000 CR3: 00000000d9291000 CR4: 00000000000006e0
Call Trace:
proc_sys_call_handler+0xb3/0xc0
proc_sys_write+0x14/0x20
vfs_write+0xba/0x1e0
SyS_write+0x46/0xb0
tracesys+0xe1/0xe6
However, if the percpu_pagelist_fraction sysctl is set by the user, it
is also impossible to restore it to the kernel default since the user
cannot write 0 to the sysctl.
This patch allows the user to write 0 to restore the default behavior.
It still requires a fraction equal to or larger than 8, however, as
stated by the documentation for sanity. If a value in the range [1, 7]
is written, the sysctl will return EINVAL.
This successfully solves the divide by zero issue at the same time.
Signed-off-by: David Rientjes <rientjes@google.com>
Reported-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 80df284765 upstream.
As sysctl_hung_task_timeout_sec is unsigned long, when this value is
larger then LONG_MAX/HZ, the function schedule_timeout_interruptible in
watchdog will return immediately without sleep and with print :
schedule_timeout: wrong timeout value ffffffffffffff83
and then the funtion watchdog will call schedule_timeout_interruptible
again and again. The screen will be filled with
"schedule_timeout: wrong timeout value ffffffffffffff83"
This patch does some check and correction in sysctl, to let the function
schedule_timeout_interruptible allways get the valid parameter.
Signed-off-by: Liu Hua <sdu.liu@huawei.com>
Tested-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull core debug changes from Ingo Molnar:
"This contains mostly kernel debugging related updates:
- make hung_task detection more configurable to distros
- add final bits for x86 UV NMI debugging, with related KGDB changes
- update the mailing-list of MAINTAINERS entries I'm involved with"
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
hung_task: Display every hung task warning
sysctl: Add neg_one as a standard constraint
x86/uv/nmi, kgdb/kdb: Fix UV NMI handler when KDB not configured
x86/uv/nmi: Fix Sparse warnings
kgdb/kdb: Fix no KDB config problem
MAINTAINERS: Restore "L: linux-kernel@vger.kernel.org" entries
When khungtaskd detects hung tasks, it prints out
backtraces from a number of those tasks.
Limiting the number of backtraces being printed
out can result in the user not seeing the information
necessary to debug the issue. The hung_task_warnings
sysctl controls this feature.
This patch makes it possible for hung_task_warnings
to accept a special value to print an unlimited
number of backtraces when khungtaskd detects hung
tasks.
The special value is -1. To use this value it is
necessary to change types from ulong to int.
Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: oleg@redhat.com
Link: http://lkml.kernel.org/r/1390239253-24030-3-git-send-email-atomlin@redhat.com
[ Build warning fix. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add neg_one to the list of standard constraints - will be used by the next patch.
Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: oleg@redhat.com
Link: http://lkml.kernel.org/r/1390239253-24030-2-git-send-email-atomlin@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
For general-purpose (i.e. distro) kernel builds it makes sense to build
with CONFIG_KEXEC to allow end users to choose what kind of things they
want to do with kexec. However, in the face of trying to lock down a
system with such a kernel, there needs to be a way to disable kexec_load
(much like module loading can be disabled). Without this, it is too easy
for the root user to modify kernel memory even when CONFIG_STRICT_DEVMEM
and modules_disabled are set. With this change, it is still possible to
load an image for use later, then disable kexec_load so the image (or lack
of image) can't be altered.
The intention is for using this in environments where "perfect"
enforcement is hard. Without a verified boot, along with verified
modules, and along with verified kexec, this is trying to give a system a
better chance to defend itself (or at least grow the window of
discoverability) against attack in the face of a privilege escalation.
In my mind, I consider several boot scenarios:
1) Verified boot of read-only verified root fs loading fd-based
verification of kexec images.
2) Secure boot of writable root fs loading signed kexec images.
3) Regular boot loading kexec (e.g. kcrash) image early and locking it.
4) Regular boot with no control of kexec image at all.
1 and 2 don't exist yet, but will soon once the verified kexec series has
landed. 4 is the state of things now. The gap between 2 and 4 is too
large, so this change creates scenario 3, a middle-ground above 4 when 2
and 1 are not possible for a system.
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add a working sysctl to enable/disable automatic numa memory balancing
at runtime.
This allows us to track down performance problems with this feature and
is generally a good idea.
This was possible earlier through debugfs, but only with special
debugging options set. Also fix the boot message.
[akpm@linux-foundation.org: s/sched_numa_balancing/sysctl_numa_balancing/]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Some applications that run on HPC clusters are designed around the
availability of RAM and the overcommit ratio is fine tuned to get the
maximum usage of memory without swapping. With growing memory, the
1%-of-all-RAM grain provided by overcommit_ratio has become too coarse
for these workload (on a 2TB machine it represents no less than 20GB).
This patch adds the new overcommit_kbytes sysctl variable that allow a
much finer grain.
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix nommu build]
Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove the deadline specific sysctls for now. The problem with them is
that the interaction with the exisiting rt knobs is nearly impossible
to get right.
The current (as per before this patch) situation is that the rt and dl
bandwidth is completely separate and we enforce rt+dl < 100%. This is
undesirable because this means that the rt default of 95% leaves us
hardly any room, even though dl tasks are saver than rt tasks.
Another proposed solution was (a discarted patch) to have the dl
bandwidth be a fraction of the rt bandwidth. This is highly
confusing imo.
Furthermore neither proposal is consistent with the situation we
actually want; which is rt tasks ran from a dl server. In which case
the rt bandwidth is a direct subset of dl.
So whichever way we go, the introduction of dl controls at this point
is painful. Therefore remove them and instead share the rt budget.
This means that for now the rt knobs are used for dl admission control
and the dl runtime is accounted against the rt runtime. I realise that
this isn't entirely desirable either; but whatever we do we appear to
need to change the interface later, so better have a small interface
for now.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-zpyqbqds1r0vyxtxza1e7rdc@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In order of deadline scheduling to be effective and useful, it is
important that some method of having the allocation of the available
CPU bandwidth to tasks and task groups under control.
This is usually called "admission control" and if it is not performed
at all, no guarantee can be given on the actual scheduling of the
-deadline tasks.
Since when RT-throttling has been introduced each task group have a
bandwidth associated to itself, calculated as a certain amount of
runtime over a period. Moreover, to make it possible to manipulate
such bandwidth, readable/writable controls have been added to both
procfs (for system wide settings) and cgroupfs (for per-group
settings).
Therefore, the same interface is being used for controlling the
bandwidth distrubution to -deadline tasks and task groups, i.e.,
new controls but with similar names, equivalent meaning and with
the same usage paradigm are added.
However, more discussion is needed in order to figure out how
we want to manage SCHED_DEADLINE bandwidth at the task group level.
Therefore, this patch adds a less sophisticated, but actually
very sensible, mechanism to ensure that a certain utilization
cap is not overcome per each root_domain (the single rq for !SMP
configurations).
Another main difference between deadline bandwidth management and
RT-throttling is that -deadline tasks have bandwidth on their own
(while -rt ones doesn't!), and thus we don't need an higher level
throttling mechanism to enforce the desired bandwidth.
This patch, therefore:
- adds system wide deadline bandwidth management by means of:
* /proc/sys/kernel/sched_dl_runtime_us,
* /proc/sys/kernel/sched_dl_period_us,
that determine (i.e., runtime / period) the total bandwidth
available on each CPU of each root_domain for -deadline tasks;
- couples the RT and deadline bandwidth management, i.e., enforces
that the sum of how much bandwidth is being devoted to -rt
-deadline tasks to stay below 100%.
