linux-stable.git/kernel/sched, branch v3.16.50 Linux kernel stable tree perf: Avoid horrible stack usage 2017-11-11T13:34:02+00:00 Peter Zijlstra (Intel) peterz@infradead.org 2014-12-16T11:47:34+00:00 17b9e402f8864261628403985e0dfad96af9e459 commit 86038c5ea81b519a8a1fcfcd5e4599aab0cdd119 upstream. Both Linus (most recent) and Steve (a while ago) reported that perf related callbacks have massive stack bloat. The problem is that software events need a pt_regs in order to properly report the event location and unwind stack. And because we could not assume one was present we allocated one on stack and filled it with minimal bits required for operation. Now, pt_regs is quite large, so this is undesirable. Furthermore it turns out that most sites actually have a pt_regs pointer available, making this even more onerous, as the stack space is pointless waste. This patch addresses the problem by observing that software events have well defined nesting semantics, therefore we can use static per-cpu storage instead of on-stack. Linus made the further observation that all but the scheduler callers of perf_sw_event() have a pt_regs available, so we change the regular perf_sw_event() to require a valid pt_regs (where it used to be optional) and add perf_sw_event_sched() for the scheduler. We have a scheduler specific call instead of a more generic _noregs() like construct because we can assume non-recursion from the scheduler and thereby simplify the code further (_noregs would have to put the recursion context call inline in order to assertain which __perf_regs element to use). One last note on the implementation of perf_trace_buf_prepare(); we allow .regs = NULL for those cases where we already have a pt_regs pointer available and do not need another. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Reported-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Javi Merino <javi.merino@arm.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Petr Mladek <pmladek@suse.cz> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/20141216115041.GW3337@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 86038c5ea81b519a8a1fcfcd5e4599aab0cdd119 upstream.

Both Linus (most recent) and Steve (a while ago) reported that perf
related callbacks have massive stack bloat.

The problem is that software events need a pt_regs in order to
properly report the event location and unwind stack. And because we
could not assume one was present we allocated one on stack and filled
it with minimal bits required for operation.

Now, pt_regs is quite large, so this is undesirable. Furthermore it
turns out that most sites actually have a pt_regs pointer available,
making this even more onerous, as the stack space is pointless waste.

This patch addresses the problem by observing that software events
have well defined nesting semantics, therefore we can use static
per-cpu storage instead of on-stack.

Linus made the further observation that all but the scheduler callers
of perf_sw_event() have a pt_regs available, so we change the regular
perf_sw_event() to require a valid pt_regs (where it used to be
optional) and add perf_sw_event_sched() for the scheduler.

We have a scheduler specific call instead of a more generic _noregs()
like construct because we can assume non-recursion from the scheduler
and thereby simplify the code further (_noregs would have to put the
recursion context call inline in order to assertain which __perf_regs
element to use).

