On Thu, 2018-07-19 at 16:53 +0200, Christoph Hellwig wrote: > On Thu, Jul 19, 2018 at 12:08:41PM +0100, Al Viro wrote: > > > Well, there has never been a promise that it will grab *all* > > > pages in the > > > iter AFAIK. Practically, I think that it was just too hairy to > > > implement in > > > the macro magic that iter processing is... Al might know more > > > (added to > > > CC). > > > > Not really - it's more that VM has every right to refuse letting > > you pin > > an arbitrary amount of pages anyway. > > In which case the code after this patch isn't going to help either, > because > it still tries to pin it all, just in multiple calls to > get_user_pages(). That was not the point of the patch. It's not about a situation in which MM refuses to pin more pages. The patch is about the "iterator::next()" nature of bio_iov_iter_get_pages(). If it can't pin the pages, bio_iov_iter_get_pages() returns an error code (elsewhere in this thread we discussed how to treat that right). Otherwise, it always adds _some_ data to the bio, but the amount added depends on the segment structure of the input iov_iter. If the input iovec has just a single segment, it fills the bio in a single call. With multiple segments, it just returns the page(s) of the first segment. The point of my patch is to make no difference between single- segment and multi-segment IOs. Regards Martin -- Dr. Martin Wilck <mwilck@xxxxxxxx>, Tel. +49 (0)911 74053 2107 SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg)
