On systems with 4096 cores attemping to read /proc/sched_debug 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 and treat each cpu as an individual record.
The output should be identical to previous version.
Signed-off-by: Nathan Zimmer <nzim...@sgi.com>
CC: Ingo Molnar <mi...@redhat.com>
CC: Peter Zijlstra <pet...@infradead.org>
CC: linux-kernel@vger.kernel.org
CC: Al Viro <v...@zeniv.linux.org.uk>
---
kernel/sched/debug.c | 73 ++++++++++++++++++++++++++++++++++++++++++++-----
1 files changed, 65 insertions(+), 8 deletions(-)
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 6f79596..d519cc7 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -322,11 +322,10 @@ static const char *sched_tunable_scaling_names[] = {
"linear"
};
-static int sched_debug_show(struct seq_file *m, void *v)
+static void sched_debug_header(struct seq_file *m)
{
u64 ktime, sched_clk, cpu_clk;
unsigned long flags;
- int cpu;
local_irq_save(flags);
ktime = ktime_to_ns(ktime_get());
@@ -368,14 +367,22 @@ static int sched_debug_show(struct seq_file *m, void *v)
#undef PN
#undef P
- SEQ_printf(m, " .%-40s: %d (%s)\n", "sysctl_sched_tunable_scaling",
+ SEQ_printf(m, " .%-40s: %d (%s)\n",
+ "sysctl_sched_tunable_scaling",
sysctl_sched_tunable_scaling,
sched_tunable_scaling_names[sysctl_sched_tunable_scaling]);
+}
- for_each_online_cpu(cpu)
+static int sched_debug_show(struct seq_file *m, void *v)
+{
+ int cpu = (unsigned long)(v - 2);
+
+ if (cpu != -1 && cpu != nr_cpu_ids)
print_cpu(m, cpu);
-
- SEQ_printf(m, "\n");
+ else if (cpu == -1)
+ sched_debug_header(m);
+ else
+ SEQ_printf(m, "\n");
return 0;
}
@@ -385,16 +392,66 @@ void sysrq_sched_debug_show(void)
sched_debug_show(NULL, NULL);
}
+static void *sched_debug_start(struct seq_file *file, loff_t *offset)
+{
+ unsigned long n = *offset;
+
+ if (n == 0)
+ return (void *) 1;
+
+ n--;
+
+ if (n > 0)
+ n = cpumask_next(n - 1, cpu_online_mask);
+ else
+ n = cpumask_first(cpu_online_mask);
+
+ *offset = n + 1;
+
+ if (n <= nr_cpu_ids)
+ return (void *)(unsigned long)(n + 2);
+ return NULL;
+}
+
+static void *sched_debug_next(struct seq_file *file, void *data, loff_t
*offset)
+{
+ (*offset)++;
+ return sched_debug_start(file, offset);
+}
+
+static void sched_debug_stop(struct seq_file *file, void *data)
+{
+}
+
+
+static const struct seq_operations sched_debug_sops = {
+ .start = sched_debug_start,
+ .next = sched_debug_next,
+ .stop = sched_debug_stop,
+ .show = sched_debug_show,
+};
+
+static int sched_debug_release(struct inode *inode, struct file *file)
+{
+ seq_release(inode, file);
+
+ return 0;
+}
+
static int sched_debug_open(struct inode *inode, struct file *filp)
{
- return single_open(filp, sched_debug_show, NULL);
+ int ret = 0;
+
+ ret = seq_open(filp, &sched_debug_sops);
+
+ return ret;
}
static const struct file_operations sched_debug_fops = {
.open = sched_debug_open,
.read = seq_read,
.llseek = seq_lseek,
- .release = single_release,
+ .release = sched_debug_release,
};
static int __init init_sched_debug_procfs(void)
--
1.6.0.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