This means that, for a root_domain comprising M CPUs, -deadline tasks
can be created until the sum of their bandwidths stay below:
M * (sched_dl_runtime_us / sched_dl_period_us)
It is also possible to disable this bandwidth management logic, and
be thus free of oversubscribing the system up to any arbitrary level.
Signed-off-by: Dario Faggioli <raistlin@linux.it>
Signed-off-by: Juri Lelli <juri.lelli@gmail.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1383831828-15501-12-git-send-email-juri.lelli@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
commit 887c290e (sched/numa: Decide whether to favour task or group weights
based on swap candidate relationships) drop the check against
sysctl_numa_balancing_settle_count, this patch remove the sysctl.
Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Link: http://lkml.kernel.org/r/1386833006-6600-1-git-send-email-liwanp@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull core locking changes from Ingo Molnar:
"The biggest changes:
- add lockdep support for seqcount/seqlocks structures, this
unearthed both bugs and required extra annotation.
- move the various kernel locking primitives to the new
kernel/locking/ directory"
* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
block: Use u64_stats_init() to initialize seqcounts
locking/lockdep: Mark __lockdep_count_forward_deps() as static
lockdep/proc: Fix lock-time avg computation
locking/doc: Update references to kernel/mutex.c
ipv6: Fix possible ipv6 seqlock deadlock
cpuset: Fix potential deadlock w/ set_mems_allowed
seqcount: Add lockdep functionality to seqcount/seqlock structures
net: Explicitly initialize u64_stats_sync structures for lockdep
locking: Move the percpu-rwsem code to kernel/locking/
locking: Move the lglocks code to kernel/locking/
locking: Move the rwsem code to kernel/locking/
locking: Move the rtmutex code to kernel/locking/
locking: Move the semaphore core to kernel/locking/
locking: Move the spinlock code to kernel/locking/
locking: Move the lockdep code to kernel/locking/
locking: Move the mutex code to kernel/locking/
hung_task debugging: Add tracepoint to report the hang
x86/locking/kconfig: Update paravirt spinlock Kconfig description
lockstat: Report avg wait and hold times
lockdep, x86/alternatives: Drop ancient lockdep fixup message
...
Need to check the return value of proc_put_char(), as was done in
__do_proc_doulongvec_minmax().
Signed-off-by: Chen Gang <gang.chen@asianux.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull scheduler changes from Ingo Molnar:
"The main changes in this cycle are:
- (much) improved CONFIG_NUMA_BALANCING support from Mel Gorman, Rik
van Riel, Peter Zijlstra et al. Yay!
- optimize preemption counter handling: merge the NEED_RESCHED flag
into the preempt_count variable, by Peter Zijlstra.
- wait.h fixes and code reorganization from Peter Zijlstra
- cfs_bandwidth fixes from Ben Segall
- SMP load-balancer cleanups from Peter Zijstra
- idle balancer improvements from Jason Low
- other fixes and cleanups"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (129 commits)
ftrace, sched: Add TRACE_FLAG_PREEMPT_RESCHED
stop_machine: Fix race between stop_two_cpus() and stop_cpus()
sched: Remove unnecessary iteration over sched domains to update nr_busy_cpus
sched: Fix asymmetric scheduling for POWER7
sched: Move completion code from core.c to completion.c
sched: Move wait code from core.c to wait.c
sched: Move wait.c into kernel/sched/
sched/wait: Fix __wait_event_interruptible_lock_irq_timeout()
sched: Avoid throttle_cfs_rq() racing with period_timer stopping
sched: Guarantee new group-entities always have weight
sched: Fix hrtimer_cancel()/rq->lock deadlock
sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining
sched: Fix race on toggling cfs_bandwidth_used
sched: Remove extra put_online_cpus() inside sched_setaffinity()
sched/rt: Fix task_tick_rt() comment
sched/wait: Fix build breakage
sched/wait: Introduce prepare_to_wait_event()
sched/wait: Add ___wait_cond_timeout() to wait_event*_timeout() too
sched: Remove get_online_cpus() usage
sched: Fix race in migrate_swap_stop()
...
Pull perf updates from Ingo Molnar:
"As a first remark I'd like to note that the way to build perf tooling
has been simplified and sped up, in the future it should be enough for
you to build perf via:
cd tools/perf/
make install
(ie without the -j option.) The build system will figure out the
number of CPUs and will do a parallel build+install.
The various build system inefficiencies and breakages Linus reported
against the v3.12 pull request should now be resolved - please
(re-)report any remaining annoyances or bugs.
Main changes on the perf kernel side:
* Performance optimizations:
. perf ring-buffer code optimizations, by Peter Zijlstra
. perf ring-buffer code optimizations, by Oleg Nesterov
. x86 NMI call-stack processing optimizations, by Peter Zijlstra
. perf context-switch optimizations, by Peter Zijlstra
. perf sampling speedups, by Peter Zijlstra
. x86 Intel PEBS processing speedups, by Peter Zijlstra
* Enhanced hardware support:
. for Intel Ivy Bridge-EP uncore PMUs, by Zheng Yan
. for Haswell transactions, by Andi Kleen, Peter Zijlstra
* Core perf events code enhancements and fixes by Oleg Nesterov:
. for uprobes, if fork() is called with pending ret-probes
. for uprobes platform support code
* New ABI details by Andi Kleen:
. Report x86 Haswell TSX transaction abort cost as weight
Main changes on the perf tooling side (some of these tooling changes
utilize the above kernel side changes):
* 'perf report/top' enhancements:
. Convert callchain children list to rbtree, greatly reducing the
time taken for callchain processing, from Namhyung Kim.
. Add new COMM infrastructure, further improving histogram
processing, from Frédéric Weisbecker, one fix from Namhyung Kim.
. Add /proc/kcore based live-annotation improvements, including
build-id cache support, multi map 'call' instruction navigation
fixes, kcore address validation, objdump workarounds. From
Adrian Hunter.
. Show progress on histogram collapsing, that can take a long
time, from Namhyung Kim.
. Add --max-stack option to limit callchain stack scan in 'top'
and 'report', improving callchain processing when reducing the
stack depth is an option, from Waiman Long.
. Add new option --ignore-vmlinux for perf top, from Willy
Tarreau.
* 'perf trace' enhancements:
. 'perf trace' now can can use a 'perf probe' dynamic tracepoints
to hook into the userspace -> kernel pathname copy so that it
can map fds to pathnames without reading /proc/pid/fd/ symlinks.
From Arnaldo Carvalho de Melo.
. Show VFS path associated with fd in live sessions, using a
'vfs_getname' 'perf probe' created dynamic tracepoint or by
looking at /proc/pid/fd, from Arnaldo Carvalho de Melo.
. Add 'trace' beautifiers for lots of syscall arguments, from
Arnaldo Carvalho de Melo.
. Implement more compact 'trace' output by suppressing zeroed
args, from Arnaldo Carvalho de Melo.
. Show thread COMM by default in 'trace', from Arnaldo Carvalho de
Melo.
. Add option to show full timestamp in 'trace', from David Ahern.
. Add 'record' command in 'trace', to record raw_syscalls:*, from
David Ahern.
