Skip to content

C
clone-Linux

gsplhtlxg / clone-Linux

通知 2
Star 0
Fork 1
C
clone-Linux

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 19 5月, 2009 1 次提交
  3. 15 5月, 2009 2 次提交
    • T
      sched, timers: cleanup avenrun users · 2d02494f
      由 Thomas Gleixner 提交于 
      avenrun is an rough estimate so we don't have to worry about
consistency of the three avenrun values. Remove the xtime lock
dependency and provide a function to scale the values. Cleanup the
users.

[ Impact: cleanup ]
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
      2d02494f
    • T
      sched, timers: move calc_load() to scheduler · dce48a84
      由 Thomas Gleixner 提交于 
      Dimitri Sivanich noticed that xtime_lock is held write locked across
calc_load() which iterates over all online CPUs. That can cause long
latencies for xtime_lock readers on large SMP systems. 

The load average calculation is an rough estimate anyway so there is
no real need to protect the readers vs. the update. It's not a problem
when the avenrun array is updated while a reader copies the values.

Instead of iterating over all online CPUs let the scheduler_tick code
update the number of active tasks shortly before the avenrun update
happens. The avenrun update itself is handled by the CPU which calls
do_timer().

[ Impact: reduce xtime_lock write locked section ]
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
      dce48a84
  5. 15 4月, 2009 1 次提交
  7. 09 4月, 2009 1 次提交
    • N
      sched: do not count frozen tasks toward load · e3c8ca83
      由 Nathan Lynch 提交于 
      Freezing tasks via the cgroup freezer causes the load average to climb
because the freezer's current implementation puts frozen tasks in
uninterruptible sleep (D state).

Some applications which perform job-scheduling functions consult the
load average when making decisions.  If a cgroup is frozen, the load
average does not provide a useful measure of the system's utilization
to such applications.  This is especially inconvenient if the job
scheduler employs the cgroup freezer as a mechanism for preempting low
priority jobs.  Contrast this with using SIGSTOP for the same purpose:
the stopped tasks do not count toward system load.

Change task_contributes_to_load() to return false if the task is
frozen.  This results in /proc/loadavg behavior that better meets
users' expectations.
Signed-off-by: NNathan Lynch <ntl@pobox.com>
Acked-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NNigel Cunningham <nigel@tuxonice.net>
Tested-by: NNigel Cunningham <nigel@tuxonice.net>
Cc: <stable@kernel.org>
Cc: containers@lists.linux-foundation.org
Cc: linux-pm@lists.linux-foundation.org
Cc: Matt Helsley <matthltc@us.ibm.com>
LKML-Reference: <20090408194512.47a99b95@manatee.lan>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      e3c8ca83
  9. 03 4月, 2009 5 次提交
  11. 01 4月, 2009 2 次提交
  13. 24 3月, 2009 2 次提交
    • S
      function-graph: ignore times across schedule · 8aef2d28
      由 Steven Rostedt 提交于 
      Impact: more accurate timings

The current method of function graph tracing does not take into
account the time spent when a task is not running. This shows functions
that call schedule have increased costs:

 3) + 18.664 us   |      }
 ------------------------------------------
 3)    <idle>-0    =>  kblockd-123
 ------------------------------------------

 3)               |      finish_task_switch() {
 3)   1.441 us    |        _spin_unlock_irq();
 3)   3.966 us    |      }
 3) ! 2959.433 us |    }
 3) ! 2961.465 us |  }

This patch uses the tracepoint in the scheduling context switch to
account for time that has elapsed while a task is scheduled out.
Now we see:

 ------------------------------------------
 3)    <idle>-0    =>  edac-po-1067
 ------------------------------------------

 3)               |      finish_task_switch() {
 3)   0.685 us    |        _spin_unlock_irq();
 3)   2.331 us    |      }
 3) + 41.439 us   |    }
 3) + 42.663 us   |  }
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
      8aef2d28
    • T
      genirq: add threaded interrupt handler support · 3aa551c9
      由 Thomas Gleixner 提交于 
      Add support for threaded interrupt handlers:

A device driver can request that its main interrupt handler runs in a
thread. To achive this the device driver requests the interrupt with
request_threaded_irq() and provides additionally to the handler a
thread function. The handler function is called in hard interrupt
context and needs to check whether the interrupt originated from the
device. If the interrupt originated from the device then the handler
can either return IRQ_HANDLED or IRQ_WAKE_THREAD. IRQ_HANDLED is
returned when no further action is required. IRQ_WAKE_THREAD causes
the genirq code to invoke the threaded (main) handler. When
IRQ_WAKE_THREAD is returned handler must have disabled the interrupt
on the device level. This is mandatory for shared interrupt handlers,
but we need to do it as well for obscure x86 hardware where disabling
an interrupt on the IO_APIC level redirects the interrupt to the
legacy PIC interrupt lines.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>
      3aa551c9
  15. 12 3月, 2009 1 次提交
  17. 27 2月, 2009 2 次提交
  19. 15 2月, 2009 1 次提交
    • N
      lockdep: annotate reclaim context (__GFP_NOFS) · cf40bd16
      由 Nick Piggin 提交于 
      Here is another version, with the incremental patch rolled up, and
added reclaim context annotation to kswapd, and allocation tracing
to slab allocators (which may only ever reach the page allocator
in rare cases, so it is good to put annotations here too).

