Linus has a very nice write up about __nocast vs __bitwise. Can any one help to integrate the write up to a patch against man page sparse.1? I would much rather hitting the apply button than hacking documents. But if nobody wants to do it, it would have to be me then. Yes, I am a lazy baster :-) Thanks Chris ---------- Forwarded message ---------- From: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Date: Thu, Mar 22, 2012 at 3:08 PM Subject: Re: [PATCH 00/16] mm: prepare for converting vm->vm_flags to 64-bit To: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>, Konstantin Khlebnikov <khlebnikov@xxxxxxxxxx>, Minchan Kim <minchan@xxxxxxxxxx>, "linux-mm@xxxxxxxxx" <linux-mm@xxxxxxxxx>, "linux-kernel@xxxxxxxxxxxxxxx" <linux-kernel@xxxxxxxxxxxxxxx>, Hugh Dickins <hughd@xxxxxxxxxx>, KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>, Ben Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>, "linux@xxxxxxxxxxxxxxxx" <linux@xxxxxxxxxxxxxxxx> On Thu, Mar 22, 2012 at 2:41 PM, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: >> >> Use __bitwise for that - check how gfp_t is handled. > > So what does __nocast do? __nocast warns about explicit or implicit casting to different types. HOWEVER, it doesn't consider two 32-bit integers to be different types, so a __nocast 'int' type may be returned as a regular 'int' type and then the __nocast is lost. So "__nocast" on integer types is usually not that powerful. It just gets lost too easily. It's more useful for things like pointers. It also doesn't warn about the mixing: you can add integers to __nocast integer types, and it's not really considered anything wrong. __bitwise ends up being a "stronger integer separation". That one doesn't allow you to mix with non-bitwise integers, so now it's much harder to lose the type by mistake. So basic rules is: - "__nocast" on its own tends to be more useful for *big* integers that still need to act like integers, but you want to make it much less likely that they get truncated by mistake. So a 64-bit integer that you don't want to mistakenly/silently be returned as "int", for example. But they mix well with random integer types, so you can add to them etc without using anything special. However, that mixing also means that the __nocast really gets lost fairly easily. - "__bitwise" is for *unique types* that cannot be mixed with other types, and that you'd never want to just use as a random integer (the integer 0 is special, though, and gets silently accepted iirc - it's kind of like "NULL" for pointers). So "gfp_t" or the "safe endianness" types would be __bitwise: you can only operate on them by doing specific operations that know about *that* particular type. Generally, you want __bitwise if you are looking for type safety. "__nocast" really is pretty weak. Linus -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
