![]() Int (*pr_register)(struct block_device *bdev, u64 old_key, u64 new_key, ![]() READ_FULL_STATUS is sometimes broken for SCSI devices. ![]() Userspace interface we may want something like SCSI's READ_FULL_STATUS and read_keysĪnd read_reservation are exactly what dm-multipath and LIO need, but for a Not sure if we wanted to export the interface to userspace yet. Note: This only initially adds the struct definitions in the kernel as I'm To optimize it's error handling so it can check if it's getting an errorīecause there's an existing reservation or if we need to retry different The READ_KEYS and READ_RESERVATION commands and will allow dm-multipath 23:19 ` " Mike 23:19 ` Mike Christie -1 siblings, 0 replies 114+ messages in threadĪdd callouts for reading keys and reservations. * block: Add PR callouts for read keys and reservation.* Use block pr_ops in LIO 23:19 ` Mike Christie 0 siblings, 0 replies 114+ messages in threadĭm-devel mailing ^ raw reply 114+ messages in thread.Added patches to allow pr_ops to pass up a BLK_STS so we could returnĪ RESERVATION_CONFLICT status when a pr_ops callout fails. Fixed sd pr_ops UA handling issue found when running libiscsi PGR tests. Fixed bug in target_core_iblock where a variable was not initialized Style for the code and I didn't want a mix and match. Style that it complained about because it looked like it was the preferred Use generic SCSI passthrough error handling interface. Add more info to patch subjects about why we need to add blk_status Add Extended Data Structure control flag enum and use in code for checks. Move pr type conversion to array and add nvme_pr_type. Rearrange patches so helpers are added with users to avoid compilation This patchset has been tested with the libiscsi PGR ops and with window's Support for target_core_iblock to bypass the emulate PR code and call Add callouts to read a reservation and it's keys. The patches are separated in the following groups: And then iblock will use the pr_ops to pass the PRĬommand to the real devices similar to what we do for unmap today. So these patchesĪllow iblock to work like pscsi/tcmu where they can pass a PR command to Nice for specific use cases, but iblock gives you the best performance andĪllows you to use stacked devices like dm-multipath. Setting up a cluster FS/framework is pain and waste when your realīackend device is already a distributed device, and pscsi and tcmu are To use tcmu or pscsi or use a cluster aware FS/framework for the LIO prįile. Target_core_iblock module to support cluster applications in VMs.Ĭurrently, to use windows clustering or linux clustering (pacemaker +Ĭluster labs scsi fence agents) in VMs with LIO and vhost-scsi, you have The patches in this thread allow us to use the block pr_ops with LIO's The following patches were built over Linus's tree and this allows the SCSI layer passthrough users to control retries for Linux-block, dm-devel, snitzer, axboe, linux-nvme, chaitanyak, To: bvanassche, hch, martin.petersen, linux-scsi, james.bottomley, Use block pr_ops in LIO All of help / color / mirror / Atom feed * Use block pr_ops in LIO 23:19 ` Mike Christie 0 siblings, 0 replies 114+ messages in threadįrom: Mike Christie 23:19 UTC ( / raw)
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |