summaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)AuthorFilesLines
2017-11-11block, scsi: Make SCSI quiesce and resume work reliablyBart Van Assche1-12/+30
The contexts from which a SCSI device can be quiesced or resumed are: * Writing into /sys/class/scsi_device/*/device/state. * SCSI parallel (SPI) domain validation. * The SCSI device power management methods. See also scsi_bus_pm_ops. It is essential during suspend and resume that neither the filesystem state nor the filesystem metadata in RAM changes. This is why while the hibernation image is being written or restored that SCSI devices are quiesced. The SCSI core quiesces devices through scsi_device_quiesce() and scsi_device_resume(). In the SDEV_QUIESCE state execution of non-preempt requests is deferred. This is realized by returning BLKPREP_DEFER from inside scsi_prep_state_check() for quiesced SCSI devices. Avoid that a full queue prevents power management requests to be submitted by deferring allocation of non-preempt requests for devices in the quiesced state. This patch has been tested by running the following commands and by verifying that after each resume the fio job was still running: for ((i=0; i<10; i++)); do ( cd /sys/block/md0/md && while true; do [ "$(<sync_action)" = "idle" ] && echo check > sync_action sleep 1 done ) & pids=($!) for d in /sys/class/block/sd*[a-z]; do bdev=${d#/sys/class/block/} hcil=$(readlink "$d/device") hcil=${hcil#../../../} echo 4 > "$d/queue/nr_requests" echo 1 > "/sys/class/scsi_device/$hcil/device/queue_depth" fio --name="$bdev" --filename="/dev/$bdev" --buffered=0 --bs=512 \ --rw=randread --ioengine=libaio --numjobs=4 --iodepth=16 \ --iodepth_batch=1 --thread --loops=$((2**31)) & pids+=($!) done sleep 1 echo "$(date) Hibernating ..." >>hibernate-test-log.txt systemctl hibernate sleep 10 kill "${pids[@]}" echo idle > /sys/block/md0/md/sync_action wait echo "$(date) Done." >>hibernate-test-log.txt done Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name> References: "I/O hangs after resuming from suspend-to-ram" (https://marc.info/?l=linux-block&m=150340235201348). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11ide, scsi: Tell the block layer at request allocation time about preempt ↵Bart Van Assche2-5/+5
requests Convert blk_get_request(q, op, __GFP_RECLAIM) into blk_get_request_flags(q, op, BLK_MQ_PREEMPT). This patch does not change any functionality. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Acked-by: David S. Miller <davem@davemloft.net> [ for IDE ] Acked-by: Martin K. Petersen <martin.petersen@oracle.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet: kill nvmet_inline_bio_initChristoph Hellwig1-14/+4
Much easier to just opencode this helper. Also use ARRAY_SIZE instead of passing the inline bvec array size manually. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Sagi Grimberg <sagi@rimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet: better data length validationChristoph Hellwig5-25/+34
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>
2017-11-11nvme-pci: avoid dereference of symbol from unloaded moduleMing Lei1-1/+2
The 'remove_work' may be scheduled to run after nvme_remove() returns since we can't simply cancel it in nvme_remove() for avoiding deadlock. Once nvme_remove() returns, this module(nvme) can be unloaded. On the other hand, nvme_put_ctrl() calls ctr->ops->free_ctrl which may point to nvme_pci_free_ctrl() in unloaded module. This patch avoids this issue by queuing 'remove_work' via 'nvme_wq', and flush this worqueue in nvme_exit() as suggested by Sagi. Suggested-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: send uevent for some asynchronous eventsKeith Busch2-0/+29
This will give udev a chance to observe and handle asynchronous event notifications and clear the log to unmask future events of the same type. The driver will create a change uevent of the asyncronuos event result before submitting the next AEN request to the device if a completed AEN event is of type error, smart, command set or vendor specific, Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Guan Junxiong <guanjunxiong@huawei.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: unexport starting async event workKeith Busch2-8/+1
Async event work is for core use only and should not be called directly from drivers. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Guan Junxiong <guanjunxiong@huawei.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: remove handling of multiple AEN requestsKeith Busch6-40/+11
