[PATCH] arm64: mm: fix location of _etext

[Mark Rutland]

On Thu, Jun 23, 2016 at 01:49:05PM +0200, Ard Biesheuvel wrote:
> On 23 June 2016 at 13:40, Mark Rutland <mark.rutland at arm.com> wrote:
> > On Thu, Jun 23, 2016 at 12:46:32PM +0200, Ard Biesheuvel wrote:
> >> @@ -232,7 +232,7 @@ void __init setup_arch(char **cmdline_p)
> >>
> >>       sprintf(init_utsname()->machine, ELF_PLATFORM);
> >>       init_mm.start_code = (unsigned long) _text;
> >> -     init_mm.end_code   = (unsigned long) _etext;
> >> +     init_mm.end_code   = (unsigned long) __init_begin;
> >
> > Why does this need to be __init_begin?
> >
> > We don't have code in .rodata, the exception tables, or notes, so
> > shouldn't this be left pointing at _etext if we're tightening things up?
> >
> 
> You are right, and I am happy to change that in the same patch.
> Strictly speaking, it is a separate change, but that applies equally
> to not updating the _etext references in kaslr.c

Ok. FWIW, with that and the comment fixup:

Reviewed-by: Mark Rutland <mark.rutland at arm.com>

Thanks,
Mark.