On Tue, Oct 18, 2016 at 01:09:24PM +0200, Eric Botcazou wrote:
> > > No, "mode" is the mode of the MEM, not that of the SYMBOL_REF.
> > I still don't see it, could you explain a bit more?
> MODE is the mode of operands before:
> operands = force_const_mem (mode, operands);
> and after. But the test is on the address of the MEM, not on the MEM itself:
> if (TARGET_TOC
> && GET_CODE (XEXP (operands, 0)) == SYMBOL_REF
> && use_toc_relative_ref (XEXP (operands, 0), Pmode))
> because it's the mode of SYMBOL_REF we are interesting in (and
> guarantees that it's Pmode).
We need to pass the mode of the actual datum we would put in the TOC to
the use_toc_relative_ref function, not the mode of its address.
I must be missing something...