[PATCH 3/5] tracing: Have eprobes use filtering logic of trace events

From: Steven Rostedt
Date: Mon Nov 29 2021 - 21:43:31 EST


From: "Steven Rostedt (VMware)" <rostedt@xxxxxxxxxxx>

The eprobes open code the reserving of the event on the ring buffer for
ftrace instead of using the ftrace event wrappers, which means that it
doesn't get affected by the filters, breaking the filtering logic on user
space.

Signed-off-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
---
kernel/trace/trace_eprobe.c | 16 +++++-----------
1 file changed, 5 insertions(+), 11 deletions(-)

diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c
index 928867f527e7..88487752d307 100644
--- a/kernel/trace/trace_eprobe.c
+++ b/kernel/trace/trace_eprobe.c
@@ -489,18 +489,12 @@ __eprobe_trace_func(struct eprobe_data *edata, void *rec)
if (trace_trigger_soft_disabled(edata->file))
return;

- fbuffer.trace_ctx = tracing_gen_ctx();
- fbuffer.trace_file = edata->file;
-
dsize = get_eprobe_size(&edata->ep->tp, rec);
- fbuffer.regs = NULL;
-
- fbuffer.event =
- trace_event_buffer_lock_reserve(&fbuffer.buffer, edata->file,
- call->event.type,
- sizeof(*entry) + edata->ep->tp.size + dsize,
- fbuffer.trace_ctx);
- if (!fbuffer.event)
+
+ entry = trace_event_buffer_reserve(&fbuffer, edata->file,
+ sizeof(*entry) + edata->ep->tp.size + dsize);
+
+ if (!entry)
return;

entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event);
--
2.33.0