Implement a yield_type for MuQSS with a tunable in sysctl and documentation, bumping version to 0.150

3 files changed, 40 insertions, 7 deletions

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index 01e0a4a861e3..2a82a37242a5 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -95,6 +95,7 @@ show up in /proc/sys/kernel:
 - unknown_nmi_panic
 - watchdog
 - watchdog_thresh
+- yield_type
 - version
 
 ==============================================================
@@ -1082,3 +1083,13 @@ The softlockup threshold is (2 * watchdog_thresh). Setting this
 tunable to zero will disable lockup detection altogether.
 
 ==============================================================
+
+yield_type: (MuQSS CPU scheduler only)
+
+This determines what type of yield calls to sched_yield will perform.
+
+ 0: No yield.
+ 1: Yield only to better priority/deadline tasks. (default)
+ 2: Expire timeslice and recalculate deadline.
+
+==============================================================
diff --git a/kernel/sched/MuQSS.c b/kernel/sched/MuQSS.c
index 18cb55ded3b7..7617ae49ab09 100644
--- a/kernel/sched/MuQSS.c
+++ b/kernel/sched/MuQSS.c
@@ -137,19 +137,20 @@
 
 void print_scheduler_version(void)
 {
-	printk(KERN_INFO "MuQSS CPU scheduler v0.144 by Con Kolivas.\n");
+	printk(KERN_INFO "MuQSS CPU scheduler v0.150 by Con Kolivas.\n");
 }
 
 /*
  * This is the time all tasks within the same priority round robin.
- * Value is in ms and set to a minimum of 6ms. Scales with number of cpus.
+ * Value is in ms and set to a minimum of 6ms.
  * Tunable via /proc interface.
  */
 int rr_interval __read_mostly = 6;
 
-/* Tunable to choose whether to prioritise latency or throughput, simple
- * binary yes or no */
-
+/*
+ * Tunable to choose whether to prioritise latency or throughput, simple
+ * binary yes or no
+ */
 int sched_interactive __read_mostly = 1;
 
 /*
@@ -160,6 +161,14 @@ int sched_interactive __read_mostly = 1;
 int sched_iso_cpu __read_mostly = 70;
 
 /*
+ * sched_yield_type - Choose what sort of yield sched_yield will perform.
+ * 0: No yield.
+ * 1: Yield only to better priority/deadline tasks. (default)
+ * 2: Expire timeslice and recalculate deadline.
+ */
+int sched_yield_type __read_mostly = 1;
+
+/*
  * The relative length of deadline for each priority(nice) level.
  */
 static int prio_ratios[NICE_WIDTH] __read_mostly;
@@ -5077,9 +5086,12 @@ SYSCALL_DEFINE0(sched_yield)
 	struct task_struct *p;
 	struct rq *rq;
 
+	if (!sched_yield_type)
+		goto out;
 	p = current;
 	rq = this_rq_lock();
-	time_slice_expired(p, rq);
+	if (sched_yield_type > 1)
+		time_slice_expired(p, rq);
 	schedstat_inc(rq->yld_count);
 
 	/*
@@ -5092,7 +5104,7 @@ SYSCALL_DEFINE0(sched_yield)
 	sched_preempt_enable_no_resched();
 
 	schedule();
-
+out:
 	return 0;
 }
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index d7a4df09cf58..2557e03c3c0d 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -131,6 +131,7 @@ static int __read_mostly one_thousand = 1000;
 extern int rr_interval;
 extern int sched_interactive;
 extern int sched_iso_cpu;
+extern int sched_yield_type;
 #endif
 #ifdef CONFIG_PRINTK
 static int ten_thousand = 10000;
@@ -1048,6 +1049,15 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.procname	= "yield_type",
+		.data		= &sched_yield_type,
+		.maxlen		= sizeof (int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.extra1		= &zero,
+		.extra2		= &two,
+	},
 #endif
 #if defined(CONFIG_S390) && defined(CONFIG_SMP)
 	{