Currently we test up to 512KB I/O sizes, and this I/O
size could fragment to 129 IOVs if split on 4KB
boundaries if the buffer start is not 4KB aligned.
This is likely the cause of a recent slew of failures
in the test pool which coincide with updating to
fio 3.3.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I189157af578e75e025ca8e3420712739e604fca7
Reviewed-on: https://review.gerrithub.io/395872
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This allows a channel's request_set resources to be
used for queuing I/O requests. This is needed
for upcoming thin provisioning functionality,
where we must queue I/O requests that need to
allocate a cluster, if another cluster allocation
is in progress.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie8d3e799afc0b56bc95ba5ecab11253d8bc8608f
Reviewed-on: https://review.gerrithub.io/395037
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Now all operations (both single buffer and iov-based
payloads) which span a cluster boundary get split into
separate blob calls for each cluster.
This will simplify upcoming patches that need to do
special operations if a cluster needs to be allocated.
This code can now be added to just the single cluster
operations and not have to worry about splits. It
will also simplify the code that will eventually queue
requests which require a cluster allocation if an
allocation is already in progress.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I22a850e6e23e3b3be31183b28d01d57c163b25b1
Reviewed-on: https://review.gerrithub.io/395035
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
A future patch will queue an operation in
_spdk_blob_request_submit_op_single if the cluster is not allocated
and another allocation is already in progress. If the queueing fails
because of no channel resources, we want to fail the callback and
need to do this before creating the batch.
This causes a bit of code duplication, but in the end should make the
code easier to read.
While here, pass spdk_blob instead of spdk_blob_data to
_spdk_blob_request_submit_op_single. This simplifies a future patch
which will need the spdk_blob when queueing an operation. It also
incidentally makes it consistent with _spdk_blob_request_submit_op_split.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib0ec0e5138eac5bf208fcde6676cd2a77a1a663f
Reviewed-on: https://review.gerrithub.io/395196
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Make it non vhost-user specific. The same
code will be later reused by upcoming
construct_virtio_pci_bdev RPC call.
Change-Id: I6964fedea2a1d52c9b06d5fd8cafe745e06df0ab
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394504
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
The default name remains as sprintf("VirtioScsi%d", counter++).
Change-Id: I244c1389f8dfac16f61aaf1d610a7888dd55e56c
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394446
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This is just the API introduction. See subsequent
patches for its RPC usage.
Change-Id: Iadb7c9bf6a56ab4330c9f2215c6006a2935d208d
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394445
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Currently setting cpumask to portal is possible only through
iSCSI.conf. This patch makes possible for any user to set cpumask
through JSON-RPC too.
The following are done in this patch:
- To keep compatibility, cpumask parameter handled as optional.
- Python test code is added.
- Current python script for JSON-RPC does not work correctly for
IPv6 and the issue is fixed.
Change-Id: I42ef397ce95040a36db4430417a35e9e97527477
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/391728
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This concept was added very early in the SPDK iSCSI target
development process when there was a high focus on maximizing
throughput on a single iSCSI connection with 512-byte reads.
Realistically, in multi-connection environments focused on
predominantly 4KB (or more) workloads, this concept loses its
effectiveness - it is relatively rare that PDUs from multiple
I/O would coalesce within the default 8us flush timeout period.
There were no users of flush timeout in the SPDK tree - it was
not even documented in the iSCSI configuration file example and
there was no way to modify it at run time. So leave the change
out of the CHANGELOG.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I08c3e959fb3945dc2c9cb89248305d0c88aa778f
Reviewed-on: https://review.gerrithub.io/395520
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Currently cpumask cannot be specified for each portal when it is
created by JSON-RPC and portal group creation is not unified
between iSCSI.conf and JSON-RPC.
This patch does the following:
- cpumask string is decoded in spdk_iscsi_portal_create() which
is common between iSCSI.conf and JSON-RPC.
- parsing configline of portal is difficult to understand and
hence it is refactored.
