From mboxrd@z Thu Jan  1 00:00:00 1970
Content-Type: multipart/mixed; boundary="===============9052203325938734276=="
MIME-Version: 1.0
From: Roland McGrath <roland@hack.frob.com>
To: elfutils-devel@lists.fedorahosted.org
Subject: Re: [PATCH] Do without on-stack variable length arrays.
Date: Fri, 04 Sep 2015 15:14:22 -0700
Message-ID: <20150904221423.177EA2C3B1E@topped-with-meat.com>
In-Reply-To: CAEk9T_3L+5XA4=jGBtw78-0+O97Su_VN=_sKk8pr78HKr_F1Gg@mail.gmail.com

--===============9052203325938734276==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

> Is there some simple way that I can test elfutils with fortify or bound
> checking tools?

Depending on your system and build setup, _FORTIFY_SOURCE might be the
default.  It's easy enough to enable it with CPPFLAGS=3D-D_FORTIFY_SOURCE=
=3D2
to configure AFAIK.

> How about using pointers and malloc or alloca, and indexing through the
> pointers? That should avoid complaints from stronger bound checkers,
> although static bound checking probably won't find bound error either.

Right, that's what I think it will have to be.  The cleanest way to do the
32/64 variant switching is not immediately clear to me in the abstract,
though in each particular instance you'll probably be able to decide easily
enough what feels clean.

> Or is there other pattern preferred by elfutils?

I don't think we've figured anything out.

--===============9052203325938734276==--