Avoid needlessly costly migrations. CONFIG_SCHED_CPU_AFFINITY

Given a choice between several available CPUs (unlinked) on which
to schedule a task, let the scheduler select the CPU closest to
where that task was previously scheduled.  Hopefully, this will
reduce cache migration penalties.

Notes: SCHED_CPU_AFFINITY is dependent upon x86 (only x86 is
supported at this time). Also PFair/PD^2 does not make use of
this feature.

Signed-off-by: Andrea Bastoni <bastoni@cs.unc.edu>

1 files changed, 8 insertions, 0 deletions

diff --git a/litmus/litmus.c b/litmus/litmus.c
index bb8c6c7e9dd1..73af6c3010d6 100644
--- a/litmus/litmus.c
+++ b/litmus/litmus.c
@@ -17,6 +17,10 @@
 #include <litmus/litmus_proc.h>
 #include <litmus/sched_trace.h>
 
+#ifdef CONFIG_SCHED_CPU_AFFINITY
+#include <litmus/affinity.h>
+#endif
+
 /* Number of RT tasks that exist in the system */
 atomic_t rt_task_count          = ATOMIC_INIT(0);
 static DEFINE_RAW_SPINLOCK(task_transition_lock);
@@ -540,6 +544,10 @@ static int __init _init_litmus(void)
 
         init_litmus_proc();
 
+#ifdef CONFIG_SCHED_CPU_AFFINITY
+        init_topology();
+#endif
+
         return 0;
 }