. Add summary option to dump syscall statistics in 'trace', from
David Ahern.
. Improve error messages in 'trace', providing hints about system
configuration steps needed for using it, from Ramkumar
Ramachandra.
. 'perf trace' now emits hints as to why tracing is not possible,
helping the user to setup the system to allow tracing in the
desired permission granularity, telling if the problem is due to
debugfs not being mounted or with not enough permission for
!root, /proc/sys/kernel/perf_event_paranoit value, etc. From
Arnaldo Carvalho de Melo.
* 'perf record' enhancements:
. Check maximum frequency rate for record/top, emitting better
error messages, from Jiri Olsa.
. 'perf record' code cleanups, from David Ahern.
. Improve write_output error message in 'perf record', from Adrian
Hunter.
. Allow specifying B/K/M/G unit to the --mmap-pages arguments,
from Jiri Olsa.
. Fix command line callchain attribute tests to handle the new
-g/--call-chain semantics, from Arnaldo Carvalho de Melo.
* 'perf kvm' enhancements:
. Disable live kvm command if timerfd is not supported, from David
Ahern.
. Fix detection of non-core features, from David Ahern.
* 'perf list' enhancements:
. Add usage to 'perf list', from David Ahern.
. Show error in 'perf list' if tracepoints not available, from
Pekka Enberg.
* 'perf probe' enhancements:
. Support "$vars" meta argument syntax for local variables,
allowing asking for all possible variables at a given probe
point to be collected when it hits, from Masami Hiramatsu.
* 'perf sched' enhancements:
. Address the root cause of that 'perf sched' stack initialization
build slowdown, by programmatically setting a big array after
moving the global variable back to the stack. Fix from Adrian
Hunter.
* 'perf script' enhancements:
. Set up output options for in-stream attributes, from Adrian
Hunter.
. Print addr by default for BTS in 'perf script', from Adrian
Juntmer
* 'perf stat' enhancements:
. Improved messages when doing profiling in all or a subset of
CPUs using a workload as the session delimitator, as in:
'perf stat --cpu 0,2 sleep 10s'
from Arnaldo Carvalho de Melo.
. Add units to nanosec-based counters in 'perf stat', from David
Ahern.
. Remove bogus info when using 'perf stat' -e cycles/instructions,
from Ramkumar Ramachandra.
* 'perf lock' enhancements:
. 'perf lock' fixes and cleanups, from Davidlohr Bueso.
* 'perf test' enhancements:
. Fixup PERF_SAMPLE_TRANSACTION handling in sample synthesizing
and 'perf test', from Adrian Hunter.
. Clarify the "sample parsing" test entry, from Arnaldo Carvalho
de Melo.
. Consider PERF_SAMPLE_TRANSACTION in the "sample parsing" test,
from Arnaldo Carvalho de Melo.
. Memory leak fixes in 'perf test', from Felipe Pena.
* 'perf bench' enhancements:
. Change the procps visible command-name of invididual benchmark
tests plus cleanups, from Ingo Molnar.
* Generic perf tooling infrastructure/plumbing changes:
. Separating data file properties from session, code
reorganization from Jiri Olsa.
. Fix version when building out of tree, as when using one of
these:
$ make help | grep perf
perf-tar-src-pkg - Build perf-3.12.0.tar source tarball
perf-targz-src-pkg - Build perf-3.12.0.tar.gz source tarball
perf-tarbz2-src-pkg - Build perf-3.12.0.tar.bz2 source tarball
perf-tarxz-src-pkg - Build perf-3.12.0.tar.xz source tarball
$
from David Ahern.
. Enhance option parse error message, showing just the help lines
of the options affected, from Namhyung Kim.
. libtraceevent updates from upstream trace-cmd repo, from Steven
Rostedt.
. Always use perf_evsel__set_sample_bit to set sample_type, from
Adrian Hunter.
. Memory and mmap leak fixes from Chenggang Qin.
. Assorted build fixes for from David Ahern and Jiri Olsa.
. Speed up and prettify the build system, from Ingo Molnar.
. Implement addr2line directly using libbfd, from Roberto Vitillo.
. Separate the GTK support in a separate libperf-gtk.so DSO, that
is only loaded when --gtk is specified, from Namhyung Kim.
. perf bash completion fixes and improvements from Ramkumar
Ramachandra.
. Support for Openembedded/Yocto -dbg packages, from Ricardo
Ribalda Delgado.
And lots and lots of other fixes and code reorganizations that did not
make it into the list, see the shortlog, diffstat and the Git log for
details!"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (300 commits)
uprobes: Fix the memory out of bound overwrite in copy_insn()
uprobes: Fix the wrong usage of current->utask in uprobe_copy_process()
perf tools: Remove unneeded include
perf record: Remove post_processing_offset variable
perf record: Remove advance_output function
perf record: Refactor feature handling into a separate function
perf trace: Don't relookup fields by name in each sample
perf tools: Fix version when building out of tree
perf evsel: Ditch evsel->handler.data field
uprobes: Export write_opcode() as uprobe_write_opcode()
uprobes: Introduce arch_uprobe->ixol
uprobes: Kill module_init() and module_exit()
uprobes: Move function declarations out of arch
perf/x86/intel: Add Ivy Bridge-EP uncore IRP box support
perf/x86/intel/uncore: Add filter support for IvyBridge-EP QPI boxes
perf: Factor out strncpy() in perf_event_mmap_event()
tools/perf: Add required memory barriers
perf: Fix arch_perf_out_copy_user default
perf: Update a stale comment
perf: Optimize perf_output_begin() -- address calculation
...
Conflicts:
kernel/Makefile
There are conflicts in kernel/Makefile due to file moving in the
scheduler tree - resolve them.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Turn the initial value of sysctl kernel.sysrq (SYSRQ_DEFAULT_ENABLE)
into a Kconfig variable.
Original version by Bastian Blank <waldi@debian.org>.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Shared faults can lead to lots of unnecessary page migrations,
slowing down the system, and causing private faults to hit the
per-pgdat migration ratelimit.
This patch adds sysctl numa_balancing_migrate_deferred, which specifies
how many shared page migrations to skip unconditionally, after each page
migration that is skipped because it is a shared fault.
This reduces the number of page migrations back and forth in
shared fault situations. It also gives a strong preference to
the tasks that are already running where most of the memory is,
and to moving the other tasks to near the memory.
Testing this with a much higher scan rate than the default
still seems to result in fewer page migrations than before.
Memory seems to be somewhat better consolidated than previously,
with multi-instance specjbb runs on a 4 node system.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-62-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With scan rate adaptions based on whether the workload has properly
converged or not there should be no need for the scan period reset
hammer. Get rid of it.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-60-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch favours moving tasks towards NUMA node that recorded a higher
number of NUMA faults during active load balancing. Ideally this is
self-reinforcing as the longer the task runs on that node, the more faults
it should incur causing task_numa_placement to keep the task running on that
node. In reality a big weakness is that the nodes CPUs can be overloaded
and it would be more efficient to queue tasks on an idle node and migrate
to the new node. This would require additional smarts in the balancer so
for now the balancer will simply prefer to place the task on the preferred
node for a PTE scans which is controlled by the numa_balancing_settle_count
sysctl. Once the settle_count number of scans has complete the schedule
is free to place the task on an alternative node if the load is imbalanced.
