On Tue, Aug 3, 2021 at 10:57 PM Al Viro [off-list ref] wrote:

On Tue, Aug 03, 2021 at 09:18:09PM +0200, Andreas Gruenbacher wrote:

quoted

Turn fault_in_pages_{readable,writeable} into versions that return the number
of bytes faulted in instead of returning a non-zero value when any of the
requested pages couldn't be faulted in.  This supports the existing users that
require all pages to be faulted in, but also new users that are happy if any
pages can be faulted in.

Neither of these functions is entirely trivial and it doesn't seem useful to
inline them, so move them to mm/gup.c.

Rename the functions to fault_in_{readable,writeable} to make sure that code
that uses them can be fixed instead of breaking silently.

Sigh...  We'd already discussed that; it's bloody pointless.  Making short
fault-in return success - absolutely.  Reporting exact number of bytes is
not going to be of any use to callers.

I'm not actually convinced that you're right about this. Let's see
what we'll end up with; we can always strip things down in the end.

Thanks,
Andreas