The user space event FD from perf_event_open() syscall is converted
to the pointer to struct perf event and stored in map.
Signed-off-by: kaixu xia <xiaka...@huawei.com>
---
include/linux/perf_event.h | 2 ++
kernel/bpf/syscall.c | 68 ++++++++++++++++++++++++++++++++++++++++++++
kernel/events/core.c | 22 ++++++++++++++
3 files changed, 92 insertions(+)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 2027809..2ea4067 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -641,6 +641,7 @@ extern int perf_event_init_task(struct task_struct *child);
extern void perf_event_exit_task(struct task_struct *child);
extern void perf_event_free_task(struct task_struct *task);
extern void perf_event_delayed_put(struct task_struct *task);
+extern struct perf_event *perf_event_get(unsigned int fd);
extern void perf_event_print_debug(void);
extern void perf_pmu_disable(struct pmu *pmu);
extern void perf_pmu_enable(struct pmu *pmu);
@@ -979,6 +980,7 @@ static inline int perf_event_init_task(struct task_struct
*child) { return 0; }
static inline void perf_event_exit_task(struct task_struct *child) { }
static inline void perf_event_free_task(struct task_struct *task) { }
static inline void perf_event_delayed_put(struct task_struct *task) { }
+static struct perf_event *perf_event_get(unsigned int fd) {
return NULL; }
static inline void perf_event_print_debug(void)
{ }
static inline int perf_event_task_disable(void)
{ return -EINVAL; }
static inline int perf_event_task_enable(void) {
return -EINVAL; }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 4c2d9e6..ac76792 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -17,6 +17,7 @@
#include <linux/license.h>
#include <linux/filter.h>
#include <linux/version.h>
+#include <linux/perf_event.h>
static LIST_HEAD(bpf_map_types);
@@ -65,6 +66,19 @@ void bpf_map_put(struct bpf_map *map)
}
}
+static int bpf_map_perf_event_put(void *value)
+{
+ struct perf_event *event;
+
+ event = (struct perf_event *)(*(unsigned long *)value);
+ if (!event)
+ return -EBADF;
+
+ perf_event_release_kernel(event);
+
+ return 0;
+}
+
static int bpf_map_release(struct inode *inode, struct file *filp)
{
struct bpf_map *map = filp->private_data;
@@ -75,6 +89,13 @@ static int bpf_map_release(struct inode *inode, struct file
*filp)
*/
bpf_prog_array_map_clear(map);
+ if (map->flags & BPF_MAP_FLAG_PERF_EVENT) {
+ rcu_read_lock();
+ if (map->ops->map_traverse_elem(bpf_map_perf_event_put, map) <
0)
+ return -EINVAL;
+ rcu_read_unlock();
+ }
+
bpf_map_put(map);
return 0;
}
@@ -176,6 +197,10 @@ static int map_lookup_elem(union bpf_attr *attr)
if (IS_ERR(map))
return PTR_ERR(map);
+ if (map->flags & BPF_MAP_FLAG_PERF_EVENT)
+ /* prevent user space from reading elem for PMU map */
+ return -EACCES;
+
err = -ENOMEM;
key = kmalloc(map->key_size, GFP_USER);
if (!key)
@@ -215,6 +240,39 @@ err_put:
return err;
}
+static int replace_map_with_perf_event(void *value)
+{
+ struct perf_event *event;
+ u32 fd;
+
+ fd = *(u32 *)value;
+
+ event = perf_event_get(fd);
+ if (IS_ERR(event))
+ return PTR_ERR(event);
+
+ if (atomic_long_inc_not_zero(&event->refcount))
+ memcpy(value, &event, sizeof(struct perf_event *));
+ else
+ return -ENOENT;
+
+ return 0;
+}
+
+static bool check_map_perf_event_stored(struct bpf_map *map, void *key)
+{
+ void *value;
+ bool is_stored = false;
+
+ rcu_read_lock();
+ value = map->ops->map_lookup_elem(map, key);
+ if (value && (*(unsigned long *)value))
+ is_stored = true;
+ rcu_read_unlock();
+
+ return is_stored;
+}
+
#define BPF_MAP_UPDATE_ELEM_LAST_FIELD flags
static int map_update_elem(union bpf_attr *attr)
@@ -252,6 +310,16 @@ static int map_update_elem(union bpf_attr *attr)
if (copy_from_user(value, uvalue, map->value_size) != 0)
goto free_value;
+ if (map->flags & BPF_MAP_FLAG_PERF_EVENT) {
+ err = -EINVAL;
+ if (check_map_perf_event_stored(map, key))
+ goto free_value;
+
+ err = -EBADF;
+ if (replace_map_with_perf_event(value) != 0)
+ goto free_value;
+ }
+
/* eBPF program that use maps are running under rcu_read_lock(),
* therefore all map accessors rely on this fact, so do the same here
*/
diff --git a/kernel/events/core.c b/kernel/events/core.c
index e965cfa..c4e34b7 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8582,6 +8582,28 @@ void perf_event_delayed_put(struct task_struct *task)
WARN_ON_ONCE(task->perf_event_ctxp[ctxn]);
}
+struct perf_event *perf_event_get(unsigned int fd)
+{
+ struct perf_event *event;
+ struct fd f;
+
+ f = fdget(fd);
+
+ if (!f.file)
+ return ERR_PTR(-EBADF);
+
+ if (f.file->f_op != &perf_fops) {
+ fdput(f);
+ return ERR_PTR(-EINVAL);
+ }
+
+ event = f.file->private_data;
+
+ fdput(f);
+
+ return event;
+}
+
/*
* inherit a event from parent task to child task:
*/
--
1.7.10.4
--
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/
