Re: [RFC][PATCH 15/16] objtool: Implement noinstr validation

[Peter Zijlstra]

On Mon, Mar 16, 2020 at 11:19:04AM -0500, Josh Poimboeuf wrote:
> On Mon, Mar 16, 2020 at 02:24:19PM +0100, Peter Zijlstra wrote:
> > > And "read_instr_hints" reads as "read_instruction_hints()".
> > > 
> > > Can we come up with a more readable name?  Why not just "notrace"?
> > > 
> > > The trace begin/end annotations could be
> > > 
> > >   trace_allow_begin()
> > >   trace_allow_end()
> > 
> > notrace already exists and we didn't want to confuse things further.
> 
> Um, why would it confuse things to call a section of notrace code
> ".notrace.text"???

Because it is strictly stronger than the notrace attribute is. And we
certainly don't want all that is now marked notrace to end up there.

Our section also very much excludes kprobes and hardware breakpoints for
starters.