Re: [PATCH v10 00/36] tracing: fprobe: function_graph: Multi-function graph and fprobe on fgraph

[Masami Hiramatsu (Google) <mhiramat@kernel.org>]

On Fri, 24 May 2024 18:41:56 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> On Tue,  7 May 2024 23:08:00 +0900
> "Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
> 
> > Steven Rostedt (VMware) (15):
> >       function_graph: Convert ret_stack to a series of longs
> >       fgraph: Use BUILD_BUG_ON() to make sure we have structures divisible by long
> >       function_graph: Add an array structure that will allow multiple callbacks
> >       function_graph: Allow multiple users to attach to function graph
> >       function_graph: Remove logic around ftrace_graph_entry and return
> >       ftrace/function_graph: Pass fgraph_ops to function graph callbacks
> >       ftrace: Allow function_graph tracer to be enabled in instances
> >       ftrace: Allow ftrace startup flags exist without dynamic ftrace
> >       function_graph: Have the instances use their own ftrace_ops for filtering
> >       function_graph: Add "task variables" per task for fgraph_ops
> >       function_graph: Move set_graph_function tests to shadow stack global var
> >       function_graph: Move graph depth stored data to shadow stack global var
> >       function_graph: Move graph notrace bit to shadow stack global var
> >       function_graph: Implement fgraph_reserve_data() and fgraph_retrieve_data()
> >       function_graph: Add selftest for passing local variables
> 
> Hi Masami,
> 
> While reviewing these patches, I realized there's several things I dislike
> about the patches I wrote. So I took these patches and started cleaning
> them up a little. Mostly renaming functions and adding comments.

Thanks for cleaning up the patches!!

> 
> As this is a major change to the function graph tracer, and I feel nervous
> about building something on top of this, how about I take over these
> patches and push them out for the next merge window. I'm hoping to get them
> into linux-next by v6.10-rc2 (I spent the day working on them, and it's
> mostly minor tweaks).

OK.

> Then I can push it out to 6.11 and get some good testing against it. Then
> we can add your stuff on top and get that merged in 6.12.

Yeah, it is reasonable plan. I also concerns about the stability. Especially,
this involves fprobe side changes too. If we introduce both at once, it may
mess up many things.

> 
> If all goes well, I'm hoping to get a series on just these patches (and
> your selftest addition) by tonight.
> 
> Thoughts?

I agree with you.

Thank you,

> 
> -- Steve


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>