to avoid renumbering test cases in the future.
SIGTERM should be called as last test case, because
it kills vhost instance. If it is called in the middle
another test case should start vhost again.
Want to avoid this.
Change-Id: Idd3e1492bd83e911d17e730c680187cd98f0c125
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/399131
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Rather than using 'cd' and then ./-relative paths to launch the vhost
test scripts, just specify the full path relative to $WORKDIR. This
should allow print_backtrace to find the script sources for these tests
when failures occur.
Change-Id: I17ca1ebea393df4da427668645ba6f99356a57da
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398972
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>
Also tweak the 'scanbuild_make' timing marker so that it is only called
'scanbuild_make' if scan-build is in use.
Change-Id: I413b8a7cfd4eb2c1090dd73e78aef4ba563d0cc5
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398995
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
QEMU overrides this value anyway, but other
virtio implementation may not. Setting
blkcfg->num_queues to its target value
won't break anything, and will help out
SPDK virtio-blk driver implementation.
For equivalent field in virtio-scsi, the
Virtio 1.0 spec says:
```
num_queues
is the total number of request virtqueues exposed by the device. The
driver MAY use only one request queue, or it can use more to achieve
better performance.
```
Multiqueue for virtio-blk is oficially still
unsupported, so the blkcfg->num_queues remains
undocumented. Let's implement it adequately
to the virtio-scsi spec.
Change-Id: Ief520385997bb7e745ed17501a972c55955346d2
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/399121
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
New vhost user messages GET_CONFIG/SET_CONFIG can be used for
vhost-blk for the purpose to get configuration parameter such
as: Capacity and block size. This commit enable this feature,
users don't need to append capacity any more when started
QEMU. Also event notifier is added for the purpose to change
capacity of block device while QEMU is running.
Also re-enable the vhost-blk tests.
Change-Id: I06ef697984eeea3abbbd655bdcaccaa3b7aa72d7
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/386546
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>
Install the new custom qemu branch in a different location
(/usr/local/qemu/spdk-2.12-pre) so it can coexist with the current qemu
branch.
Change-Id: I6986dcaef804dcefc4e77c83c019ba0f3be8dbd6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398983
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>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
For the purpose switching to a new QEMU branch smoothly, disable
vhost-blk autotest temporarily, and we will enable it at the next
commit.
Change-Id: I0368edcbea2468d10f1e95da39b09199d39c7216
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/395885
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Problem
--------------------
Even though rpc.py has the support of IPV6 it was unable to parse IPV6 address family
Solution
-------------------
Instead of creating the key value pair based on all occurrence of “:” operand .
The key value pair is created on the first occurrence of “:” operand.
The above solution works for IPV4 as well as IPV6.
modified: scripts/rpc/nvmf.py
Change-Id: Ibdc1c695eed724374fefe265e07958c698b5fb25
Signed-off-by: suman chakraborty <suman.chakraborty@wdc.com>
Reviewed-on: https://review.gerrithub.io/399132
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
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>
Add vhost-scsi and vhost-blk multi-os test case
which used Ubuntu and Centos VM in nightly
Fix vhost nightly readonly case for vm shutdown timeout
The 10s isn't long enough for the VM to shutdown
so the script trap error in line "((timeo-=1))"
in common.sh script and exit
Change-Id: I5a44a2b1bf6b3247383603d5896b8bdde16a9a45
Signed-off-by: Chen Wang <chenx.wang@intel.com>
Reviewed-on: https://review.gerrithub.io/393602
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: qun wan <uniqueanna@hotmail.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
No functional change - this just reduces an upcoming
patch that will make further changes to this function.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic186f315b09ebadfba6120a70fac03ebb9beaffb
Reviewed-on: https://review.gerrithub.io/398856
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
The socket-related code was already broken out into
lib/net/sock.c, so break out the header portions
from include/spdk/net.h into its own sock.h.
This prepares for some upcoming changes in how
TCP sockets are abstracted, to enable alternative
userspace TCP stack implementations to be used with
SPDK.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I40b162e72ea80c235b49f10b17c2085fcfb385d4
Reviewed-on: https://review.gerrithub.io/398851
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie76019c2d5a82e24f0ec93b0c046fdbeea8bd337
Reviewed-on: https://review.gerrithub.io/398850
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8a698531b261c303babac3c5c1594919430c0bac
Reviewed-on: https://review.gerrithub.io/398855
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
DPDK can take a long time to build - even when nothing
in DPDK has changed. So for users who are using SPDK's
DPDK submodule, add a SKIP_DPDK_BUILD option. When
this is set to 1, the build will skip building the
DPDK submodule if it is being used.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib48238fe607276fb69e495f8b7318e5b82754276
Reviewed-on: https://review.gerrithub.io/398849
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
The "core" parameter is no longer accepted for construct_nvmf_subsystem,
and it is also no longer returned as part of the get_nvmf_subsystems
result. Update the documentation to match.
