On Tue, Apr 27, 2021 at 12:18:44AM +0200, Eric Botcazou wrote:
> > At least sizetype is for GIMPLE compatible with size_t or unsigned long (or
> > whatever unsigned type has the same precision), so this looks wrong to me.
>
> It's for bitsizetype here, as it's a conversion from bits into bytes, so
> larger than DWARF2_ADDR_SIZE.
Yes, but even bitsizetype is undistinguishable from other (usually 2 *
pointer size) precision integral types.
> > For non-strict DWARF or DWARF5 and above, I don't see why we can't use
> > typed DWARF ops instead though, see what e.g. mem_loc_descriptor does for
> > UDIV.
>
> case UDIV:
> if ((!dwarf_strict || dwarf_version >= 5)
> && is_a <scalar_int_mode> (mode, &int_mode))
> {
> if (GET_MODE_SIZE (int_mode) > DWARF2_ADDR_SIZE)
> {
> op = DW_OP_div;
> goto do_binop;
> }
> mem_loc_result = typed_binop (DW_OP_div, rtl,
> base_type_for_mode (int_mode, 1),
> int_mode, mem_mode);
> }
> break;
>
> This also looks incorrect if the size is larger than DWARF2_ADDR_SIZE then,
> isn't there a typo in the code?
>
> if (GET_MODE_SIZE (int_mode) < DWARF2_ADDR_SIZE)
> {
> op = DW_OP_div;
> goto do_binop;
> }
I think you're right.
Jakub
