On Mon, 8 Feb 2016, Eric Botcazou wrote:
> > Not sure what to make of this comment ;)
>
> I guess it was a variant of the usual can-of-worms-opening thing. :-)
>
> > Did you see where we create the SSA name with the overaligned type?
>
> In tree-ssa-pre.c:insert_into_preds_of_block:
>
> (gdb) p debug_pre_expr(expr)
> {mem_ref<0B>,yyvsp_1}@.MEM_2
>
> (gdb) p debug_tree(type)
> <pointer_type 0x7ffff6d335e8
> type <record_type 0x7ffff6cea498 Tnode type_0 BLK
> size <integer_cst 0x7ffff6c43678 constant 0>
> unit size <integer_cst 0x7ffff6c43618 constant 0>
> align 8 symtab 0 alias set -1 canonical type 0x7ffff6cea3f0 context
> <translation_unit_decl 0x7ffff6d2e0f0 D.1453>
> pointer_to_this <pointer_type 0x7ffff6ceac78> chain <type_decl
> 0x7ffff6c54a18 D.1410>>
> sizes-gimplified unsigned SI
> size <integer_cst 0x7ffff6c435e8 type <integer_type 0x7ffff6c3c150
> bitsizetype> constant 32>
> unit size <integer_cst 0x7ffff6c43600 type <integer_type 0x7ffff6c3c0a8
> sizetype> constant 4>
> align 64 symtab 0 alias set -1 canonical type 0x7ffff6cead20>
>
> /* Now build a phi for the new variable. */
> temp = make_temp_ssa_name (type, NULL, "prephitmp");
Yes, that place I just fixed. I mean for the SRA case.
Richard.