The driver can handle tracking only one AEN request, so this patch removes handling for multiple ones. Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: James Smart <james.smart@broadcom.com> Signed-off-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme-fc: remove unused "queue_size" fieldKeith Busch1-6/+3
This was being saved in a structure, but never used anywhere. The queue size is obtained through other means, so there's no reason to duplicate this without a user for it. Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Guan Junxiong <guanjunxiong@huawei.com> Reviewed-by: James Smart <james.smart@broadcom.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: centralize AEN definesKeith Busch6-58/+22
All the transports were unnecessarilly duplicating the AEN request accounting. This patch defines everything in one place. Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Guan Junxiong <guanjunxiong@huawei.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet: remove redundant local variableSagi Grimberg1-9/+4
the status is either success or some status id and we don't need a local variable for it. Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet: remove redundant memset if failed to get_smart_log failedSagi Grimberg1-3/+1
We already allocated the buffer with kzalloc. Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: fix eui_show() print formatJavier González1-1/+1
Fix print formatting, but keep the original output to prevent user breakage as suggested by Joe Perches. Signed-off-by: Javier González <javier@cnexlabs.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: compare NQN string with right sizeJavier González1-1/+1
Copy subnqns using NVMF_NQN_SIZE as it is < 256 Signed-off-by: Javier González <javier@cnexlabs.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11block: pass full fmode_t to blk_verify_commandChristoph Hellwig1-1/+1
Use the obvious calling convention. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet: fix comment typos in admin-cmd.cMinwoo Im1-2/+2
small typos fixed in admin-cmd.c Signed-off-by: Minwoo Im <minwoo.im.dev@gmail.com> Reviewed-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme-rdma: fix nvme_rdma_create_queue_ib error flowMax Gurtovoy1-1/+1
QP object is created using rdma_cm api, therefore the destruction should use the same api for symmetry. Signed-off-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvmet-rdma: update queue list during ib_device removalIsrael Rukshin1-2/+4
A NULL deref happens when nvmet_rdma_remove_one() is called more than once (e.g. while connected via 2 ports). The first call frees the queues related to the first ib_device but doesn't remove them from the queue list. While calling nvmet_rdma_remove_one() for the second ib_device it goes over the full queue list again and we get the NULL deref. Fixes: f1d4ef7d ("nvmet-rdma: register ib_client to not deadlock in device removal") Signed-off-by: Israel Rukshin <israelr@mellanox.com> Reviewed-by: Max Gurtovoy <maxg@mellanox.com> Reviewed-by: Sagi Grimberg <sagi@grmberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11lpfc: tie in to new dev_loss_tmo interface in nvme transportJames Smart1-0/+5
This patch calls the new nvme transport routine for dev_loss_tmo whenever the SCSI fc transport calls the lldd to make a dynamic change to a remote ports dev_loss_tmo. 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>
2017-11-11nvme-fc: decouple ns references from lldd referencesJames Smart1-6/+78
In the lldd api, a lldd may unregister a remoteport (loss of connectivity or driver unload) or localport (driver unload). The lldd must wait for the remoteport_delete or localport_delete before completing its actions post the unregister. The xxx_deletes currently occur only when the xxxport structure is fully freed after all references are removed. Thus the lldd may be held hostage until an app or in-kernel entity that has a namespace open finally closes so the namespace can be removed, the controller removed, thus the transport objects, thus the lldd. This patch decouples the transport and os-facing objects from the lldd and the remoteport and localport. There is a point in all deletions where the transport will no longer interact with the lldd on behalf of a controller. That point centers around the association established