[S390] qdio: optimize cache line usage of struct qdio_irq

Remove a memset hack that relied on the internal layout of the
qdio_irq struct and move the per device statistics data into an own
cache line to avoid cache line bashing between the inbound and the
outbound queue tasklets. Also reduce the number of allocated queues
from 32 to 4 which is the current maximum. That saves a cache line
in struct qdio_irq.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

drivers/s390/cio/qdio.h

@@ -208,7 +208,7 @@ struct qdio_dev_perf_stat {
 	unsigned int eqbs_partial;
 	unsigned int sqbs;
 	unsigned int sqbs_partial;
-};
+} ____cacheline_aligned;
 
 struct qdio_queue_perf_stat {
 	/*
@@ -329,12 +329,8 @@ struct qdio_irq {
 	struct qdio_ssqd_desc ssqd_desc;
 	void (*orig_handler) (struct ccw_device *, unsigned long, struct irb *);
 
-	struct qdio_dev_perf_stat perf_stat;
 	int perf_stat_enabled;
-	/*
-	 * Warning: Leave these members together at the end so they won't be
-	 * cleared in qdio_setup_irq.
-	 */
+
 	struct qdr *qdr;
 	unsigned long chsc_page;
 
@@ -343,6 +339,7 @@ struct qdio_irq {
 
 	debug_info_t *debug_area;
 	struct mutex setup_mutex;
+	struct qdio_dev_perf_stat perf_stat;
 };
 
 /* helper functions */