On Wed, Nov 29, 2023 at 08:48:41PM +0530, Kanchan Joshi wrote: > On 11/29/2023 3:57 AM, Keith Busch wrote: > > If the user address can't directly be used for reason, like too many > > segments or address unalignement, fallback to a copy of the user vec > > while keeping the user address pinned for the IO duration so that it > > can safely be copied on completion in any process context. > > The pinning requirement is only for read. But code keeps user-memory > pinned for write too. Is there any reason? It just makes the completion simpler. I'll split the cases so we unpin on writes after the copy during setup.
