From mboxrd@z Thu Jan 1 00:00:00 1970 From: sagi@grimberg.me (Sagi Grimberg) Date: Thu, 31 May 2018 00:37:33 +0300 Subject: [PATCH v3 1/3] nvme-rdma: correctly check for target keyed sgl support In-Reply-To: <8da47cd4-44d5-2229-ef82-26d165dfc245@opengridcomputing.com> References: <14063C7AD467DE4B82DEDB5C278E8663B38EE822@FMSMSX108.amr.corp.intel.com> <3acbea43-d777-88a5-0059-10275655d545@opengridcomputing.com> <8da47cd4-44d5-2229-ef82-26d165dfc245@opengridcomputing.com> Message-ID: >>> I can see that the 2 and 20 are defined for specific things. Since they are >>> used in several places (in the next 2 patches), is there any chance these >>> could be defined bits? >>> >>> Mike >>> >> That seems reasonable.? I would think these defines could also be shared >> across the host and target.? Perhaps include/linux/nvme.h?? I see both >> nvme/host/nvme.h and nvme/target/nvmet.h include linux/nvme.h.? So they >> could go there. >> >> Christoph, what do you think?? It seems like these are either nvme >> protocol bits or nvme/f protocol bits... >> >> Steve. >> >> > > It looks like these bits are part of the Identify Controller Data > Structure, figure 90 in the 1.2.1 spec.? The SGL Support (SGLS) field > within that struct.?? So I guess it belongs in include/linux/nvme.h? > But I don't see any of this big structure in that header. > Yea, good idea, though please add the offset type sgl as well.