Re: support array bounds checking

[kthomas at gwdg dot de (Philipp Thomas)]

On Tue, 27 Apr 1999 10:47:52 -0700, you wrote:

>It might be a good starting point.  But we need to get the correct 
>copyright assignments. That might be difficult at this late stage.

Nope, it isn't ;-) I asked Richard and he said it would be no problem as
it would be fine with him to get them into egcs. Following his request I
sent the changes disclaimer to his work address and he said he'd mail it
from there.
What might be a problem is getting a disclaimer from his employer/school =
at that time though. I've mailed Richard on that point.

>> I know it's entirely non-standard to support array bounds
>> checking in C/C++, but why not do it anyway? (I'm not
>> volunteering though ;-)... I have exams to worry about).
>
>Because somebody needs to volunteer.  (Or maybe the Cygnus sales
>staff needs to sell someone on paying for it ...)

This might be easier then you think ;-) Richard pointed out to me that
Herman ten Brugge <Haj.Ten.Brugge@net.HCC.nl> did port the patches to gcc
2.8.1 and egcs 1.1.2. Herman's patches are available from
http://web.inter.nl.net/hcc/Haj.Ten.Brugge. As his web pages says that he's
working on TI dsp support for the GNU tool chain, the FSF should have his
disclaimer, but I'll ask him.=20

I've downloaded the egcs 1.1.2 patches and will see what - if any - needs
changed to apply them to the current cvs tree. As my future disclaimer for gcc
is on its snail way to the FSF, I don't see any more obvious obstacles.
That is, besides the question of whether that code will get in, of cause.

Richard has stopped maintaining the gcc patches and is looking for someone
to take over from him and to look through patches/mail he archived. So if I
find myself fit enough to cope with that code, I will take over the maintainer
task if the patches are accepted for egcs inclusion (which of cause is a
definitive post 1.2 branching topic).

Philipp Thomas

-- 
caffeine low .... brain halted