On Tue, 14 Jan 2025 10:18:06 +0900 Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx> wrote: > > I'd rather just nuke the calltime and rettime from struct ftrace_graph_ret. > > Yeah, this one is good to me. But this will make a slightly different time > for the same function when we enable multiple function graph tracer on > different instances (mine records it once and reuse it but this will record > on each instance). We may need to notice it on the document. :) And I believe it's separate for each instance now anyway, so I was going to send this to Linus as well: diff --git a/Documentation/trace/ftrace.rst b/Documentation/trace/ftrace.rst index 272464bb7c60..2b74f96d09d5 100644 --- a/Documentation/trace/ftrace.rst +++ b/Documentation/trace/ftrace.rst @@ -810,6 +810,12 @@ Here is the list of current tracers that may be configured. to draw a graph of function calls similar to C code source. + Note that the function graph calculates the timings of when the + function starts and returns internally and for each instance. If + there are two instances that run function graph tracer and traces + the same functions, the length of the timings may be slightly off as + each read the timestamp separately and not at the same time. + "blk" The block tracer. The tracer used by the blktrace user > > Reviewed-by: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx> Thanks, -- Steve
