pipe_max_size is assigned directly via procfs sysctl:

  static struct ctl_table fs_table[] = {
          ...
          {
                  .procname       = "pipe-max-size",
                  .data           = &pipe_max_size,
                  .maxlen         = sizeof(int),
                  .mode           = 0644,
                  .proc_handler   = &pipe_proc_fn,
                  .extra1         = &pipe_min_size,
          },
          ...

  int pipe_proc_fn(struct ctl_table *table, int write, void __user *buf,
                   size_t *lenp, loff_t *ppos)
  {
          ...
          ret = proc_dointvec_minmax(table, write, buf, lenp, ppos)
          ...

and then later rounded in-place a few statements later:

          ...
          pipe_max_size = round_pipe_size(pipe_max_size);
          ...

This leaves a window of time between initial assignment and rounding
that may be visible to other threads.  (For example, one thread sets a
non-rounded value to pipe_max_size while another reads its value.)

Similar reads of pipe_max_size are potentially racy:

  pipe.c :: alloc_pipe_info()
  pipe.c :: pipe_set_size()

Add a new proc_dopipe_max_size() that consolidates reading the new value
from the user buffer, verifying bounds, and calling round_pipe_size()
with a single assignment to pipe_max_size.

Link: http://lkml.kernel.org/r/1507658689-11669-4-git-send-email-joe.lawrence@redhat.comSigned-off-by: NJoe Lawrence <joe.lawrence@redhat.com>
Reported-by: NMikulas Patocka <mpatocka@redhat.com>
Reviewed-by: NMikulas Patocka <mpatocka@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Patch series "A few round_pipe_size() and pipe-max-size fixups", v3.

While backporting Michael's "pipe: fix limit handling" patchset to a
distro-kernel, Mikulas noticed that current upstream pipe limit handling
contains a few problems:

  1 - procfs signed wrap: echo'ing a large number into
      /proc/sys/fs/pipe-max-size and then cat'ing it back out shows a
      negative value.

  2 - round_pipe_size() nr_pages overflow on 32bit:  this would
      subsequently try roundup_pow_of_two(0), which is undefined.

  3 - visible non-rounded pipe-max-size value: there is no mutual
      exclusion or protection between the time pipe_max_size is assigned
      a raw value from proc_dointvec_minmax() and when it is rounded.

  4 - unsigned long -> unsigned int conversion makes for potential odd
      return errors from do_proc_douintvec_minmax_conv() and
      do_proc_dopipe_max_size_conv().

This version underwent the same testing as v1:
https://marc.info/?l=linux-kernel&m=150643571406022&w=2

This patch (of 4):

pipe_max_size is defined as an unsigned int:

  unsigned int pipe_max_size = 1048576;

but its procfs/sysctl representation is an integer:

  static struct ctl_table fs_table[] = {
          ...
          {
                  .procname       = "pipe-max-size",
                  .data           = &pipe_max_size,
                  .maxlen         = sizeof(int),
                  .mode           = 0644,
                  .proc_handler   = &pipe_proc_fn,
                  .extra1         = &pipe_min_size,
          },
          ...

that is signed:

  int pipe_proc_fn(struct ctl_table *table, int write, void __user *buf,
                   size_t *lenp, loff_t *ppos)
  {
          ...
          ret = proc_dointvec_minmax(table, write, buf, lenp, ppos)

This leads to signed results via procfs for large values of pipe_max_size:

  % echo 2147483647 >/proc/sys/fs/pipe-max-size
  % cat /proc/sys/fs/pipe-max-size
  -2147483648

Use unsigned operations on this variable to avoid such negative values.

Link: http://lkml.kernel.org/r/1507658689-11669-2-git-send-email-joe.lawrence@redhat.comSigned-off-by: NJoe Lawrence <joe.lawrence@redhat.com>
Reported-by: NMikulas Patocka <mpatocka@redhat.com>
Reviewed-by: NMikulas Patocka <mpatocka@redhat.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This is the second step which introduces a tunable interface that allow
numa stats configurable for optimizing zone_statistics(), as suggested
by Dave Hansen and Ying Huang.

=========================================================================

When page allocation performance becomes a bottleneck and you can
tolerate some possible tool breakage and decreased numa counter
precision, you can do:

	echo 0 > /proc/sys/vm/numa_stat

In this case, numa counter update is ignored.  We can see about
*4.8%*(185->176) drop of cpu cycles per single page allocation and
reclaim on Jesper's page_bench01 (single thread) and *8.1%*(343->315)
drop of cpu cycles per single page allocation and reclaim on Jesper's
page_bench03 (88 threads) running on a 2-Socket Broadwell-based server
(88 threads, 126G memory).

Benchmark link provided by Jesper D Brouer (increase loop times to
10000000):

  https://github.com/netoptimizer/prototype-kernel/tree/master/kernel/mm/bench

=========================================================================

When page allocation performance is not a bottleneck and you want all
tooling to work, you can do:

	echo 1 > /proc/sys/vm/numa_stat

This is system default setting.

Many thanks to Michal Hocko, Dave Hansen, Ying Huang and Vlastimil Babka
for comments to help improve the original patch.

[keescook@chromium.org: make sure mutex is a global static]
  Link: http://lkml.kernel.org/r/20171107213809.GA4314@beast
Link: http://lkml.kernel.org/r/1508290927-8518-1-git-send-email-kemi.wang@intel.comSigned-off-by: NKemi Wang <kemi.wang@intel.com>
Signed-off-by: NKees Cook <keescook@chromium.org>
Reported-by: NJesper Dangaard Brouer <brouer@redhat.com>
Suggested-by: NDave Hansen <dave.hansen@intel.com>
Suggested-by: NYing Huang <ying.huang@intel.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: "Luis R . Rodriguez" <mcgrof@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Christopher Lameter <cl@linux.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Tim Chen <tim.c.chen@intel.com>
Cc: Andi Kleen <andi.kleen@intel.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix up makefiles, remove references, and git rm kmemcheck.

