3 files changed, 264 insertions, 0 deletions

diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 14d9505178ca..2a22e46390d3 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -106,6 +106,15 @@ config CONTEXT_SWITCH_TRACER
           This tracer gets called from the context switch and records
           all switching of tasks.
 
+config STACK_TRACER
+        bool "Trace max stack"
+        depends on HAVE_FTRACE
+        select FTRACE
+        select STACKTRACE
+        help
+          This tracer records the max stack of the kernel, and displays
+          it in debugfs/tracing/stack_trace
+
 config DYNAMIC_FTRACE
         bool "enable/disable ftrace tracepoints dynamically"
         depends on FTRACE
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 71d17de17288..58ec61c44bd6 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_FTRACE) += trace_functions.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
+obj-$(CONFIG_STACK_TRACER) += trace_stack.o
 obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o
 
 libftrace-y := ftrace.o
diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c
new file mode 100644
index 000000000000..4d1e522e3fe8
--- /dev/null
+++ b/kernel/trace/trace_stack.c
@@ -0,0 +1,254 @@
+/*
+ * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com>
+ *
+ */
+#include <linux/stacktrace.h>
+#include <linux/kallsyms.h>
+#include <linux/seq_file.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/debugfs.h>
+#include <linux/ftrace.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include "trace.h"
+
+#define STACK_TRACE_ENTRIES 500
+
+static unsigned long stack_dump_trace[STACK_TRACE_ENTRIES] =
+        { [0 ... (STACK_TRACE_ENTRIES-1)] = ULONG_MAX };
+static struct stack_trace max_stack_trace = {
+        .max_entries            = STACK_TRACE_ENTRIES,
+        .entries                = stack_dump_trace,
+};
+
+static unsigned long max_stack_size;
+static raw_spinlock_t max_stack_lock =
+        (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED;
+
+static int stack_trace_disabled __read_mostly;
+static DEFINE_PER_CPU(int, trace_active);
+
+static inline void check_stack(void)
+{
+        unsigned long this_size;
+        unsigned long flags;
+
+        this_size = ((unsigned long)&this_size) & (THREAD_SIZE-1);
+        this_size = THREAD_SIZE - this_size;
+
+        if (this_size <= max_stack_size)
+                return;
+
+        raw_local_irq_save(flags);
+        __raw_spin_lock(&max_stack_lock);
+
+        /* a race could have already updated it */
+        if (this_size <= max_stack_size)
+                goto out;
+
+        max_stack_size = this_size;
+
+        max_stack_trace.nr_entries      = 0;
+        max_stack_trace.skip            = 1;
+
+        save_stack_trace(&max_stack_trace);
+
+ out:
+        __raw_spin_unlock(&max_stack_lock);
+        raw_local_irq_restore(flags);
+}
+
+static void
+stack_trace_call(unsigned long ip, unsigned long parent_ip)
+{
+        int cpu, resched;
+
+        if (unlikely(!ftrace_enabled || stack_trace_disabled))
+                return;
+
+        resched = need_resched();
+        preempt_disable_notrace();
+
+        cpu = raw_smp_processor_id();
+        /* no atomic needed, we only modify this variable by this cpu */
+        if (per_cpu(trace_active, cpu)++ != 0)
+                goto out;
+
+        check_stack();
+
+ out:
+        per_cpu(trace_active, cpu)--;
+        /* prevent recursion in schedule */
+        if (resched)
+                preempt_enable_no_resched_notrace();
+        else
+                preempt_enable_notrace();
+}
+
+static struct ftrace_ops trace_ops __read_mostly =
+{
+        .func = stack_trace_call,
+};
+
+static ssize_t
+stack_max_size_read(struct file *filp, char __user *ubuf,
+                    size_t count, loff_t *ppos)
+{
+        unsigned long *ptr = filp->private_data;
