This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Small SH fix
- From: Kaz Kojima <kkojima at rr dot iij4u dot or dot jp>
- To: bonzini at gnu dot org
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 29 Oct 2010 13:06:50 +0900 (JST)
- Subject: Re: [PATCH] Small SH fix
- References: <1288294147-32401-3-git-send-email-bonzini@gnu.org>
Paolo Bonzini <bonzini@gnu.org> wrote:
> I found this by inspection; I haven't tested it but it is obviously
> the right thing to do here to handle for_return == 2.
>
> Index: gcc/config/sh/sh.c
> ===================================================================
> --- gcc/config/sh/sh.c (branch diag)
> +++ gcc/config/sh/sh.c (working copy)
> @@ -8149,12 +8149,13 @@ sh_dwarf_register_span (rtx reg)
> static enum machine_mode
> sh_promote_function_mode (const_tree type, enum machine_mode mode,
> int *punsignedp, const_tree funtype,
> - int for_return ATTRIBUTE_UNUSED)
> + int for_return)
> {
> if (sh_promote_prototypes (funtype))
> return promote_mode (type, mode, punsignedp);
> else
> - return mode;
> + return default_promote_function_mode (type, mode, punsignedp, funtype,
> + for_return);
> }
>
> static bool
Ah, yet another "why did it work previously".
Ok with ChangeLog entry. Thanks for finding this!
Regards,
kaz