Re: [RFC PATCH v3] perf/x86: make perf callchain work without CONFIG_FRAME_POINTER

[Josh Poimboeuf <jpoimboe@redhat.com>]

On Fri, Apr 19, 2019 at 07:47:18PM +0800, Kairui Song wrote:
> On Fri, Apr 19, 2019 at 5:43 PM Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > On Fri, Apr 19, 2019 at 10:17:49AM +0800, Kairui Song wrote:
> > > On Fri, Apr 19, 2019 at 8:58 AM Josh Poimboeuf <jpoimboe@redhat.com> wrote:
> > > >
> > > > I still don't like using regs->bp because it results in different code
> > > > paths for FP and ORC.  In the FP case, the regs are treated like real
> > > > regs even though they're fake.
> > > >
> > > > Something like the below would be much simpler.  Would this work?  I don't
> > > > know if any other code relies on the fake regs->bp or regs->sp.
> > >
> > > Works perfectly. My only concern is that FP path used to work very
> > > well, not sure it's a good idea to change it, and this may bring some
> > > extra overhead for FP path.
> >
> > Given Josh wrote all that code, I'm fairly sure it is still OK :-)
> >
> > But also looking at the code in unwind_frame.c, __unwind_start() seems
> > to pretty much do what the removed caller_frame_pointer() did (when
> > .regs=NULL) but better.
> >
> 
> OK, with FP we will also need to do a few more extra unwinding,
> previously it start directly from the frame of the trace point, now
> have to trace back to the trace point first.
> If that's fine I could post another update (that will be pretty much
> just copy&paste from the Josh's code he posted :P , is this OK?)

You're right that FP will need to unwind a few extra frames, but I doubt
that will make much of a difference performance-wise.  I prefer the
simpler approach.

If you use that patch for the next version then you can add

  Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>

Thanks.

-- 
Josh