Fixes: 7b397f6281 ("nvmf: No longer tie subsystems to CPU cores")
Change-Id: Ibec85de6942f8da99f195b747423040c54a44959
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398873
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Change-Id: I351c7bda1b32dc161c49d47fb2887595099e4be6
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/396696
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 fixes the new nvmf_subsystem_add_listener RPC method to use "nqn"
as the parameter name to be consistent with the other NVMe-oF target RPC
methods.
Since this method is brand new, no provision for compatibility with the
old name is provided.
Fixes: 6336217e16 ("nvmf: Add rpc to add listeners to subsystems")
Change-Id: I86263fb1c53ed6ab48fe70917686d78a4a7ce28d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398870
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Change-Id: I4c75bb8e44e2dadd7b9046beb8ec3b4a39707626
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/392176
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will be reused in an upcoming patch to remove listen addresses.
Change-Id: Ia168ae1741f998fc7085ae05fb08406856aff6aa
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398689
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Change-Id: I53ffdd061ba068070f66a752780229ecfe53e737
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398688
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Don't allow a duplicate listener to be added to a subsystem.
Change-Id: I1a5f2ae09146a31859921430fb716a36d7f2048e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398652
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>
Change-Id: I04820b2da1e75d1e84d3a5f4a274a11c9f154b82
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/398758
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>
The multithread testing framework is well positioned
for the QoS testing. As QoS will initiate poller for
the rate limiting. Added the related poller functions
and proper poll_thread handling here.
Change-Id: I253a72d3f85af381703dde4e4d0d7366fe682d78
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/393012
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The type of a queue is purely a function of its queue ID.
Add a helper function, spdk_nvmf_qpair_is_admin_queue(), to make the
logic more obvious (akin to nvme_qpair_is_admin_queue() in the NVMe
library).
Change-Id: I7cf5a82d1e3dc50834cc2ff03f5f88c5719c6952
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398663
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
vm_setup was creating setup with only one disk type defined.
With this patch it can set disk type per disk by adding disk type
after disk name separated by coma
e.g.
vm_setup --disks="Nvme0n1,spdk_vhost_scsi"
Change-Id: Ib89f440210468f4c9082aadcf07ce461e34c4ff8
Signed-off-by: Pawel Niedzwiecki <pawelx.niedzwiecki@intel.com>
Reviewed-on: https://review.gerrithub.io/395980
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
The kernel vhost target enables notifying after processing
each interrupt in a following manner.
* unset NO_NOTIFY flag
* mb()
* check avail ring for slipped requests
And we do the following after issuing each I/O.
* update avail ring
* check NO_NOTIFY
If NO_NOTIFY check is reordered and read first, we might read
old `true` value, and avail ring might be updated after the
kernel has already done its check. This easily leads to deadlock.
Change-Id: I6bb4490775dfed6fb2987e97c39b713054ae26ad
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396499
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 didn't make much sense anyway.
Removed them so there's less code
Change-Id: I3247c640b13e5847b5f6d822072b63cad269ac8c
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/398467
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>
Change-Id: Id1e2a7da0db319c4d43988ec4991462fc956dde2
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/398542
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>
Will follow this up with a doc change but want to make sure we're
all good with it first. This is meant to not only beef up
blobstore testing but provide the 'match' utilitiy for all test
cases where we are currently calling an executable and either
counting only on a return code to determine success or failure
or worse yet we're just running it and if it doesn't explode we
assume its a pass.
The 'match' util was borrowed from the PMDK folks after first
adding the "ignore" feature upstream to make it easier to use
in SPDK. It works like this:
When the developer checks in a test they create and check in
the output of the test with two different file extensions:
.ignore: should include a string per line for output lines
that we want to totally ignore typically because they're
platform specific so the output could be different from
machine to machine. In this case I'm ignoring all output
lines with 'DPDK' or 'EAL' or '...' in them. The first
few are obvious, the last is because the test tool will
print a varrying number of these as progress indicators.
.match: this is a copy of the output that the developer
'fixes' up by replacing platform specific output strings
with replacable tokens as described in the 'match' help.
This is where you'd want to match an entire line minus
something like a CPU count or free block count or
something. The 'ignore' feature was added simply so we
wouldn't have to edit every single line of an output
file that had DPDK or EAL in it.
