This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Immediate uses checked in
- From: Andrew MacLeod <amacleod at redhat dot com>
- To: Kazu Hirata <kazu at cs dot umass dot edu>
- Cc: gcc-patches <gcc-patches at gcc dot gnu dot org>, Diego Novillo <dnovillo at redhat dot com>
- Date: Fri, 08 Apr 2005 13:58:14 -0400
- Subject: Re: Immediate uses checked in
- References: <1112728031.3733.3.camel@pain> <20050408.134617.127197700.kazu@cs.umass.edu>
On Fri, 2005-04-08 at 13:46, Kazu Hirata wrote:
> Hi Andrew,
>
> > I have just checked in the integrated immediate uses patch.
>
> Would you care to document the new update_stmt? A quick grep for
> modify_stmt shows:
>
> doc/tree-ssa.texi:@findex modify_stmt
> doc/tree-ssa.texi:modified by a call to @code{modify_stmt}. So, if your pass replaces
> doc/tree-ssa.texi:operands in a statement, make sure to call @code{modify_stmt}.
> tree-cfg.c:/* A list of all the noreturn calls passed to modify_stmt.
>
> If you could also document what BREAK_FROM_SAFE_IMM_USE is for either
> somewhere in the source code or in some texi, that would be great. I
> am a bit worried that somebody might just use "break;" instead of
> BREAK_FROM_SAFE_IMM_USE.
>
yeah, I should upate the documentation.
If someone uses 'break', they will get an abort in verify_ssa at the end
of their optimization when the placeholder node is detected. They won't
make that mistake for long :-)
> Or maybe we could put a link to your post?
>
> http://gcc.gnu.org/ml/gcc-patches/2005-04/msg00164.html
>
> which I find very useful. (But then "refer to this message for how to
> use ..." sounds a bit ugly.)
I should also document the entire immediate uses interface.
Andrew