Re: [RFC PATCH v2 12/31] function_graph: Have the instances use their own ftrace_ops for filtering

2023-11-09 Thread Google
On Thu, 9 Nov 2023 21:18:48 -0500 Steven Rostedt wrote: > On Fri, 10 Nov 2023 10:51:54 +0900 > Masami Hiramatsu (Google) wrote: > > > So this patch registers ftrace_ops for each fgraph_ops to ftrace. > > This means that the ftrace_graph_func() will be called twice or more > > on the same

Re: [RFC PATCH v2 12/31] function_graph: Have the instances use their own ftrace_ops for filtering

2023-11-09 Thread Steven Rostedt
On Fri, 10 Nov 2023 10:51:54 +0900 Masami Hiramatsu (Google) wrote: > So this patch registers ftrace_ops for each fgraph_ops to ftrace. > This means that the ftrace_graph_func() will be called twice or more > on the same function. > Thus should I call ftrace_startup() once when the first

Re: [RFC PATCH v2 12/31] function_graph: Have the instances use their own ftrace_ops for filtering

2023-11-09 Thread Google
Hi, I found strange behavior that if we set two or more probes on the same function, its callback called twice or more. Also, the stacktrace failed. And I found the root cause is here; On Wed, 8 Nov 2023 23:26:42 +0900 "Masami Hiramatsu (Google)" wrote: > @@ -910,9 +917,10 @@ int

[RFC PATCH v2 12/31] function_graph: Have the instances use their own ftrace_ops for filtering

2023-11-08 Thread Masami Hiramatsu (Google)
From: Steven Rostedt (VMware) Allow for instances to have their own ftrace_ops part of the fgraph_ops that makes the funtion_graph tracer filter on the set_ftrace_filter file of the instance and not the top instance. Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Masami Hiramatsu