On Fri, Oct 11, 2019 at 03:14:09PM +0100, Richard Sandiford wrote:
> Marek Polacek <pola...@redhat.com> writes:
> > On Thu, Oct 10, 2019 at 08:00:53PM +0100, Richard Sandiford wrote:
> >> Ping
> >>
> >> Richard Sandiford <richard.sandif...@arm.com> writes:
> >> > The current aka diagnostics can sometimes leak internal details that
> >> > seem more likely to be distracting than useful. E.g. on aarch64:
> >> >
> >> > void f (va_list *va) { *va = 1; }
> >> >
> >> > gives:
> >> >
> >> > incompatible types when assigning to type ‘va_list’ {aka ‘__va_list’}
> >> > from type ‘int’
> >> >
> >> > where __va_list isn't something the user is expected to know about.
> >> > A similar thing happens for C++ on the arm_neon.h-based:
> >> >
> >> > float x;
> >> > int8x8_t y = x;
> >> >
> >> > which gives:
> >> >
> >> > cannot convert ‘float’ to ‘int8x8_t’ {aka ‘__Int8x8_t’} in
> >> > initialization
> >> >
> >> > This is accurate -- and __Int8x8_t is defined by the AArch64 PCS --
> >> > but it's not going to be meaningful to most users.
> >
> > Agreed.
> >
> >> +/* Return true if it is worth exposing the DECL_ORIGINAL_TYPE of TYPE to
> >> + the user in diagnostics, false if it would be better to use TYPE
> >> itself.
> >> + TYPE is known to satisfy typedef_variant_p. */
> >> +
> >> +bool
> >> +user_facing_original_type_p (const_tree type)
> >> +{
> >> + gcc_assert (typedef_variant_p (type));
> >> + tree decl = TYPE_NAME (type);
> >> +
> >> + /* Look through any typedef in "user" code. */
> >> + if (!DECL_IN_SYSTEM_HEADER (decl) && !DECL_IS_BUILTIN (decl))
> >> + return true;
> >> +
> >> + /* If the original type is also named and is in the user namespace,
> >> + assume it too is a user-facing type. */
> >> + tree orig_type = DECL_ORIGINAL_TYPE (decl);
> >> + if (tree orig_id = TYPE_IDENTIFIER (orig_type))
> >> + {
> >> + const char *name = IDENTIFIER_POINTER (orig_id);
> >> + if (name[0] != '_' || (name[1] != '_' && !ISUPPER (name[1])))
> >> + return true;
> >
> > This looks like name_reserved_for_implementation_p.
> >
> > The rest looks fine to me!
>
> Ah, nice! I'd looked for a helper but missed that one.
>
> Here's just the C parts, with that change. Tested on aarch64-linux-gnu
> and x86_64-linux-gnu. OK to install?
Ok, thanks!
--
Marek Polacek • Red Hat, Inc. • 300 A St, Boston, MA
