[PATCH] ELF/ARM: Ignore R_ARM_V4BX but allow linking

[Adhemerval Zanella]

This was my very question in the first patch about it I sent:

I am not certainly that it is the right approach to handle this cases in arch-neutral
code, so I am more than willing to hear better suggestions.

And I am not aware of a similar relocation with same constraints,
so I am not sure which model will fit better.  Some options I can
think is to make it reference to itself or to an special symbol.

On 15-04-2015 17:25, Rui Ueyama wrote:
> It feels to me that creating a dangling reference having a null pointer is wrong in the first place. I think in most places we don't expect a reference having a null pointer. Conceptually it's a directed edge of a graph, so there must be a vertex when you follow an edge.
> 
> Does this really need to be treated that way? Is there any other way to handle this in the same manner as other symbols/references?
> 
> 
> http://reviews.llvm.org/D9035
> 
> EMAIL PREFERENCES
>   http://reviews.llvm.org/settings/panel/emailpreferences/
> 
>