[srikar@linux.vnet.ibm.com: Fixed statistics]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
[ Tunable and use higher faults instead of preferred. ]
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-23-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
/proc/sys/kernel/perf_event_max_sample_rate will accept
negative values as well as 0.
Negative values are unreasonable, and 0 causes a
divide by zero exception in perf_proc_update_handler.
This patch enforces a lower limit of 1.
Signed-off-by: Knut Petersen <Knut_Petersen@t-online.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/5242DB0C.4070005@t-online.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
As 'sysctl_hung_task_check_count' is 'unsigned long' when this
value is assigned to max_count in check_hung_uninterruptible_tasks(),
it's truncated to 'int' type.
This causes a minor artifact: if we write 2^32 to sysctl.hung_task_check_count,
hung task detection will be effectively disabled.
With this fix, it will still truncate the user input to 32 bits, but
reading sysctl.hung_task_check_count reflects the actual truncated value.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/523FFF4E.9050401@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull vfs pile 4 from Al Viro:
"list_lru pile, mostly"
This came out of Andrew's pile, Al ended up doing the merge work so that
Andrew didn't have to.
Additionally, a few fixes.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (42 commits)
super: fix for destroy lrus
list_lru: dynamically adjust node arrays
shrinker: Kill old ->shrink API.
shrinker: convert remaining shrinkers to count/scan API
staging/lustre/libcfs: cleanup linux-mem.h
staging/lustre/ptlrpc: convert to new shrinker API
staging/lustre/obdclass: convert lu_object shrinker to count/scan API
staging/lustre/ldlm: convert to shrinkers to count/scan API
hugepage: convert huge zero page shrinker to new shrinker API
i915: bail out earlier when shrinker cannot acquire mutex
drivers: convert shrinkers to new count/scan API
fs: convert fs shrinkers to new scan/count API
xfs: fix dquot isolation hang
xfs-convert-dquot-cache-lru-to-list_lru-fix
xfs: convert dquot cache lru to list_lru
xfs: rework buffer dispose list tracking
xfs-convert-buftarg-lru-to-generic-code-fix
xfs: convert buftarg LRU to generic code
fs: convert inode and dentry shrinking to be node aware
vmscan: per-node deferred work
...
Now hugepage migration is enabled, although restricted on pmd-based
hugepages for now (due to lack of testing.) So we should allocate
migratable hugepages from ZONE_MOVABLE if possible.
This patch makes GFP flags in hugepage allocation dependent on migration
support, not only the value of hugepages_treat_as_movable. It provides no
change on the behavior for architectures which do not support hugepage
migration,
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Hillf Danton <dhillf@gmail.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This series reworks our current object cache shrinking infrastructure in
two main ways:
* Noticing that a lot of users copy and paste their own version of LRU
lists for objects, we put some effort in providing a generic version.
It is modeled after the filesystem users: dentries, inodes, and xfs
(for various tasks), but we expect that other users could benefit in
the near future with little or no modification. Let us know if you
have any issues.
* The underlying list_lru being proposed automatically and
transparently keeps the elements in per-node lists, and is able to
manipulate the node lists individually. Given this infrastructure, we
are able to modify the up-to-now hammer called shrink_slab to proceed
with node-reclaim instead of always searching memory from all over like
it has been doing.
Per-node lru lists are also expected to lead to less contention in the lru
locks on multi-node scans, since we are now no longer fighting for a
global lock. The locks usually disappear from the profilers with this
change.
Although we have no official benchmarks for this version - be our guest to
independently evaluate this - earlier versions of this series were
performance tested (details at
http://permalink.gmane.org/gmane.linux.kernel.mm/100537) yielding no
visible performance regressions while yielding a better qualitative
behavior in NUMA machines.
With this infrastructure in place, we can use the list_lru entry point to
provide memcg isolation and per-memcg targeted reclaim. Historically,
those two pieces of work have been posted together. This version presents
only the infrastructure work, deferring the memcg work for a later time,
so we can focus on getting this part tested. You can see more about the
history of such work at http://lwn.net/Articles/552769/
Dave Chinner (18):
dcache: convert dentry_stat.nr_unused to per-cpu counters
dentry: move to per-sb LRU locks
dcache: remove dentries from LRU before putting on dispose list
mm: new shrinker API
shrinker: convert superblock shrinkers to new API
list: add a new LRU list type
inode: convert inode lru list to generic lru list code.
dcache: convert to use new lru list infrastructure
list_lru: per-node list infrastructure
shrinker: add node awareness
fs: convert inode and dentry shrinking to be node aware
xfs: convert buftarg LRU to generic code
xfs: rework buffer dispose list tracking
xfs: convert dquot cache lru to list_lru
fs: convert fs shrinkers to new scan/count API
drivers: convert shrinkers to new count/scan API
shrinker: convert remaining shrinkers to count/scan API
shrinker: Kill old ->shrink API.
Glauber Costa (7):
fs: bump inode and dentry counters to long
super: fix calculation of shrinkable objects for small numbers
list_lru: per-node API
vmscan: per-node deferred work
i915: bail out earlier when shrinker cannot acquire mutex
hugepage: convert huge zero page shrinker to new shrinker API
list_lru: dynamically adjust node arrays
This patch:
There are situations in very large machines in which we can have a large
quantity of dirty inodes, unused dentries, etc. This is particularly true
when umounting a filesystem, where eventually since every live object will
eventually be discarded.
Dave Chinner reported a problem with this while experimenting with the
shrinker revamp patchset. So we believe it is time for a change. This
patch just moves int to longs. Machines where it matters should have a
big long anyway.
Signed-off-by: Glauber Costa <glommer@openvz.org>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Carlos Maiolino <cmaiolino@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Gleb Natapov <gleb@redhat.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: J. Bruce Fields <bfields@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Kent Overstreet <koverstreet@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
When (integer) sysctl values are expressed in ms and have to be
represented internally as jiffies. The msecs_to_jiffies function
returns an unsigned long, which gets assigned to the integer.
This patch prevents the value to be assigned if bigger than
INT_MAX, done in a similar way as in cba9f3 ("Range checking in
do_proc_dointvec_(userhz_)jiffies_conv").
Signed-off-by: Francesco Fusco <ffusco@redhat.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: linux-kernel@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
were added to 3.10, which includes several bug fixes that have been
marked for stable.
As for new features, there were a few, but nothing to write to LWN about.
These include:
New function trigger called "dump" and "cpudump" that will cause
ftrace to dump its buffer to the console when the function is called.
The difference between "dump" and "cpudump" is that "dump" will dump
the entire contents of the ftrace buffer, where as "cpudump" will only
dump the contents of the ftrace buffer for the CPU that called the function.
Another small enhancement is a new sysctl switch called "traceoff_on_warning"
which, when enabled, will disable tracing if any WARN_ON() is triggered.
This is useful if you want to debug what caused a warning and do not
want to risk losing your trace data by the ring buffer overwriting the
data before you can disable it. There's also a kernel command line
option that will make this enabled at boot up called the same thing.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
iQEcBAABAgAGBQJR1uF2AAoJEOdOSU1xswtMJ1IH/2LSiZAKTA2QaRgGQC/5Bb9c
XSOI1HfD/78lmUvTyb0AX8sLpkzZlvIONEQ/WaZUFo1Zjbrl45zJUwMkTE9uImEg
ZqI5x8OiiN6j4XrRbfYn3Ti060H/Jq41pZXa+shh961Vv51ilv/1yyLkoRmnjzuO
JTloPdXDV7icOqqiSdgxSdtUSv59Ef1ZdHgvvsb3aqzMC5btVQPi4kIys0ST1Tr1
pMWBY+UgvH0xYm3gvTR+W6jjDlkVZEH2alkmcinfr+uC1tm9DDqK2HA17Pd5yZ5z
HNdT76lCzf9iqRF5F8HUvUt+PIp76dNNxAt2qpB6APqAuJTojyguxXHDbY/0kzs=
=UvLi
-----END PGP SIGNATURE-----
Merge tag 'trace-3.11' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing changes from Steven Rostedt:
"The majority of the changes here are cleanups for the large changes
that were added to 3.10, which includes several bug fixes that have
been marked for stable.
