Hi,
On 07/20/2018 01:39 AM, Yonghong Song wrote:
On 7/10/18 10:14 AM, Vince Weaver wrote:
The PERF_EVENT_IOC_QUERY_BPF ioctl was introduced in Linux 4.16.
Signed-off-by: Vince Weaver <vincent.weaver@xxxxxxxxx>
diff --git a/man2/perf_event_open.2 b/man2/perf_event_open.2
index b598acf17..a464e95a2 100644
--- a/man2/perf_event_open.2
+++ b/man2/perf_event_open.2
@@ -2945,6 +2945,47 @@ even though the ring-buffer remains empty.
The argument is an unsigned 32-bit integer.
A nonzero value pauses the ring-buffer, while a
zero value resumes the ring-buffer.
+.TP
+.BR PERF_EVENT_IOC_QUERY_BPF " (since Linux 4.16)"
+.\" commit f371b304f12e31fe30207c41ca7754564e0ea4dc
+This allows querying which Berkeley Packet Filter (BPF)
+programs are attached to an existing kprobe tracepoint.
The query is not limited to kprobe tracepoint. It can be done
on ukprobe, tracepoint and syscall trace event. The underlying
PMU are "tracepoint", "kprobe" and "uprobe".
So, maybe "kprobe tracepoint" => "kprobe, uprobe, or tracepoint"?
Sounds like a small fix is in order, Vince? Can you send a patch?
Thanks,
Michael