- UT code is added.
JSON-RPC will be added by the next patch.
Change-Id: I13b9989263fae5facff260de32a55ec99f5d5a06
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/392447
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For I/O that do not span a cluster boundary, just issue
a single batch command to underlying block device.
For I/O that do span a cluster boundary, issue a batch
command for each against the blob (not the block device)
for each cluster accessed by the I/O.
This is all in preparation for upcoming patches which
enable thin provisioning and hence cluster allocation
in the I/O path. It will simplify implementation of
the cluster allocation path since now that code only
needs to be concerned with a single allocation at once.
Splitting for readv/writev will be handled in a
later patch.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia2341abbda599dace3357c4eec06ab6602ef81a8
Reviewed-on: https://review.gerrithub.io/395027
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This breaks out the logic for building a batch for
non-iov operations to a separate function. Future
patches will do further modifications on this
new function - separating it out into two separate
functions, one for operations that span a cluster boundary
and one for those that do not.
No functional change here - this is just moving code
around to reduce size of upcoming patches.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id1e67e6305d7ba2317700e1477e12c749ebf664c
Reviewed-on: https://review.gerrithub.io/395026
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
These will be used to implement user request splitting
on cluster boundaries.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I29e00ae9555fdd8a149e92be3cf88a2e528f5c0e
Reviewed-on: https://review.gerrithub.io/395021
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This clarifies that the read/write/etc. operation is
being performaned on the block device. This
clarification will be important in some upcoming
patches which will batch similar operations on a
blob (as part of splitting a user request into smaller
operations if it spans a cluster boundary).
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic574804c15f769ee80c1b6f68ca3b77ec910f0f6
Reviewed-on: https://review.gerrithub.io/395017
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
These were unused in blobstore so remove them for now.
This reduces number of code changes in some upcoming
patches.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I02e04f7d84fff1d2a70f371d222425f373a5d4ff
Reviewed-on: https://review.gerrithub.io/395016
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
There existing some cases like in our UT code, the
start_poller_fn and stop_poller_fn is not configured.
Add a check here and properly handle these cases.
Change-Id: Iaac83d78547432584f6de0c54e9d2e06b9b35741
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/392996
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When there are two bdevs built on the same io_device,
it is possible that one bdev entirely saturates
underlying queue, not letting the second bdev issue
a single I/O. The second bdev will silently fail any
subsequent I/O and append it to the nomem_io list.
However, since we resend I/O only from I/O completion
callback and there's no outstanding I/O for that bdev
(io_outstanding==0), the I/O will never be resent.
It'll be stuck in nomem_io forever.
This patch makes nomem_io list to be shared between
bdevs built on the same device. It is now possible
that I/O completion callback from one bdev will retry
sending I/O from other bdev.
The shared bdev data is based on thread-local
bdev_mgmt_channel, so doesn't need any external
synchronization.
Change-Id: Ia5ac3a1627ce3de4087e43907c329aa7d07ed7c7
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394658
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
We would still like to keep up this Macro, then if users
want to change this value with their own ones,
they can change it.
Change-Id: Ic13c9ba0889955cc41a75f700c58822728782f6c
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/394311
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Individual threads can now pause and resume subsystems
as needed, so the master_thread isn't used.
Change-Id: I49ccb1f87d3e5956db7bd8fc9da0542c5c487923
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/394260
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use the general purpose mechanism to pause a subsystem when a bdev
is hot removed.
Change-Id: I3cd4fb15140d6682707b7840c6a9c3f63fe48e55
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/394259
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
When the state change is known to fail immediately, use
a return code instead of calling the callback.
Most of the callbacks didn't actually check for errors,
so this patch also doesn't check for errors on return
codes. That should be done in the future.
Change-Id: I67e03f93d7f53892473dfc073f4150e7e620cad2
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/394281
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Eventually this should probably queue, but for now
just fail these commands.