with the target/subsystem. It will access the lldd whenever it attempts to create an association and while the association is active. New associations may only be created if the remoteport is live (thus the localport is live). It will not access the lldd after deleting the association. Therefore, the patch tracks the count of active controllers - those with associations being created or that are active - on a remoteport. It also tracks the number of remoteports that have active controllers, on a a localport. When a remoteport is unregistered, as soon as there are no active controllers, the lldd's remoteport_delete may be called and the lldd may continue. Similarly, when a localport is unregistered, as soon as there are no remoteports with active controllers, the localport_delete callback may be made. This significantly speeds up unregistration with the lldd. The transport objects continue in suspended status with reconnect timers running, and upon expiration, normal ref-counting will occur and the objects will be freed. The transport object may still be held hostage by the application/kernel module, but that is acceptable. With this change, the lldd may be fully unloaded and reloaded, and if registrations occur prior to the timeouts, the nvme controller and namespaces will resume normally as if a link bounce. 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>
2017-11-11nvme-fc: fix localport resume using stale valuesJames Smart1-2/+10
The localport resume was not updating the lldd ops structure. If the lldd is unloaded and reloaded, the ops pointers will differ. Additionally, as there are device references taken by the localport, ensure that resume only resumes if the device matches as well. 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>
2017-11-11nvme: check admin passthru command effectsKeith Busch2-0/+108
The NVMe standard provides a command effects log page so the host may be aware of special requirements it may need to do for a particular command. For example, the command may need to run with IO quiesced to prevent timeouts or undefined behavior, or it may change the logical block formats that determine how the host needs to construct future commands. This patch saves the nvme command effects log page if the controller supports it, and performs appropriate actions before and after an admin passthrough command is completed. If the controller does not support the command effects log page, the driver will define the effects for known opcodes. The nvme format and santize are the only commands in this patch with known effects. Signed-off-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: factor get log into a helperKeith Busch1-6/+13
And fix the warning on a successful firmware log. Reviewed-by: Javier González <javier@cnexlabs.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: fix and clarify the check for missing metadataChristoph Hellwig1-13/+18
Update the check in nvme_setup_rw for missing metadata so that it is together with the other metadata handling, does not contain impossible to reach conditions and warns if we get an impossible requests for a (non-PI) metadata-enabled namespace when CONFIG_BLK_DEV_INTEGRITY is not set. Also add a little helper that checks if a given metadata configuration contains protection information Signed-off-by: Christoph Hellwig <hch@lst.de> Reported-by: Javier González <jg@lightnvm.io> Reviewed-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: split __nvme_revalidate_diskChristoph Hellwig1-23/+26
Split out the code that applies the calculate value to a given disk/queue into new helper that can be reused by the multipath code. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: set the chunk size before freezing the queueChristoph Hellwig1-2/+3
We don't need a frozen queue to update the chunk_size, which just is a hint, and moving it a little earlier will allow for some better code reuse with the multipath code. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: don't pass struct nvme_ns to nvme_config_discardChristoph Hellwig1-16/+17
To allow reusing this function for the multipath node. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: don't pass struct nvme_ns to nvme_init_integrityChristoph Hellwig1-7/+7
To allow reusing this function for the multipath node. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: always unregister the integrity profile in __nvme_revalidate_diskChristoph Hellwig1-30/+10