One last note on the implementation of perf_trace_buf_prepare(); we
allow .regs = NULL for those cases where we already have a pt_regs
pointer available and do not need another.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Javi Merino <javi.merino@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Petr Mladek <pmladek@suse.cz>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Cc: Vaibhav Nagarnaik <vnagarnaik@google.com>
Link: http://lkml.kernel.org/r/20141216115041.GW3337@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/topology: Fix overlapping sched_group_capacity 2017-10-12T14:27:43+00:00 Peter Zijlstra peterz@infradead.org 2017-04-25T12:31:11+00:00 5a7aba3f27f7262cc5fad1b85804ea0ad9f4275c commit 1676330ecfa840113a37b25a49afda068380d19c upstream. When building the overlapping groups we need to attach a consistent sched_group_capacity structure. That is, all 'identical' sched_group's should have the _same_ sched_group_capacity. This can (once again) be demonstrated with a topology like: node 0 1 2 3 0: 10 20 30 20 1: 20 10 20 30 2: 30 20 10 20 3: 20 30 20 10 But we need at least 2 CPUs per node for this to show up, after all, if there is only one CPU per node, our CPU @i is per definition a unique CPU that reaches this domain (aka balance-cpu). Given the above NUMA topo and 2 CPUs per node: [] CPU0 attaching sched-domain(s): [] domain-0: span=0,4 level=DIE [] groups: 0:{ span=0 }, 4:{ span=4 } [] domain-1: span=0-1,3-5,7 level=NUMA [] groups: 0:{ span=0,4 mask=0,4 cap=2048 }, 1:{ span=1,5 mask=1,5 cap=2048 }, 3:{ span=3,7 mask=3,7 cap=2048 } [] domain-2: span=0-7 level=NUMA [] groups: 0:{ span=0-1,3-5,7 mask=0,4 cap=6144 }, 2:{ span=1-3,5-7 mask=2,6 cap=6144 } [] CPU1 attaching sched-domain(s): [] domain-0: span=1,5 level=DIE [] groups: 1:{ span=1 }, 5:{ span=5 } [] domain-1: span=0-2,4-6 level=NUMA [] groups: 1:{ span=1,5 mask=1,5 cap=2048 }, 2:{ span=2,6 mask=2,6 cap=2048 }, 4:{ span=0,4 mask=0,4 cap=2048 } [] domain-2: span=0-7 level=NUMA [] groups: 1:{ span=0-2,4-6 mask=1,5 cap=6144 }, 3:{ span=0,2-4,6-7 mask=3,7 cap=6144 } Observe how CPU0-domain1-group0 and CPU1-domain1-group4 are the 'same' but have a different id (0 vs 4). To fix this, use the group balance CPU to select the SGC. This means we have to compute the full mask for each CPU and require a second temporary mask to store the group mask in (it otherwise lives in the SGC). The fixed topology looks like: [] CPU0 attaching sched-domain(s): [] domain-0: span=0,4 level=DIE [] groups: 0:{ span=0 }, 4:{ span=4 } [] domain-1: span=0-1,3-5,7 level=NUMA [] groups: 0:{ span=0,4 mask=0,4 cap=2048 }, 1:{ span=1,5 mask=1,5 cap=2048 }, 3:{ span=3,7 mask=3,7 cap=2048 } [] domain-2: span=0-7 level=NUMA [] groups: 0:{ span=0-1,3-5,7 mask=0,4 cap=6144 }, 2:{ span=1-3,5-7 mask=2,6 cap=6144 } [] CPU1 attaching sched-domain(s): [] domain-0: span=1,5 level=DIE [] groups: 1:{ span=1 }, 5:{ span=5 } [] domain-1: span=0-2,4-6 level=NUMA [] groups: 1:{ span=1,5 mask=1,5 cap=2048 }, 2:{ span=2,6 mask=2,6 cap=2048 }, 0:{ span=0,4 mask=0,4 cap=2048 } [] domain-2: span=0-7 level=NUMA [] groups: 1:{ span=0-2,4-6 mask=1,5 cap=6144 }, 3:{ span=0,2-4,6-7 mask=3,7 cap=6144 } Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com> Signed-off-by: Peter 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> Cc: linux-kernel@vger.kernel.org Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans") Signed-off-by: Ingo Molnar <mingo@kernel.org> [bwh: Backported to 3.16: adjust filename, context] Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 1676330ecfa840113a37b25a49afda068380d19c upstream.

When building the overlapping groups we need to attach a consistent
sched_group_capacity structure. That is, all 'identical' sched_group's
should have the _same_ sched_group_capacity.

This can (once again) be demonstrated with a topology like:

  node   0   1   2   3
    0:  10  20  30  20
    1:  20  10  20  30
    2:  30  20  10  20
    3:  20  30  20  10

But we need at least 2 CPUs per node for this to show up, after all,
if there is only one CPU per node, our CPU @i is per definition a
unique CPU that reaches this domain (aka balance-cpu).

