This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: INSERT_ATTRIBUTES and PRAGMA_INSERT_ATTRIBUTES
- To: Mumit Khan <khan at NanoTech dot Wisc dot EDU>
- Subject: Re: INSERT_ATTRIBUTES and PRAGMA_INSERT_ATTRIBUTES
- From: "Joseph S. Myers" <jsm28 at cam dot ac dot uk>
- Date: Sat, 30 Jun 2001 08:20:11 +0100 (BST)
- cc: <gcc at gcc dot gnu dot org>, <amylaar at redhat dot com>, <aoliva at redhat dot com>, <nickc at redhat dot com>, <scox at redhat dot com>, <m dot hayes at elec dot canterbury dot ac dot nz>, <meissner at redhat dot com>
On Fri, 29 Jun 2001, Mumit Khan wrote:
> Please see if http://gcc.gnu.org/ml/gcc-patches/1999-06n/msg00709.html
> answers your questions. If not, I'll hopefully dig out the info on what
> made the changes necessary.
It explains why that definition of PRAGMA_INSERT_ATTRIBUTES was removed.
Two concrete questions:
* Should PRAGMA_INSERT_ATTRIBUTES go away, or should the mechanism stay in
case it is used again?
* Should SH be using INSERT_ATTRIBUTES instead?
A supplementary question on SET_DEFAULT_DECL_ATTRIBUTES, which does a
similar job to INSERT_ATTRIBUTES: is that mechanism (which applies at
slightly different times to INSERT_ATTRIBUTES, sometimes before, sometimes
after) necessary as well as INSERT_ATTRIBUTES, or could it go away in
favour of INSERT_ATTRIBUTES? Only c4x and v850 define
SET_DEFAULT_DECL_ATTRIBUTES; the c4x definition of the macro takes the
address of its attributes argument, which makes it seem like a badly
specified macro (the attributes argument isn't documented) in the first
place.
--
Joseph S. Myers
jsm28@cam.ac.uk