This is safe because the queue is always frozen when we revalidate, and it simplifies both the existing code as well as the multipath implementation. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-11nvme: move the dying queue check from cancel to completionChristoph Hellwig1-6/+3
With multipath we don't want a hard DNR bit on a request that is cancelled by a controller reset, but instead want to be able to retry it on another patch. To archive this don't always set the DNR bit when the queue is dying in nvme_cancel_request, but defer that decision to nvme_req_needs_retry. Note that it applies to any command there and not just cancelled commands, but one the queue is dying that is the right thing to do anyway. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-07nbd: don't start req until after the dead connection logicJosef Bacik1-13/+7
We can end up sleeping for a while waiting for the dead timeout, which means we could get the per request timer to fire. We did handle this case, but if the dead timeout happened right after we submitted we'd either tear down the connection or possibly requeue as we're handling an error and race with the endio which can lead to panics and other hilarity. Fixes: 560bc4b39952 ("nbd: handle dead connections") Cc: stable@vger.kernel.org Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-07nbd: wait uninterruptible for the dead timeoutJosef Bacik1-3/+3
If we have a pending signal or the user kills their application then it'll bring down the whole device, which is less than awesome. Instead wait uninterruptible for the dead timeout so we're sure we gave it our best shot. Fixes: 560bc4b39952 ("nbd: handle dead connections") Cc: stable@vger.kernel.org Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-04blk-mq: don't handle failure in .get_budgetMing Lei1-8/+3
It is enough to just check if we can get the budget via .get_budget(). And we don't need to deal with device state change in .get_budget(). For SCSI, one issue to be fixed is that we have to call scsi_mq_uninit_cmd() to free allocated ressources if SCSI device fails to handle the request. And it isn't enough to simply call blk_mq_end_request() to do that if this request is marked as RQF_DONTPREP. Fixes: 0df21c86bdbf(scsi: implement .get_budget and .put_budget for blk-mq) Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-04SCSI: don't get target/host busy_count in scsi_mq_get_budget()Ming Lei1-16/+13
It is very expensive to atomic_inc/atomic_dec the host wide counter of host->busy_count, and it should have been avoided via blk-mq's mechanism of getting driver tag, which uses the more efficient way of sbitmap queue. Also we don't check atomic_read(&sdev->device_busy) in scsi_mq_get_budget() and don't run queue if the counter becomes zero, so IO hang may be caused if all requests are completed just before the current SCSI device is added to shost->starved_list. Fixes: 0df21c86bdbf(scsi: implement .get_budget and .put_budget for blk-mq) Reported-by: Bart Van Assche <bart.vanassche@wdc.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-03cdrom: hide CONFIG_CDROM menu selectionJens Axboe1-8/+1
We don't need to expose this. The point is that drivers select the uniform CDROM layer, if they need it, the user should not have to make a conscious decision on whether to include this separately or not. Fixes: 2a750166a5be ("block: Rework drivers/cdrom/Makefile") Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-03block: add a poll_fn callback to struct request_queueChristoph Hellwig1-1/+1
That we we can also poll non blk-mq queues. Mostly needed for the NVMe multipath code, but could also be useful elsewhere. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-03Merge branch 'nvme-4.15' of git://git.infradead.org/nvme into for-4.15/blockJens Axboe12-491/+1008
Pull NVMe changes from Christoph: "Below are the currently queue nvme updates for Linux 4.15. There are a few more things that could make it for this merge window, but I'd like to get things into linux-next, especially for the unlikely case that Linus decided to cut -rc8. Highlights: - support for SGLs in the PCIe driver (Chaitanya Kulkarni) - disable I/O schedulers for the admin queue (Israel Rukshin) - various Fibre Channel fixes and enhancements (James Smart) - various refactoring for better code sharing between transports (Sagi Grimberg and me) as well as lots of little bits from various contributors."
