Hi, On Wed, Dec 1, 2010 at 4:57 PM, Nicolas Palix <[email protected]> wrote: >> Hi! >> >> Hope I'm right here (or someone can point me in the right direction). >> >> I've been trying to integrate the software analysis tool developed [0] >> here in Erlangen and your herodotos to get some information about >> development of the found errors over time -- I'm currently running the >> lot on a linux kernel as well as the drivers subtree. >> >> While the whole-kernel thing already works fine, I'm having >> trouble passing the "false positives check". As soon as I mark any of >> the findings as 'BUG' make web fails with some error message [1]. As >> all the files in ./results apart from the correl.org have been >> regenerated from scratch and the path in both the *.exist as well as >> the *.new.org seem to match the right existing file (and are the same >> (with the relevant modification) to the working example on the >> whole-kernel side. >> >> Are there some more cache files directly in the source tree I >> don't find? Or do you have some hint for debugging my Problem? I'll >> attach some (hopefully the relevant) files (fragments). > > You should have absolute path in the Org files. Herodotos > is otherwise unable to correctly manipulate the paths.
If you still have trouble after that, can you send me the Herodotos configuration and the Herdotos version you are using. Regards. > >> >> Regards >> >> Christoph Egger >> >> [0] http://www4.cs.fau.de/Research/VAMOS/ >> [1] *** ERROR *** Processing Drivers: >> ./results/drivers/Drivers_CodeUndead.new.org >> Unable to compute file existence information for >> staging/bcm/Misc.c. >> Is the source code accessible or your cache files uptodate? >> _______________________________________________ >> Cocci mailing list >> [email protected] >> http://lists.diku.dk/mailman/listinfo/cocci >> (Web access from inside DIKUs LAN only) >> >> > > > > -- > Nicolas Palix > http://proton.inrialpes.fr/~npalix/ > -- Nicolas Palix http://proton.inrialpes.fr/~npalix/ _______________________________________________ Cocci mailing list [email protected] http://lists.diku.dk/mailman/listinfo/cocci (Web access from inside DIKUs LAN only)
