Re: BUG elf32-i386 R_386_PC32 done wrong

[Eric Botcazou]

> The ABI states linkers cannot make -4 themselves, they have to
> read it from a file?  Heck, let's break it!

Come on, be serious, the ABI gives the reloc's formula, which de facto 
prescribes assembler's and linker's behaviour.

> I gave the current definition in my previous emails (the only
> valid one is based on how microprocessors work) and yet ld fails
> to link as stated above.  [hint: microprocessors don't know what
> we are saying or what ABI says; ld has no need whatsoever to get
> -4 from input files; ld writers should know that.]

There is no "valid one".  The ABI is a convention, agreed upon by OS and 
compiler vendors.  You don't like it, fine, but it's here to stay.

> ABI again?  Are you saying ABI doesn't know how to do rel
> relocs?  Again, the location must contain the offset to the
> symbol from the current contents of the CPU eip register.  Are
> you saying ABI contradicts that?

Yes, it's not the formula of R_386_PC32.  Read the docs on the Sun site.

-- 
Eric Botcazou