On 10/18/2023 8:48 PM, Keith Busch wrote: > From: Keith Busch <kbusch@xxxxxxxxxx> > > Map user metadata buffers directly instead of maintaining a complicated > copy buffer. > > Now that the bio tracks the metadata through its bip, nvme doesn't need > special metadata handling, callbacks, or additional fields in the pdu. > This greatly simplifies passthrough handling and avoids a "might_fault" > copy_to_user in the completion path. This also creates pdu space to > track the original request separately from its bio, further simplifying > polling without relying on special iouring fields. > > The downside is that nvme requires the metadata buffer be physically > contiguous, so user space will need to utilize huge pages if the buffer > needs to span multiple pages. I did not notice where this downside part is getting checked in the code.