2017-11-03nvme: comment typo fixed in clearing AERMinwoo Im1-1/+1
a tiny typo fixed in a comment. Signed-off-by: Minwoo Im <minwoo.im.dev@gmail.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-02skd: use ktime_get_real_seconds()Arnd Bergmann1-1/+2
Like many storage drivers, skd uses an unsigned 32-bit number for interchanging the current time with the firmware. This will overflow in y2106 and is otherwise safe. However, the get_seconds() function is generally considered deprecated since the behavior is different between 32-bit and 64-bit architectures, and using it may indicate a bigger problem. To annotate that we've thought about this, let's add a comment here and migrate to the ktime_get_real_seconds() function that consistently returns a 64-bit number. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-02block: fix CDROM dependency on BLK_DEVArnd Bergmann2-1/+2
After the cdrom cleanup, I get randconfig warnings for some configurations: warning: (BLK_DEV_IDECD && BLK_DEV_SR) selects CDROM which has unmet direct dependencies (BLK_DEV) This adds an explicit BLK_DEV dependency for both drivers. The other drivers that select 'CDROM' already have this and don't need a change. Fixes: 2a750166a5be ("block: Rework drivers/cdrom/Makefile") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-01nvme: Remove unused headersKeith Busch1-4/+0
Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvmet: fix fatal_err_work deadlockJames Smart1-2/+14
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>
2017-11-01nvme-fc: add dev_loss_tmo timeout and remoteport resume supportJames Smart1-39/+239
When a remoteport is unregistered (connectivity lost), the following actions are taken: - the remoteport is marked DELETED - the time when dev_loss_tmo would expire is set in the remoteport - all controllers on the remoteport are reset. After a controller resets, it will stall in a RECONNECTING state waiting for one of the following: - the controller will continue to attempt reconnect per max_retries and reconnect_delay. As no remoteport connectivity, the reconnect attempt will immediately fail. If max reconnects has not been reached, a new reconnect_delay timer will be schedule. If the current time plus another reconnect_delay exceeds when dev_loss_tmo expires on the remote port, then the reconnect_delay will be shortend to schedule no later than when dev_loss_tmo expires. If max reconnect attempts are reached (e.g. ctrl_loss_tmo reached) or dev_loss_tmo ix exceeded without connectivity, the controller is deleted. - the remoteport is re-registered prior to dev_loss_tmo expiring. The resume of the remoteport will immediately attempt to reconnect each of its suspended controllers. Signed-off-by: James Smart <james.smart@broadcom.com> Reviewed-by: Hannes Reinecke <hare@suse.com> [hch: updated to use nvme_delete_ctrl] Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme: allow controller RESETTING to RECONNECTING transitionJames Smart1-0/+1
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: James Smart <james.smart@broadcom.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme-fc: check connectivity before initiating reconnectsJames Smart1-7/+16
Check remoteport connectivity before initiating reconnects Signed-off-by: James Smart <james.smart@broadcom.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme-fc: add a dev_loss_tmo field to the remoteportJames Smart1-0/+31
Add a dev_loss_tmo value, paralleling the SCSI FC transport, for device connectivity loss. The transport initializes the value in the nvme_fc_register_remoteport() call. If the value is not set, a default of 60s is set. Add a new routine to the api, nvme_fc_set_remoteport_devloss() routine, which allows the lldd to dynamically update the value on an existing remoteport. Signed-off-by: James Smart <james.smart@broadcom.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme-fc: change ctlr state assignments during reset/reconnectJames Smart1-15/+13
Clean up some of the controller state checks and add the RESETTING->RECONNECTING state transition. Specifically: - the movement of the RESETTING state change and schedule of reset_work to core doesn't work wiht nvme_fc_error_recovery setting state to RECONNECTING before attempting to reset. Remove the state change as the reset request does it. - In the rare cases where an error occurs right as we're transitioning to LIVE, defer the controller start actions. - In error handling on teardown of associations while performing initial controller creation - avoid quiesce calls on the admin_q. They are unneeded. - Add the RESETTING->RECONNECTING transition in the reset handler. Signed-off-by: James Smart <james.smart@broadcom.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme: flush reset_work before safely continuing with delete operationSagi Grimberg2-1/+1
Prevent racing controller reset and delete flows. reset_work must not ever self-requeue so flushing it suffices. Reported-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme-rdma: reuse nvme_delete_ctrl when reconnect attempts expireSagi Grimberg1-1/+1
instead of just queueing delete work Reported-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-11-01nvme: consolidate common code from ->reset_workChristoph Hellwig4-21/+4
No change in behavior except that the FC code cancels two work items a little later now. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: James Smart <james.smart@broadcom.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
generated by cgit v1.2.3 (git 2.39.0) at 2024-10-18 02:34:08 +0300