Link: http://lkml.kernel.org/r/20171007030159.22241-4-alexander.levin@verizon.comSigned-off-by: NSasha Levin <alexander.levin@verizon.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Vegard Nossum <vegardno@ifi.uio.no>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Tim Hansen <devtimhansen@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This tunable has been obsolete since 2.6.32, and writes to the
file have been failing and complaining in dmesg since then:

nr_pdflush_threads exported in /proc is scheduled for removal

That was 8 years ago. Remove the file ABI obsolete notice, and
the sysfs file.
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

do_proc_douintvec_conv() has two UINT_MAX checks, we can remove one.
This has no functional changes other than fixing a compiler warning:

  kernel/sysctl.c:2190]: (warning) Identical condition '*lvalp>UINT_MAX', second condition is always false

Fixes: 4f2fec00 ("sysctl: simplify unsigned int support")
Link: http://lkml.kernel.org/r/20170919072918.12066-1-mcgrof@kernel.orgSigned-off-by: NLuis R. Rodriguez <mcgrof@kernel.org>
Reported-by: NDavid Binderman <dcb314@hotmail.com>
Acked-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

System will hang if user set sysctl_sched_time_avg to 0:

  [root@XXX ~]# sysctl kernel.sched_time_avg_ms=0

  Stack traceback for pid 0
  0xffff883f6406c600 0 0 1 3 R 0xffff883f6406cf50 *swapper/3
  ffff883f7ccc3ae8 0000000000000018 ffffffff810c4dd0 0000000000000000
  0000000000017800 ffff883f7ccc3d78 0000000000000003 ffff883f7ccc3bf8
  ffffffff810c4fc9 ffff883f7ccc3c08 00000000810c5043 ffff883f7ccc3c08
  Call Trace:
  <IRQ> [<ffffffff810c4dd0>] ? update_group_capacity+0x110/0x200
  [<ffffffff810c4fc9>] ? update_sd_lb_stats+0x109/0x600
  [<ffffffff810c5507>] ? find_busiest_group+0x47/0x530
  [<ffffffff810c5b84>] ? load_balance+0x194/0x900
  [<ffffffff810ad5ca>] ? update_rq_clock.part.83+0x1a/0xe0
  [<ffffffff810c6d42>] ? rebalance_domains+0x152/0x290
  [<ffffffff810c6f5c>] ? run_rebalance_domains+0xdc/0x1d0
  [<ffffffff8108a75b>] ? __do_softirq+0xfb/0x320
  [<ffffffff8108ac85>] ? irq_exit+0x125/0x130
  [<ffffffff810b3a17>] ? scheduler_ipi+0x97/0x160
  [<ffffffff81052709>] ? smp_reschedule_interrupt+0x29/0x30
  [<ffffffff8173a1be>] ? reschedule_interrupt+0x6e/0x80
   <EOI> [<ffffffff815bc83c>] ? cpuidle_enter_state+0xcc/0x230
  [<ffffffff815bc80c>] ? cpuidle_enter_state+0x9c/0x230
  [<ffffffff815bc9d7>] ? cpuidle_enter+0x17/0x20
  [<ffffffff810cd6dc>] ? cpu_startup_entry+0x38c/0x420
  [<ffffffff81053373>] ? start_secondary+0x173/0x1e0

Because divide-by-zero error happens in function:

update_group_capacity()
  update_cpu_capacity()
    scale_rt_capacity()
     {
          ...
          total = sched_avg_period() + delta;
          used = div_u64(avg, total);
          ...
     }

To fix this issue, check user input value of sysctl_sched_time_avg, keep
it unchanged when hitting invalid input, and set the minimum limit of
sysctl_sched_time_avg to 1 ms.
Reported-by: NJames Puthukattukaran <james.puthukattukaran@oracle.com>
Signed-off-by: NEthan Zhao <ethan.zhao@oracle.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: efault@gmx.de
Cc: ethan.kernel@gmail.com
Cc: keescook@chromium.org
Cc: mcgrof@kernel.org
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/1504504774-18253-1-git-send-email-ethan.zhao@oracle.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

Reflect that these variables are user interface related and remove the
whitespace damage in the sysctl table while at it.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NDon Zickus <dzickus@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Chris Metcalf <cmetcalf@mellanox.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Ulrich Obergfell <uobergfe@redhat.com>
Link: http://lkml.kernel.org/r/20170912194147.783210221@linutronix.deSigned-off-by: NIngo Molnar <mingo@kernel.org>

The sysctl of the nmi_watchdog file prevents writes by setting:

    min = max = 0

if none of the users is enabled. That involves ifdeffery and is competely
non obvious.

If none of the facilities is enabeld, then the file can simply be made read
only. Move the ifdeffery into the header and use a constant for file
permissions.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NDon Zickus <dzickus@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Chris Metcalf <cmetcalf@mellanox.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Ulrich Obergfell <uobergfe@redhat.com>
Link: http://lkml.kernel.org/r/20170912194147.706073616@linutronix.deSigned-off-by: NIngo Molnar <mingo@kernel.org>

Split SOFTLOCKUP_DETECTOR from LOCKUP_DETECTOR, and split
HARDLOCKUP_DETECTOR_PERF from HARDLOCKUP_DETECTOR.