Haven't tested this version as such, but it should be getting closer
to merge worthy ;)

--
After noticing some code in mm/filemap.c accidentally perform a __GFP_FS
allocation when it should not have been, I thought it might be a good idea to
try to catch this kind of thing with lockdep.

I coded up a little idea that seems to work. Unfortunately the system has to
actually be in __GFP_FS page reclaim, then take the lock, before it will mark
it. But at least that might still be some orders of magnitude more common
(and more debuggable) than an actual deadlock condition, so we have some
improvement I hope (the concept is no less complete than discovery of a lock's
interrupt contexts).

I guess we could even do the same thing with __GFP_IO (normal reclaim), and
even GFP_NOIO locks too... but filesystems will have the most locks and fiddly
code paths, so let's start there and see how it goes.

It *seems* to work. I did a quick test.

=================================
[ INFO: inconsistent lock state ]
2.6.28-rc6-00007-ged313489-dirty #26
---------------------------------
inconsistent {in-reclaim-W} -> {ov-reclaim-W} usage.
modprobe/8526 [HC0[0]:SC0[0]:HE1:SE1] takes:
 (testlock){--..}, at: [<ffffffffa0020055>] brd_init+0x55/0x216 [brd]
{in-reclaim-W} state was registered at:
  [<ffffffff80267bdb>] __lock_acquire+0x75b/0x1a60
  [<ffffffff80268f71>] lock_acquire+0x91/0xc0
  [<ffffffff8070f0e1>] mutex_lock_nested+0xb1/0x310
  [<ffffffffa002002b>] brd_init+0x2b/0x216 [brd]
  [<ffffffff8020903b>] _stext+0x3b/0x170
  [<ffffffff80272ebf>] sys_init_module+0xaf/0x1e0
  [<ffffffff8020c3fb>] system_call_fastpath+0x16/0x1b
  [<ffffffffffffffff>] 0xffffffffffffffff
irq event stamp: 3929
hardirqs last  enabled at (3929): [<ffffffff8070f2b5>] mutex_lock_nested+0x285/0x310
hardirqs last disabled at (3928): [<ffffffff8070f089>] mutex_lock_nested+0x59/0x310
softirqs last  enabled at (3732): [<ffffffff8061f623>] sk_filter+0x83/0xe0
softirqs last disabled at (3730): [<ffffffff8061f5b6>] sk_filter+0x16/0xe0

other info that might help us debug this:
1 lock held by modprobe/8526:
 #0:  (testlock){--..}, at: [<ffffffffa0020055>] brd_init+0x55/0x216 [brd]

stack backtrace:
Pid: 8526, comm: modprobe Not tainted 2.6.28-rc6-00007-ged313489-dirty #26
Call Trace:
 [<ffffffff80265483>] print_usage_bug+0x193/0x1d0
 [<ffffffff80266530>] mark_lock+0xaf0/0xca0
 [<ffffffff80266735>] mark_held_locks+0x55/0xc0
 [<ffffffffa0020000>] ? brd_init+0x0/0x216 [brd]
 [<ffffffff802667ca>] trace_reclaim_fs+0x2a/0x60
 [<ffffffff80285005>] __alloc_pages_internal+0x475/0x580
 [<ffffffff8070f29e>] ? mutex_lock_nested+0x26e/0x310
 [<ffffffffa0020000>] ? brd_init+0x0/0x216 [brd]
 [<ffffffffa002006a>] brd_init+0x6a/0x216 [brd]
 [<ffffffffa0020000>] ? brd_init+0x0/0x216 [brd]
 [<ffffffff8020903b>] _stext+0x3b/0x170
 [<ffffffff8070f8b9>] ? mutex_unlock+0x9/0x10
 [<ffffffff8070f83d>] ? __mutex_unlock_slowpath+0x10d/0x180
 [<ffffffff802669ec>] ? trace_hardirqs_on_caller+0x12c/0x190
 [<ffffffff80272ebf>] sys_init_module+0xaf/0x1e0
 [<ffffffff8020c3fb>] system_call_fastpath+0x16/0x1b
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      cf40bd16
  21. 12 2月, 2009 1 次提交
  23. 11 2月, 2009 2 次提交
  25. 09 2月, 2009 1 次提交
  27. 06 2月, 2009 1 次提交
  29. 05 2月, 2009 2 次提交
    • P
      timers: split process wide cpu clocks/timers · 4cd4c1b4
      由 Peter Zijlstra 提交于 
      Change the process wide cpu timers/clocks so that we:

 1) don't mess up the kernel with too many threads,
 2) don't have a per-cpu allocation for each process,
 3) have no impact when not used.

