Re: [PATCH v6 2/4] bpf: Add new bpf map type to store the pointer to struct perf_event

From: Peter Zijlstra
Date: Wed Aug 05 2015 - 05:40:30 EST


On Tue, Aug 04, 2015 at 08:58:14AM +0000, Kaixu Xia wrote:

> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 2027809..81fc99e 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -641,6 +641,8 @@ 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 struct perf_event_attr *perf_event_attrs(struct perf_event *event);

const struct perf_event_attr *perf_event_attrs();


> +static void *perf_event_fd_array_get_ptr(struct bpf_map *map, int fd)
> +{
> + struct perf_event *event;
> + struct perf_event_attr *attr;

Also const

> +
> + event = perf_event_get(fd);
> + if (IS_ERR(event))
> + return event;
> +
> + attr = perf_event_attrs(event);
> + if (IS_ERR(attr))
> + return attr;
> +
> + if (attr->type != PERF_TYPE_RAW &&
> + attr->type != PERF_TYPE_HARDWARE) {
> + perf_event_release_kernel(event);
> + return ERR_PTR(-EINVAL);
> + }
> + return event;
> +}

> +struct perf_event_attr *perf_event_attrs(struct perf_event *event)

Also const

> +{
> + if (!event)
> + return ERR_PTR(-EINVAL);
> +
> + return &event->attr;
> +}

It doesn't make sense (ever) to allow users of this function to change
attributes of the event, inspecting them is fine.
--
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/