LOCKUP_DETECTOR implies the general boot, sysctl, and programming
interfaces for the lockup detectors.

An architecture that wants to use a hard lockup detector must define
HAVE_HARDLOCKUP_DETECTOR_PERF or HAVE_HARDLOCKUP_DETECTOR_ARCH.

Alternatively an arch can define HAVE_NMI_WATCHDOG, which provides the
minimum arch_touch_nmi_watchdog, and it otherwise does its own thing and
does not implement the LOCKUP_DETECTOR interfaces.

sparc is unusual in that it has started to implement some of the
interfaces, but not fully yet.  It should probably be converted to a full
HAVE_HARDLOCKUP_DETECTOR_ARCH.

[npiggin@gmail.com: fix]
  Link: http://lkml.kernel.org/r/20170617223522.66c0ad88@roar.ozlabs.ibm.com
Link: http://lkml.kernel.org/r/20170616065715.18390-4-npiggin@gmail.comSigned-off-by: NNicholas Piggin <npiggin@gmail.com>
Reviewed-by: NDon Zickus <dzickus@redhat.com>
Reviewed-by: NBabu Moger <babu.moger@oracle.com>
Tested-by: Babu Moger <babu.moger@oracle.com>	[sparc]
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

To keep parity with regular int interfaces provide the an unsigned int
proc_douintvec_minmax() which allows you to specify a range of allowed
valid numbers.

Adding proc_douintvec_minmax_sysadmin() is easy but we can wait for an
actual user for that.

Link: http://lkml.kernel.org/r/20170519033554.18592-6-mcgrof@kernel.orgSigned-off-by: NLuis R. Rodriguez <mcgrof@kernel.org>
Acked-by: NKees Cook <keescook@chromium.org>
Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit e7d316a0 ("sysctl: handle error writing UINT_MAX to u32
fields") added proc_douintvec() to start help adding support for
unsigned int, this however was only half the work needed.  Two fixes
have come in since then for the following issues:

  o Printing the values shows a negative value, this happens since
    do_proc_dointvec() and this uses proc_put_long()

This was fixed by commit 5380e564 ("sysctl: don't print negative
flag for proc_douintvec").

  o We can easily wrap around the int values: UINT_MAX is 4294967295, if
    we echo in 4294967295 + 1 we end up with 0, using 4294967295 + 2 we
    end up with 1.
  o We echo negative values in and they are accepted

This was fixed by commit 425fffd8 ("sysctl: report EINVAL if value
is larger than UINT_MAX for proc_douintvec").

It still also failed to be added to sysctl_check_table()...  instead of
adding it with the current implementation just provide a proper and
simplified unsigned int support without any array unsigned int support
with no negative support at all.

Historically sysctl proc helpers have supported arrays, due to the
complexity this adds though we've taken a step back to evaluate array
users to determine if its worth upkeeping for unsigned int.  An
evaluation using Coccinelle has been done to perform a grammatical
search to ask ourselves:

  o How many sysctl proc_dointvec() (int) users exist which likely
    should be moved over to proc_douintvec() (unsigned int) ?
	Answer: about 8
	- Of these how many are array users ?
		Answer: Probably only 1
  o How many sysctl array users exist ?
	Answer: about 12

This last question gives us an idea just how popular arrays: they are not.
Array support should probably just be kept for strings.

The identified uint ports are:

  drivers/infiniband/core/ucma.c - max_backlog
  drivers/infiniband/core/iwcm.c - default_backlog
  net/core/sysctl_net_core.c - rps_sock_flow_sysctl()
  net/netfilter/nf_conntrack_timestamp.c - nf_conntrack_timestamp -- bool
  net/netfilter/nf_conntrack_acct.c nf_conntrack_acct -- bool
  net/netfilter/nf_conntrack_ecache.c - nf_conntrack_events -- bool
  net/netfilter/nf_conntrack_helper.c - nf_conntrack_helper -- bool
  net/phonet/sysctl.c proc_local_port_range()

The only possible array users is proc_local_port_range() but it does not
seem worth it to add array support just for this given the range support
works just as well.  Unsigned int support should be desirable more for
when you *need* more than INT_MAX or using int min/max support then does
not suffice for your ranges.

If you forget and by mistake happen to register an unsigned int proc
entry with an array, the driver will fail and you will get something as
follows:

sysctl table check failed: debug/test_sysctl//uint_0002 array now allowed
CPU: 2 PID: 1342 Comm: modprobe Tainted: G        W   E <etc>
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS <etc>
Call Trace:
 dump_stack+0x63/0x81
 __register_sysctl_table+0x350/0x650
 ? kmem_cache_alloc_trace+0x107/0x240
 __register_sysctl_paths+0x1b3/0x1e0
 ? 0xffffffffc005f000
 register_sysctl_table+0x1f/0x30
 test_sysctl_init+0x10/0x1000 [test_sysctl]
 do_one_initcall+0x52/0x1a0
 ? kmem_cache_alloc_trace+0x107/0x240
 do_init_module+0x5f/0x200
 load_module+0x1867/0x1bd0
 ? __symbol_put+0x60/0x60
 SYSC_finit_module+0xdf/0x110
 SyS_finit_module+0xe/0x10
 entry_SYSCALL_64_fastpath+0x1e/0xad
RIP: 0033:0x7f042b22d119
<etc>

Fixes: e7d316a0 ("sysctl: handle error writing UINT_MAX to u32 fields")
Link: http://lkml.kernel.org/r/20170519033554.18592-5-mcgrof@kernel.orgSigned-off-by: NLuis R. Rodriguez <mcgrof@kernel.org>
Suggested-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Liping Zhang <zlpnobody@gmail.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The mode sysctl_writes_strict positional checks keep being copy and pasted
as we add new proc handlers.  Just add a helper to avoid code duplication.

