[PATCH] arm: ftrace: Adds support for ftrace operations

Steven Rostedt rostedt at goodmis.org
Thu Nov 27 06:58:25 PST 2014


On Thu, 27 Nov 2014 11:23:45 +0100
Jean-Jacques Hiblot <jjhiblot at traphandler.com> wrote:

> 2014-11-26 0:20 GMT+01:00 Steven Rostedt <rostedt at goodmis.org>:
> > On Mon, 24 Nov 2014 10:59:24 +0100
> > Jean-Jacques Hiblot <jjhiblot at traphandler.com> wrote:
> >
> >> Hi Steve,
> >> did you get a chance of running your tests suite on real hardware with
> >> this patch? I'd be really interested in the results.
> >
> > Yes I finally got it working on my beaglebone white. It passed all the
> > tests that passed without it. Although I'm not sure this patch itself
> > was tested as I didn't have kprobes which might use it.
> >
> Thanks for testing it. At least it doesn't cause regression.
> What do you mean by not having krpobes which might use it ? Are you
> thinking of the kprobe optimization based on ftrace ?
> 

Yep. My tests use that. But as ARM doesn't have an mfentry where
functions probed by kprobes at the start are optimized with ftrace, my
tests wont test it.

I also didn't have kprobes enabled on this board. Is it fully supported
in ARM, and does kprobes use it there too?

-- Steve



More information about the linux-arm-kernel mailing list