On 9/7/14, 2:23 PM, Nico Weber wrote:
On Wed, Sep 3, 2014 at 6:25 AM, Jonathan Roelofs <[email protected]
<mailto:[email protected]>> wrote:
On 9/3/14, 5:09 AM, Renato Golin wrote:
On 3 September 2014 01:02, Antoine Labour <[email protected]
<mailto:[email protected]>> wrote:
Where in the EHABI spec is that defined? AFAICT all it says in
section 6.2
is that there's data for the PR, but it doesn't say anything about
the
contents of that data.
I concur. The EHABI doesn't define what's in the personality data
block, let alone that the length must be the first byte. I think these
are premature assumptions.
Hmm, re-reading those sections, I see that it doesn't. I suppose all that
can be done is to check that the prel31 points at __gxx_personality_v0, and
fail to unwind otherwise.
Did this happen? If not, should we back out the change until someone has time to
implement this correctly?
Not yet. I'll implement it at some point tomorrow, and send it here for review.
This was already broken before this patch, so I'm not so worried about backing
it out.
Jon
Jon
cheers,
--renato
_________________________________________________
cfe-commits mailing list
[email protected] <mailto:[email protected]>
http://lists.cs.uiuc.edu/__mailman/listinfo/cfe-commits
<http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits>
_________________________________________________
cfe-commits mailing list
[email protected] <mailto:[email protected]>
http://lists.cs.uiuc.edu/__mailman/listinfo/cfe-commits
<http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits>
--
Jon Roelofs
[email protected]
CodeSourcery / Mentor Embedded
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