Link: http://lkml.kernel.org/r/20170519033554.18592-4-mcgrof@kernel.orgSigned-off-by: NLuis R. Rodriguez <mcgrof@kernel.org>
Suggested-by: NKees Cook <keescook@chromium.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Document the different sysctl_writes_strict modes in code.

Link: http://lkml.kernel.org/r/20170519033554.18592-3-mcgrof@kernel.orgSigned-off-by: NLuis R. Rodriguez <mcgrof@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

do_proc_dointvec_jiffies_conv() uses LONG_MAX/HZ as the max value to
avoid overflow.  But actually the *valp is int type, so it still causes
overflow.

For example,

  echo 2147483647 > ./sys/net/ipv4/tcp_keepalive_time

Then,

  cat ./sys/net/ipv4/tcp_keepalive_time

The output is "-1", it is not expected.

Now use INT_MAX/HZ as the max value instead LONG_MAX/HZ to fix it.

Link: http://lkml.kernel.org/r/1490109532-9228-1-git-send-email-fgao@ikuai8.comSigned-off-by: NGao Feng <fgao@ikuai8.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

timer_migration sysctl acts as a boolean switch, so the allowed values
should be restricted to 0 and 1.

Add the necessary extra fields to the sysctl table entry to enforce that.

[ tglx: Rewrote changelog ]
Signed-off-by: NMyungho Jung <mhjungk@gmail.com>
Link: http://lkml.kernel.org/r/1492640690-3550-1-git-send-email-mhjungk@gmail.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

Currently, inputting the following command will succeed but actually the
value will be truncated:

  # echo 0x12ffffffff > /proc/sys/net/ipv4/tcp_notsent_lowat

This is not friendly to the user, so instead, we should report error
when the value is larger than UINT_MAX.

Fixes: e7d316a0 ("sysctl: handle error writing UINT_MAX to u32 fields")
Signed-off-by: NLiping Zhang <zlpnobody@gmail.com>
Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I saw some very confusing sysctl output on my system:
  # cat /proc/sys/net/core/xfrm_aevent_rseqth
  -2
  # cat /proc/sys/net/core/xfrm_aevent_etime
  -10
  # cat /proc/sys/net/ipv4/tcp_notsent_lowat
  -4294967295

Because we forget to set the *negp flag in proc_douintvec, so it will
become a garbage value.

Since the value related to proc_douintvec is always an unsigned integer,
so we can set *negp to false explictily to fix this issue.

Fixes: e7d316a0 ("sysctl: handle error writing UINT_MAX to u32 fields")
Signed-off-by: NLiping Zhang <zlpnobody@gmail.com>
Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We are going to split <linux/sched/coredump.h> out of <linux/sched.h>, which
will have to be picked up from other headers and a couple of .c files.

Create a trivial placeholder <linux/sched/coredump.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.

Include the new header in the files that are going to need it.
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NIngo Molnar <mingo@kernel.org>

We added the 'sched_rr_timeslice_ms' SCHED_RR tuning knob in this commit:

  ce0dbbbb ("sched/rt: Add a tuning knob to allow changing SCHED_RR timeslice")

... which name suggests to users that it's in milliseconds, while in reality
it's being set in milliseconds but the result is shown in jiffies.

This is obviously confusing when HZ is not 1000, it makes it appear like the
value set failed, such as HZ=100:

  root# echo 100 > /proc/sys/kernel/sched_rr_timeslice_ms
  root# cat /proc/sys/kernel/sched_rr_timeslice_ms
  10

Fix this to be milliseconds all around.
Signed-off-by: NShile Zhang <shile.zhang@nokia.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1485612049-20923-1-git-send-email-shile.zhang@nokia.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

We perform the conversion between kernel jiffies and ms only when
exporting kernel value to user space.

We need to do the opposite operation when value is written by user.

Only matters when HZ != 1000
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This was entirely automated, using the script by Al:

  PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*<asm/uaccess.h>'
  sed -i -e "s!$PATT!#include <linux/uaccess.h>!" \
        $(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h)

to do the replacement at the end of the merge window.
Requested-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I was amused to find "unsafe core_pattern" warning having these lines in
/etc/sysctl.conf:

	fs.suid_dumpable=2
	kernel.core_pattern=/core/core-%e-%p-%E
	kernel.core_uses_pid=0

Turns out kernel is formally right.  Default core_pattern is just "core",
which doesn't qualify for secure path while setting suid.dumpable.

Hint admins about solution, clarify sysctl names, delete unnecessary '\'
characters (string literals are concatenated regardless) and reformat for
easier grepping.

Link: http://lkml.kernel.org/r/20161029152124.GA1258@avx2Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Acked-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently, when tracepoint_printk is set (enabled by the "tp_printk" kernel
command line), it causes trace events to print via printk(). This is a very
dangerous operation, but is useful for debugging.

The issue is, it's seldom used, but it is always checked even if it's not
enabled by the kernel command line. Instead of having this feature called by
a branch against a variable, turn that variable into a static key, and this
will remove the test and jump.

