Whenever a cmd is received a reference is taken while looking up the
queue. The reference is removed after the cmd is done as the iod is
returned for reuse. The fod may be reused for a deferred (recevied but
no job context) cmd. Existing code removes the reference only if the
fod is not reused for another command. Given the fod may be used for
one or more ios, although a reference was taken per io, it won't be
matched on the frees.
Remove the reference on every fod free. This pairs the references to
each io.
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reorganize nvmet_fc_handle_fcp_rqst() so that the nvmet req.transfer_len
field is set after the call nvmet_req_init(). An update to nvmet now
has nvmet_req_init() clearing the field, thus the fc transport was losing
the value.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently the NVMe target stores the expexted data length in req->data_len
and uses that for data transfer decisions, but that does not take the
actual transfer length in the SGLs into account. So this adds a new
transfer_len field, into which the transport drivers store the actual
transfer length. We then check the two match before actually executing
the command.
The FC transport driver already had such a field, which is removed in
favour of the common one.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Below is a stack trace for an issue that was reported.
What's happening is that the nvmet layer had it's controller kato
timeout fire, which causes it to schedule its fatal error handler
via the fatal_err_work element. The error handler is invoked, which
calls the transport delete_ctrl() entry point, and as the transport
tears down the controller, nvmet_sq_destroy ends up doing the final
put on the ctlr causing it to enter its free routine. The ctlr free
routine does a cancel_work_sync() on fatal_err_work element, which
then does a flush_work and wait_for_completion. But, as the wait is
in the context of the work element being flushed, its in a catch-22
and the thread hangs.
[ 326.903131] nvmet: ctrl 1 keep-alive timer (15 seconds) expired!
[ 326.909832] nvmet: ctrl 1 fatal error occurred!
[ 327.643100] lpfc 0000:04:00.0: 0:6313 NVMET Defer ctx release xri
x114 flg x2
[ 494.582064] INFO: task kworker/0:2:243 blocked for more than 120
seconds.
[ 494.589638] Not tainted 4.14.0-rc1.James+ #1
[ 494.594986] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message.
[ 494.603718] kworker/0:2 D 0 243 2 0x80000000
[ 494.609839] Workqueue: events nvmet_fatal_error_handler [nvmet]
[ 494.616447] Call Trace:
[ 494.619177] __schedule+0x28d/0x890
[ 494.623070] schedule+0x36/0x80
[ 494.626571] schedule_timeout+0x1dd/0x300
[ 494.631044] ? dequeue_task_fair+0x592/0x840
[ 494.635810] ? pick_next_task_fair+0x23b/0x5c0
[ 494.640756] wait_for_completion+0x121/0x180
[ 494.645521] ? wake_up_q+0x80/0x80
[ 494.649315] flush_work+0x11d/0x1a0
[ 494.653206] ? wake_up_worker+0x30/0x30
[ 494.657484] __cancel_work_timer+0x10b/0x190
[ 494.662249] cancel_work_sync+0x10/0x20
[ 494.666525] nvmet_ctrl_put+0xa3/0x100 [nvmet]
[ 494.671482] nvmet_sq_:q+0x64/0xd0 [nvmet]
[ 494.676540] nvmet_fc_delete_target_queue+0x202/0x220 [nvmet_fc]
[ 494.683245] nvmet_fc_delete_target_assoc+0x6d/0xc0 [nvmet_fc]
[ 494.689743] nvmet_fc_delete_ctrl+0x137/0x1a0 [nvmet_fc]
[ 494.695673] nvmet_fatal_error_handler+0x30/0x40 [nvmet]
[ 494.701589] process_one_work+0x149/0x360
[ 494.706064] worker_thread+0x4d/0x3c0
[ 494.710148] kthread+0x109/0x140
[ 494.713751] ? rescuer_thread+0x380/0x380
[ 494.718214] ? kthread_park+0x60/0x60
Correct by having the fc transport convert to a different workq context
for the actual controller teardown which may call the cancel_work_sync.
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
When searching for queue id's ensure they are within the expected range.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Avoid calling the put routine, as it may traverse to free routines while
holding the target lock.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
fc transport is treating NVMET_NR_QUEUES as maximum queue count, e.g.
admin queue plus NVMET_NR_QUEUES-1 io queues. But NVMET_NR_QUEUES is
the number of io queues, so maximum queue count is really
NVMET_NR_QUEUES+1.
Fix the handling in the target fc transport
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The FC-NVME target transport used the FC-specific error codes in
return codes when the transport or lldd failed. Instead of using the
FC-specific values, now use a generic value (NVME_SC_INTERNAL).
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The existing nvmet_fc sg list handling has 2 faults:
a) the request between LLDD and transport has too large of an sg
list (256 elements), which is normally 256k (64 elements).
b) sglist handling doesn't optimize on the fact that each element
is a page.
This patch removes the static sg list in the request and uses the
dynamic list already present in the nvmet_fc transport. It also
simplies the handling of the sg list on multiple sequences to
take advantage of the per-page divisions.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
There were 2 statics introduced that were bogus. Removed the static
designations.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Use list_for_each_entry_safe to prevent list handling from referencing
next pointers directly after list_del's
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
At queue creation, the transport allocates a local job struct
(struct nvmet_fc_fcp_iod) for each possible element of the queue.
When a new CMD is received from the wire, a jobs struct is allocated
from the queue and then used for the duration of the command.
