Commit Graph

11 Commits

Author SHA1 Message Date
Daniel Verkamp
249a68e92b bdev: add API to claim block devices
Claim the block devices used by iSCSI LUNs and NVMe-oF subsystems so
they can't accidentally be reused.

This will also be used by virtual block devices to allow layering of
bdevs.

Change-Id: I5384923fbf24f13f4ce720a797c5a628053d49f4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-01-11 16:49:39 -07:00
Jim Harris
292c9c42aa scsi: simplify lun task execution
An old prototype SPDK AHCI driver would return
TASK_SET_FULL if all NCQ slots were full on a given
disk.  This would kick the SCSI task back to the LUN
to be retried later.  Since then, we have pushed
responsibility onto the bdev modules themselves
to handle this kind of queueing/retry logic.

Removing this logic allows us to make some additional
changes that enable tasks to get completed inline without
an extra event callback to handle completion.  We also
no longer need to worry about checking if pending tasks
need to be executed in the complete_task() routine, since
the execute() routine will now always exhaust the pending_task
list.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If2dc3ab017e0dbc225c8f627e1f87c5a8e9b1e3e
2017-01-09 11:37:25 -07:00
Daniel Verkamp
c3ede774c7 event: remove spdk_event_t typedef
Use 'struct spdk_event *' directly for consistency with the rest of the
API.

Change-Id: Ib41a9bf47f5b18f4aebf5f4dee055455cb12ef7d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-01-05 11:57:18 -07:00
Tsuyoshi Uchida
3560b00f20 scsi: no check lun->dev in spdk_scsi_lun_deletable (#82)
We don't need check lun->dev in spdk_scsi_lun_deletable.
Whichever dev is null or not null, spdk_scsi_lun_delete works
appropriately.
2016-12-08 18:41:03 -07:00
Pawel Wodkowski
a1948352a3 lib/scsi: handle scattered input/output buffers for non IO commands
Fix buffer overflow/underflow for commands with alloc length scattered
into multiple preallocated buffers (eg. INQUIRY)

Change-Id: If6f7cabc7a6a7fb384bb015e14dc38548f484d0f
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2016-11-21 10:52:35 -07:00
Pawel Wodkowski
f30f0c76f1 scsi: refactor usage of iov from spdk_scsi_task
This patch is preparation for fixing alloc_len overrun in SENSE 6/10 and
READCAP 6/10. To simplify code forbid usage of iov outside of
scsi/task.c.

This also drop SPDK_SCSI_TASK_ALLOC_BUFFER flag that obfuscate code. As
a replacement assume that if field alloc_len is non zero it mean that
iov.buffer is internally allocated. Functions
spdk_scsi_task_free_data(), spdk_scsi_task_set_data() and
spdk_scsi_task_alloc_data() manage this field.

Change-Id: Ife357a5bc36121f93a4c5d259b9a5a01559e7708
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2016-11-09 10:19:53 -07:00
Tsuyoshi Uchida
a5f0327897 scsi: translate nvme error to scsi error (#54)
This patch translates NVMe status code to SCSI sense according to NVM
Express: SCSI Translation Reference.

http://nvmexpress.org/wp-content/uploads/NVM_Express_-_SCSI_Translation_Reference-1_5_20150624_Gold.pdf
2016-10-28 13:06:45 -07:00
Piotr Pelplinski
a2d0fa5351 spdk: add support for readv and writev in scsi layer
This patch adds support for spdk_bdev_readv in scsi layer.
It also fixes write so that it uses multiple iov's instead of one.
Currently we should use only task->iov (for single vector operation)
or task->iovs (for multiple vector operations).

Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: Ia3b2f6d18fd212b11d7b63b11dc46ec5bbc74788
2016-10-26 04:24:08 -07:00
Ziye Yang
6c4a07ecbe scsi: change spdk_put_task to spdk_scsi_task_put
Change-Id: I2598627523b2a78b68d7e7fdadbb7d8ef44592b9
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2016-09-29 15:42:29 -07:00
Jim Harris
06ef766bc0 scsi: add spdk_io_channel support
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I189ea68c8550bbaeab79be36cb3c4faba7cd2f69
2016-09-26 14:02:07 -07:00
Daniel Verkamp
bd4ac74eaf scsi: import SCSI/blockdev translation layer
Change-Id: Ie96943f40ea8be4156d55bc5eeacc567743cf9d6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2016-08-01 10:35:01 -07:00