1. 10 11月, 2007 12 次提交
    • I
      sched: remove PREEMPT_RESTRICT · 3e3e13f3
      Ingo Molnar 提交于
      remove PREEMPT_RESTRICT. (this is a separate commit so that any
      regression related to the removal itself is bisectable)
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      3e3e13f3
    • I
      sched: turn off PREEMPT_RESTRICT · 52d3da1a
      Ingo Molnar 提交于
      PREEMPT_RESTRICT was a method aimed at reducing the amount of wakeup
      related preemption. It has a disadvantage though, it can prevent
      legitimate wakeups if a task is 'unlucky' to be hit too early by a tick
      that clears peer_preempt.
      
      Now that the wakeup preemption has been cleaned up we dont seem to have
      excessive preemptions anymore, so this feature can be turned off. (and
      removed in the next patch)
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      52d3da1a
    • I
      KVM: fix !SMP build error · a5fbb6d1
      Ingo Molnar 提交于
      fix a !SMP build error:
      
      drivers/kvm/kvm_main.c: In function 'kvm_flush_remote_tlbs':
      drivers/kvm/kvm_main.c:220: error: implicit declaration of function 'smp_call_function_mask'
      
      (and also avoid unused function warning related to up_smp_call_function()
      not making use of the 'func' parameter.)
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      a5fbb6d1
    • I
      x86: make nmi_cpu_busy() always defined · 0492007e
      Ingo Molnar 提交于
      nmi_cpu_busy() must be available on !SMP too.
      
      this is in preparation to a smp_call_function_mask() fix.
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      0492007e
    • I
      x86: make ipi_handler() always defined · 4e2947f1
      Ingo Molnar 提交于
      prepare for up_smp_call_function() to ensure that the 'func'
      pointer is unused. (which is related to a KVM build fix)
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      4e2947f1
    • E
      sched: cleanup, use NSEC_PER_MSEC and NSEC_PER_SEC · d6322faf
      Eric Dumazet 提交于
      1) hardcoded 1000000000 value is used five times in places where
         NSEC_PER_SEC might be more readable.
      
      2) A conversion from nsec to msec uses the hardcoded 1000000 value,
         which is a candidate for NSEC_PER_MSEC.
      
      no code changed:
      
          text    data     bss     dec     hex filename
         44359    3326      36   47721    ba69 sched.o.before
         44359    3326      36   47721    ba69 sched.o.after
      Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      d6322faf
    • I
      sched: reintroduce SMP tunings again · 19978ca6
      Ingo Molnar 提交于
      Yanmin Zhang reported an aim7 regression and bisected it down to:
      
       |  commit 38ad464d
       |  Author: Ingo Molnar <mingo@elte.hu>
       |  Date:   Mon Oct 15 17:00:02 2007 +0200
       |
       |     sched: uniform tunings
       |
       |     use the same defaults on both UP and SMP.
      
      fix this by reintroducing similar SMP tunings again. This resolves
      the regression.
      
      (also update the comments to match the ilog2(nr_cpus) tuning effect)
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      19978ca6
    • P
      sched: restore deterministic CPU accounting on powerpc · fa13a5a1
      Paul Mackerras 提交于
      Since powerpc started using CONFIG_GENERIC_CLOCKEVENTS, the
      deterministic CPU accounting (CONFIG_VIRT_CPU_ACCOUNTING) has been
      broken on powerpc, because we end up counting user time twice: once in
      timer_interrupt() and once in update_process_times().
      
      This fixes the problem by pulling the code in update_process_times
      that updates utime and stime into a separate function called
      account_process_tick.  If CONFIG_VIRT_CPU_ACCOUNTING is not defined,
      there is a version of account_process_tick in kernel/timer.c that
      simply accounts a whole tick to either utime or stime as before.  If
      CONFIG_VIRT_CPU_ACCOUNTING is defined, then arch code gets to
      implement account_process_tick.
      
      This also lets us simplify the s390 code a bit; it means that the s390
      timer interrupt can now call update_process_times even when
      CONFIG_VIRT_CPU_ACCOUNTING is turned on, and can just implement a
      suitable account_process_tick().
      
      account_process_tick() now takes the task_struct * as an argument.
      Tested both with and without CONFIG_VIRT_CPU_ACCOUNTING.
      Signed-off-by: NPaul Mackerras <paulus@samba.org>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      fa13a5a1
    • B
      sched: fix delay accounting regression · 9a41785c
      Balbir Singh 提交于
      Fix the delay accounting regression introduced by commit
      75d4ef16. rq no longer has sched_info
      data associated with it. task_struct sched_info structure is used by delay
      accounting to provide back statistics to user space.
      
      also remove direct use of sched_clock() (which is not a valid thing to
      do anymore) and use rq->clock instead.
      Signed-off-by: NBalbir Singh <balbir@linux.vnet.ibm.com>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      9a41785c
    • P
      sched: reintroduce the sched_min_granularity tunable · b2be5e96
      Peter Zijlstra 提交于
      we lost the sched_min_granularity tunable to a clever optimization
      that uses the sched_latency/min_granularity ratio - but the ratio
      is quite unintuitive to users and can also crash the kernel if the
      ratio is set to 0. So reintroduce the min_granularity tunable,
      while keeping the ratio maintained internally.
      
      no functionality changed.
      
      [ mingo@elte.hu: some fixlets. ]
      Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      b2be5e96
    • P
      sched: documentation: place_entity() comments · 2cb8600e
      Peter Zijlstra 提交于
      Add a few comments to place_entity(). No code changed.
      Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      2cb8600e
    • P
      sched: fix vslice · 10b77724
      Peter Zijlstra 提交于
      vslice was missing a factor NICE_0_LOAD, as weight is in
      weight*NICE_0_LOAD units.
      
      the effect of this bug was larger initial slices and
      thus latency-noisier forks.
      Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
      Signed-off-by: NIngo Molnar <mingo@elte.hu>
      10b77724
  2. 07 11月, 2007 2 次提交
  3. 06 11月, 2007 26 次提交