lkml.org 
[lkml]   [2018]   [Jun]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    SubjectRe: [PATCH v2 1/2] locking: Implement an algorithm choice for Wound-Wait mutexes
    From
    Date
    On 06/14/2018 01:36 PM, Peter Zijlstra wrote:
    > On Thu, Jun 14, 2018 at 09:29:21AM +0200, Thomas Hellstrom wrote:
    >
    >> __ww_mutex_wakeup_for_backoff(struct mutex *lock, struct ww_acquire_ctx *ww_ctx)
    >> {
    >> struct mutex_waiter *cur;
    >> + unsigned int is_wait_die = ww_ctx->ww_class->is_wait_die;
    >>
    >> lockdep_assert_held(&lock->wait_lock);
    >>
    >> @@ -310,13 +348,14 @@ __ww_mutex_wakeup_for_backoff(struct mutex *lock, struct ww_acquire_ctx *ww_ctx)
    >> if (!cur->ww_ctx)
    >> continue;
    >>
    >> - if (cur->ww_ctx->acquired > 0 &&
    >> + if (is_wait_die && cur->ww_ctx->acquired > 0 &&
    >> __ww_ctx_stamp_after(cur->ww_ctx, ww_ctx)) {
    >> debug_mutex_wake_waiter(lock, cur);
    >> wake_up_process(cur->task);
    >> }
    >>
    >> - break;
    >> + if (is_wait_die || __ww_mutex_wound(lock, cur->ww_ctx, ww_ctx))
    >> + break;
    >> }
    >> }
    > I ended up with:
    >
    >
    > static void __sched
    > __ww_mutex_check_waiters(struct mutex *lock, struct ww_acquire_ctx *ww_ctx)
    > {
    > bool is_wait_die = ww_ctx->ww_class->is_wait_die;
    > struct mutex_waiter *cur;
    >
    > lockdep_assert_held(&lock->wait_lock);
    >
    > list_for_each_entry(cur, &lock->wait_list, list) {
    > if (!cur->ww_ctx)
    > continue;
    >
    > if (is_wait_die) {
    > /*
    > * Because __ww_mutex_add_waiter() and
    > * __ww_mutex_check_stamp() wake any but the earliest
    > * context, this can only affect the first waiter (with
    > * a context).
    > */
    > if (cur->ww_ctx->acquired > 0 &&
    > __ww_ctx_stamp_after(cur->ww_ctx, ww_ctx)) {
    > debug_mutex_wake_waiter(lock, cur);
    > wake_up_process(cur->task);
    > }
    >
    > break;
    > }
    >
    > if (__ww_mutex_wound(lock, cur->ww_ctx, ww_ctx))
    > break;
    > }
    > }

    Looks OK to me.

    >
    > Currently you don't allow mixing WD and WW contexts (which is not
    > immediately obvious from the above code), and the above hard relies on
    > that. Are there sensible use cases for mixing them? IOW will your
    > current restriction stand without hassle?

    Contexts _must_ agree on the algorithm used to resolve deadlocks. With
    Wait-Die, for example, older transactions will wait if a lock is held by
    a younger transaction and with Wound-Wait, younger transactions will
    wait if a lock is held by an older transaction so there is no way of
    mixing them.

    Thanks,

    /Thomas


    \
     
     \ /
      Last update: 2018-06-14 13:55    [W:3.029 / U:0.188 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site