uprobes/perf: Teach trace_uprobe/perf code to use UPROBE_HANDLER_REMOVE
Change uprobe_trace_func() and uprobe_perf_func() to return "int". Change
uprobe_dispatcher() to return "trace_ret | perf_ret" although this is not
needed, currently TP_FLAG_TRACE/TP_FLAG_PROFILE are mutually exclusive.
The only functional change is that uprobe_perf_func() checks the filtering
too and returns UPROBE_HANDLER_REMOVE if nobody wants to trace current.
Testing:
# perf probe -x /lib/libc.so.6 syscall
# perf record -e probe_libc:syscall -i perl -e 'fork; syscall -1 for 1..10; wait'
# perf report --show-total-period
100.00% 10 perl libc-2.8.so [.] syscall
Before this patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 20
A child process doesn't have a counter, but still it hits this breakoint
"copied" by dup_mmap().
After the patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 11
The child process hits this int3 only once and does unapply_uprobe().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index b7850f5..2399f14 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -486,7 +486,7 @@
};
/* uprobe handler */
-static void uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
+static int uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
{
struct uprobe_trace_entry_head *entry;
struct ring_buffer_event *event;
@@ -504,7 +504,7 @@
event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
size, irq_flags, pc);
if (!event)
- return;
+ return 0;
entry = ring_buffer_event_data(event);
entry->ip = instruction_pointer(task_pt_regs(current));
@@ -514,6 +514,8 @@
if (!filter_current_check_discard(buffer, call, entry, event))
trace_buffer_unlock_commit(buffer, event, irq_flags, pc);
+
+ return 0;
}
/* Event entry printers */
@@ -721,7 +723,7 @@
}
/* uprobe profile handler */
-static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
+static int uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
{
struct ftrace_event_call *call = &tu->call;
struct uprobe_trace_entry_head *entry;
@@ -730,11 +732,14 @@
int size, __size, i;
int rctx;
+ if (!uprobe_perf_filter(&tu->consumer, 0, current->mm))
+ return UPROBE_HANDLER_REMOVE;
+
__size = sizeof(*entry) + tu->size;
size = ALIGN(__size + sizeof(u32), sizeof(u64));
size -= sizeof(u32);
if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough"))
- return;
+ return 0;
preempt_disable();
@@ -752,6 +757,7 @@
out:
preempt_enable();
+ return 0;
}
#endif /* CONFIG_PERF_EVENTS */
@@ -792,18 +798,19 @@
static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs)
{
struct trace_uprobe *tu;
+ int ret = 0;
tu = container_of(con, struct trace_uprobe, consumer);
tu->nhit++;
if (tu->flags & TP_FLAG_TRACE)
- uprobe_trace_func(tu, regs);
+ ret |= uprobe_trace_func(tu, regs);
#ifdef CONFIG_PERF_EVENTS
if (tu->flags & TP_FLAG_PROFILE)
- uprobe_perf_func(tu, regs);
+ ret |= uprobe_perf_func(tu, regs);
#endif
- return 0;
+ return ret;
}
static struct trace_event_functions uprobe_funcs = {
