This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Fix handling of calls wrt EH
- From: Richard Henderson <rth at redhat dot com>
- To: Jan Hubicka <jh at suse dot cz>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 13 May 2005 11:22:11 -0700
- Subject: Re: Fix handling of calls wrt EH
- References: <20050513032654.GF18936@kam.mff.cuni.cz>
On Fri, May 13, 2005 at 05:26:54AM +0200, Jan Hubicka wrote:
> + /* ??? For the benefit of calls.c, converting all this to rtl,
> + we need to record the call expression, not just the outer
> + modify statement. */
> + if ((TREE_CODE (t) == MODIFY_EXPR
> + || TREE_CODE (t) == RETURN_EXPR)
> + && (t = get_call_expr_in (t)))
> + add_stmt_to_eh_region_fn (ifun, t, num);
If we're going to do this, I think we should rearrange add_stmt_eh_region
and record_stmt_eh_region (and their callers) such that we only do this in
one place.
And did you really need to check RETURN_EXPR? We're not testing for that
in lower_eh_constructs_1. Either your test is unnecessary or we've got a
latent bug.
r~