On Mon, Jan 23, 2023 at 12:35:28PM +0100, David Hildenbrand wrote: > On 17.01.23 16:58, Jason Gunthorpe wrote: > > Now that NULL locked doesn't have a special meaning we can just make it > > non-NULL in all cases and remove the special tests. > > > > get_user_pages() and pin_user_pages() can safely pass in a locked = 1 > > > > get_user_pages_remote) and pin_user_pages_remote() can swap in a local > > variable for locked if NULL is passed. > > > > Remove all the NULL checks. > > > > Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxx> > > --- > > mm/gup.c | 30 ++++++++++++++++++++---------- > > 1 file changed, 20 insertions(+), 10 deletions(-) > > ... doesn't really look like a real cleanup. Especially with the > > 2 files changed, 20 insertions(+), 12 deletions(-) > > on the previous patch and a new internal flag .... > > What's the benefit? There are all kinds of unnecessary branches on the faster paths, inside loops, etc to check for NULL when all we really needed was a single bit flag. It isalos much clearer to understand that a FOLL flag changes the behavior of how GUP works rather than some weirdo NULL argument. Jason
