On Mon, Nov 05, 2018 at 12:50:50PM +0100, Johannes Thumshirn wrote: > On 05/11/2018 11:55, Ming Lei wrote: > > On Mon, Nov 05, 2018 at 11:23:01AM +0100, Johannes Thumshirn wrote: > >> With drivers like iSer we are seeing a lot of bio splitting and smaller I/Os > >> being submitted to the driver. > >> > >> The root cause of this issue that the virtual boundary mask code does not take > >> into consideration that some of the memory segments in the SG list may have > >> come from a huge memory page that is being managed in the SG list as 4K > > > > I guess you mean something like 64K PAGE_SIZE, instead of huge page. > > No I mean like a 2M page from an upper layer. If you mean the real huge page, this patch shouldn't have made a difference because bio_vec->bv_offset is in [0, PAGE_SIZE), and iSer sets virt boundary as 4K - 1. However, things will change after multipage bvec is introduced. Thanks, Ming
