diff options
author | James Smart <jsmart2021@gmail.com> | 2017-04-11 11:32:31 -0700 |
---|---|---|
committer | Christoph Hellwig <hch@lst.de> | 2017-04-21 16:41:51 +0200 |
commit | a97ec51b37efacb84f286979876675a8143035b0 (patch) | |
tree | 49ce2509063d78d3d8699c538e74443d922ab0c7 /include/linux/nvme-fc-driver.h | |
parent | ce79bfc2c7b415f0f296adc59a54c9c781d63617 (diff) | |
download | talos-op-linux-a97ec51b37efacb84f286979876675a8143035b0.tar.gz talos-op-linux-a97ec51b37efacb84f286979876675a8143035b0.zip |
nvmet_fc: Rework target side abort handling
target transport:
----------------------
There are cases when there is a need to abort in-progress target
operations (writedata) so that controller termination or errors can
clean up. That can't happen currently as the abort is another target
op type, so it can't be used till the running one finishes (and it may
not). Solve by removing the abort op type and creating a separate
downcall from the transport to the lldd to request an io to be aborted.
The transport will abort ios on queue teardown or io errors. In general
the transport tries to call the lldd abort only when the io state is
idle. Meaning: ops that transmit data (readdata or rsp) will always
finish their transmit (or the lldd will see a state on the
link or initiator port that fails the transmit) and the done call for
the operation will occur. The transport will wait for the op done
upcall before calling the abort function, and as the io is idle, the
io can be cleaned up immediately after the abort call; Similarly, ios
that are not waiting for data or transmitting data must be in the nvmet
layer being processed. The transport will wait for the nvmet layer
completion before calling the abort function, and as the io is idle,
the io can be cleaned up immediately after the abort call; As for ops
that are waiting for data (writedata), they may be outstanding
indefinitely if the lldd doesn't see a condition where the initiatior
port or link is bad. In those cases, the transport will call the abort
function and wait for the lldd's op done upcall for the operation, where
it will then clean up the io.
Additionally, if a lldd receives an ABTS and matches it to an outstanding
request in the transport, A new new transport upcall was created to abort
the outstanding request in the transport. The transport expects any
outstanding op call (readdata or writedata) will completed by the lldd and
the operation upcall made. The transport doesn't act on the reported
abort (e.g. clean up the io) until an op done upcall occurs, a new op is
attempted, or the nvmet layer completes the io processing.
fcloop:
----------------------
Updated to support the new target apis.
On fcp io aborts from the initiator, the loopback context is updated to
NULL out the half that has completed. The initiator side is immediately
called after the abort request with an io completion (abort status).
On fcp io aborts from the target, the io is stopped and the initiator side
sees it as an aborted io. Target side ops, perhaps in progress while the
initiator side is done, continue but noop the data movement as there's no
structure on the initiator side to reference.
patch also contains:
----------------------
Revised lpfc to support the new abort api
commonized rsp buffer syncing and nulling of private data based on
calling paths.
errors in op done calls don't take action on the fod. They're bad
operations which implies the fod may be bad.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Diffstat (limited to 'include/linux/nvme-fc-driver.h')
-rw-r--r-- | include/linux/nvme-fc-driver.h | 25 |
1 files changed, 16 insertions, 9 deletions
diff --git a/include/linux/nvme-fc-driver.h b/include/linux/nvme-fc-driver.h index d98ddb2feabc..0db37158a61d 100644 --- a/include/linux/nvme-fc-driver.h +++ b/include/linux/nvme-fc-driver.h @@ -533,9 +533,6 @@ enum { * rsp as well */ NVMET_FCOP_RSP = 4, /* send rsp frame */ - NVMET_FCOP_ABORT = 5, /* abort exchange via ABTS */ - NVMET_FCOP_BA_ACC = 6, /* send BA_ACC */ - NVMET_FCOP_BA_RJT = 7, /* send BA_RJT */ }; /** @@ -572,8 +569,6 @@ enum { * upon compeletion of the operation. The nvmet-fc layer will also set a * private pointer for its own use in the done routine. * - * Note: the LLDD must never fail a NVMET_FCOP_ABORT request !! - * * Values set by the NVMET-FC layer prior to calling the LLDD fcp_op * entrypoint. * @op: Indicates the FCP IU operation to perform (see NVMET_FCOP_xxx) @@ -784,10 +779,6 @@ struct nvmet_fc_target_port { * or upon success/failure of FCP_CONF if it is supported, the * LLDD is to set the nvmefc_tgt_fcp_req fcp_error field and * consider the operation complete. - * NVMET_FCOP_ABORT: the LLDD is to terminate the exchange - * corresponding to the fcp operation. The LLDD shall send - * ABTS and follow FC exchange abort-multi rules, including - * ABTS retries and possible logout. * Upon completing the indicated operation, the LLDD is to set the * status fields for the operation (tranferred_length and fcp_error * status) in the request, then call the "done" routine @@ -808,6 +799,17 @@ struct nvmet_fc_target_port { * Returns 0 on success, -<errno> on failure (Ex: -EIO) * Entrypoint is Mandatory. * + * @fcp_abort: Called by the transport to abort an active command. + * The command may be in-between operations (nothing active in LLDD) + * or may have an active WRITEDATA operation pending. The LLDD is to + * initiate the ABTS process for the command and return from the + * callback. The ABTS does not need to be complete on the command. + * The fcp_abort callback inherently cannot fail. After the + * fcp_abort() callback completes, the transport will wait for any + * outstanding operation (if there was one) to complete, then will + * call the fcp_req_release() callback to return the command's + * exchange context back to the LLDD. + * * @fcp_req_release: Called by the transport to return a nvmefc_tgt_fcp_req * to the LLDD after all operations on the fcp operation are complete. * This may be due to the command completing or upon completion of @@ -848,6 +850,8 @@ struct nvmet_fc_target_template { struct nvmefc_tgt_ls_req *tls_req); int (*fcp_op)(struct nvmet_fc_target_port *tgtport, struct nvmefc_tgt_fcp_req *fcpreq); + void (*fcp_abort)(struct nvmet_fc_target_port *tgtport, + struct nvmefc_tgt_fcp_req *fcpreq); void (*fcp_req_release)(struct nvmet_fc_target_port *tgtport, struct nvmefc_tgt_fcp_req *fcpreq); @@ -877,4 +881,7 @@ int nvmet_fc_rcv_fcp_req(struct nvmet_fc_target_port *tgtport, struct nvmefc_tgt_fcp_req *fcpreq, void *cmdiubuf, u32 cmdiubuf_len); +void nvmet_fc_rcv_fcp_abort(struct nvmet_fc_target_port *tgtport, + struct nvmefc_tgt_fcp_req *fcpreq); + #endif /* _NVME_FC_DRIVER_H */ |