In order to accomplish this we're going to split it into two parts:

 - clocks; which can take all the time they want since they run
           from user context -- ie. sys_clock_gettime(CLOCK_PROCESS_CPUTIME_ID)

 - timers; which need constant time sampling but since they're
           explicity used, the user can pay the overhead.

The clock readout will go back to a full sum of the thread group, while the
timers will run of a global 'clock' that only runs when needed, so only
programs that make use of the facility pay the price.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      4cd4c1b4
    • P
      signal: re-add dead task accumulation stats. · 32bd671d
      由 Peter Zijlstra 提交于 
      We're going to split the process wide cpu accounting into two parts:

 - clocks; which can take all the time they want since they run
           from user context.

 - timers; which need constant time tracing but can affort the overhead
           because they're default off -- and rare.

The clock readout will go back to a full sum of the thread group, for this
we need to re-add the exit stats that were removed in the initial itimer
rework (f06febc9: timers: fix itimer/many thread hang).

Furthermore, since that full sum can be rather slow for large thread groups
and we have the complete dead task stats, revert the do_notify_parent time
computation.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      32bd671d
  31. 03 2月, 2009 1 次提交
  33. 01 2月, 2009 1 次提交
  35. 30 1月, 2009 1 次提交
  37. 23 1月, 2009 1 次提交
    • S
      trace, lockdep: manual preempt count adding for local_bh_disable · 7e49fcce
      由 Steven Rostedt 提交于 
      Impact: fix to preempt trace triggering lockdep check_flag failure

In local_bh_disable, the use of add_preempt_count causes the
preempt tracer to start recording the time preemption is off.
But because it already modified the preempt_count to show
softirqs disabled, and before it called the lockdep code to
handle this, it causes a state that lockdep can not handle.

The preempt tracer will reset the ring buffer on start of a trace,
and the ring buffer reset code does a spin_lock_irqsave. This
calls into lockdep and lockdep will fail when it detects the
invalid state of having softirqs disabled but the internal
current->softirqs_enabled is still set.

The fix is to manually add the SOFTIRQ_OFFSET to preempt count
and call the preempt tracer code outside the lockdep critical
area.

Thanks to Peter Zijlstra for suggesting this solution.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      7e49fcce
  39. 16 1月, 2009 2 次提交
    • M
      softlockup: decouple hung tasks check from softlockup detection · e162b39a
      由 Mandeep Singh Baines 提交于 
      Decoupling allows:

* hung tasks check to happen at very low priority

* hung tasks check and softlockup to be enabled/disabled independently
  at compile and/or run-time

* individual panic settings to be enabled disabled independently
  at compile and/or run-time

* softlockup threshold to be reduced without increasing hung tasks
  poll frequency (hung task check is expensive relative to softlock watchdog)

* hung task check to be zero over-head when disabled at run-time
Signed-off-by: NMandeep Singh Baines <msb@google.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      e162b39a
    • I
      sched: fix !CONFIG_SCHEDSTATS build failure · 34cb6135
      由 Ingo Molnar 提交于 
      Stephen Rothwell reported this linux-next build failure with !CONFIG_SCHEDSTATS:

| In file included from kernel/sched.c:1703:
| kernel/sched_fair.c: In function 'adaptive_gran':
| kernel/sched_fair.c:1324: error: 'struct sched_entity' has no member named 'avg_wakeup'

The start_runtime and avg_wakeup metrics are now not just for statistics,
but also for scheduling - so they always need to be available. (Also
move out the nr_migrations fields - for future perfcounters usage.)
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      34cb6135
  41. 15 1月, 2009 3 次提交
    • P
      sched: introduce avg_wakeup · 831451ac
      由 Peter Zijlstra 提交于 
      Introduce a new avg_wakeup statistic.

avg_wakeup is a measure of how frequently a task wakes up other tasks, it
represents the average time between wakeups, with a limit of avg_runtime
for when it doesn't wake up anybody.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NMike Galbraith <efault@gmx.de>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      831451ac
    • P
      mutex: implement adaptive spinning · 0d66bf6d
      由 Peter Zijlstra 提交于 
      Change mutex contention behaviour such that it will sometimes busy wait on
acquisition - moving its behaviour closer to that of spinlocks.

This concept got ported to mainline from the -rt tree, where it was originally
implemented for rtmutexes by Steven Rostedt, based on work by Gregory Haskins.

