From mboxrd@z Thu Jan  1 00:00:00 1970
Content-Type: multipart/mixed; boundary="===============3595467992178937142=="
MIME-Version: 1.0
From: Mark Wielaard <mjw@redhat.com>
To: elfutils-devel@lists.fedorahosted.org
Subject: Re: [PATCH] libebl: Add ebl_unwind_ret_mask.
Date: Tue, 17 Jun 2014 08:56:02 +0200
Message-ID: <1402988162.4374.4.camel@bordewijk.wildebeest.org>
In-Reply-To: 20140616222329.1BA782C397B@topped-with-meat.com

--===============3595467992178937142==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

On Mon, 2014-06-16 at 15:23 -0700, Roland McGrath wrote:
> What does "actual return value" mean?  With the low bit intact, it is the
> actual value that must be used in a bx instruction.  So why is this a new
> hook rather than being the same as the "func_value" hook?

Good question. They are indeed used for the exact same thing. I just
didn't realize they were till now. Somehow I was working on the function
table issue and the unwind issue separately. It certainly makes sense to
have just one hook. Especially because ARM is the only backend using
them at the moment. I'll rework both patches/hooks into one.

Thanks,

Mark


--===============3595467992178937142==--