Given the above NUMA topo and 2 CPUs per node:

  [] CPU0 attaching sched-domain(s):
  []  domain-0: span=0,4 level=DIE
  []   groups: 0:{ span=0 }, 4:{ span=4 }
  []   domain-1: span=0-1,3-5,7 level=NUMA
  []    groups: 0:{ span=0,4 mask=0,4 cap=2048 }, 1:{ span=1,5 mask=1,5 cap=2048 }, 3:{ span=3,7 mask=3,7 cap=2048 }
  []    domain-2: span=0-7 level=NUMA
  []     groups: 0:{ span=0-1,3-5,7 mask=0,4 cap=6144 }, 2:{ span=1-3,5-7 mask=2,6 cap=6144 }
  [] CPU1 attaching sched-domain(s):
  []  domain-0: span=1,5 level=DIE
  []   groups: 1:{ span=1 }, 5:{ span=5 }
  []   domain-1: span=0-2,4-6 level=NUMA
  []    groups: 1:{ span=1,5 mask=1,5 cap=2048 }, 2:{ span=2,6 mask=2,6 cap=2048 }, 4:{ span=0,4 mask=0,4 cap=2048 }
  []    domain-2: span=0-7 level=NUMA
  []     groups: 1:{ span=0-2,4-6 mask=1,5 cap=6144 }, 3:{ span=0,2-4,6-7 mask=3,7 cap=6144 }

Observe how CPU0-domain1-group0 and CPU1-domain1-group4 are the
'same' but have a different id (0 vs 4).

To fix this, use the group balance CPU to select the SGC. This means
we have to compute the full mask for each CPU and require a second
temporary mask to store the group mask in (it otherwise lives in the
SGC).

The fixed topology looks like:

  [] CPU0 attaching sched-domain(s):
  []  domain-0: span=0,4 level=DIE
  []   groups: 0:{ span=0 }, 4:{ span=4 }
  []   domain-1: span=0-1,3-5,7 level=NUMA
  []    groups: 0:{ span=0,4 mask=0,4 cap=2048 }, 1:{ span=1,5 mask=1,5 cap=2048 }, 3:{ span=3,7 mask=3,7 cap=2048 }
  []    domain-2: span=0-7 level=NUMA
  []     groups: 0:{ span=0-1,3-5,7 mask=0,4 cap=6144 }, 2:{ span=1-3,5-7 mask=2,6 cap=6144 }
  [] CPU1 attaching sched-domain(s):
  []  domain-0: span=1,5 level=DIE
  []   groups: 1:{ span=1 }, 5:{ span=5 }
  []   domain-1: span=0-2,4-6 level=NUMA
  []    groups: 1:{ span=1,5 mask=1,5 cap=2048 }, 2:{ span=2,6 mask=2,6 cap=2048 }, 0:{ span=0,4 mask=0,4 cap=2048 }
  []    domain-2: span=0-7 level=NUMA
  []     groups: 1:{ span=0-2,4-6 mask=1,5 cap=6144 }, 3:{ span=0,2-4,6-7 mask=3,7 cap=6144 }

Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com>
Signed-off-by: Peter 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>
Cc: linux-kernel@vger.kernel.org
Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[bwh: Backported to 3.16: adjust filename, context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/topology: Fix overlapping sched_group_mask 2017-10-12T14:27:42+00:00 Peter Zijlstra peterz@infradead.org 2017-04-25T12:00:49+00:00 00c978eada13e2cd1bc7da485e99ab5fb7c3418c commit 73bb059f9b8a00c5e1bf2f7ca83138c05d05e600 upstream. The point of sched_group_mask is to select those CPUs from sched_group_cpus that can actually arrive at this balance domain. The current code gets it wrong, as can be readily demonstrated with a topology like: node 0 1 2 3 0: 10 20 30 20 1: 20 10 20 30 2: 30 20 10 20 3: 20 30 20 10 Where (for example) domain 1 on CPU1 ends up with a mask that includes CPU0: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 (mask: 1), 2, 0 [] domain 1: span 0-3 level NUMA [] groups: 0-2 (mask: 0-2) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072) This causes sched_balance_cpu() to compute the wrong CPU and consequently should_we_balance() will terminate early resulting in missed load-balance opportunities. The fixed topology looks like: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 (mask: 1), 2, 0 [] domain 1: span 0-3 level NUMA [] groups: 0-2 (mask: 1) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072) (note: this relies on OVERLAP domains to always have children, this is true because the regular topology domains are still here -- this is before degenerate trimming) Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com> Signed-off-by: Peter 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> Cc: linux-kernel@vger.kernel.org Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans") Signed-off-by: Ingo Molnar <mingo@kernel.org> [bwh: Backported to 3.16: - Use span, not sg_span - Adjust filename context] Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 73bb059f9b8a00c5e1bf2f7ca83138c05d05e600 upstream.

