Hi,
is there any documentation or paper somewhere that describes in more details
the points-to analysis module of CIL ?
In particular, I'm interested in the differences between Olf and Golf with
respect to performance, and the possible ways to tune the analysis to make
it faster.
Thanks,
Olivier
icial distribution.
Any opinion on that ?
Olivier
On Tue, Jun 22, 2010 at 11:27 AM, Gabriel Kerneis wrote:
> Marco,
>
> On Tue, Jun 22, 2010 at 11:06:58AM +0200, Marco Trudel wrote:
> > Oh, actually Olivier Crameri already reported that two months ago...
> > Does this mean th
I had to deal with the same exception a while ago, and if I remember well,
this happens when simplemem tries to simplify an expression that's outside
of a function body.
Creating temporary variables does not work in that context, as those are
meant to be local variables. If you can paste the C cod
Hi,
did something change with CIL's repository ?
It used to work fine, but I'm now getting the following error:
Acil/Bootstrap
Acil/install-sh
U cil
Fetching external item into 'cil/ocamlutil'
cram...@hal.cs.berkeley.edu's password:
svn: Connection closed unexpectedly
Thanks,
Olivie
I just found this email with the exact patch I was looking for. Thank
you Gabriel, this is very helpful !
Does anyone know if it's going to be included in the official CIL distribution ?
Olivier
On Fri, Jan 15, 2010 at 6:22 PM, Gabriel Kerneis wrote:
> A bit hackish, but it has to since atomic
Hi,
in my CIL module I'm trying to keep track precisely of the line
numbers for the initialization code of the globals (like arrays or
structures).
I'm using for that the cilVisitor and in particular the vinit
function. However, if I'm not mistaken, the only "location" available
in this context i
Hi,
if this helps, there is a good example of how to do this in a module
included with CIL. It catches calls to malloc, calloc and such. The
module is in src/ext/sfi.ml
Best regards,
Olivier
On Fri, May 29, 2009 at 11:10 PM, Mark Hills wrote:
> Divya,
>
> I'm not giving too much detail, but hop
(fun fi -> isCompleteType
fi.ftype) comp.cfields
| _ -> true
Regards,
Olivier
On Mon, Apr 6, 2009 at 10:19 PM, Elnatan Reisner wrote:
> I think you are looking for the 'cdefined' field of Cil.compinfo.
>
> On Mon, 2009-04-06 at 15:44 +0200, Olivier Crameri wrote:
>> H
Hi all,
I've been having a few problems with incomplete types in CIL.
Apparently, Cil.isCompleteType considers a structure with no fields as
being a complete type. I don't know if this is on purpose, or if this
is a bug. It seems to be valid C, but I can't see any use of having a
structure define
