Transport will typically transition from LIVE to RESETTING when initially
performing a reset or recovering from an error.  Adding this transition
allows a transport to transition to RECONNECTING when it checks/waits for
connectivity then creates new transport connections and reinits the
controller.
Signed-off-by: NJames Smart <james.smart@broadcom.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Prevent racing controller reset and delete flows. reset_work must not
ever self-requeue so flushing it suffices.
Reported-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

No change in behavior except that the FC code cancels two work items a
little later now.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NJames Smart <james.smart@broadcom.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Move the ->delete_work and the associated helpers to common code instead
of duplicating them in every driver.  This also adds the missing reference
get/put for the loop driver.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NJames Smart <james.smart@broadcom.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

This is a much more sensible check than just the admin queue.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@rimbeg.me>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>

Use the core chrdev code to set up the link between the character device
and the nvme controller.  This allows us to get rid of the global list
of all controllers, and also ensures that we have both a reference to
the controller and the transport module before the open method of the
character device is called.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sgi@grimberg.me>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NHannes Reinecke <hare@suse.com>

Instead of allocating a separate struct device for the character device
handle embedd it into struct nvme_ctrl and use it for the main controller
refcounting.  This removes double refcounting and gets us an automatic
reference for the character device operations.  We keep ctrl->device as a
pointer for now to avoid chaning printks all over, but in the future we
could look into message printing helpers that take a controller structure
similar to what other subsystems do.

Note the delete_ctrl operation always already has a reference (either
through sysfs due this change, or because every open file on the
/dev/nvme-fabrics node has a refernece) when it is entered now, so we
don't need to do the unless_zero variant there.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NHannes Reinecke <hare@suse.com>

Now that we are protected against lookup vs free races for the namespace
by using kref_get_unless_zero we don't need the hack of NULLing out the
disk private data during removal.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>

For kref_get_unless_zero to protect against lookup vs free races we need
to use it in all places where we aren't guaranteed to already hold a
reference.  There is no such guarantee in nvme_find_get_ns, so switch to
kref_get_unless_zero in this function.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>

Switch to the ida_simple_* helpers instead of opencoding them.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>

Move blk_mq_reinit_tagset from blk-mq to nvme core
as the only user of it. Current transports that use
it (rdma, fc) simply implement .reinit_request op.

This patch does not change any functionality.
Reviewed-by: NJens Axboe <axboe@kernel.dk>
Reviewed-by: NMax Gurtovoy <maxg@mellanox.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

We can just use our normal ioctl handler for the compat case and remove
the boilerplate code for it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NMax Gurtovoy <maxg@mellanox.com>

The underlying blk_mq_tag_set, and request timeout parameters support an
unsigned int. Extend the size of the nvme module parameters for io and
admin commands to match.
Signed-off-by: NMarc Olson <marcolso@amazon.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

async_event_work might race as it is executed from two different
workqueues at the moment.
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Currently the nvme_req_needs_retry() applies several checks to see if
a retry is allowed. On of those is whether the current time has exceeded
the start time of the io plus the timeout length. This check, if an io
times out, means there is never a retry allowed for the io. Which means
applications see the io failure.

Remove this check and allow the io to timeout, like it does on other
protocols, and retries to be made.

On the FC transport, a frame can be lost for an individual io, and there
may be no other errors that escalate for the connection/association.
The io will timeout, which causes the transport to escalate into creating
a new association, but the io that timed out, due to this retry logic, has
already failed back to the application and things are hosed.
Signed-off-by: NJames Smart <james.smart@broadcom.com>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

If an nvme async_event command completes, in most cases, a new
async event is posted. However, if the controller enters a
resetting or reconnecting state, there is nothing to block the
scheduled work element from posting the async event again. Nor are
there calls from the transport to stop async events when an
association dies.

In the case of FC, where the association is torn down, the aer must
be aborted on the FC link and completes through the normal job
completion path. Thus the terminated async event ends up being
rescheduled even though the controller isn't in a valid state for
the aer, and the reposting gets the transport into a partially torn
down data structure.

It's possible to hit the scenario on rdma, although much less likely
due to an aer completing right as the association is terminated and
as the association teardown reclaims the blk requests via
nvme_cancel_request() so its immediate, not a link-related action
like on FC.

