On Thu, 25 Apr 2002, Neeraj wrote: > So, what would be your inference in the case of architectures where the null > pointer is not designated by 0(zero) (lot of them are mentioned there on the > C-faq page). What would then ((struct a *)0L)->field refer to and how will the > compiler know that (null-pointer)->field is required to evaluate the offset of > field in the structure "a" ? > > Any suggestions or hints are welcome. :) I'm guessing that a compiler would see that it's an offset from 0, and hence know that something 8 bytes from the start of the struct returns an 8. However, I'll admit that I may be wrong, and I don't have an architecture that I could test this on. In these cases (where I don't really know for sure) I always use some safe code: (void *)&((struct a*)0L)->field - (void *)(struct a*)0L which is similar to what you suggested in an earlier post. Note that before the (void*) cast, the first expression (the address of "field") is a pointer to the type of "field", while the second is to something of type struct. You cast to (unsigned long) to compare them, but I prefer to use pointer arithmetic (subtracting 2 pointers of the same type is perfectly legal, and since the addressed objects are different I use (void*)). Regards, Paul Gearon Software Engineer Telephone: +61 7 3876 2188 Plugged In Software Fax: +61 7 3876 4899 http://www.PIsoftware.com PGP Key available via finger Catapultam habeo. Nisi pecuniam omnem mihi dabis, ad caput tuum saxum immane mittam. (Translation from latin: "I have a catapult. Give me all the money, or I will fling an enormous rock at your head.") -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