The point of sched_group_mask is to select those CPUs from
sched_group_cpus that can actually arrive at this balance domain.

The current code gets it wrong, as can be readily demonstrated with a
topology like:

  node   0   1   2   3
    0:  10  20  30  20
    1:  20  10  20  30
    2:  30  20  10  20
    3:  20  30  20  10

Where (for example) domain 1 on CPU1 ends up with a mask that includes
CPU0:

  [] CPU1 attaching sched-domain:
  []  domain 0: span 0-2 level NUMA
  []   groups: 1 (mask: 1), 2, 0
  []   domain 1: span 0-3 level NUMA
  []    groups: 0-2 (mask: 0-2) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072)

This causes sched_balance_cpu() to compute the wrong CPU and
consequently should_we_balance() will terminate early resulting in
missed load-balance opportunities.

The fixed topology looks like:

  [] CPU1 attaching sched-domain:
  []  domain 0: span 0-2 level NUMA
  []   groups: 1 (mask: 1), 2, 0
  []   domain 1: span 0-3 level NUMA
  []    groups: 0-2 (mask: 1) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072)

(note: this relies on OVERLAP domains to always have children, this is
 true because the regular topology domains are still here -- this is
 before degenerate trimming)

Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com>
Signed-off-by: Peter 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>
Cc: linux-kernel@vger.kernel.org
Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[bwh: Backported to 3.16:
 - Use span, not sg_span
 - Adjust filename context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/topology: Fix building of overlapping sched-groups 2017-10-12T14:27:42+00:00 Peter Zijlstra peterz@infradead.org 2017-04-14T15:24:02+00:00 49d4283d847987fccbd7c7ce8f59f0fd765702ff commit 0372dd2736e02672ac6e189c31f7d8c02ad543cd upstream. When building the overlapping groups, we very obviously should start with the previous domain of _this_ @cpu, not CPU-0. This can be readily demonstrated with a topology like: node 0 1 2 3 0: 10 20 30 20 1: 20 10 20 30 2: 30 20 10 20 3: 20 30 20 10 Where (for example) CPU1 ends up generating the following nonsensical groups: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 2 0 [] domain 1: span 0-3 level NUMA [] groups: 1-3 (cpu_capacity = 3072) 0-1,3 (cpu_capacity = 3072) Where the fact that domain 1 doesn't include a group with span 0-2 is the obvious fail. With patch this looks like: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 0 2 [] domain 1: span 0-3 level NUMA [] groups: 0-2 (cpu_capacity = 3072) 0,2-3 (cpu_capacity = 3072) Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com> Signed-off-by: Peter 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> Cc: linux-kernel@vger.kernel.org Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans") Signed-off-by: Ingo Molnar <mingo@kernel.org> [bwh: Backported to 3.16: adjust filename] Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 0372dd2736e02672ac6e189c31f7d8c02ad543cd upstream.

When building the overlapping groups, we very obviously should start
with the previous domain of _this_ @cpu, not CPU-0.

This can be readily demonstrated with a topology like:

  node   0   1   2   3
    0:  10  20  30  20
    1:  20  10  20  30
    2:  30  20  10  20
    3:  20  30  20  10

Where (for example) CPU1 ends up generating the following nonsensical groups:

  [] CPU1 attaching sched-domain:
  []  domain 0: span 0-2 level NUMA
  []   groups: 1 2 0
  []   domain 1: span 0-3 level NUMA
  []    groups: 1-3 (cpu_capacity = 3072) 0-1,3 (cpu_capacity = 3072)

Where the fact that domain 1 doesn't include a group with span 0-2 is
the obvious fail.