The job struct contains buffer space for the wire command iu. Thus,
upon allocation of the job struct, the cmd iu buffer is copied to
the job struct and the LLDD may immediately free/reuse the CMD IU
buffer passed in the call.
However, in some circumstances, due to the packetized nature of FC
and the api of the FC LLDD which may issue a hw command to send the
wire response, but the LLDD may not get the hw completion for the
command and upcall the nvmet_fc layer before a new command may be
asynchronously received on the wire. In other words, its possible
for the initiator to get the response from the wire, thus believe a
command slot free, and send a new command iu. The new command iu
may be received by the LLDD and passed to the transport before the
LLDD had serviced the hw completion and made the teardown calls for
the original job struct. As such, there is no available job struct
available for the new io. E.g. it appears like the host sent more
queue elements than the queue size. It didn't based on it's
understanding.
Rather than treat this as a hard connection failure queue the new
request until the job struct does free up. As the buffer isn't
copied as there's no job struct, a special return value must be
returned to the LLDD to signify to hold off on recycling the cmd
iu buffer. And later, when a job struct is allocated and the
buffer copied, a new LLDD callback is introduced to notify the
LLDD and allow it to recycle it's command iu buffer.
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The FC-NVME spec hasn't locked down on the format string for TRADDR.
Currently the spec is lobbying for "nn-<16hexdigits>:pn-<16hexdigits>"
where the wwn's are hex values but not prefixed by 0x.
Most implementations so far expect a string format of
"nn-0x<16hexdigits>:pn-0x<16hexdigits>" to be used. The transport
uses the match_u64 parser which requires a leading 0x prefix to set
the base properly. If it's not there, a match will either fail or return
a base 10 value.
The resolution in T11 is pushing out. Therefore, to fix things now and
to cover any eventuality and any implementations already in the field,
this patch adds support for both formats.
The change consists of replacing the token matching routine with a
routine that validates the fixed string format, and then builds
a local copy of the hex name with a 0x prefix before calling
the system parser.
Note: the same parser routine exists in both the initiator and target
transports. Given this is about the only "shared" item, we chose to
replicate rather than create an interdendency on some shared code.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
We always need to do non-equal comparisms on the native endian versions
to get the correct result.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Target validation of the Create Association LS revised to accept any
LS as long as all non-pad data has been received. This allows a (newer)
target to accept the LS from older initiators with varying pad lengths.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
if a nvme command is issued with an opcode that is not supported by
the target (example: opcode 21 - detach namespace), the target
crashes due to a null pointer.
nvmet_req_init() detects the bad opcode and immediately calls the nvme
command done routine with an error status, allowing the transport to
send the response. However, the FC transport was aborting the command
on error, so the abort freed the lldd point, but the rsp transmit path
referenced it psot the free.
Fix by removing the abort call on nvmet_req_init() failure.
The completion response will be sent with an error status code.
As the completion path will terminate the io, ensure the data_sg
lists show an unused state so that teardown paths are successful.
Signed-off-by: Paul Ely <Paul.Ely@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Found by sparse.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
That's what it's used as.
Found by sparse.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
The passed in desc_len is a big endian value, so mark it as such.
Found by sparse.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Found by sparse.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
The FC-NVME spec revised syntax to avoid comma separators.
Sync with the change in the parser for traddr on port attachments.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
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>
With the advent of the opdone calls changing context, the lldd can no
longer assume that once the op->done call returns for RSP operations
that the request struct is no longer being accessed.
As such, revise the lldd api for a req_release callback that the
transport will call when the job is complete. This will also be used
with abort cases.
Fixed text in api header for change in io complete semantics.
Revised lpfc to support the new req_release api.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Two new feature flags were added to control whether upcalls to the
transport result in context switches or stay in the calling context.
NVMET_FCTGTFEAT_CMD_IN_ISR:
By default, if the flag is not set, the transport assumes the
lldd is in a non-isr context and in the cpu context it should be
for the io queue. As such, the cmd handler is called directly in the
calling context.
If the flag is set, indicating the upcall is an isr context, the
transport mandates a transition to a workqueue. The workqueue assigned
to the queue is used for the context.
NVMET_FCTGTFEAT_OPDONE_IN_ISR
By default, if the flag is not set, the transport assumes the
lldd is in a non-isr context and in the cpu context it should be
for the io queue. As such, the fcp operation done callback is called
directly in the calling context.
If the flag is set, indicating the upcall is an isr context, the
transport mandates a transition to a workqueue. The workqueue assigned
to the queue is used for the context.
Updated lpfc for flags
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Clear SG list to avoid double frees of payload page list
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@fb.com>
Cleanup of abort flag processing in fcp_op_done.
References were unnecessary
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@fb.com>
Correct logic in disconnect queue LS handling.
Rework so that queue searching and error reporting is above the
section to send back a ls rjt
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Implements the FC-NVME T11 definition of how nvme fabric capsules are
performed on an FC fabric. Utilizes a lower-layer API to FC host adapters
to send/receive FC-4 LS operations and perform the FCP transactions
necessary to perform and FCP IO request for NVME.
The T11 definitions for FC-4 Link Services are implemented which create
NVMeOF connections. Implements the hooks with nvmet layer to pass NVME
commands to it for processing and posting of data/response base to the
host via the different connections.
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Jay Freyensee <james_p_freyensee@linux.intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
James Smart