+        char buf[64];
+        int r;
+
+        r = snprintf(buf, sizeof(buf), "%ld\n", *ptr);
+        if (r > sizeof(buf))
+                r = sizeof(buf);
+        return simple_read_from_buffer(ubuf, count, ppos, buf, r);
+}
+
+static ssize_t
+stack_max_size_write(struct file *filp, const char __user *ubuf,
+                     size_t count, loff_t *ppos)
+{
+        long *ptr = filp->private_data;
+        unsigned long val, flags;
+        char buf[64];
+        int ret;
+
+        if (count >= sizeof(buf))
+                return -EINVAL;
+
+        if (copy_from_user(&buf, ubuf, count))
+                return -EFAULT;
+
+        buf[count] = 0;
+
+        ret = strict_strtoul(buf, 10, &val);
+        if (ret < 0)
+                return ret;
+
+        raw_local_irq_save(flags);
+        __raw_spin_lock(&max_stack_lock);
+        *ptr = val;
+        __raw_spin_unlock(&max_stack_lock);
+        raw_local_irq_restore(flags);
+
+        return count;
+}
+
+static struct file_operations stack_max_size_fops = {
+        .open           = tracing_open_generic,
+        .read           = stack_max_size_read,
+        .write          = stack_max_size_write,
+};
+
+static void *
+t_next(struct seq_file *m, void *v, loff_t *pos)
+{
+        unsigned long *t = m->private;
+
+        (*pos)++;
+
+        if (!t || *t == ULONG_MAX)
+                return NULL;
+
+        t++;
+        m->private = t;
+
+        return t;
+}
+
+static void *t_start(struct seq_file *m, loff_t *pos)
+{
+        unsigned long *t = m->private;
+        loff_t l = 0;
+
+        local_irq_disable();
+        __raw_spin_lock(&max_stack_lock);
+
+        for (; t && l < *pos; t = t_next(m, t, &l))
+                ;
+
+        return t;
+}
+
+static void t_stop(struct seq_file *m, void *p)
+{
+        __raw_spin_unlock(&max_stack_lock);
+        local_irq_enable();
+}
+
+static int trace_lookup_stack(struct seq_file *m, unsigned long addr)
+{
+#ifdef CONFIG_KALLSYMS
+        char str[KSYM_SYMBOL_LEN];
+
+        sprint_symbol(str, addr);
+
+        return seq_printf(m, "[<%p>] %s\n", (void*)addr, str);
+#else
+        return seq_printf(m, "%p\n", (void*)addr);
+#endif
+}
+
+static int t_show(struct seq_file *m, void *v)
+{
+        unsigned long *t = v;
+
+        if (!t || *t == ULONG_MAX)
+                return 0;
+
+        trace_lookup_stack(m, *t);
+
+        return 0;
+}
+
+static struct seq_operations stack_trace_seq_ops = {
+        .start          = t_start,
+        .next           = t_next,
+        .stop           = t_stop,
+        .show           = t_show,
+};
+
+static int stack_trace_open(struct inode *inode, struct file *file)
+{
+        int ret;
+
+        ret = seq_open(file, &stack_trace_seq_ops);
+        if (!ret) {
+                struct seq_file *m = file->private_data;
+                m->private = stack_dump_trace;
+        }
+
+        return ret;
+}
+
+static struct file_operations stack_trace_fops = {
+        .open           = stack_trace_open,
+        .read           = seq_read,
+        .llseek         = seq_lseek,
+};
+
+static __init int stack_trace_init(void)
+{
+        struct dentry *d_tracer;
+        struct dentry *entry;
+
+        d_tracer = tracing_init_dentry();
+
+        entry = debugfs_create_file("stack_max_size", 0644, d_tracer,
+                                    &max_stack_size, &stack_max_size_fops);
+        if (!entry)
+                pr_warning("Could not create debugfs 'stack_max_size' entry\n");
+
+        entry = debugfs_create_file("stack_trace", 0444, d_tracer,
+                                    NULL, &stack_trace_fops);
+        if (!entry)
+                pr_warning("Could not create debugfs 'stack_trace' entry\n");
+
+        register_ftrace_function(&trace_ops);
+
+        return 0;
+}
+
+device_initcall(stack_trace_init);