Change the way jobs are tracked in the CE.

The new code tracks CE jobs so that we can detect if a CE task falls
behind due to jobs overrunning their budgets.

There are also various bug fixes related to timers.

We may not want to arm timers until the task release, as well.

1 files changed, 0 insertions, 2 deletions

diff --git a/include/litmus/sched_mc.h b/include/litmus/sched_mc.h
index 069198ab7e28..d29796298701 100644
--- a/include/litmus/sched_mc.h
+++ b/include/litmus/sched_mc.h
@@ -20,8 +20,6 @@ struct mc_task {
 struct mc_job {
         int is_ghost:1;
         lt_t ghost_budget;
-        /* which invocation of the CE is this */
-        unsigned int expected_job;
 };
 
 #ifdef __KERNEL__