TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy

TEAM-ADA@LISTSERV.ACM.ORG

Options: Use Classic View

Use Proportional Font
Show HTML Part by Default
Condense Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Sender: "Team Ada: Ada Advocacy Issues (83 & 95)" <[log in to unmask]>
Date: Mon, 10 Jul 2000 16:12:19 -0400
Reply-To: Michael Feldman <[log in to unmask]>
From: Michael Feldman <[log in to unmask]>
Content-Transfer-Encoding: 7bit
In-Reply-To: <[log in to unmask]> from "Tucker Taft" at Jul 10, 2000 03:08:40 PM
Content-Type: text/plain; charset=us-ascii
MIME-Version: 1.0
Parts/Attachments: text/plain (32 lines)
[said Tucker]
>
> Our Ada 95 compiler for the SPARC does not require the stack to be
> executable.  That implies that the one from Green Hills and the one
> from Aonix do not either.  In fact, I suspect that GNAT is
> one of the few Ada 95 compilers that does put executable code on
> the stack.  I believe it does this because GCC uses "trampolines"
> to support pointers to nested procedures.
>
> -Tucker Taft   [log in to unmask]
>
We (GWU) and a few other educational sites have had this problem.
Our sysadmins have insisted on disabling stack execution. ACT has
provided a patch for earlier GNAT versions that would enable you
to rebuild the compiler from sources such that no trampolining/
stack execution is done. This makes both the compiler and the code
it generates compliant with the security fix.

As I recall, ACT actually sent me a patched Solaris binary, so I
didn't have to rebuild it myself.

ACT has told me they were preparing a "better" solution for 3.13,
but I have not seen it to date. You might want to ask ACT if you can
get a copy of the patched 3.12p, and tell them you are interested in
the forthcoming solution for 3.13.

Actually, you might want to re-post your question to [log in to unmask]
to open the issue again, and maybe even send it as a bug report
to [log in to unmask]

Mike Feldman

ATOM RSS1 RSS2