Testing with Ingo's test-mutex application (http://lkml.org/lkml/2006/1/8/50)
gave a 345% boost for VFS scalability on my testbox:

 # ./test-mutex-shm V 16 10 | grep "^avg ops"
 avg ops/sec:               296604

 # ./test-mutex-shm V 16 10 | grep "^avg ops"
 avg ops/sec:               85870

The key criteria for the busy wait is that the lock owner has to be running on
a (different) cpu. The idea is that as long as the owner is running, there is a
fair chance it'll release the lock soon, and thus we'll be better off spinning
instead of blocking/scheduling.

Since regular mutexes (as opposed to rtmutexes) do not atomically track the
owner, we add the owner in a non-atomic fashion and deal with the races in
the slowpath.

Furthermore, to ease the testing of the performance impact of this new code,
there is means to disable this behaviour runtime (without having to reboot
the system), when scheduler debugging is enabled (CONFIG_SCHED_DEBUG=y),
by issuing the following command:

 # echo NO_OWNER_SPIN > /debug/sched_features

This command re-enables spinning again (this is also the default):

 # echo OWNER_SPIN > /debug/sched_features
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      0d66bf6d
    • P
      mutex: preemption fixes · 41719b03
      由 Peter Zijlstra 提交于 
      The problem is that dropping the spinlock right before schedule is a voluntary
preemption point and can cause a schedule, right after which we schedule again.

Fix this inefficiency by keeping preemption disabled until we schedule, do this
by explicity disabling preemption and providing a schedule() variant that
assumes preemption is already disabled.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      41719b03
  43. 14 1月, 2009 1 次提交
  45. 08 1月, 2009 1 次提交
    • P
      itimers: remove the per-cpu-ish-ness · 490dea45
      由 Peter Zijlstra 提交于 
      Either we bounce once cacheline per cpu per tick, yielding n^2 bounces
or we just bounce a single..

Also, using per-cpu allocations for the thread-groups complicates the
per-cpu allocator in that its currently aimed to be a fixed sized
allocator and the only possible extention to that would be vmap based,
which is seriously constrained on 32 bit archs.

So making the per-cpu memory requirement depend on the number of
processes is an issue.

Lastly, it didn't deal with cpu-hotplug, although admittedly that might
be fixable.
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      490dea45
  47. 07 1月, 2009 1 次提交
  49. 05 1月, 2009 1 次提交
  51. 31 12月, 2008 1 次提交
    • M
      [PATCH] idle cputime accounting · 79741dd3
      由 Martin Schwidefsky 提交于 
      The cpu time spent by the idle process actually doing something is
currently accounted as idle time. This is plain wrong, the architectures
that support VIRT_CPU_ACCOUNTING=y can do better: distinguish between the
time spent doing nothing and the time spent by idle doing work. The first
is accounted with account_idle_time and the second with account_system_time.
The architectures that use the account_xxx_time interface directly and not
the account_xxx_ticks interface now need to do the check for the idle
process in their arch code. In particular to improve the system vs true
idle time accounting the arch code needs to measure the true idle time
instead of just testing for the idle process.
To improve the tick based accounting as well we would need an architecture
primitive that can tell us if the pt_regs of the interrupted context
points to the magic instruction that halts the cpu.

In addition idle time is no more added to the stime of the idle process.
This field now contains the system time of the idle process as it should
be. On systems without VIRT_CPU_ACCOUNTING this will always be zero as
every tick that occurs while idle is running will be accounted as idle
time.

This patch contains the necessary common code changes to be able to
distinguish idle system time and true idle time. The architectures with
support for VIRT_CPU_ACCOUNTING need some changes to exploit this.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>
      79741dd3
  53. 30 12月, 2008 1 次提交
    • J
      sched: sched.c declare variables before they get used · 47fea2ad
      由 Jaswinder Singh Rajput 提交于 
      Impact: cleanup, avoid sparse warnings

In linux/sched.h moved out sysctl_sched_latency, sysctl_sched_min_granularity,
sysctl_sched_wakeup_granularity, sysctl_sched_shares_ratelimit and
sysctl_sched_shares_thresh from #ifdef CONFIG_SCHED_DEBUG as these variables
are common for both.

Fixes these sparse warnings:
  kernel/sched.c:825:14: warning: symbol 'sysctl_sched_shares_ratelimit' was not declared. Should it be static?
  kernel/sched.c:832:14: warning: symbol 'sysctl_sched_shares_thresh' was not declared. Should it be static?
  kernel/sched_fair.c:37:14: warning: symbol 'sysctl_sched_latency' was not declared. Should it be static?
  kernel/sched_fair.c:43:14: warning: symbol 'sysctl_sched_min_granularity' was not declared. Should it be static?
  kernel/sched_fair.c:72:14: warning: symbol 'sysctl_sched_wakeup_granularity' was not declared. Should it be static?
Signed-off-by: NJaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
      47fea2ad