From mboxrd@z Thu Jan 1 00:00:00 1970 From: martin.petersen@oracle.com (Martin K. Petersen) Date: Tue, 27 Feb 2018 22:46:32 -0500 Subject: [RFC PATCH] nvme-pci: Bounce buffer for interleaved metadata In-Reply-To: <5662c6d9-0c87-6074-12b8-39db53ce3c7f@grimberg.me> (Sagi Grimberg's message of "Sun, 25 Feb 2018 19:30:48 +0200") References: <20180224000547.7252-1-keith.busch@intel.com> <5662c6d9-0c87-6074-12b8-39db53ce3c7f@grimberg.me> Message-ID: Sagi, > It makes sense for FC (and few RDMA devices) that already get > interleaved metadata from the wire to keep it as is instead of > scattering it if the backend nvme device supports interleaved mode... Yeah, assuming that the PI doesn't have to get translated. > I would say that this support for this is something that belongs in > the block layer. IIRC mkp also expressed interest in using > preadv2/pwritev2 to for user-space to use DIF with some accounting on > the iovec Indeed. > so maybe we can add a flag for interleaved metadata. What would the use case be for this? Userspace target driver? I know lots of widgets that rely on interleaved but they are all using SPDK. -- Martin K. Petersen Oracle Linux Engineering