On Thu, Jan 9, 2020 at 10:48 AM Alexei Starovoitov
<alexei.starovoitov@xxxxxxxxx> wrote:
>
> On Thu, Dec 26, 2019 at 01:18:55PM -0800, Andrii Nakryiko wrote:
> > Streamline BPF_TRACE_x macro by moving out return type and section attribute
> > definition out of macro itself. That makes those function look in source code
> > similar to other BPF programs. Additionally, simplify its usage by determining
> > number of arguments automatically (so just single BPF_TRACE vs a family of
> > BPF_TRACE_1, BPF_TRACE_2, etc). Also, allow more natural function argument
> > syntax without commas inbetween argument type and name.
> >
> > Given this helper is useful not only for tracing tp_btf/fenty/fexit programs,
> > but could be used for LSM programs and others following the same pattern,
> > rename BPF_TRACE macro into more generic BPF_HANDLER. Existing BPF_TRACE_x
> > usages in selftests are converted to new BPF_HANDLER macro.
> >
> > Following the same pattern, define BPF_KPROBE and BPF_KRETPROBE macros for
> > nicer usage of kprobe/kretprobe arguments, respectively. BPF_KRETPROBE, adopts
> > same convention used by fexit programs, that last defined argument is probed
> > function's return result.
> >
> > Signed-off-by: Andrii Nakryiko <andriin@xxxxxx>
> ...
> > +
> > +#define BPF_HANDLER(name, args...) \
> > +name(unsigned long long *ctx); \
> > +static __always_inline typeof(name(0)) ____##name(args); \
> > +typeof(name(0)) name(unsigned long long *ctx) \
> > +{ \
> > + _Pragma("GCC diagnostic push") \
> > + _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
> > + return ____##name(___bpf_ctx_cast(args)); \
> > + _Pragma("GCC diagnostic pop") \
>
> It says "GCC ..", but does it actually work with gcc?
> If the answer is no, I think it's still ok, but would be good to document.
I'll write a simple test program to verify. I expect it to work, of course :)
>
> Other than the above BPF_HANDLER, BPF_KPROBE, BPF_KRETPROBE distinction make sense.
> Please document it. It's not obvious when to use each one.
Yep, will do.
>
> Also the intent is do let users do s/BPF_KRETPROBE/BPF_HANDLER/ conversion
> when they transition from kretprobe to fexit without changing anything else
> in the function body and function declaration? That's neat if that can work.
Yep, it should be a trivial s/BPF_KPROBE/BPF_HANDLER/ change.
