Re: suspend2 merge (was Re: [Suspend2-devel] Re: CFS and suspend2: hang in atomic copy)

[Linus Torvalds <torvalds@linux-foundation.org>]

On Thu, 26 Apr 2007, Pavel Machek wrote:
> 
> #define SNAPSHOT_ATOMIC_SNAPSHOT	_IOW(SNAPSHOT_IOC_MAGIC, 3, void *)
> #define SNAPSHOT_SET_IMAGE_SIZE		_IOW(SNAPSHOT_IOC_MAGIC, 6, unsigned long)
> #define SNAPSHOT_AVAIL_SWAP		_IOR(SNAPSHOT_IOC_MAGIC, 7, void *)
> #define SNAPSHOT_GET_SWAP_PAGE		_IOR(SNAPSHOT_IOC_MAGIC, 8, void *)
> #define SNAPSHOT_SET_SWAP_FILE		_IOW(SNAPSHOT_IOC_MAGIC, 10, unsigned int)
> #define SNAPSHOT_PMOPS			_IOW(SNAPSHOT_IOC_MAGIC, 12, unsigned int)
> 
> Are these a problem? Do we need to just use u32 as a argument to keep
> ioctl numbers same between 32 and 64bit versions?

No, you need to use the *proper* type as an argument, and assuming that 
type has the same representation in both 32-bit and 64-bit world, the 
numbers will automatically match.

Using "void *" is totally bogus. It's supposed to be the actual argument 
you pass in, not the pointer to it. If your argument doesn't have a 
"struct xyz" kind of format, then you could use "int" (or u32 or 
something: but realistically int is 32-bit for the forseeable future), but 
it's always wrong to pass in "void *" or "unsigned long", since either of 
those are just a sign of the interface being either (a) misunderstood or 
(b) broken.

		Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/