Re: RFH: MIPS ld breaking .eh_frame data ...

[David Daney <ddaney at avtrex dot com>]

Richard Sandiford wrote:
> David Daney <ddaney@avtrex.com> writes:
>   
> > Is is sufficient to just prevent the linker from converting to the PC 
> > relative augmentation?  If so I may prepare a patch to do that.
> >     
>
> I was responsible for the guilty linker code, so I'll volunteer to fix it.
> Hopefully this weekend...
>
> Richard
>   
Thanks Richard,

It seems that by using PC relative augmentation, the number of 
relocations in a shared object (in .eh_frame) is probably greatly 
reduced.  Is this the reason for converting to PC relative addressing?

From looking at the code it looks like the FDE header is being properly 
relocated, all that needs to be done is to fix any DW_CFA_set_loc 
offsets as well and it would work.

Is this the fix you are thinking of?

If we had just kept allowing gas to emit relocations for the distance 
between symbols in different sections, GCC could generate the PC 
relative form directly and ld would not have to do any of this...

Thanks again,
David Daney