Then you modify the test script to save the output and
smply run the match util providing the name of the
match file and if it fails to match the actual output
with the saved output that's been token'ized the script
will error.
The obvious advantage here is that now we can confirm all
of the output from a test executable is as we expect.
Change-Id: I701d36ee83d37b6193e16ed3171e114f814e5eb3
Signed-off-by: paul luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/397027
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>
if vhost retuns error on exit using spdk_vhost_kill, the tests
would continue instead failing
Change-Id: I923d579f9e8614c39b27186418439d185770b3a7
Signed-off-by: Pawel Niedzwiecki <pawelx.niedzwiecki@intel.com>
Reviewed-on: https://review.gerrithub.io/398622
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>
Tree-wide cleanup of all instances of printf()-style functions where a
format string contains a space before a newline character.
Change-Id: Ib5b5861e97bed9e9d62db03875547e3f771f4769
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/397031
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
This section was deprecated in the previous release, and its removal
was planned for v18.04.
Users may configure the RPC listen address via the command-line '-r'
parameter.
Change-Id: Ic714417b87d422492b8742852c16983eb61793c2
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/398239
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This patch adds two internal functions for the channel
create and destroy.
Change-Id: I3c5ca7a0633e5a5b3f95a36ed03f2b4cb4792e4f
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/395677
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
In the different cases to creat a bdev channel, added
a common cleanup function to destroy the resource.
The same function is also called at the bdev channel
destruction.
Change-Id: I81b60cab9df3a8975b0a9982482c9d27899d8a79
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/398351
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This field was only required to check
if we can safely upcast vdev object.
We can just as well check vdev->backend
instead. The vdev->type is not needed here.
Change-Id: I525350957406d4299151e0557b9025ca7bea5371
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396584
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When a target is removed, spdk_scsi_dev_destruct
removes its luns, consequently closing bdev descriptors
and in vhost case - triggering target hotremove again.
This doesn't really have any negative consequences, as
the second hotremove just fails silently. But let's
cleanup this for sanity.
Change-Id: I47c76814696b49905b9fb00667e9529db3f58c12
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396575
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Instead of:
* spdk_vhost_scsi_dev_remove(vdev)
* spdk_vhost_blk_dev_remove(vdev)
we now have
* spdk_vhost_dev_remove(vdev)
All the logic is already handled internally. This patch only
changes the API. Also, previous vhost_dev_construct()/remove()
functions have been renamed to vhost_dev_register()/unregister()
because that's what they really do.
Change-Id: I7dd0d77bc5b633bec075e0a71345ddbed62697b4
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396574
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
We already have a vhost subsystem and have to do
its fini anyway, so let's remove application fini.
Technically, any SPDK app can use vhost library
and create vhost devices, so we need to delete them
on shutdown in all cases - in module fini.
Change-Id: I013b45b4e73fd19288b9c10bc995dc24894ecf35
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/396573
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Track individual parts of the setup_vm phase in the vhost initiator test
that builds SPDK inside the VM.
Change-Id: I25c924808db0f934dd53043b06668da96099cf62
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/397285
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Checking uniqueness of initiator group is done without mutex and
before register/unregister. This is not thread-safe.
This patch is a preparation to dynamic addition of initiator
information to existing initiator groups.
Change-Id: I44f48c857210522eee70d14bc3735ec73b0c5c6f
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/397032
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>
Orderings of portal group create operation are diferrent between
config file and JSON-RPC. Unification is necessary for correct
concurrency control and the global accept poller like NVMf-tgt.
Hence unify the ordering of operations in this patch.
Common ordering of portal group create operation between configuration
file and JSON-RPC after this patch is the following:
- create a portal group
- create portals
- add the portals to the portal group
- open the portals of the portal group
- add the portal group to the global portal group list
After this patch, the gap between listening socket and accepting socket
will be increased a little when portals groups are creted by config file.
However this will cause no issue because of the TCP backlog and resend
mechanism.
Besides, necessary concurrency control is added and minor refactoring
is done.
About portal group delete operation, orderings of application shutdown
and JSON-RPC are already unified.
Change-Id: I1db7ef4400388127134d7734c68e138a4573b734
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/396848
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>
Checking uniqueness of portal group is done without mutex and
before register/unregister. This is not thread-safe.
Hence this patch is added to ensure PG uniqueness.
A little related refactoring is also done.
Change-Id: Iaa3b5e380f2be5cfdaa2d69f9f2763c98954b0c3
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/396847
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>
Network portal must be unique globally but mutex is not added yet.
This patch is added to ensure it.
Change-Id: I3cdd85fd524b0da767d3cd83022e0637f3a32bc9
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/396846
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>
