This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: __nodebug__ attribute for use on SSE intrinsic wrappers
- From: Geoff Keating <geoffk at apple dot com>
- To: Devang Patel <dpatel at apple dot com>
- Cc: Mike Stump <mrs at apple dot com>, Daniel Jacobowitz <drow at false dot org>, Fariborz Jahanian <fjahanian at apple dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 29 Jul 2005 14:39:26 -0700
- Subject: Re: PATCH: __nodebug__ attribute for use on SSE intrinsic wrappers
- References: <78E4E2DA-0070-11DA-BCD4-003065BDF310@apple.com> <7815E539-C34C-4716-9F4D-B5F3011B5744@apple.com>
On 29/07/2005, at 1:45 PM, Devang Patel wrote:
We're converging towards one solution (that is, support new
attribute, decorate intrinsic headers, subtract in stabs and use
DW_AT_ARTIFICIAL in DWARF). Now, once 1) one of the GWP OK it in
principal and 2) we select name of new attribute, Fariborz (or I)
can prepare updated patch.
I think this is OK in principle, and there seems to be agreement on
this (somewhat violent agreement, but I'm pretty sure it's
agreement). I'm not sure if DW_AT_artificial is an exact match, but
the only alternative I see would be a GNU-specific attribute, and I
think we should try DW_AT_artificial before going to such lengths.
Marking these functions as DECL_ARTIFICIAL has other consequences,
but I don't *think* any of them are undesirable, so I suggest trying
that as an implementation technique.
When you document the new attribute, I'd suggest saying that it
should only be used in header files that are part of the compiler, or
something like that.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature