[PATCH] dl: Use "adr" assembler command to get proper load address

[H.J. Lu]

On Thu, Oct 7, 2021 at 9:53 AM Adhemerval Zanella
<adhemerval.zanella@linaro.org> wrote:
>
>
>
> On 07/10/2021 13:22, H.J. Lu wrote:
> > On Thu, Oct 7, 2021 at 8:57 AM Szabolcs Nagy <szabolcs.nagy@arm.com> wrote:
> >>
> >> The 10/07/2021 07:29, H.J. Lu via Libc-alpha wrote:
> >>> On Thu, Oct 7, 2021 at 7:18 AM Adhemerval Zanella via Libc-alpha
> >>> <libc-alpha@sourceware.org> wrote:
> >>>> On 07/10/2021 07:00, Lukasz Majewski wrote:
> >>>>> On Thu, 7 Oct 2021 11:19:26 +0200
> >>>>> Lukasz Majewski <lukma@denx.de> wrote:
> >>>>> And indeed it is the case. Yocto/OE by default perform prelinking (use
> >>>>> prelink program) to speedup start time of dynamic program.
> >>>>>
> >>>>> The prelink [1] itself assigns some virtual addresses to all required
> >>>>> shared objects (in our case for /sbin/init), so no clashes are
> >>>>> encountered.
> >>>>>
> >>>>> And using prelink is a _default_ behaviour in Yocto/OE poky distro.
> >>>>
> >>>> Does it work without prelink? Also, does it fail with prelink in real
> >>>> hardware?
> >>>>
> >>>> It indeed might be a prelink issue in fact.
> >>>
> >>> This will fail everywhere if prelink is used.
> >>
> >> i thought the point of prelinking is that the vaddr in
> >> the elf image is the runtime address so you don't have
> >> to process relative relocs or adjust pointers in the
> >> dynamic array with += l_addr at all (and then l_addr
> >> does not have to be correct, ld.so would still work).
> >>
> >> but here ld.so is loaded to some random offset on top
> >> of the static prelink offset. is this expected?
> >> does it make sense to prelink ld.so this way?
> >
> > ld.so is loaded by the kernel.  It makes no sense to
> > prelink ld.so.
> >
>
> The question is whether we consider this a regression or if
> we don't support it and instruct consumers to not do it.

I consider it a pilot error.  We can add a check in ld.so
to error out if it has been prelinked.

-- 
H.J.