On Tue, Jan 11, 2022 at 01:23:53PM -0500, Jeff Moyer wrote: > Maybe I have bad taste, but the patches didn't look like cruft to me. > :) They do to me. The extend the corner case of request on request stacking that already is a bit of mess even more by adding yet another special case in the block layer. > > I'm not sure why we'd prevent users from using dm-mpath on nvmeof. I > think there's agreement that the nvme native multipath implementation is > the preferred way (that's the default in rhel9, even), but I don't think > that's a reason to nack this patch set. > > Or have I missed your point entirely? No you have not missed the point. nvme-multipath exists longer than the nvme-tcp driver and is the only supported one for it upstream for a good reason. If RedHat wants to do their own weirdo setups they can patch their kernels, but please leave the upstrem code alone.