As for new features, there were a few, but nothing to write to LWN
about. These include:
New function trigger called "dump" and "cpudump" that will cause
ftrace to dump its buffer to the console when the function is called.
The difference between "dump" and "cpudump" is that "dump" will dump
the entire contents of the ftrace buffer, where as "cpudump" will only
dump the contents of the ftrace buffer for the CPU that called the
function.
Another small enhancement is a new sysctl switch called
"traceoff_on_warning" which, when enabled, will disable tracing if any
WARN_ON() is triggered. This is useful if you want to debug what
caused a warning and do not want to risk losing your trace data by the
ring buffer overwriting the data before you can disable it. There's
also a kernel command line option that will make this enabled at boot
up called the same thing"
* tag 'trace-3.11' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (34 commits)
tracing: Make tracing_open_generic_{tr,tc}() static
tracing: Remove ftrace() function
tracing: Remove TRACE_EVENT_TYPE enum definition
tracing: Make tracer_tracing_{off,on,is_on}() static
tracing: Fix irqs-off tag display in syscall tracing
uprobes: Fix return value in error handling path
tracing: Fix race between deleting buffer and setting events
tracing: Add trace_array_get/put() to event handling
tracing: Get trace_array ref counts when accessing trace files
tracing: Add trace_array_get/put() to handle instance refs better
tracing: Protect ftrace_trace_arrays list in trace_events.c
tracing: Make trace_marker use the correct per-instance buffer
ftrace: Do not run selftest if command line parameter is set
tracing/kprobes: Don't pass addr=ip to perf_trace_buf_submit()
tracing: Use flag buffer_disabled for irqsoff tracer
tracing/kprobes: Turn trace_probe->files into list_head
tracing: Fix disabling of soft disable
tracing: Add missing syscall_metadata comment
tracing: Simplify code for showing of soft disabled flag
tracing/kprobes: Kill probe_enable_lock
...
Pull nohz updates/fixes from Frederic Weisbecker:
' Note that "watchdog: Boot-disable by default on full dynticks" is a temporary
solution to solve the issue with the watchdog that prevents the tick from
stopping. This is to make sure that 3.11 doesn't have that problem as several
people complained about it.
A proper and longer term solution has been proposed by Peterz:
http://lkml.kernel.org/r/20130618103632.GO3204@twins.programming.kicks-ass.net
'
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch keeps track of how long perf's NMI handler is taking,
and also calculates how many samples perf can take a second. If
the sample length times the expected max number of samples
exceeds a configurable threshold, it drops the sample rate.
This way, we don't have a runaway sampling process eating up the
CPU.
This patch can tend to drop the sample rate down to level where
perf doesn't work very well. *BUT* the alternative is that my
system hangs because it spends all of its time handling NMIs.
I'll take a busted performance tool over an entire system that's
busted and undebuggable any day.
BTW, my suspicion is that there's still an underlying bug here.
Using the HPET instead of the TSC is definitely a contributing
factor, but I suspect there are some other things going on.
But, I can't go dig down on a bug like that with my machine
hanging all the time.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus@samba.org
Cc: acme@ghostprotocols.net
Cc: Dave Hansen <dave@sr71.net>
[ Prettified it a bit. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We have two very conflicting state variable names in the
watchdog:
* watchdog_enabled: This one reflects the user interface. It's
set to 1 by default and can be overriden with boot options
or sysctl/procfs interface.
* watchdog_disabled: This is the internal toggle state that
tells if watchdog threads, timers and NMI events are currently
running or not. This state mostly depends on the user settings.
It's a convenient state latch.
Now we really need to find clearer names because those
are just too confusing to encourage deep review.
watchdog_enabled now becomes watchdog_user_enabled to reflect
its purpose as an interface.
watchdog_disabled becomes watchdog_running to suggest its
role as a pure internal state.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Anish Singh <anish198519851985@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Don Zickus <dzickus@redhat.com>
Add a traceoff_on_warning option in both the kernel command line as well
as a sysctl option. When set, any WARN*() function that is hit will cause
the tracing_on variable to be cleared, which disables writing to the
ring buffer.
This is useful especially when tracing a bug with function tracing. When
a warning is hit, the print caused by the warning can flood the trace with
the functions that producing the output for the warning. This can make the
resulting trace useless by either hiding where the bug happened, or worse,
by overflowing the buffer and losing the trace of the bug totally.
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In old kernels, it's allowed to set softlockup_thresh to -1 or 0
to disable softlockup detection. However watchdog_thresh only
uses 0 to disable detection, and setting it to -1 just froze my
box and nothing I can do but reboot.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Acked-by: Don Zickus <dzickus@redhat.com>
Link: http://lkml.kernel.org/r/51959668.9040106@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add an admin_reserve_kbytes knob to allow admins to change the hardcoded
memory reserve to something other than 3%, which may be multiple
gigabytes on large memory systems. Only about 8MB is necessary to
enable recovery in the default mode, and only a few hundred MB are
required even when overcommit is disabled.
This affects OVERCOMMIT_GUESS and OVERCOMMIT_NEVER.
admin_reserve_kbytes is initialized to min(3% free pages, 8MB)
I arrived at 8MB by summing the RSS of sshd or login, bash, and top.
Please see first patch in this series for full background, motivation,
testing, and full changelog.
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: make init_admin_reserve() static]
Signed-off-by: Andrew Shewmaker <agshew@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add user_reserve_kbytes knob.
Limit the growth of the memory reserved for other user processes to
min(3% current process size, user_reserve_pages). Only about 8MB is
necessary to enable recovery in the default mode, and only a few hundred
MB are required even when overcommit is disabled.
user_reserve_pages defaults to min(3% free pages, 128MB)
I arrived at 128MB by taking the max VSZ of sshd, login, bash, and top ...
then adding the RSS of each.
This only affects OVERCOMMIT_NEVER mode.
Background
1. user reserve
__vm_enough_memory reserves a hardcoded 3% of the current process size for
other applications when overcommit is disabled. This was done so that a
user could recover if they launched a memory hogging process. Without the
reserve, a user would easily run into a message such as:
bash: fork: Cannot allocate memory
2. admin reserve
Additionally, a hardcoded 3% of free memory is reserved for root in both
overcommit 'guess' and 'never' modes. This was intended to prevent a
scenario where root-cant-log-in and perform recovery operations.
Note that this reserve shrinks, and doesn't guarantee a useful reserve.
Motivation
The two hardcoded memory reserves should be updated to account for current
memory sizes.
Also, the admin reserve would be more useful if it didn't shrink too much.
When the current code was originally written, 1GB was considered
"enterprise". Now the 3% reserve can grow to multiple GB on large memory
systems, and it only needs to be a few hundred MB at most to enable a user
or admin to recover a system with an unwanted memory hogging process.