To simplify things, the functions output_printk() and
trace_event_buffer_commit() were moved from trace_events.c to trace.c.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

For mostly historical reasons, the x86 oops dump shows the raw stack
values:

  ...
  [registers]
  Stack:
   ffff880079af7350 ffff880079905400 0000000000000000 ffffc900008f3ae0
   ffffffffa0196610 0000000000000001 00010000ffffffff 0000000087654321
   0000000000000002 0000000000000000 0000000000000000 0000000000000000
  Call Trace:
  ...

This seems to be an artifact from long ago, and probably isn't needed
anymore.  It generally just adds noise to the dump, and it can be
actively harmful because it leaks kernel addresses.

Linus says:

  "The stack dump actually goes back to forever, and it used to be
   useful back in 1992 or so. But it used to be useful mainly because
   stacks were simpler and we didn't have very good call traces anyway. I
   definitely remember having used them - I just do not remember having
   used them in the last ten+ years.

   Of course, it's still true that if you can trigger an oops, you've
   likely already lost the security game, but since the stack dump is so
   useless, let's aim to just remove it and make games like the above
   harder."

This also removes the related 'kstack=' cmdline option and the
'kstack_depth_to_print' sysctl.
Suggested-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/e83bd50df52d8fe88e94d2566426ae40d813bf8f.1477405374.git.jpoimboe@redhat.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

The last user of this tunable was removed in 2012 in commit:

  82958366 ("sched: Replace update_shares weight distribution with per-entity computation")

Delete it since its very existence confuses people.
Signed-off-by: NMatt Fleming <matt@codeblueprint.co.uk>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20161019141059.26408-1-matt@codeblueprint.co.ukSigned-off-by: NIngo Molnar <mingo@kernel.org>

CAI Qian <caiqian@redhat.com> pointed out that the semantics
of shared subtrees make it possible to create an exponentially
increasing number of mounts in a mount namespace.

    mkdir /tmp/1 /tmp/2
    mount --make-rshared /
    for i in $(seq 1 20) ; do mount --bind /tmp/1 /tmp/2 ; done

Will create create 2^20 or 1048576 mounts, which is a practical problem
as some people have managed to hit this by accident.

As such CVE-2016-6213 was assigned.

Ian Kent <raven@themaw.net> described the situation for autofs users
as follows:

> The number of mounts for direct mount maps is usually not very large because of
> the way they are implemented, large direct mount maps can have performance
> problems. There can be anywhere from a few (likely case a few hundred) to less
> than 10000, plus mounts that have been triggered and not yet expired.
>
> Indirect mounts have one autofs mount at the root plus the number of mounts that
> have been triggered and not yet expired.
>
> The number of autofs indirect map entries can range from a few to the common
> case of several thousand and in rare cases up to between 30000 and 50000. I've
> not heard of people with maps larger than 50000 entries.
>
> The larger the number of map entries the greater the possibility for a large
> number of active mounts so it's not hard to expect cases of a 1000 or somewhat
> more active mounts.

So I am setting the default number of mounts allowed per mount
namespace at 100,000.  This is more than enough for any use case I
know of, but small enough to quickly stop an exponential increase
in mounts.  Which should be perfect to catch misconfigurations and
malfunctioning programs.

For anyone who needs a higher limit this can be changed by writing
to the new /proc/sys/fs/mount-max sysctl.
Tested-by: NCAI Qian <caiqian@redhat.com>
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>

After 7e8e385a ("x86/compat: Remove sys32_vm86_warning"), this
function has become unused, so we can remove it as well.

Link: http://lkml.kernel.org/r/20160617142903.3070388-1-arnd@arndb.deSigned-off-by: NArnd Bergmann <arnd@arndb.de>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Propagate unsignedness for grand total of 149 bytes:

	$ ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux
	add/remove: 0/0 grow/shrink: 0/10 up/down: 0/-149 (-149)
	function                                     old     new   delta
	set_close_on_exec                             99      98      -1
	put_files_struct                             201     200      -1
	get_close_on_exec                             59      58      -1
	do_prlimit                                   498     497      -1
	do_execveat_common.isra                     1662    1661      -1
	__close_fd                                   178     173      -5
	do_dup2                                      219     204     -15
	seq_show                                     685     660     -25
	__alloc_fd                                   384     357     -27
	dup_fd                                       718     646     -72

It mostly comes from converting "unsigned int" to "long" for bit operations.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

We have scripts which write to certain fields on 3.18 kernels but this
seems to be failing on 4.4 kernels.  An entry which we write to here is
xfrm_aevent_rseqth which is u32.

  echo 4294967295  > /proc/sys/net/core/xfrm_aevent_rseqth

Commit 230633d1 ("kernel/sysctl.c: detect overflows when converting
to int") prevented writing to sysctl entries when integer overflow
occurs.  However, this does not apply to unsigned integers.

Heinrich suggested that we introduce a new option to handle 64 bit
limits and set min as 0 and max as UINT_MAX.  This might not work as it
leads to issues similar to __do_proc_doulongvec_minmax.  Alternatively,
we would need to change the datatype of the entry to 64 bit.

  static int __do_proc_doulongvec_minmax(void *data, struct ctl_table
  {
      i = (unsigned long *) data;   //This cast is causing to read beyond the size of data (u32)
      vleft = table->maxlen / sizeof(unsigned long); //vleft is 0 because maxlen is sizeof(u32) which is lesser than sizeof(unsigned long) on x86_64.

Introduce a new proc handler proc_douintvec.  Individual proc entries
will need to be updated to use the new handler.

