浏览代码

[POWERPC] spusched: Fix timeslice calculations

The current timeslice code mixes 'jiffies' up with 'spesched ticks'. This
change correctly defines the number of time slices each SPE contexts is
given, and clarifies the comment.

This brings the default timeslice for SPE contexts into a reasonable
range.

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Jeremy Kerr 18 年之前
父节点
当前提交
60e2423933
共有 1 个文件被更改,包括 4 次插入4 次删除
  1. 4 4
      arch/powerpc/platforms/cell/spufs/sched.c

+ 4 - 4
arch/powerpc/platforms/cell/spufs/sched.c

@@ -70,11 +70,11 @@ static struct timer_list spusched_timer;
 /*
  * These are the 'tuning knobs' of the scheduler:
  *
- * Minimum timeslice is 5 msecs (or 10 jiffies, whichever is larger),
- * default timeslice is 100 msecs, maximum timeslice is 800 msecs.
+ * Minimum timeslice is 5 msecs (or 1 spu scheduler tick, whichever is
+ * larger), default timeslice is 100 msecs, maximum timeslice is 800 msecs.
  */
-#define MIN_SPU_TIMESLICE	max(5 * HZ / 100, 10)
-#define DEF_SPU_TIMESLICE	(100 * HZ / 100)
+#define MIN_SPU_TIMESLICE	max(5 * HZ / (1000 * SPUSCHED_TICK), 1)
+#define DEF_SPU_TIMESLICE	(100 * HZ / (1000 * SPUSCHED_TICK))
 
 #define MAX_USER_PRIO		(MAX_PRIO - MAX_RT_PRIO)
 #define SCALE_PRIO(x, prio) \