With patch this looks like:

  [] CPU1 attaching sched-domain:
  []  domain 0: span 0-2 level NUMA
  []   groups: 1 0 2
  []   domain 1: span 0-3 level NUMA
  []    groups: 0-2 (cpu_capacity = 3072) 0,2-3 (cpu_capacity = 3072)

Debugged-by: Lauro Ramos Venancio <lvenanci@redhat.com>
Signed-off-by: Peter 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>
Cc: linux-kernel@vger.kernel.org
Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[bwh: Backported to 3.16: adjust filename]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/topology: Refactor function build_overlap_sched_groups() 2017-10-12T14:27:42+00:00 Lauro Ramos Venancio lvenanci@redhat.com 2017-04-13T13:56:07+00:00 07a1b64820763f257ba5099bd691c22f1e9a15b4 commit 8c0334697dc37eb3d6d7632304d3a3662248daac upstream. Create functions build_group_from_child_sched_domain() and init_overlap_sched_group(). No functional change. Signed-off-by: Lauro Ramos Venancio <lvenanci@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> 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/1492091769-19879-2-git-send-email-lvenanci@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org> [bwh: Backported to 3.16: - Ccode being moved is slightly different - Adjust filename] Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 8c0334697dc37eb3d6d7632304d3a3662248daac upstream.

Create functions build_group_from_child_sched_domain() and
init_overlap_sched_group(). No functional change.

Signed-off-by: Lauro Ramos Venancio <lvenanci@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Rik van Riel <riel@redhat.com>
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/1492091769-19879-2-git-send-email-lvenanci@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[bwh: Backported to 3.16:
 - Ccode being moved is slightly different
 - Adjust filename]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched: Rename a misleading variable in build_overlap_sched_groups() 2017-10-12T14:27:42+00:00 Zhihui Zhang zzhsuny@gmail.com 2014-08-02T01:18:03+00:00 7b565013a51c217a648c067d12f7a17186f172f9 commit aaecac4ad46b35ad308245384d019633fb9bc21b upstream. The child variable in build_overlap_sched_groups() actually refers to the peer or sibling domain of the given CPU. Rename it to sibling to be consistent with the naming in build_group_mask(). Signed-off-by: Zhihui Zhang <zzhsuny@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1406942283-18249-1-git-send-email-zzhsuny@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit aaecac4ad46b35ad308245384d019633fb9bc21b upstream.

The child variable in build_overlap_sched_groups() actually refers to the
peer or sibling domain of the given CPU. Rename it to sibling to be consistent
with the naming in build_group_mask().

Signed-off-by: Zhihui Zhang <zzhsuny@gmail.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/1406942283-18249-1-git-send-email-zzhsuny@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/loadavg: Avoid loadavg spikes caused by delayed NO_HZ accounting 2017-07-18T17:40:15+00:00 Matt Fleming matt@codeblueprint.co.uk 2017-02-17T12:07:30+00:00 b6695dba0c1bd720bb1d22b93dc07769db05ee7d commit 6e5f32f7a43f45ee55c401c0b9585eb01f9629a8 upstream. If we crossed a sample window while in NO_HZ we will add LOAD_FREQ to the pending sample window time on exit, setting the next update not one window into the future, but two. This situation on exiting NO_HZ is described by: this_rq->calc_load_update < jiffies < calc_load_update In this scenario, what we should be doing is: this_rq->calc_load_update = calc_load_update [ next window ] But what we actually do is: this_rq->calc_load_update = calc_load_update + LOAD_FREQ [ next+1 window ] This has the effect of delaying load average updates for potentially up to ~9seconds. This can result in huge spikes in the load average values due to per-cpu uninterruptible task counts being out of sync when accumulated across all CPUs. It's safe to update the per-cpu active count if we wake between sample windows because any load that we left in 'calc_load_idle' will have been zero'd when the idle load was folded in calc_global_load(). This issue is easy to reproduce before, commit 9d89c257dfb9 ("sched/fair: Rewrite runnable load and utilization average tracking") just by forking short-lived process pipelines built from ps(1) and grep(1) in a loop. I'm unable to reproduce the spikes after that commit, but the bug still seems to be present from code review. Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Mike Galbraith <umgwanakikbuti@gmail.com> Cc: Morten Rasmussen <morten.rasmussen@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vincent Guittot <vincent.guittot@linaro.org> Fixes: commit 5167e8d ("sched/nohz: Rewrite and fix load-avg computation -- again") Link: http://lkml.kernel.org/r/20170217120731.11868-2-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar <mingo@kernel.org> [bwh: Backported to 3.16: adjust filename] Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 6e5f32f7a43f45ee55c401c0b9585eb01f9629a8 upstream.