[akpm@linux-foundation.org: coding-style fixes]
Fixes: 230633d1 ("kernel/sysctl.c:detect overflows when converting to int")
Link: http://lkml.kernel.org/r/1471479806-5252-1-git-send-email-subashab@codeaurora.orgSigned-off-by: NSubash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add a "printk.devkmsg" kernel command line parameter which controls how
userspace writes into /dev/kmsg.  It has three options:

 * ratelimit - ratelimit logging from userspace.
 * on  - unlimited logging from userspace
 * off - logging from userspace gets ignored

The default setting is to ratelimit the messages written to it.

This changes the kernel default setting of "on" to "ratelimit" and we do
that because we want to keep userspace spamming /dev/kmsg to sane
levels.  This is especially moot when a small kernel log buffer wraps
around and messages get lost.  So the ratelimiting setting should be a
sane setting where kernel messages should have a bit higher chance of
survival from all the spamming.

It additionally does not limit logging to /dev/kmsg while the system is
booting if we haven't disabled it on the command line.

Furthermore, we can control the logging from a lower priority sysctl
interface - kernel.printk_devkmsg.

That interface will succeed only if printk.devkmsg *hasn't* been
supplied on the command line.  If it has, then printk.devkmsg is a
one-time setting which remains for the duration of the system lifetime.
This "locking" of the setting is to prevent userspace from changing the
logging on us through sysctl(2).

This patch is based on previous patches from Linus and Steven.

[bp@suse.de: fixes]
  Link: http://lkml.kernel.org/r/20160719072344.GC25563@nazgul.tnic
Link: http://lkml.kernel.org/r/20160716061745.15795-3-bp@alien8.deSigned-off-by: NBorislav Petkov <bp@suse.de>
Cc: Dave Young <dyoung@redhat.com>
Cc: Franck Bui <fbui@suse.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

As reclaim is now per-node based, convert zone_reclaim to be
node_reclaim.  It is possible that a node will be reclaimed multiple
times if it has multiple zones but this is unavoidable without caching
all nodes traversed so far.  The documentation and interface to
userspace is the same from a configuration perspective and will will be
similar in behaviour unless the node-local allocation requests were also
limited to lower zones.

Link: http://lkml.kernel.org/r/1467970510-21195-24-git-send-email-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It is not always easy to determine the cause of an RCU stall just by
analysing the RCU stall messages, mainly when the problem is caused
by the indirect starvation of rcu threads. For example, when preempt_rcu
is not awakened due to the starvation of a timer softirq.

We have been hard coding panic() in the RCU stall functions for
some time while testing the kernel-rt. But this is not possible in
some scenarios, like when supporting customers.

This patch implements the sysctl kernel.panic_on_rcu_stall. If
set to 1, the system will panic() when an RCU stall takes place,
enabling the capture of a vmcore. The vmcore provides a way to analyze
all kernel/tasks states, helping out to point to the culprit and the
solution for the stall.

The kernel.panic_on_rcu_stall sysctl is disabled by default.

Changes from v1:
- Fixed a typo in the git log
- The if(sysctl_panic_on_rcu_stall) panic() is in a static function
- Fixed the CONFIG_TINY_RCU compilation issue
- The var sysctl_panic_on_rcu_stall is now __read_mostly

Cc: Jonathan Corbet <corbet@lwn.net>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Acked-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>
Reviewed-by: NArnaldo Carvalho de Melo <acme@kernel.org>
Tested-by: N"Luis Claudio R. Goncalves" <lgoncalv@redhat.com>
Signed-off-by: NDaniel Bristot de Oliveira <bristot@redhat.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

Provide /proc/sys/vm/stat_refresh to force an immediate update of
per-cpu into global vmstats: useful to avoid a sleep(2) or whatever
before checking counts when testing.  Originally added to work around a
bug which left counts stranded indefinitely on a cpu going idle (an
inaccuracy magnified when small below-batch numbers represent "huge"
amounts of memory), but I believe that bug is now fixed: nonetheless,
this is still a useful knob.

Its schedule_on_each_cpu() is probably too expensive just to fold into
reading /proc/meminfo itself: give this mode 0600 to prevent abuse.
Allow a write or a read to do the same: nothing to read, but "grep -h
Shmem /proc/sys/vm/stat_refresh /proc/meminfo" is convenient.  Oh, and
since global_page_state() itself is careful to disguise any underflow as
0, hack in an "Invalid argument" and pr_warn() if a counter is negative
after the refresh - this helped to fix a misaccounting of
NR_ISOLATED_FILE in my migration code.

But on recent kernels, I find that NR_ALLOC_BATCH and NR_PAGES_SCANNED
often go negative some of the time.  I have not yet worked out why, but
have no evidence that it's actually harmful.  Punt for the moment by
just ignoring the anomaly on those.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andres Lagar-Cavilla <andreslc@google.com>
Cc: Yang Shi <yang.shi@linaro.org>
Cc: Ning Qu <quning@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Andres Lagar-Cavilla <andreslc@google.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The perf_sample->ip_callchain->nr value includes all the entries in the
ip_callchain->ip[] array, real addresses and PERF_CONTEXT_{KERNEL,USER,etc},
while what the user expects is that what is in the kernel.perf_event_max_stack
sysctl or in the upcoming per event perf_event_attr.sample_max_stack knob be
honoured in terms of IP addresses in the stack trace.

So allocate a bunch of extra entries for contexts, and do the accounting
via perf_callchain_entry_ctx struct members.

A new sysctl, kernel.perf_event_max_contexts_per_stack is also
introduced for investigating possible bugs in the callchain
implementation by some arch.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zefan Li <lizefan@huawei.com>
Link: http://lkml.kernel.org/n/tip-3b4wnqk340c4sg4gwkfdi9yk@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

So that it can be used for other stack related knobs, such as the
upcoming one to tweak the max number of of contexts per stack sample.

