lkml.org 
[lkml]   [2014]   [Jun]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    Patch in this message
    /
    From
    Subject[PATCH] hrtimers: conditionally lock/unlock spinlock in hrtimer_get_next_event
    Date
    In hrtimer_get_next_event we unconditionally lock/unlock spinlock, even if it's
    not required (hrtimer_hres_active() != 0). This patch moves
    locking/unlocking and mindelta range check inside the if clause,
    so we don't execute unnecessary operations.

    Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
    ---
    kernel/hrtimer.c | 13 +++++++------
    1 file changed, 7 insertions(+), 6 deletions(-)

    diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
    index 39f339dcbe93..ce21e5f6bcf0 100644
    --- a/kernel/hrtimer.c
    +++ b/kernel/hrtimer.c
    @@ -1166,9 +1166,9 @@ ktime_t hrtimer_get_next_event(void)
    unsigned long flags;
    int i;

    - raw_spin_lock_irqsave(&cpu_base->lock, flags);
    -
    if (!hrtimer_hres_active()) {
    + raw_spin_lock_irqsave(&cpu_base->lock, flags);
    +
    for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) {
    struct hrtimer *timer;
    struct timerqueue_node *next;
    @@ -1183,12 +1183,13 @@ ktime_t hrtimer_get_next_event(void)
    if (delta.tv64 < mindelta.tv64)
    mindelta.tv64 = delta.tv64;
    }
    - }

    - raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
    + raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
    +
    + if (mindelta.tv64 < 0)
    + mindelta.tv64 = 0;
    + }

    - if (mindelta.tv64 < 0)
    - mindelta.tv64 = 0;
    return mindelta;
    }
    #endif
    --
    1.8.3.2


    \
     
     \ /
      Last update: 2014-06-09 14:21    [W:2.704 / U:0.040 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site