I've found that reducing these reserves is especially beneficial for a
specific type of application load:
* single application system
* one or few processes (e.g. one per core)
* allocating all available memory
* not initializing every page immediately
* long running
I've run scientific clusters with this sort of load. A long running job
sometimes failed many hours (weeks of CPU time) into a calculation. They
weren't initializing all of their memory immediately, and they weren't
using calloc, so I put systems into overcommit 'never' mode. These
clusters run diskless and have no swap.
However, with the current reserves, a user wishing to allocate as much
memory as possible to one process may be prevented from using, for
example, almost 2GB out of 32GB.
The effect is less, but still significant when a user starts a job with
one process per core. I have repeatedly seen a set of processes
requesting the same amount of memory fail because one of them could not
allocate the amount of memory a user would expect to be able to allocate.
For example, Message Passing Interfce (MPI) processes, one per core. And
it is similar for other parallel programming frameworks.
Changing this reserve code will make the overcommit never mode more useful
by allowing applications to allocate nearly all of the available memory.
Also, the new admin_reserve_kbytes will be safer than the current behavior
since the hardcoded 3% of available memory reserve can shrink to something
useless in the case where applications have grabbed all available memory.
Risks
* "bash: fork: Cannot allocate memory"
The downside of the first patch-- which creates a tunable user reserve
that is only used in overcommit 'never' mode--is that an admin can set
it so low that a user may not be able to kill their process, even if
they already have a shell prompt.
Of course, a user can get in the same predicament with the current 3%
reserve--they just have to launch processes until 3% becomes negligible.
* root-cant-log-in problem
The second patch, adding the tunable rootuser_reserve_pages, allows
the admin to shoot themselves in the foot by setting it too small. They
can easily get the system into a state where root-can't-log-in.
However, the new admin_reserve_kbytes will be safer than the current
behavior since the hardcoded 3% of available memory reserve can shrink
to something useless in the case where applications have grabbed all
available memory.
Alternatives
* Memory cgroups provide a more flexible way to limit application memory.
Not everyone wants to set up cgroups or deal with their overhead.
* We could create a fourth overcommit mode which provides smaller reserves.
The size of useful reserves may be drastically different depending
on the whether the system is embedded or enterprise.
* Force users to initialize all of their memory or use calloc.
Some users don't want/expect the system to overcommit when they malloc.
Overcommit 'never' mode is for this scenario, and it should work well.
The new user and admin reserve tunables are simple to use, with low
overhead compared to cgroups. The patches preserve current behavior where
3% of memory is less than 128MB, except that the admin reserve doesn't
shrink to an unusable size under pressure. The code allows admins to tune
for embedded and enterprise usage.
FAQ
* How is the root-cant-login problem addressed?
What happens if admin_reserve_pages is set to 0?
Root is free to shoot themselves in the foot by setting
admin_reserve_kbytes too low.
On x86_64, the minimum useful reserve is:
8MB for overcommit 'guess'
128MB for overcommit 'never'
admin_reserve_pages defaults to min(3% free memory, 8MB)
So, anyone switching to 'never' mode needs to adjust
admin_reserve_pages.
* How do you calculate a minimum useful reserve?
A user or the admin needs enough memory to login and perform
recovery operations, which includes, at a minimum:
sshd or login + bash (or some other shell) + top (or ps, kill, etc.)
For overcommit 'guess', we can sum resident set sizes (RSS)
because we only need enough memory to handle what the recovery
programs will typically use. On x86_64 this is about 8MB.
For overcommit 'never', we can take the max of their virtual sizes (VSZ)
and add the sum of their RSS. We use VSZ instead of RSS because mode
forces us to ensure we can fulfill all of the requested memory allocations--
even if the programs only use a fraction of what they ask for.
On x86_64 this is about 128MB.
When swap is enabled, reserves are useful even when they are as
small as 10MB, regardless of overcommit mode.
When both swap and overcommit are disabled, then the admin should
tune the reserves higher to be absolutley safe. Over 230MB each
was safest in my testing.
* What happens if user_reserve_pages is set to 0?
Note, this only affects overcomitt 'never' mode.
Then a user will be able to allocate all available memory minus
admin_reserve_kbytes.
However, they will easily see a message such as:
"bash: fork: Cannot allocate memory"
And they won't be able to recover/kill their application.
The admin should be able to recover the system if
admin_reserve_kbytes is set appropriately.
* What's the difference between overcommit 'guess' and 'never'?
"Guess" allows an allocation if there are enough free + reclaimable
pages. It has a hardcoded 3% of free pages reserved for root.
"Never" allows an allocation if there is enough swap + a configurable
percentage (default is 50) of physical RAM. It has a hardcoded 3% of
free pages reserved for root, like "Guess" mode. It also has a
hardcoded 3% of the current process size reserved for additional
applications.
* Why is overcommit 'guess' not suitable even when an app eventually
writes to every page? It takes free pages, file pages, available
swap pages, reclaimable slab pages into consideration. In other words,
these are all pages available, then why isn't overcommit suitable?
Because it only looks at the present state of the system. It
does not take into account the memory that other applications have
malloced, but haven't initialized yet. It overcommits the system.
Test Summary
There was little change in behavior in the default overcommit 'guess'
mode with swap enabled before and after the patch. This was expected.
Systems run most predictably (i.e. no oom kills) in overcommit 'never'
mode with swap enabled. This also allowed the most memory to be allocated
to a user application.
Overcommit 'guess' mode without swap is a bad idea. It is easy to
crash the system. None of the other tested combinations crashed.
This matches my experience on the Roadrunner supercomputer.
Without the tunable user reserve, a system in overcommit 'never' mode
and without swap does not allow the admin to recover, although the
admin can.
With the new tunable reserves, a system in overcommit 'never' mode
and without swap can be configured to:
1. maximize user-allocatable memory, running close to the edge of
recoverability
2. maximize recoverability, sacrificing allocatable memory to
ensure that a user cannot take down a system
Test Description
Fedora 18 VM - 4 x86_64 cores, 5725MB RAM, 4GB Swap
System is booted into multiuser console mode, with unnecessary services
turned off. Caches were dropped before each test.
Hogs are user memtester processes that attempt to allocate all free memory
as reported by /proc/meminfo
In overcommit 'never' mode, memory_ratio=100
Test Results
3.9.0-rc1-mm1
Overcommit | Swap | Hogs | MB Got/Wanted | OOMs | User Recovery | Admin Recovery
---------- ---- ---- ------------- ---- ------------- --------------
guess yes 1 5432/5432 no yes yes
guess yes 4 5444/5444 1 yes yes
guess no 1 5302/5449 no yes yes
guess no 4 - crash no no
never yes 1 5460/5460 1 yes yes
never yes 4 5460/5460 1 yes yes
never no 1 5218/5432 no no yes
never no 4 5203/5448 no no yes
3.9.0-rc1-mm1-tunablereserves
User and Admin Recovery show their respective reserves, if applicable.