In all those cases we can only change the value if there are no perf
sessions collecting stacks, so they need to grab that mutex, etc.

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-8t3fk94wuzp8m2z1n4gc0s17@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

The default remains 127, which is good for most cases, and not even hit
most of the time, but then for some cases, as reported by Brendan, 1024+
deep frames are appearing on the radar for things like groovy, ruby.

And in some workloads putting a _lower_ cap on this may make sense. One
that is per event still needs to be put in place tho.

The new file is:

  # cat /proc/sys/kernel/perf_event_max_stack
  127

Chaging it:

  # echo 256 > /proc/sys/kernel/perf_event_max_stack
  # cat /proc/sys/kernel/perf_event_max_stack
  256

But as soon as there is some event using callchains we get:

  # echo 512 > /proc/sys/kernel/perf_event_max_stack
  -bash: echo: write error: Device or resource busy
  #

Because we only allocate the callchain percpu data structures when there
is a user, which allows for changing the max easily, its just a matter
of having no callchain users at that point.
Reported-and-Tested-by: NBrendan Gregg <brendan.d.gregg@gmail.com>
Reviewed-by: NFrederic Weisbecker <fweisbec@gmail.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDavid Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zefan Li <lizefan@huawei.com>
Link: http://lkml.kernel.org/r/20160426002928.GB16708@kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

In machines with 140G of memory and enterprise flash storage, we have
seen read and write bursts routinely exceed the kswapd watermarks and
cause thundering herds in direct reclaim.  Unfortunately, the only way
to tune kswapd aggressiveness is through adjusting min_free_kbytes - the
system's emergency reserves - which is entirely unrelated to the
system's latency requirements.  In order to get kswapd to maintain a
250M buffer of free memory, the emergency reserves need to be set to 1G.
That is a lot of memory wasted for no good reason.

On the other hand, it's reasonable to assume that allocation bursts and
overall allocation concurrency scale with memory capacity, so it makes
sense to make kswapd aggressiveness a function of that as well.

Change the kswapd watermark scale factor from the currently fixed 25% of
the tunable emergency reserve to a tunable 0.1% of memory.

Beyond 1G of memory, this will produce bigger watermark steps than the
current formula in default settings.  Ensure that the new formula never
chooses steps smaller than that, i.e.  25% of the emergency reserve.

On a 140G machine, this raises the default watermark steps - the
distance between min and low, and low and high - from 16M to 143M.
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMel Gorman <mgorman@suse.de>
Acked-by: NRik van Riel <riel@redhat.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

schedstats is very useful during debugging and performance tuning but it
incurs overhead to calculate the stats. As such, even though it can be
disabled at build time, it is often enabled as the information is useful.

This patch adds a kernel command-line and sysctl tunable to enable or
disable schedstats on demand (when it's built in). It is disabled
by default as someone who knows they need it can also learn to enable
it when necessary.

The benefits are dependent on how scheduler-intensive the workload is.
If it is then the patch reduces the number of cycles spent calculating
the stats with a small benefit from reducing the cache footprint of the
scheduler.

These measurements were taken from a 48-core 2-socket
machine with Xeon(R) E5-2670 v3 cpus although they were also tested on a
single socket machine 8-core machine with Intel i7-3770 processors.

netperf-tcp
                           4.5.0-rc1             4.5.0-rc1
                             vanilla          nostats-v3r1
Hmean    64         560.45 (  0.00%)      575.98 (  2.77%)
Hmean    128        766.66 (  0.00%)      795.79 (  3.80%)
Hmean    256        950.51 (  0.00%)      981.50 (  3.26%)
Hmean    1024      1433.25 (  0.00%)     1466.51 (  2.32%)
Hmean    2048      2810.54 (  0.00%)     2879.75 (  2.46%)
Hmean    3312      4618.18 (  0.00%)     4682.09 (  1.38%)
Hmean    4096      5306.42 (  0.00%)     5346.39 (  0.75%)
Hmean    8192     10581.44 (  0.00%)    10698.15 (  1.10%)
Hmean    16384    18857.70 (  0.00%)    18937.61 (  0.42%)

Small gains here, UDP_STREAM showed nothing intresting and neither did
the TCP_RR tests. The gains on the 8-core machine were very similar.

tbench4
                                 4.5.0-rc1             4.5.0-rc1
                                   vanilla          nostats-v3r1
Hmean    mb/sec-1         500.85 (  0.00%)      522.43 (  4.31%)
Hmean    mb/sec-2         984.66 (  0.00%)     1018.19 (  3.41%)
Hmean    mb/sec-4        1827.91 (  0.00%)     1847.78 (  1.09%)
Hmean    mb/sec-8        3561.36 (  0.00%)     3611.28 (  1.40%)
Hmean    mb/sec-16       5824.52 (  0.00%)     5929.03 (  1.79%)
Hmean    mb/sec-32      10943.10 (  0.00%)    10802.83 ( -1.28%)
Hmean    mb/sec-64      15950.81 (  0.00%)    16211.31 (  1.63%)
Hmean    mb/sec-128     15302.17 (  0.00%)    15445.11 (  0.93%)
Hmean    mb/sec-256     14866.18 (  0.00%)    15088.73 (  1.50%)
Hmean    mb/sec-512     15223.31 (  0.00%)    15373.69 (  0.99%)
Hmean    mb/sec-1024    14574.25 (  0.00%)    14598.02 (  0.16%)
Hmean    mb/sec-2048    13569.02 (  0.00%)    13733.86 (  1.21%)
Hmean    mb/sec-3072    12865.98 (  0.00%)    13209.23 (  2.67%)

