On Tue, Oct 25, 2022 at 10:51:02AM +0800, Guo Ren wrote:
> On Mon, Oct 24, 2022 at 8:14 PM Mark Rutland <mark.rutland@xxxxxxx> wrote:
> >
> > On Mon, Oct 24, 2022 at 02:06:04PM +0200, Peter Zijlstra wrote:
> > > On Mon, Oct 24, 2022 at 12:56:03PM +0100, Mark Rutland wrote:
> > >
> > > > How about we split this like:
> > > >
> > > > | /*
> > > > | * Prevent the compiler from instrumenting this code in any way
> > > > | * This does not prevent instrumentation via KPROBES, which must be
> > > > | * prevented through other means if necessary.
> > >
> > > Perhaps point to NOINSTR_TEXT in vmlinux.lds.h
> >
> > Makes sense, will do.
> Do I need to update the comment with NOINSTR_TEXT? eg:
>
> * Prevent the compiler from instrumenting this code in any way
> * This does not prevent instrumentation via KPROBES, which must be
> * prevented through other means if necessary. See NOINSTR_TEXT
> * in vmlinux.lds.h.
I think given Peter's reply we can leave the patch as-is for now, and we can
leave commentary or other changes to a later follow up. I'm happy to propose
patches for that once the existing bits are merged.
Sorry for confusing matters!
> > > > | */
> > > > | #define __no_compiler_instrument \
> > > > | noinline notrace noinline notrace __no_kcsan \
> > > > | __no_sanitize_address __no_sanitize_coverage
> > > > |
> > > > | /*
> > > > | * Section for code which can't be instrumented at all.
> > > > | * Any code in this section cannot be instrumented with KPROBES.
> > > > | */
> > > > | #define noinstr __no_compiler_instrument section(".noinstr.text")
> > > >
> > > > ... then we don't need __noinstr_section(), and IMO the split is
> > > > clearer.
> > >
> > > Yeah, perhaps, no strong feelings. Note I have this in the sched-idle
> > > series as well (which I still need to rebase and repost :/).
> >
> > Ah; I'll sit on this for now then, and once that's all in I can send a
> > cleanup/rework patch. Sorry for the noise!
> We still keep __noinstr_section(), right?
Yes -- for now this patch can stay as-is, and __noinstr_section() will remain.
Thanks,
Mark.
