On Mon, 30 Aug 2010, Vasiliy Kulikov wrote: > Julia, > > On Mon, Aug 30, 2010 at 12:53 +0200, Julia Lawall wrote: > > I can send a patch shortly that will make it possible to parse this file, > > but it seems that it then crashes later. The backtrace is not telling me > > where the problem is, and I'm not sure it is really worth tracking it > > down, because the file seems to be just one enormous array, which is > > probably not very useful to do anything with. > > I guess you are running > > coccinelle with the -dir option, and in the case of this stack overflow, > > it just goes on to the next file? > > No, I used it this way: > > find -name '*.c' | xargs -L 1 spatch ... > > because with -dir it runs enormously slower.
That is quite strange > As I understand, with -dir spatch may process all the files as the > whole, right? No, it should parse the semantic patch one (your approach parses it many times), and then do a find for the .c files, and then apply the already parsed semantic patch to each file. There should not be any accumulated state between files. If it is very slow, then perhaps there is a memory leak somewhere, and that would be something to look into. Is it slow on gdb or glibc specifically? > > If that is the case, perhaps we can > > just leave this one as is. > > This one is not very interesting, but it seems to me that there are a > few poorly written programs those are able to drop coccinelle. IMO it is > not good. > > > If I find such 'faulty' non-array files, I'll tell you. That would be very helpful, thanks. julia _______________________________________________ Cocci mailing list [email protected] http://lists.diku.dk/mailman/listinfo/cocci (Web access from inside DIKUs LAN only)
