[PATCH v4 09/17] perf/core: Use static_call to optimize perf_guest_info_callbacks

[Peter Zijlstra]

On Fri, Feb 04, 2022 at 09:35:49AM -0800, Sami Tolvanen wrote:
> On Wed, Feb 2, 2022 at 10:43 AM Sean Christopherson <seanjc at google.com> wrote:
> > > +DEFINE_STATIC_CALL_RET0(__perf_guest_state, *perf_guest_cbs->state);
> > > +DEFINE_STATIC_CALL_RET0(__perf_guest_get_ip, *perf_guest_cbs->get_ip);
> > > +DEFINE_STATIC_CALL_RET0(__perf_guest_handle_intel_pt_intr, *perf_guest_cbs->handle_intel_pt_intr);
> >
> > Using __static_call_return0() makes clang's CFI sad on arm64 due to the resulting
> > function prototype mistmatch, which IIUC, is verified by clang's __cfi_check()
> > for indirect calls, i.e. architectures without CONFIG_HAVE_STATIC_CALL.
> >
> > We could fudge around the issue by using stubs, massaging prototypes, etc..., but
> > that means doing that for every arch-agnostic user of __static_call_return0().
> >
> > Any clever ideas?  Can we do something like generate a unique function for every
> > DEFINE_STATIC_CALL_RET0 for CONFIG_HAVE_STATIC_CALL=n, e.g. using typeof() to
> > get the prototype?
> 
> I'm not sure there's a clever fix for this. On architectures without
> HAVE_STATIC_CALL, this is an indirect call to a function with a
> mismatching type, which CFI is intended to catch.
> 
> The obvious way to solve the problem would be to use a stub function
> with the correct type, which I agree, isn't going to scale. You can
> alternatively check if .func points to __static_call_return0 and not
> make the indirect call if it does. If neither of these options are
> feasible, you can disable CFI checking in the functions that have
> these static calls using the __nocfi attribute.

There's also the tp_stub_func() thing that does basically the same
thing.