If we crossed a sample window while in NO_HZ we will add LOAD_FREQ to
the pending sample window time on exit, setting the next update not
one window into the future, but two.

This situation on exiting NO_HZ is described by:

  this_rq->calc_load_update < jiffies < calc_load_update

In this scenario, what we should be doing is:

  this_rq->calc_load_update = calc_load_update		     [ next window ]

But what we actually do is:

  this_rq->calc_load_update = calc_load_update + LOAD_FREQ   [ next+1 window ]

This has the effect of delaying load average updates for potentially
up to ~9seconds.

This can result in huge spikes in the load average values due to
per-cpu uninterruptible task counts being out of sync when accumulated
across all CPUs.

It's safe to update the per-cpu active count if we wake between sample
windows because any load that we left in 'calc_load_idle' will have
been zero'd when the idle load was folded in calc_global_load().

This issue is easy to reproduce before,

  commit 9d89c257dfb9 ("sched/fair: Rewrite runnable load and utilization average tracking")

just by forking short-lived process pipelines built from ps(1) and
grep(1) in a loop. I'm unable to reproduce the spikes after that
commit, but the bug still seems to be present from code review.

Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Fixes: commit 5167e8d ("sched/nohz: Rewrite and fix load-avg computation -- again")
Link: http://lkml.kernel.org/r/20170217120731.11868-2-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[bwh: Backported to 3.16: adjust filename]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
locking/static_keys: Add static_key_{en,dis}able() helpers 2017-06-05T20:17:18+00:00 Peter Zijlstra peterz@infradead.org 2015-07-24T13:03:40+00:00 0331b8df8465955329d55a419ddbe4532be53de6 commit e33886b38cc82a9fc3b2d655dfc7f50467594138 upstream. Add two helpers to make it easier to treat the refcount as boolean. Suggested-by: Jason Baron <jasonbaron0@gmail.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit e33886b38cc82a9fc3b2d655dfc7f50467594138 upstream.

Add two helpers to make it easier to treat the refcount as boolean.

Suggested-by: Jason Baron <jasonbaron0@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/cputime: Fix invalid gtime in proc 2017-03-16T02:27:15+00:00 Hiroshi Shimamoto h-shimamoto@ct.jp.nec.com 2015-11-19T15:47:28+00:00 0c8a7400036ee86884375ac8856a3d6addf11361 commit 2541117b0cf79977fa11a0d6e17d61010677bd7b upstream. /proc/stats shows invalid gtime when the thread is running in guest. When vtime accounting is not enabled, we cannot get a valid delta. The delta is calculated with now - tsk->vtime_snap, but tsk->vtime_snap is only updated when vtime accounting is runtime enabled. This patch makes task_gtime() just return gtime without computing the buggy non-existing tickless delta when vtime accounting is not enabled. Use context_tracking_is_enabled() to check if vtime is accounting on some cpu, in which case only we need to check the tickless delta. This way we fix the gtime value regression on machines not running nohz full. The kernel config contains CONFIG_VIRT_CPU_ACCOUNTING_GEN=y and CONFIG_NO_HZ_FULL_ALL=n and boot without nohz_full. I ran and stop a busy loop in VM and see the gtime in host. Dump the 43rd field which shows the gtime in every second: # while :; do awk '{print $3" "$43}' /proc/3955/task/4014/stat; sleep 1; done S 4348 R 7064566 R 7064766 R 7064967 R 7065168 S 4759 S 4759 During running busy loop, it returns large value. After applying this patch, we can see right gtime. # while :; do awk '{print $3" "$43}' /proc/10913/task/10956/stat; sleep 1; done S 5338 R 5365 R 5465 R 5566 R 5666 S 5726 S 5726 Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Chris Metcalf <cmetcalf@ezchip.com> Cc: Christoph Lameter <cl@linux.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Luiz Capitulino <lcapitulino@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul E . McKenney <paulmck@linux.vnet.ibm.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1447948054-28668-2-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 2541117b0cf79977fa11a0d6e17d61010677bd7b upstream.