Fix by putting controller state checks in both the async event
completion routine where it schedules the async event and in the
async event work routine before it calls into the transport. It's
effectively a "stop_async_events()" behavior.  The transport, when
it creates a new association with the subsystem will transition
the state back to live and is already restarting the async event
posting.
Signed-off-by: NJames Smart <james.smart@broadcom.com>
[hch: remove taking a lock over reading the controller state]
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Adds support for the new Host Memory Buffer Minimum Descriptor Entry Size
and Host Memory Maximum Descriptors Entries field that were added in
TP 4002 HMB Enhancements.  These allow the controller to advertise
limits for the usual number of segments in the host memory buffer, as
well as a minimum usable per-segment size.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>

nvme_nvm_ns_supported assumes every device is a pci_dev, which leads to
reading an incorrect field, or possible even a dereference of unallocated
memory for fabrics controllers.

Fix this by introducing a quirk for lighnvm capable devices instead.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMatias Bjørling <mb@lightnvm.io>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>

The ioctls' struct allows the user to provide a metadata address and
length for a passthrough command. This patch uses these values that were
previously ignored and deletes the now unused wrapper function.
Signed-off-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NMax Gurtovoy <maxg@mellanox.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

These functions are used only locally in the nvme core.
Signed-off-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NMax Gurtovoy <maxg@mellanox.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Keep the metadata code in a separate helper instead of making the
main function more complicated.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NMax Gurtovoy <maxg@mellanox.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Instead validate that these identifiers do not change, as that is
prohibited by the specification.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>

The function is used in two places, and the shared code for those will
diverge later in this series.

Instead factor out a new helper to get the ids for a namespace, simplify
the calling conventions for nvme_identify_ns and just open code the
sequence.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>

If an NVMe controller reports RTD3 Entry Latency larger than
shutdown_timeout, up to a maximum of 60 seconds, use that value to set
the shutdown timer. Otherwise fall back to the module parameter which
defaults to 5 seconds.
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
[hch: removed do_div, made transition time local scope]
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Signed-off-by: NMax Gurtovoy <maxg@mellanox.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Use ctrl->device and lose the func name.
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

NVME's Timestamp feature allows controllers to be aware of the epoch
time in milliseconds. This patch adds the set features hook for various
transports through the identify path, so that resets and resumes can
update the controller as necessary.
Signed-off-by: NJon Derrick <jonathan.derrick@intel.com>
[hch: rebased on top of nvme-4.13 error handling changes,
      changed nvme_configure_timestamp to return the status]
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Define the constant "0xffffffff" (used as nsid for all namespaces)
as NVME_NSID_ALL.
Signed-off-by: NArnav Dawn <a.dawn@samsung.com>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>

This patch adds support for handling Fw activation without reset
On completion of FW-activation-starting AER, all queues are
paused till CSTS.PP is cleared or timed out (exceeds max time for
fw activtion MTFA). If device fails to clear CSTS.PP within MTFA,
driver issues reset controller.
Signed-off-by: NArnav Dawn <a.dawn@samsung.com>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>

This way we don't need a block_device structure to submit I/O.  The
block_device has different life time rules from the gendisk and
request_queue and is usually only available when the block device node
is open.  Other callers need to explicitly create one (e.g. the lightnvm
passthrough code, or the new nvme multipathing code).

For the actual I/O path all that we need is the gendisk, which exists
once per block device.  But given that the block layer also does
partition remapping we additionally need a partition index, which is
used for said remapping in generic_make_request.

Note that all the block drivers generally want request_queue or
sometimes the gendisk, so this removes a layer of indirection all
over the stack.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

The numd field of directive receive command takes number of dwords to
transfer. This fix has the correct calculation for numd.
Signed-off-by: NKwan (Hingkwan) Huen-SSI <kwan.huen@samsung.com>
Reviewed-by: NJens Axboe <axboe@kernel.dk>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

We need to return an error if a timeout occurs on any NVMe command during
initialization. Without this, the nvme reset work will be stuck. A timeout
will have a negative error code, meaning we need to stop initializing
the controller. All postitive returns mean the controller is still usable.

bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196325Signed-off-by: NKeith Busch <keith.busch@intel.com>
Cc: Martin Peres <martin.peres@intel.com>
[jth consolidated cleanup path ]
Signed-off-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Some broken controllers (such as earlier Linux targets) pad model or
serial fields with 0-bytes rather than spaces. The NVMe spec disallows
0 bytes in "ASCII" fields.  Thus strip trailing 0-bytes, too. Also make
sure that we get no underflow for pathological input.
Signed-off-by: NMartin Wilck <mwilck@suse.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

With a misbehaving controller it's possible we'll never
enter the live state and create an admin queue. When we
fail out of reset work it's possible we failed out early
enough without setting up the admin queue. We tear down
queues after a failed reset, but needed to do some more
sanitization.

