Re: [PATCH -v2] memdup_user(): introduce

[Roland Dreier <rdreier@cisco.com>]

 > I would like to question the use of the gfp argument here;
 > copy_from_user sleeps, so you can't use GFP_ATOMIC anyway.
 > You can't use GFP_NOFS etc, because the pagefault path will happily do
 > things that are equivalent, if not identical, to GFP_KERNEL.

That's a convincing argument, and furthermore, strndup_user() does not
take a gfp parameter, so interface consistency also argues that the
function prototype should just be

void *memdup_user(const void __user *src, size_t len);

(By the way, the len parameter of strndup_user() is declared as long,
which seems strange, since it matches neither the userspace strndup()
nor the kernel kstrndup(), which both use size_t.  So using size_t for
memdup_user() and possibly fixing strndup_user() to use size_t as well
seems like the sanest thing)

 - R.