Re: [PATCH 6/8] tracing: Have persistent trace instances save module addresses

[Google]

On Thu, 6 Feb 2025 10:29:49 -0500
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> On Thu, 6 Feb 2025 17:26:42 +0900
> Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx> wrote:
> 
> > This maybe too frequently scan the module because the update_last_data() is
> > called when;
> > - change tracer (this maybe OK)
> > - update "set_event"
> > - write 1 to "enable" under events
> > - change pid filter
> > - etc.
> > 
> > Once it is scanned, it should not scan again, but update by module
> > callback, because usually events are enabled individually (thus
> > write 1 to "event" can happen multiple time online).
> > 
> > I think we can move this after TRACE_ARRAY_FL_LAST_BOOT check,
> > if TRACE_ARRAY_FL_LAST_BOOT flag is set, that flag should be cleared
> > with updating the tscratch, and the flag is not set, we can skip updating
> > the scratch.
> 
> No, because we could enable it, disable it, and then the user could be
> adding and removing modules all day long, and we only add and do not remove
> (I do have a patch that does remove them).

I think TRACE_ARRAY_FL_LAST_BOOT is something like one-way flag, it is set
if the buffer is persistent, but cleared when it is reused (start writing)
and never be back. Isn't it?

> 
> But we can do the same check as patch 8 does, and only do this if the
> tracer wasn't already active.

Yeah, so I think this change needs patch 8 change.

Thank you,

> 
> -- Steve


-- 
Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>