/proc/stats shows invalid gtime when the thread is running in guest.
When vtime accounting is not enabled, we cannot get a valid delta.
The delta is calculated with now - tsk->vtime_snap, but tsk->vtime_snap
is only updated when vtime accounting is runtime enabled.

This patch makes task_gtime() just return gtime without computing the
buggy non-existing tickless delta when vtime accounting is not enabled.

Use context_tracking_is_enabled() to check if vtime is accounting on
some cpu, in which case only we need to check the tickless delta. This
way we fix the gtime value regression on machines not running nohz full.

The kernel config contains CONFIG_VIRT_CPU_ACCOUNTING_GEN=y and
CONFIG_NO_HZ_FULL_ALL=n and boot without nohz_full.

I ran and stop a busy loop in VM and see the gtime in host.
Dump the 43rd field which shows the gtime in every second:

	 # while :; do awk '{print $3" "$43}' /proc/3955/task/4014/stat; sleep 1; done
	S 4348
	R 7064566
	R 7064766
	R 7064967
	R 7065168
	S 4759
	S 4759

During running busy loop, it returns large value.

After applying this patch, we can see right gtime.

	 # while :; do awk '{print $3" "$43}' /proc/10913/task/10956/stat; sleep 1; done
	S 5338
	R 5365
	R 5465
	R 5566
	R 5666
	S 5726
	S 5726

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E . McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1447948054-28668-2-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
sched/core: Fix a race between try_to_wake_up() and a woken up task 2016-11-20T01:17:22+00:00 Balbir Singh bsingharora@gmail.com 2016-09-05T03:16:40+00:00 1a06e10495c23d353f5cf6642f590e52367402fc commit 135e8c9250dd5c8c9aae5984fde6f230d0cbfeaf upstream. The origin of the issue I've seen is related to a missing memory barrier between check for task->state and the check for task->on_rq. The task being woken up is already awake from a schedule() and is doing the following: do { schedule() set_current_state(TASK_(UN)INTERRUPTIBLE); } while (!cond); The waker, actually gets stuck doing the following in try_to_wake_up(): while (p->on_cpu) cpu_relax(); Analysis: The instance I've seen involves the following race: CPU1 CPU2 while () { if (cond) break; do { schedule(); set_current_state(TASK_UN..) } while (!cond); wakeup_routine() spin_lock_irqsave(wait_lock) raw_spin_lock_irqsave(wait_lock) wake_up_process() } try_to_wake_up() set_current_state(TASK_RUNNING); .. list_del(&waiter.list); CPU2 wakes up CPU1, but before it can get the wait_lock and set current state to TASK_RUNNING the following occurs: CPU3 wakeup_routine() raw_spin_lock_irqsave(wait_lock) if (!list_empty) wake_up_process() try_to_wake_up() raw_spin_lock_irqsave(p->pi_lock) .. if (p->on_rq && ttwu_wakeup()) .. while (p->on_cpu) cpu_relax() .. CPU3 tries to wake up the task on CPU1 again since it finds it on the wait_queue, CPU1 is spinning on wait_lock, but immediately after CPU2, CPU3 got it. CPU3 checks the state of p on CPU1, it is TASK_UNINTERRUPTIBLE and the task is spinning on the wait_lock. Interestingly since p->on_rq is checked under pi_lock, I've noticed that try_to_wake_up() finds p->on_rq to be 0. This was the most confusing bit of the analysis, but p->on_rq is changed under runqueue lock, rq_lock, the p->on_rq check is not reliable without this fix IMHO. The race is visible (based on the analysis) only when ttwu_queue() does a remote wakeup via ttwu_queue_remote. In which case the p->on_rq change is not done uder the pi_lock. The result is that after a while the entire system locks up on the raw_spin_irqlock_save(wait_lock) and the holder spins infintely Reproduction of the issue: The issue can be reproduced after a long run on my system with 80 threads and having to tweak available memory to very low and running memory stress-ng mmapfork test. It usually takes a long time to reproduce. I am trying to work on a test case that can reproduce the issue faster, but thats work in progress. I am still testing the changes on my still in a loop and the tests seem OK thus far. Big thanks to Benjamin and Nick for helping debug this as well. Ben helped catch the missing barrier, Nick caught every missing bit in my theory. Signed-off-by: Balbir Singh <bsingharora@gmail.com> [ Updated comment to clarify matching barriers. Many architectures do not have a full barrier in switch_to() so that cannot be relied upon. ] Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Alexey Kardashevskiy <aik@ozlabs.ru> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Nicholas Piggin <nicholas.piggin@gmail.com> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/e02cce7b-d9ca-1ad0-7a61-ea97c7582b37@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 
commit 135e8c9250dd5c8c9aae5984fde6f230d0cbfeaf upstream.

