On Fri, Nov 06, 2015 at 12:47:49PM -0500, Waiman Long wrote:
On 11/06/2015 09:50 AM, Peter Zijlstra wrote:Right; and while I doubt it hurts the native case (you did benchmark it
*urgh*, last time we had:I want to reuse as much of the existing native code as possible instead of
+ if (pv_wait_head_or_steal())
+ goto stolen;
while ((val = smp_load_acquire(&lock->val.counter))& _Q_LOCKED_PENDING_MASK)
cpu_relax();
...
+stolen:
while (!(next = READ_ONCE(node->next)))
cpu_relax();
...
Now you completely overhaul the native code.. what happened?
duplicating that in the PV function. The only difference now is that the PV
function will acquire that lock.
I hope), I'm not too keen on the end result code wise.
Maybe just keep the above.
Semantically, I don't want to call the lockFair enough I suppose, pv_wait_head_or_lock() then?
acquisition as lock stealing as the queue head is entitled to get the lock
next.
I can rename pv_queued_spin_trylock_unfair() toOK.
pv_queued_spin_steal_lock() to emphasize the fact that this is the routine
where lock stealing happens.