[gdal-dev] "Banning" use of underflow/overflow with unsigned integer arithmetic ?

Andrew Bell andrew.bell.ia at gmail.com
Fri Aug 16 11:40:01 PDT 2019


Is this just on when running tests?

On Fri, Aug 16, 2019 at 1:28 PM Even Rouault <even.rouault at spatialys.com>
wrote:

> Hi,
>
> I've a PR ready that might be a bit controversial among contributors as it
> effectively "bans" a well-defined behaviour of C/C++, that is underflow/
> overflow on unsigned integer arithmetic. However there are potential for
> real
> bugs to be hidden by such a behaviour (I or other contributors fixed a few
> instances of that in a few previous commits), so it seems reasonable to
> try to
> avoid it when possible, and when not possible, to tag the functions where
> this
> is explicitly allowed with CPL_NOSANITIZE_UNSIGNED_INT_OVERFLOW
>
> See
> https://github.com/OSGeo/gdal/pull/1782
>
> Even
>
> --
> Spatialys - Geospatial professional services
> http://www.spatialys.com
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/gdal-dev



-- 
Andrew Bell
andrew.bell.ia at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20190816/fe89f9ec/attachment.html>


More information about the gdal-dev mailing list