* Load-balancing at migrate-enable time?
@ 2011-10-26 18:30 Paul E. McKenney
0 siblings, 0 replies; only message in thread
From: Paul E. McKenney @ 2011-10-26 18:30 UTC (permalink / raw)
To: rostedt, peterz; +Cc: linux-rt-users
Hello!
So I am getting the second round of the model beaten into shape, but
have a question...
Assume the following state:
1. Task 0 preempted in a migrate-disable region on CPU 0.
2. Task 1 preempted, but not in a migrate-disable region,
also on CPU 0.
3. Task 2 running in a migrate-disable region also on CPU 0
Suppose that Task 2 enables migration. Will it actually migrate
in order to allow Task 1 to run? (My guess is "no", but...)
Thanx, Paul
------------------------------------------------------------------------
PS. We get to the above state by starting as follows:
1. Task 0 blocked in a migrate-disable region on CPU 0.
2. Task 1 running on CPU 0 without migration disabled.
3. Task 2 blocked in a migrate-disable region also on CPU 0
Then Task 2 awakens, preempting Task 1. (In the more-aggressive
model, this would trigger migration of Task 1, but I am trying to
match what the load balancer actually does.)
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2011-10-27 7:46 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-10-26 18:30 Load-balancing at migrate-enable time? Paul E. McKenney
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.