Change-Id: Ie1b1088aa38198c17062ac9874539ad6d99b01eb
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/394125
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
These commands can now pause a subsystem if they need to operate
on it. We don't currently implement any of the NVMe commands
that would need to pause, so this patch is simpler than most
would expect.
Change-Id: I25bfdf8e7577cda2bb0ce248d2889447032b9b4c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/394121
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch makes possible to set and get Header Digest and Data
Digest of target.
Change-Id: I1d1b892f9dfb747c0f5ad8fc4fddef40929b4143
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/394482
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
When a target is created by iSCSI.conf, only valid CHAP params
are passed to spdk_iscsi_tgt_node_construct().
When a target is created by JSON-RPC, help information encourages
users to specify valid CHAP params but
spdk_iscsi_tgt_node_construct() does not check CHAP params and
users can create targets whose CHAP params are invalid.
Change-Id: I7e9057a982f21f04782481cda74208a139c1fdad
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/394481
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add JSON-RPC to add an LUN to an existing target. The parameter
lun_id is optinal and if skipped, the lowest free LUN ID will be
assigned.
This JSON-RPC is tested in iscsi_tgt/rpc_config.
Change-Id: I010619f2d4ccec89c589bb0618466b4d15949ebb
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/385181
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Removing an LUN from an existing iSCSI target is possible by
removing the corresponding BDEV. However adding an LUN to an
existing iSCSI target is not possible yet.
Add a new function spdk_iscsi_tgt_node_add_lun() and related
functions first toward supporting this function.
JSON-RPC for this operation will be submitted an another patch.
Informing the newly added LUN to the initiator is not included
in this patch. Hence this operation is possible only for any
inactive target.
Change-Id: I3a28f4d75a17126e49c9d12ce64c3ad68f231840
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/385180
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This prepares for some future blobstore.c code that will use these
bs_dev functions directly making possible to read from different device
than the one in the current context.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I55b15544ea4b53475c4b72ee5c92ff1c0a2b8c88
Reviewed-on: https://review.gerrithub.io/393781
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I35779dc547e0c084086ec6d9bf44f86850cb7f05
Reviewed-on: https://review.gerrithub.io/393780
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Iadb29a8ce8dcebfea68d4feeb5f3de1bb3124f16
Reviewed-on: https://review.gerrithub.io/392286
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Ib3470fbac49e92308ed14e20ccde6655354f2580
Reviewed-on: https://review.gerrithub.io/389577
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
bdev hotremove event is send directly to hotremove callback given
during spdk_scsi_dev_construct() call. So any bdev hotremove might be
promoted to whole SCSI device removal (like in vhost) wich will trigger
LUN removal. But after returning from hotremove callback LUN and device
might be still referenced (eg to register poller or by outstanding IO).
Even worse: spdk_scsi_dev object is not dynamicaly allocated but
returned from static array which mean there is no way to detect
use-after-free error on spdk_scsi_dev by any tool. This might lead to
using SCSI device that is freed or assigned to different device.
To fix this:
- always delete LUN using hotremove path
- defer spdk_scsi_dev delete/removal after all LUNS are really
deleted.
Change-Id: I65598bf42cd507f620095dff5d32509a0424d060
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/393674
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
In current bdev_virtio_scsi PCI enumerate callback
implementation we rely on a global variable - a global
list of virtio devices. We do not need any opaque
context data inside this callback just yet. It will
be required to add virtio devices in runtime. See the
next patch for details.
Change-Id: I116cbd3bd633f56922eedcc7c07b8c0310e51d49
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394444
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
`rpc.py remove_virtio_scsi_bdev <name>`
The RPC will remove entire Virtio SCSI device together
with all underlying bdevs.