Small gains of 2-4% at low thread counts and otherwise flat.  The
gains on the 8-core machine were slightly different

tbench4 on 8-core i7-3770 single socket machine
Hmean    mb/sec-1        442.59 (  0.00%)      448.73 (  1.39%)
Hmean    mb/sec-2        796.68 (  0.00%)      794.39 ( -0.29%)
Hmean    mb/sec-4       1322.52 (  0.00%)     1343.66 (  1.60%)
Hmean    mb/sec-8       2611.65 (  0.00%)     2694.86 (  3.19%)
Hmean    mb/sec-16      2537.07 (  0.00%)     2609.34 (  2.85%)
Hmean    mb/sec-32      2506.02 (  0.00%)     2578.18 (  2.88%)
Hmean    mb/sec-64      2511.06 (  0.00%)     2569.16 (  2.31%)
Hmean    mb/sec-128     2313.38 (  0.00%)     2395.50 (  3.55%)
Hmean    mb/sec-256     2110.04 (  0.00%)     2177.45 (  3.19%)
Hmean    mb/sec-512     2072.51 (  0.00%)     2053.97 ( -0.89%)

In constract, this shows a relatively steady 2-3% gain at higher thread
counts. Due to the nature of the patch and the type of workload, it's
not a surprise that the result will depend on the CPU used.

hackbench-pipes
                         4.5.0-rc1             4.5.0-rc1
                           vanilla          nostats-v3r1
Amean    1        0.0637 (  0.00%)      0.0660 ( -3.59%)
Amean    4        0.1229 (  0.00%)      0.1181 (  3.84%)
Amean    7        0.1921 (  0.00%)      0.1911 (  0.52%)
Amean    12       0.3117 (  0.00%)      0.2923 (  6.23%)
Amean    21       0.4050 (  0.00%)      0.3899 (  3.74%)
Amean    30       0.4586 (  0.00%)      0.4433 (  3.33%)
Amean    48       0.5910 (  0.00%)      0.5694 (  3.65%)
Amean    79       0.8663 (  0.00%)      0.8626 (  0.43%)
Amean    110      1.1543 (  0.00%)      1.1517 (  0.22%)
Amean    141      1.4457 (  0.00%)      1.4290 (  1.16%)
Amean    172      1.7090 (  0.00%)      1.6924 (  0.97%)
Amean    192      1.9126 (  0.00%)      1.9089 (  0.19%)

Some small gains and losses and while the variance data is not included,
it's close to the noise. The UMA machine did not show anything particularly
different

pipetest
                             4.5.0-rc1             4.5.0-rc1
                               vanilla          nostats-v2r2
Min         Time        4.13 (  0.00%)        3.99 (  3.39%)
1st-qrtle   Time        4.38 (  0.00%)        4.27 (  2.51%)
2nd-qrtle   Time        4.46 (  0.00%)        4.39 (  1.57%)
3rd-qrtle   Time        4.56 (  0.00%)        4.51 (  1.10%)
Max-90%     Time        4.67 (  0.00%)        4.60 (  1.50%)
Max-93%     Time        4.71 (  0.00%)        4.65 (  1.27%)
Max-95%     Time        4.74 (  0.00%)        4.71 (  0.63%)
Max-99%     Time        4.88 (  0.00%)        4.79 (  1.84%)
Max         Time        4.93 (  0.00%)        4.83 (  2.03%)
Mean        Time        4.48 (  0.00%)        4.39 (  1.91%)
Best99%Mean Time        4.47 (  0.00%)        4.39 (  1.91%)
Best95%Mean Time        4.46 (  0.00%)        4.38 (  1.93%)
Best90%Mean Time        4.45 (  0.00%)        4.36 (  1.98%)
Best50%Mean Time        4.36 (  0.00%)        4.25 (  2.49%)
Best10%Mean Time        4.23 (  0.00%)        4.10 (  3.13%)
Best5%Mean  Time        4.19 (  0.00%)        4.06 (  3.20%)
Best1%Mean  Time        4.13 (  0.00%)        4.00 (  3.39%)

Small improvement and similar gains were seen on the UMA machine.

The gain is small but it stands to reason that doing less work in the
scheduler is a good thing. The downside is that the lack of schedstats and
tracepoints may be surprising to experts doing performance analysis until
they find the existence of the schedstats= parameter or schedstats sysctl.
It will be automatically activated for latencytop and sleep profiling to
alleviate the problem. For tracepoints, there is a simple warning as it's
not safe to activate schedstats in the context when it's known the tracepoint
may be wanted but is unavailable.
Signed-off-by: NMel Gorman <mgorman@techsingularity.net>
Reviewed-by: NMatt Fleming <matt@codeblueprint.co.uk>
Reviewed-by: NSrikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <mgalbraith@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1454663316-22048-1-git-send-email-mgorman@techsingularity.netSigned-off-by: NIngo Molnar <mingo@kernel.org>

SYSCTL_WRITES_WARN was added in commit f4aacea2 ("sysctl: allow for
strict write position handling"), and released in v3.16 in August of
2014.  Since then I can find only 1 instance of non-zero offset
writing[1], and it was fixed immediately in CRIU[2].  As such, it
appears safe to flip this to the strict state now.

[1] https://www.google.com/search?q="when%20file%20position%20was%20not%200"
[2] http://lists.openvz.org/pipermail/criu/2015-April/019819.htmlSigned-off-by: NKees Cook <keescook@chromium.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>