On 10/24/2013 09:11 AM, Richard Biener wrote:
On Thu, Oct 24, 2013 at 3:06 PM, Andrew MacLeod <amacl...@redhat.com> wrote:
I originally shuffled std_gimplify_va_arg_expr () to gimplify.c, but i
believe the suggestion was to eventually move it to targhooks.c.
build_va_arg_indirect_ref seems to hang around with it, and used by all the
same things that use std_gimplify_va_arg_expr, so this patch currently
lumps it in targhooks.c as well (which means they will get it with the same
header file). Really, I suppose it should go elsewhere.. and I'd guess
tree.c... but its nots 100% obvious... suggestions welcome. I'd think
someday this file will get split into gimple-targhooks and rtl-targhooks,
but not today i think. :-)
It also seems appropriate to move gimplify_va_arg_expr from builtins.c to
gimplify.c since its a gimplification routiine.. This showed up as I was
splitting gimple.h into gimplify.h... the prototype is in gimple.h and it
made sense to become part of gimplify.h as I try to move towards an
interface from front-ends that only requires gimplfy.h and not gimple.h or
other middle end things.
bootstraps on x86_64-unknown-linux-gnu with no new regressions. OK?
Ok.
Err ...
Index: targhooks.c
===================================================================
*** targhooks.c (revision 203915)
--- targhooks.c (working copy)
*************** along with GCC; see the file COPYING3.
*** 71,76 ****
--- 71,77 ----
#include "tree-ssa.h"
#include "tree-ssa-alias.h"
#include "insn-codes.h"
+ #include "tree-mudflap.h"
we were supposed to remove mudflap for 4.9, no?
Really? I guess it hasn't been removed yet since the include is still
there? who is doing that?
so I guess should remove this bit too then..?
tree
build_va_arg_indirect_ref (tree addr)
{
addr = build_simple_mem_ref_loc (EXPR_LOCATION (addr), addr);
- if (flag_mudflap) /* Don't instrument va_arg INDIRECT_REF. */
- mf_mark (addr);
-
return addr;
}