Change-Id: I87f349b0e7543955d54a34406cf73c4a91d1495f
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394170
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Consider the following scenario:
* user deletes a Virtio-SCSI bdev (spdk_bdev_unregister)
* user deletes entire Virtio-SCSI controller (virtio_scsi_dev_remove)
If there were any descriptors open with asynchronous
remove callback specified, the first bdev unregister
won't delete the bdev and won't notify the Virtio-SCSI
controller in any way. Subsequent Virtio-SCSI controller
deletion might result in unregistering the same bdev
again. This patch makes Virtio-SCSI controller open a
dummy descriptor on each of its bdevs, so that it's
notified of any started bdev_unregister
Change-Id: I9a8c841d08393ef1940c4cebc9dfcb58e9b0ac4a
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/394169
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Previous nbd implementation only processes one I/O at a time.
This patch increases spdk_nbd_disk io queue-depth.
lists are added to accommodate and coordinate overplapped nbd_io.
Change-Id: I3bda2c957a561b884ebfe9550b7cb6f3c991aef8
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/392609
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Apply a state for nbd io instead of using flag
bits like payload_in_progress and resp_in_progress.
Change-Id: I331567f7c4d7a6a886a145390a8b1e4db0f6040c
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/392175
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This avoids having to dereference the spdk_bdev_io's
channel in the spdk_bdev_free_io() path.
Cleaning up after hotplug events should ensure that
all associated bdev_ios have been freed (not just
completed) before the bdev channels have been freed,
but this patch gives us some more wiggle room in this
area.
Results in a small (1-2%) performance degradation on
a bdevperf microbenchmark, but should result in no
noticeable difference on any real world workload.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id8b88300fc53e8c0b83309a738a4c3bd2aeaff52
Reviewed-on: https://review.gerrithub.io/394399
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
There is no need to keep a lun name anymore - we always
use the bdev name as the lun name so it is not providing any
additional value. This also keeps us from associating
the same bdev with different LUNs on different iSCSI target
nodes or vhost-scsi controllers.
Side effect of this change is:
1) Use "bdev_name" across the APIs to make it more clear
what these names refer to.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I3d42fde22087352ce1d5dc80178bd8c5cac8cb7c
Reviewed-on: https://review.gerrithub.io/390843
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
If the file have cache data and not flushed, the file length will be
not correct.
Change-Id: Idde23e158c1d010e67579b5d6a8d87d3cfbfed2d
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/393784
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This patch provides logic for returning errors instead of
assert when size is larger than blobstore size.
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: I16d12338e2b682c39bd33d507d57ea126501a0d7
Reviewed-on: https://review.gerrithub.io/392749
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
when test hotplug case with large write command which need r2t, after the
device hot removed, the outstanding r2t commands are waitting for data out PDU,
and the primary commands probably already return with error, then the lun
could be deleted, later the related data out PDU will received and try to send
other r2t, this patch make sure the r2t command sent smoothly even the device
hot removed.
Change-Id: I49061d1258e65cb0964526f755d870ebe7095920
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/394321
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Recovery code did not claim clusters taken by metadata.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: If6726eddd22f4e1a3f9814b2348243155fb0fdb9
Reviewed-on: https://review.gerrithub.io/394173
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This `len` field describes the amount of bytes
*written* by the device. It is used to prevent
the driver from reading buffer memory that wasn't
really written to - it could contain either
leftover values from previous requests or even
be completely unitialized.
In a couple of invalid-request error conditions
we still write the VIRTIO_BLK_S_UNSUPP response while
returning used_len == 0, but that's fine. A properly
written driver will likely handle both of these cases
the same way.
Change-Id: I8e9f3a8051561096e2ee755122d534b4ffd94db1
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/393963
Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This `len` field describes the amount of bytes
*written* by the device. It is used to prevent
the driver from reading buffer memory that wasn't
really written to - it could contain either
leftover values from previous requests or even
be completely unitialized.
Change-Id: I67aee5c8b9455fc86cdeb600ea3a051a7737c66f
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/393962
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Removed task `pending` queue. All tasks were
temporarily put in this queue just to be moved
out of it yet within the same reactor cycle.
Change-Id: I32d402f7abd3cfa21c263f41149425abdc71992f
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/393911
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