Overcommit | Swap | Hogs | MB Got/Wanted | OOMs | User Recovery | Admin Recovery
---------- ---- ---- ------------- ---- ------------- --------------
guess yes 1 5419/5419 no - yes 8MB yes
guess yes 4 5436/5436 1 - yes 8MB yes
guess no 1 5440/5440 * - yes 8MB yes
guess no 4 - crash - no 8MB no
* process would successfully mlock, then the oom killer would pick it
never yes 1 5446/5446 no 10MB yes 20MB yes
never yes 4 5456/5456 no 10MB yes 20MB yes
never no 1 5387/5429 no 128MB no 8MB barely
never no 1 5323/5428 no 226MB barely 8MB barely
never no 1 5323/5428 no 226MB barely 8MB barely
never no 1 5359/5448 no 10MB no 10MB barely
never no 1 5323/5428 no 0MB no 10MB barely
never no 1 5332/5428 no 0MB no 50MB yes
never no 1 5293/5429 no 0MB no 90MB yes
never no 1 5001/5427 no 230MB yes 338MB yes
never no 4* 4998/5424 no 230MB yes 338MB yes
* more memtesters were launched, able to allocate approximately another 100MB
Future Work
- Test larger memory systems.
- Test an embedded image.
- Test other architectures.
- Time malloc microbenchmarks.
- Would it be useful to be able to set overcommit policy for
each memory cgroup?
- Some lines are slightly above 80 chars.
Perhaps define a macro to convert between pages and kb?
Other places in the kernel do this.
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: make init_user_reserve() static]
Signed-off-by: Andrew Shewmaker <agshew@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drop_caches.c provides code only invokable via sysctl, so don't compile it
in when CONFIG_SYSCTL=n.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=NKkW
-----END PGP SIGNATURE-----
Merge tag 'arc-v3.9-rc1-late' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc
Pull new ARC architecture from Vineet Gupta:
"Initial ARC Linux port with some fixes on top for 3.9-rc1:
I would like to introduce the Linux port to ARC Processors (from
Synopsys) for 3.9-rc1. The patch-set has been discussed on the public
lists since Nov and has received a fair bit of review, specially from
Arnd, tglx, Al and other subsystem maintainers for DeviceTree, kgdb...
The arch bits are in arch/arc, some asm-generic changes (acked by
Arnd), a minor change to PARISC (acked by Helge).
The series is a touch bigger for a new port for 2 main reasons:
1. It enables a basic kernel in first sub-series and adds
ptrace/kgdb/.. later
2. Some of the fallout of review (DeviceTree support, multi-platform-
image support) were added on top of orig series, primarily to
record the revision history.
This updated pull request additionally contains
- fixes due to our GNU tools catching up with the new syscall/ptrace
ABI
- some (minor) cross-arch Kconfig updates."
* tag 'arc-v3.9-rc1-late' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc: (82 commits)
ARC: split elf.h into uapi and export it for userspace
ARC: Fixup the current ABI version
ARC: gdbserver using regset interface possibly broken
ARC: Kconfig cleanup tracking cross-arch Kconfig pruning in merge window
ARC: make a copy of flat DT
ARC: [plat-arcfpga] DT arc-uart bindings change: "baud" => "current-speed"
ARC: Ensure CONFIG_VIRT_TO_BUS is not enabled
ARC: Fix pt_orig_r8 access
ARC: [3.9] Fallout of hlist iterator update
ARC: 64bit RTSC timestamp hardware issue
ARC: Don't fiddle with non-existent caches
ARC: Add self to MAINTAINERS
ARC: Provide a default serial.h for uart drivers needing BASE_BAUD
ARC: [plat-arcfpga] defconfig for fully loaded ARC Linux
ARC: [Review] Multi-platform image #8: platform registers SMP callbacks
ARC: [Review] Multi-platform image #7: SMP common code to use callbacks
ARC: [Review] Multi-platform image #6: cpu-to-dma-addr optional
ARC: [Review] Multi-platform image #5: NR_IRQS defined by ARC core
ARC: [Review] Multi-platform image #4: Isolate platform headers
ARC: [Review] Multi-platform image #3: switch to board callback
...
The existing SUID_DUMP_* defines duplicate the newer SUID_DUMPABLE_*
defines introduced in 54b501992d ("coredump: warn about unsafe
suid_dumpable / core_pattern combo"). Remove the new ones, and use the
prior values instead.
Signed-off-by: Kees Cook <keescook@chromium.org>
Reported-by: Chen Gang <gang.chen@asianux.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alan Cox <alan@linux.intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Doug Ledford <dledford@redhat.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: James Morris <james.l.morris@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lockdep, but it's a mechanical change.
Cheers,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=tiSY
-----END PGP SIGNATURE-----
Merge tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull module update from Rusty Russell:
"The sweeping change is to make add_taint() explicitly indicate whether
to disable lockdep, but it's a mechanical change."
* tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
MODSIGN: Add option to not sign modules during modules_install
MODSIGN: Add -s <signature> option to sign-file
MODSIGN: Specify the hash algorithm on sign-file command line
MODSIGN: Simplify Makefile with a Kconfig helper
module: clean up load_module a little more.
modpost: Ignore ARC specific non-alloc sections
module: constify within_module_*
taint: add explicit flag to show whether lock dep is still OK.
module: printk message when module signature fail taints kernel.
When calculating amount of dirtyable memory, min_free_kbytes should be
subtracted because it is not intended for dirty pages.
Addresses http://bugs.debian.org/695182
[akpm@linux-foundation.org: fix up min_free_kbytes extern declarations]
[akpm@linux-foundation.org: fix min() warning]
Signed-off-by: Paul Szabo <psz@maths.usyd.edu.au>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
PARISC defines /proc/sys/kernel/unaligned-trap to runtime toggle
unaligned access emulation.
The exact mechanics of enablig/disabling are still arch specific, we can
make the sysctl usable by other arches.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Acked-by: Helge Deller <deller@gmx.de>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: Helge Deller <deller@gmx.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Add a /proc/sys/kernel scheduler knob named
sched_rr_timeslice_ms that allows global changing of the
SCHED_RR timeslice value. User visable value is in milliseconds
but is stored as jiffies. Setting to 0 (zero) resets to the
default (currently 100ms).
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094704.13751796@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094659.06dced96@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix up all callers as they were before, with make one change: an
unsigned module taints the kernel, but doesn't turn off lockdep.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
IA64 defines /proc/sys/kernel/ignore-unaligned-usertrap to control
verbose warnings on unaligned access emulation.
Although the exact mechanics of what to do with sysctl (ignore/shout)
are arch specific, this change enables the sysctl to be usable cross-arch.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)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=ivQa
-----END PGP SIGNATURE-----
Merge tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma
Pull Automatic NUMA Balancing bare-bones from Mel Gorman:
"There are three implementations for NUMA balancing, this tree
(balancenuma), numacore which has been developed in tip/master and
autonuma which is in aa.git.
In almost all respects balancenuma is the dumbest of the three because
its main impact is on the VM side with no attempt to be smart about
scheduling. In the interest of getting the ball rolling, it would be
desirable to see this much merged for 3.8 with the view to building
scheduler smarts on top and adapting the VM where required for 3.9.
The most recent set of comparisons available from different people are
mel: https://lkml.org/lkml/2012/12/9/108
mingo: https://lkml.org/lkml/2012/12/7/331
tglx: https://lkml.org/lkml/2012/12/10/437
srikar: https://lkml.org/lkml/2012/12/10/397
The results are a mixed bag. In my own tests, balancenuma does
reasonably well. It's dumb as rocks and does not regress against
mainline. On the other hand, Ingo's tests shows that balancenuma is
incapable of converging for this workloads driven by perf which is bad
but is potentially explained by the lack of scheduler smarts. Thomas'
results show balancenuma improves on mainline but falls far short of
numacore or autonuma. Srikar's results indicate we all suffer on a
large machine with imbalanced node sizes.
