Hi all, this series adds support for multipathing, that is accessing nvme namespaces through multiple controllers to the nvme core driver. I think we are pretty much done with with very little changes in the last reposts. Unless I hear objections I plan to send this to Jens tomorrow with the remaining NVMe updates. It is a very thin and efficient implementation that relies on close cooperation with other bits of the nvme driver, and few small and simple block helpers. Compared to dm-multipath the important differences are how management of the paths is done, and how the I/O path works. Management of the paths is fully integrated into the nvme driver, for each newly found nvme controller we check if there are other controllers that refer to the same subsystem, and if so we link them up in the nvme driver. Then for each namespace found we check if the namespace id and identifiers match to check if we have multiple controllers that refer to the same namespaces. For now path availability is based entirely on the controller status, which at least for fabrics will be continuously updated based on the mandatory keep alive timer. Once the Asynchronous Namespace Access (ANA) proposal passes in NVMe we will also get per-namespace states in addition to that, but for now any details of that remain confidential to NVMe members. The I/O path is very different from the existing multipath drivers, which is enabled by the fact that NVMe (unlike SCSI) does not support partial completions - a controller will either complete a whole command or not, but never only complete parts of it. Because of that there is no need to clone bios or requests - the I/O path simply redirects the I/O to a suitable path. For successful commands multipath is not in the completion stack at all. For failed commands we decide if the error could be a path failure, and if yes remove the bios from the request structure and requeue them before completing the request. All together this means there is no performance degradation compared to normal nvme operation when using the multipath device node (at least not until I find a dual ported DRAM backed device :)) A git tree is available at: git://git.infradead.org/users/hch/block.git nvme-mpath gitweb: http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/nvme-mpath Changes since V6: - added slaves/holder directories (Hannes) - trivial rebase on top of the latest nvme-4.15 changes Changes since V5: - dropped various prep patches merged into the nvme tree - removed a leftover debug printk - small cleanups to blk_steal_bio - add a sysfs attribute for hidden gendisks - don't complete cancelled requests if another path is available - use the instance index for device naming - make the multipath code optional at compile time - don't use the multipath node for single-controller subsystems - add a module_option to disable multipathing (temporarily) Changes since V4: - add a refcount to release the device in struct nvme_subsystem - use the instance to name the nvme_subsystems in sysfs - remove a NULL check before nvme_put_ns_head - take a ns_head reference in ->open - improve open protection for GENHD_FL_HIDDEN - add poll support for the mpath device Changes since V3: - new block layer support for hidden gendisks - a couple new patches to refactor device handling before the actual multipath support - don't expose per-controller block device nodes - use /dev/nvmeXnZ as the device nodes for the whole subsystem. - expose subsystems in sysfs (Hannes Reinecke) - fix a subsystem leak when duplicate NQNs are found - fix up some names - don't clear current_path if freeing a different namespace Changes since V2: - don't create duplicate subsystems on reset (Keith Bush) - free requests properly when failing over in I/O completion (Keith Bush) - new devices names: /dev/nvm-sub%dn%d - expose the namespace identification sysfs files for the mpath nodes Changes since V1: - introduce new nvme_ns_ids structure to clean up identifier handling - generic_make_request_fast is now named direct_make_request and calls generic_make_request_checks - reset bi_disk on resubmission - create sysfs links between the existing nvme namespace block devices and the new share mpath device - temporarily added the timeout patches from James, this should go into nvme-4.14, though