Fixes 443bd90f: "nvme: host: unquiesce queue in nvme_kill_queues()"

[  189.650995] nvme nvme1: pci function 0000:0b:00.0
[  317.680055] nvme nvme0: Device not ready; aborting reset
[  317.680183] nvme nvme0: Removing after probe failure status: -19
[  317.681258] kasan: GPF could be caused by NULL-ptr deref or user memory access
[  317.681397] general protection fault: 0000 [#1] SMP KASAN
[  317.682984] CPU: 3 PID: 477 Comm: kworker/3:2 Not tainted 4.13.0-rc1+ #5
[  317.683112] Hardware name: Gigabyte Technology Co., Ltd. Z170X-UD5/Z170X-UD5-CF, BIOS F5 03/07/2016
[  317.683284] Workqueue: events nvme_remove_dead_ctrl_work [nvme]
[  317.683398] task: ffff8803b0990000 task.stack: ffff8803c2ef0000
[  317.683516] RIP: 0010:blk_mq_unquiesce_queue+0x2b/0xa0
[  317.683614] RSP: 0018:ffff8803c2ef7d40 EFLAGS: 00010282
[  317.683716] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 1ffff1006fbdcde3
[  317.683847] RDX: 0000000000000038 RSI: 1ffff1006f5a9245 RDI: 0000000000000000
[  317.683978] RBP: ffff8803c2ef7d58 R08: 1ffff1007bcdc974 R09: 0000000000000000
[  317.684108] R10: 1ffff1007bcdc975 R11: 0000000000000000 R12: 00000000000001c0
[  317.684239] R13: ffff88037ad49228 R14: ffff88037ad492d0 R15: ffff88037ad492e0
[  317.684371] FS:  0000000000000000(0000) GS:ffff8803de6c0000(0000) knlGS:0000000000000000
[  317.684519] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  317.684627] CR2: 0000002d1860c000 CR3: 000000045b40d000 CR4: 00000000003406e0
[  317.684758] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  317.684888] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  317.685018] Call Trace:
[  317.685084]  nvme_kill_queues+0x4d/0x170 [nvme_core]
[  317.685185]  nvme_remove_dead_ctrl_work+0x3a/0x90 [nvme]
[  317.685289]  process_one_work+0x771/0x1170
[  317.685372]  worker_thread+0xde/0x11e0
[  317.685452]  ? pci_mmcfg_check_reserved+0x110/0x110
[  317.685550]  kthread+0x2d3/0x3d0
[  317.685617]  ? process_one_work+0x1170/0x1170
[  317.685704]  ? kthread_create_on_node+0xc0/0xc0
[  317.685785]  ret_from_fork+0x25/0x30
[  317.685798] Code: 0f 1f 44 00 00 55 48 b8 00 00 00 00 00 fc ff df 48 89 e5 41 54 4c 8d a7 c0 01 00 00 53 48 89 fb 4c 89 e2 48 c1 ea 03 48 83 ec 08 <80> 3c 02 00 75 50 48 8b bb c0 01 00 00 e8 33 8a f9 00 0f ba b3
[  317.685872] RIP: blk_mq_unquiesce_queue+0x2b/0xa0 RSP: ffff8803c2ef7d40
[  317.685908] ---[ end trace a3f8704150b1e8b4 ]---
Signed-off-by: NScott Bauer <scott.bauer@intel.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

The WWID sysfs attribute can provide multiple means of a World Wide ID
for a NVMe device. It can either be a NGUID, a EUI-64 or a concatenation
of VID, Serial Number, Model and the Namespace ID in this order of
preference.

If the target also sends us a UUID use the UUID for identification and
give it the highest priority.

This eases generation of /dev/disk/by-* symlinks.
Signed-off-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJens Axboe <axboe@kernel.dk>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Usually before we teardown the controller we want to:
1. complete/cancel any ctrl inflight works
2. remove ctrl namespaces (only for removal though, resets
   shouldn't remove any namespaces).

but we do not want to destroy the controller device as
we might use it for logging during the teardown stage.

This patch adds nvme_start_ctrl() which queues inflight
controller works (aen, ns scan, queue start and keep-alive
if kato is set) and nvme_stop_ctrl() which cancels the works
namespace removal is left to the callers to handle.

Move nvme_uninit_ctrl after we are done with the
controller device.
Reviewed-by: NKeith Busch <keith.busch@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>

When we requeue a request, we can always insert the request
back to the scheduler instead of doing it when restarting
the queues and kicking the requeue work, so get rid of
the requeue kick in nvme (core and drivers).

Also, now there is no need start hw queues in nvme_kill_queues
We don't stop the hw queues anymore, so no need to
start them.
Reviewed-by: NMing Lei <ming.lei@redhat.com>
Signed-off-by: NSagi Grimberg <sagi@grimberg.me>