Re: [PATCH][3.9] ftrace: Update the kconfig for DYNAMIC_FTRACE

[Ingo Molnar <mingo@kernel.org>]

* Steven Rostedt <rostedt@goodmis.org> wrote:

> 
> Ingo,
> 
> At ELC I watched a presentation by Ezequiel about a tool he wrote to
> analyze memory wasted by kmalloc and friends. He incorrectly stated that
> you need to enable DYNAMIC_FTRACE in order to get events. I pointed out
> that this was incorrect, but he showed me that the prompt for
> DYNAMIC_FTRACE said it enabled/disabled ftrace tracepoints dynamically.
> And sure enough, it did say that. But this was written when ftrace was
> first created and hasn't been updated since. It was written well before
> tracepoints made it into the kernel. Heck, it even references that damn
> daemon!
> 
> As the merge window is still opened, and even if it wasn't, this fix
> should go mainline ASAP. It should also go into stable. This is
> definitely a bug. It may only be documentation (which means it's low
> risk to add), but it's documentation on how to set up your kernel, and
> that being wrong creates unexpected results for the user. Just like any
> other bug would.
> 
> Please pull the latest tip/perf/urgent tree, which can be found at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
> tip/perf/urgent
> 
> Head SHA1: db05021d49a994ee40a9735d9c3cb0060c9babb8
> 
> 
> Steven Rostedt (1):
>       ftrace: Update the kconfig for DYNAMIC_FTRACE
> 
> ----
>  kernel/trace/Kconfig |   24 ++++++++++++++----------
>  1 file changed, 14 insertions(+), 10 deletions(-)

Pulled, thanks Steve!

	Ingo