My own testing showed that recent numacore results have improved
dramatically, particularly in the last week but not universally.
We've butted heads heavily on system CPU usage and high levels of
migration even when it shows that overall performance is better.
There are also cases where it regresses. Of interest is that for
specjbb in some configurations it will regress for lower numbers of
warehouses and show gains for higher numbers which is not reported by
the tool by default and sometimes missed in treports. Recently I
reported for numacore that the JVM was crashing with
NullPointerExceptions but currently it's unclear what the source of
this problem is. Initially I thought it was in how numacore batch
handles PTEs but I'm no longer think this is the case. It's possible
numacore is just able to trigger it due to higher rates of migration.
These reports were quite late in the cycle so I/we would like to start
with this tree as it contains much of the code we can agree on and has
not changed significantly over the last 2-3 weeks."
* tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma: (50 commits)
mm/rmap, migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable
mm/rmap: Convert the struct anon_vma::mutex to an rwsem
mm: migrate: Account a transhuge page properly when rate limiting
mm: numa: Account for failed allocations and isolations as migration failures
mm: numa: Add THP migration for the NUMA working set scanning fault case build fix
mm: numa: Add THP migration for the NUMA working set scanning fault case.
mm: sched: numa: Delay PTE scanning until a task is scheduled on a new node
mm: sched: numa: Control enabling and disabling of NUMA balancing if !SCHED_DEBUG
mm: sched: numa: Control enabling and disabling of NUMA balancing
mm: sched: Adapt the scanning rate if a NUMA hinting fault does not migrate
mm: numa: Use a two-stage filter to restrict pages being migrated for unlikely task<->node relationships
mm: numa: migrate: Set last_nid on newly allocated page
mm: numa: split_huge_page: Transfer last_nid on tail page
mm: numa: Introduce last_nid to the page frame
sched: numa: Slowly increase the scanning period as NUMA faults are handled
mm: numa: Rate limit setting of pte_numa if node is saturated
mm: numa: Rate limit the amount of memory that is migrated between nodes
mm: numa: Structures for Migrate On Fault per NUMA migration rate limiting
mm: numa: Migrate pages handled during a pmd_numa hinting fault
mm: numa: Migrate on reference policy
...
The PTE scanning rate and fault rates are two of the biggest sources of
system CPU overhead with automatic NUMA placement. Ideally a proper policy
would detect if a workload was properly placed, schedule and adjust the
PTE scanning rate accordingly. We do not track the necessary information
to do that but we at least know if we migrated or not.
This patch scans slower if a page was not migrated as the result of a
NUMA hinting fault up to sysctl_numa_balancing_scan_period_max which is
now higher than the previous default. Once every minute it will reset
the scanner in case of phase changes.
This is hilariously crude and the numbers are arbitrary. Workloads will
converge quite slowly in comparison to what a proper policy should be able
to do. On the plus side, we will chew up less CPU for workloads that have
no need for automatic balancing.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Add a 1 second delay before starting to scan the working set of
a task and starting to balance it amongst nodes.
[ note that before the constant per task WSS sampling rate patch
the initial scan would happen much later still, in effect that
patch caused this regression. ]
The theory is that short-run tasks benefit very little from NUMA
placement: they come and go, and they better stick to the node
they were started on. As tasks mature and rebalance to other CPUs
and nodes, so does their NUMA placement have to change and so
does it start to matter more and more.
In practice this change fixes an observable kbuild regression:
# [ a perf stat --null --repeat 10 test of ten bzImage builds to /dev/shm ]
!NUMA:
45.291088843 seconds time elapsed ( +- 0.40% )
45.154231752 seconds time elapsed ( +- 0.36% )
+NUMA, no slow start:
46.172308123 seconds time elapsed ( +- 0.30% )
46.343168745 seconds time elapsed ( +- 0.25% )
+NUMA, 1 sec slow start:
45.224189155 seconds time elapsed ( +- 0.25% )
45.160866532 seconds time elapsed ( +- 0.17% )
and it also fixes an observable perf bench (hackbench) regression:
# perf stat --null --repeat 10 perf bench sched messaging
-NUMA:
-NUMA: 0.246225691 seconds time elapsed ( +- 1.31% )
+NUMA no slow start: 0.252620063 seconds time elapsed ( +- 1.13% )
+NUMA 1sec delay: 0.248076230 seconds time elapsed ( +- 1.35% )
The implementation is simple and straightforward, most of the patch
deals with adding the /proc/sys/kernel/numa_balancing_scan_delay_ms tunable
knob.
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: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
[ Wrote the changelog, ran measurements, tuned the default. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Previously, to probe the working set of a task, we'd use
a very simple and crude method: mark all of its address
space PROT_NONE.
That method has various (obvious) disadvantages:
- it samples the working set at dissimilar rates,
giving some tasks a sampling quality advantage
over others.
- creates performance problems for tasks with very
large working sets
- over-samples processes with large address spaces but
which only very rarely execute
Improve that method by keeping a rotating offset into the
address space that marks the current position of the scan,
and advance it by a constant rate (in a CPU cycles execution
proportional manner). If the offset reaches the last mapped
address of the mm then it then it starts over at the first
address.
The per-task nature of the working set sampling functionality in this tree
allows such constant rate, per task, execution-weight proportional sampling
of the working set, with an adaptive sampling interval/frequency that
goes from once per 100ms up to just once per 8 seconds. The current
sampling volume is 256 MB per interval.
As tasks mature and converge their working set, so does the
sampling rate slow down to just a trickle, 256 MB per 8
seconds of CPU time executed.
This, beyond being adaptive, also rate-limits rarely
executing systems and does not over-sample on overloaded
systems.
[ In AutoNUMA speak, this patch deals with the effective sampling
rate of the 'hinting page fault'. AutoNUMA's scanning is
currently rate-limited, but it is also fundamentally
single-threaded, executing in the knuma_scand kernel thread,
so the limit in AutoNUMA is global and does not scale up with
the number of CPUs, nor does it scan tasks in an execution
proportional manner.
So the idea of rate-limiting the scanning was first implemented
in the AutoNUMA tree via a global rate limit. This patch goes
beyond that by implementing an execution rate proportional
working set sampling rate that is not implemented via a single
global scanning daemon. ]
[ Dan Carpenter pointed out a possible NULL pointer dereference in the
first version of this patch. ]
Based-on-idea-by: Andrea Arcangeli <aarcange@redhat.com>
Bug-Found-By: Dan Carpenter <dan.carpenter@oracle.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: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
[ Wrote changelog and fixed bug. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
David Rientjes
Liu Hua
Linus Torvalds
Aaron Tomlin
Kees Cook
Andi Kleen
Jerome Marchand
Peter Zijlstra
Dario Faggioli
Wanpeng Li
Chen Gang
Ingo Molnar
Ben Hutchings
Rik van Riel
Mel Gorman
Knut Petersen
Li Zefan
Naoya Horiguchi
Glauber Costa
Francesco Fusco
Thomas Gleixner
Dave Hansen
Frederic Weisbecker
Steven Rostedt (Red Hat)
Andrew Shewmaker
Josh Triplett
Paul Szabo
Vineet Gupta
Clark Williams
Rusty Russell
Vineet Gupta
Peter Zijlstra