Re: v4.7.5 build: 0 failures 2 warnings (v4.7.5)

[Greg KH <gregkh@linuxfoundation.org>]

On Mon, Sep 26, 2016 at 01:46:51AM +0200, Arnd Bergmann wrote:
> On Sunday 25 September 2016, Greg KH wrote:
> > 
> > Hi Arnd,
> > 
> > I'm getting the following build warnings on the 4.7-stable tree, any
> > ideas what they are from:
> > 
> > In file included from ../kernel/trace/trace_irqsoff.c:15:0:
> > ../kernel/trace/trace_irqsoff.c: In function ‘stop_critical_timings’:
> > ../include/linux/ftrace.h:703:36: warning: calling ‘__builtin_return_address’ with a nonzero argument is unsafe [-Wframe-address]
> >  #  define ftrace_return_address(n) __builtin_return_address(n)
> >                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~
> > ../include/linux/ftrace.h:710:38: note: in expansion of macro ‘ftrace_return_address’
> >  #define CALLER_ADDR1 ((unsigned long)ftrace_return_address(1))
> >                                       ^~~~~~~~~~~~~~~~~~~~~
> > ../kernel/trace/trace_irqsoff.c:433:38: note: in expansion of macro ‘CALLER_ADDR1’
> >    stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
> >                                       ^~~~~~~~~~~~
> > 
> > It shows up in a few other places as well with this signature.
> > 
> > Same goes for the 4.4-stable tree
> 
> In mainline, this warning got disabled with 124a3d88fa20 ("Disable "frame-address"
> warning") and then later that change was partially reverted wtih 377ccbb48373
> ("Makefile: Mute warning for __builtin_return_address(>0) for tracing only").
> 
> It's probably fine if you apply both to the stable kernel.

Thanks, that's better, but the warning still triggers on
kernel/sched/core.c for me, as the last patch only disables it for the
tracing directory.

But it's a few less warning messages, now on to fix up the rest...

thanks,

greg k-h