The origin of the issue I've seen is related to
a missing memory barrier between check for task->state and
the check for task->on_rq.

The task being woken up is already awake from a schedule()
and is doing the following:

	do {
		schedule()
		set_current_state(TASK_(UN)INTERRUPTIBLE);
	} while (!cond);

The waker, actually gets stuck doing the following in
try_to_wake_up():

	while (p->on_cpu)
		cpu_relax();

Analysis:

The instance I've seen involves the following race:

 CPU1					CPU2

 while () {
   if (cond)
     break;
   do {
     schedule();
     set_current_state(TASK_UN..)
   } while (!cond);
					wakeup_routine()
					  spin_lock_irqsave(wait_lock)
   raw_spin_lock_irqsave(wait_lock)	  wake_up_process()
 }					  try_to_wake_up()
 set_current_state(TASK_RUNNING);	  ..
 list_del(&waiter.list);

CPU2 wakes up CPU1, but before it can get the wait_lock and set
current state to TASK_RUNNING the following occurs:

 CPU3
 wakeup_routine()
 raw_spin_lock_irqsave(wait_lock)
 if (!list_empty)
   wake_up_process()
   try_to_wake_up()
   raw_spin_lock_irqsave(p->pi_lock)
   ..
   if (p->on_rq && ttwu_wakeup())
   ..
   while (p->on_cpu)
     cpu_relax()
   ..

CPU3 tries to wake up the task on CPU1 again since it finds
it on the wait_queue, CPU1 is spinning on wait_lock, but immediately
after CPU2, CPU3 got it.

CPU3 checks the state of p on CPU1, it is TASK_UNINTERRUPTIBLE and
the task is spinning on the wait_lock. Interestingly since p->on_rq
is checked under pi_lock, I've noticed that try_to_wake_up() finds
p->on_rq to be 0. This was the most confusing bit of the analysis,
but p->on_rq is changed under runqueue lock, rq_lock, the p->on_rq
check is not reliable without this fix IMHO. The race is visible
(based on the analysis) only when ttwu_queue() does a remote wakeup
via ttwu_queue_remote. In which case the p->on_rq change is not
done uder the pi_lock.

The result is that after a while the entire system locks up on
the raw_spin_irqlock_save(wait_lock) and the holder spins infintely

Reproduction of the issue:

The issue can be reproduced after a long run on my system with 80
threads and having to tweak available memory to very low and running
memory stress-ng mmapfork test. It usually takes a long time to
reproduce. I am trying to work on a test case that can reproduce
the issue faster, but thats work in progress. I am still testing the
changes on my still in a loop and the tests seem OK thus far.

Big thanks to Benjamin and Nick for helping debug this as well.
Ben helped catch the missing barrier, Nick caught every missing
bit in my theory.

Signed-off-by: Balbir Singh <bsingharora@gmail.com>
[ Updated comment to clarify matching barriers. Many
  architectures do not have a full barrier in switch_to()
  so that cannot be relied upon. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicholas Piggin <nicholas.piggin@gmail.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/e02cce7b-d9ca-1ad0-7a61-ea97c7582b37@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>