On Fri, Jun 2, 2023 at 1:52 AM Yafang Shao <laoar.shao@xxxxxxxxx> wrote:
>
> By adding support for ->fill_link_info to the perf_event link, users will
> be able to inspect it using `bpftool link show`. While users can currently
> access this information via `bpftool perf show`, consolidating the link
> information for all link types in one place would be more convenient.
>
> Signed-off-by: Yafang Shao <laoar.shao@xxxxxxxxx>
> ---
> include/uapi/linux/bpf.h | 6 ++++++
> kernel/bpf/syscall.c | 45 ++++++++++++++++++++++++++++++++++++++++++
> tools/include/uapi/linux/bpf.h | 6 ++++++
> 3 files changed, 57 insertions(+)
>
> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
> index 22c8168..87ecf8b 100644
> --- a/include/uapi/linux/bpf.h
> +++ b/include/uapi/linux/bpf.h
> @@ -6442,6 +6442,12 @@ struct bpf_link_info {
> __u64 addrs;
> __u32 count;
> } kprobe_multi;
> + struct {
> + __aligned_u64 name; /* in/out: symbol name buffer ptr */
> + __u64 addr;
> + __u32 name_len;
> + __u32 offset;
> + } perf_event;
perf_event link could be:
a) uprobe
b) kprobe
c) tracepoint
d) generic perf_event (e.g., cpu_cycles)
This interface doesn't make it very clear which one it is. And what's
"name" for cpu_cycles event? What is the name for uprobe? Let's
actually document this, otherwise it's hard to understand how this
information can be interpreted...
> };
> } __attribute__((aligned(8)));
>
> diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> index 80c9ec0..da2de8e 100644
> --- a/kernel/bpf/syscall.c
> +++ b/kernel/bpf/syscall.c
> @@ -3303,9 +3303,54 @@ static void bpf_perf_link_dealloc(struct bpf_link *link)
> kfree(perf_link);
> }
>
[...]
