[RFC 1/2] procfs: /proc/sched_stat fails on very very large machines.

[Nathan Zimmer]

On systems with 4096 cores doing a cat /proc/sched_stat fails.
We are trying to push all the data into a single kmalloc buffer.
The issue is on these very large machines all the data will not fit in 4mb.

A better solution is to not us the single_open mechanism but to provide
our own seq_operations.

The output should be identical to previous version and thus not need the
version number.

Signed-off-by: Nathan Zimmer <nzimmer@xxxxxxx>
CC: Ingo Molnar <mingo@xxxxxxxxxx>
CC: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
CC: linux-kernel@xxxxxxxxxxxxxxx

---
 kernel/sched/stats.c |  139 +++++++++++++++++++++++++++++---------------------
 1 files changed, 81 insertions(+), 58 deletions(-)

diff --git a/kernel/sched/stats.c b/kernel/sched/stats.c
index 903ffa9..a4326a8 100644
--- a/kernel/sched/stats.c
+++ b/kernel/sched/stats.c
@@ -17,90 +17,113 @@ static int show_schedstat(struct seq_file *seq, void *v)
 	int cpu;
 	int mask_len = DIV_ROUND_UP(NR_CPUS, 32) * 9;
 	char *mask_str = kmalloc(mask_len, GFP_KERNEL);
+	cpu = *(loff_t *)v;
 
 	if (mask_str == NULL)
 		return -ENOMEM;
 
-	seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION);
-	seq_printf(seq, "timestamp %lu\n", jiffies);
-	for_each_online_cpu(cpu) {
-		struct rq *rq = cpu_rq(cpu);
+	if (!cpu) {
+		seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION);
+		seq_printf(seq, "timestamp %lu\n", jiffies);
+	}
+
+	struct rq *rq = cpu_rq(cpu);
 #ifdef CONFIG_SMP
-		struct sched_domain *sd;
-		int dcount = 0;
+	struct sched_domain *sd;
+	int dcount = 0;
 #endif
 
-		/* runqueue-specific stats */
-		seq_printf(seq,
-		    "cpu%d %u 0 %u %u %u %u %llu %llu %lu",
-		    cpu, rq->yld_count,
-		    rq->sched_count, rq->sched_goidle,
-		    rq->ttwu_count, rq->ttwu_local,
-		    rq->rq_cpu_time,
-		    rq->rq_sched_info.run_delay, rq->rq_sched_info.pcount);
+	/* runqueue-specific stats */
+	seq_printf(seq,
+	    "cpu%d %u 0 %u %u %u %u %llu %llu %lu",
+	    cpu, rq->yld_count,
+	    rq->sched_count, rq->sched_goidle,
+	    rq->ttwu_count, rq->ttwu_local,
+	    rq->rq_cpu_time,
+	    rq->rq_sched_info.run_delay, rq->rq_sched_info.pcount);
 
-		seq_printf(seq, "\n");
+	seq_printf(seq, "\n");
 
 #ifdef CONFIG_SMP
-		/* domain-specific stats */
-		rcu_read_lock();
-		for_each_domain(cpu, sd) {
-			enum cpu_idle_type itype;
-
-			cpumask_scnprintf(mask_str, mask_len,
-					  sched_domain_span(sd));
-			seq_printf(seq, "domain%d %s", dcount++, mask_str);
-			for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
-					itype++) {
-				seq_printf(seq, " %u %u %u %u %u %u %u %u",
-				    sd->lb_count[itype],
-				    sd->lb_balanced[itype],
-				    sd->lb_failed[itype],
-				    sd->lb_imbalance[itype],
-				    sd->lb_gained[itype],
-				    sd->lb_hot_gained[itype],
-				    sd->lb_nobusyq[itype],
-				    sd->lb_nobusyg[itype]);
-			}
-			seq_printf(seq,
-				   " %u %u %u %u %u %u %u %u %u %u %u %u\n",
-			    sd->alb_count, sd->alb_failed, sd->alb_pushed,
-			    sd->sbe_count, sd->sbe_balanced, sd->sbe_pushed,
-			    sd->sbf_count, sd->sbf_balanced, sd->sbf_pushed,
-			    sd->ttwu_wake_remote, sd->ttwu_move_affine,
-			    sd->ttwu_move_balance);
+	/* domain-specific stats */
+	rcu_read_lock();
+	for_each_domain(cpu, sd) {
+		enum cpu_idle_type itype;
+
+		cpumask_scnprintf(mask_str, mask_len,
+				  sched_domain_span(sd));
+		seq_printf(seq, "domain%d %s", dcount++, mask_str);
+		for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
+				itype++) {
+			seq_printf(seq, " %u %u %u %u %u %u %u %u",
+			    sd->lb_count[itype],
+			    sd->lb_balanced[itype],
+			    sd->lb_failed[itype],
+			    sd->lb_imbalance[itype],
+			    sd->lb_gained[itype],
+			    sd->lb_hot_gained[itype],
+			    sd->lb_nobusyq[itype],
+			    sd->lb_nobusyg[itype]);
 		}
-		rcu_read_unlock();
-#endif
+		seq_printf(seq,
+			   " %u %u %u %u %u %u %u %u %u %u %u %u\n",
+		    sd->alb_count, sd->alb_failed, sd->alb_pushed,
+		    sd->sbe_count, sd->sbe_balanced, sd->sbe_pushed,
+		    sd->sbf_count, sd->sbf_balanced, sd->sbf_pushed,
+		    sd->ttwu_wake_remote, sd->ttwu_move_affine,
+		    sd->ttwu_move_balance);
 	}
+	rcu_read_unlock();
+#endif
 	kfree(mask_str);
 	return 0;
 }
 
+static void *schedstat_start(struct seq_file *file, loff_t *offset)
+{
+	if (cpu_online(*offset))
+		return offset;
+	return NULL;
+}
+
+static void *schedstat_next(struct seq_file *file, void *data, loff_t *offset)
+{
+	*offset = cpumask_next(*offset, cpu_online_mask);
+	if (cpu_online(*offset))
+		return offset;
+	return NULL;
+}
+
+static void schedstat_stop(struct seq_file *file, void *data)
+{
+}
+
+static const struct seq_operations schedstat_sops = {
+	.start = schedstat_start,
+	.next  = schedstat_next,
+	.stop  = schedstat_stop,
+	.show  = show_schedstat,
+};
+
 static int schedstat_open(struct inode *inode, struct file *file)
 {
-	unsigned int size = PAGE_SIZE * (1 + num_online_cpus() / 32);
-	char *buf = kmalloc(size, GFP_KERNEL);
-	struct seq_file *m;
-	int res;
+	int res = 0;
+
+	res = seq_open(file, &schedstat_sops);
 
-	if (!buf)
-		return -ENOMEM;
-	res = single_open(file, show_schedstat, NULL);
-	if (!res) {
-		m = file->private_data;
-		m->buf = buf;
-		m->size = size;
-	} else
-		kfree(buf);
 	return res;
 }
 
+static int schedstat_release(struct inode *inode, struct file *file)
+{
+	return 0;
+};
+
 static const struct file_operations proc_schedstat_operations = {
 	.open    = schedstat_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
-	.release = single_release,
+	.release = schedstat_release,
 };
 
 static int __init proc_schedstat_init(void)
-- 
1.6.0.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/