Add an new RPC, nvmf_subsystem_get_listeners.
ANA state is per listener and per subsystem, and is stored in
subsystem listener. We can return ANA state by the existing
nvmf_get_subsystems RPC but it's confusing that listen addresses
have ANA states.
To change ANA state, we will provide a RPC to change ANA state of
only one selected subsystem listener.
To query ANA state, it will be convenient to get ANA states of all
listeners of one selected subsystem.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ic3baad6eac65d7af6e0cab2c4059e1458d41e6e2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4059
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add an new RPC, nvmf_subsystem_get_qpairs to retrieve the list of
qpairs of an NVMe-oF subsystem.
This RPC will be usable to verify if NVMe ANA works.
Pause and resume the subsystem to access the qpairs safely.
One subtle issue remains. The JSON RPC returns success even if
resuming the subsystem fails. Write FIXME to address this.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9d90a01b1117dee00d85b2e21b4f4d02d80db531
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4050
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
liburing.so is a link to actual lib created under /lib, not
/usr/lib{,64}.
Instead of checking hardcoded paths, simply check the ld cache.
Change-Id: I65954490d0b79e17de791d8045a1758cbbb381f5
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4020
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
ELRepo's elrepo-testing holds missing btrfs packages.
On Centos, system repos provide centos-release-ceph-* packages which
install necessary Ceph repositories to provide ceph package.
Also, group these repos into single array and install them in one
bulk. Use yum-config-manager to enable EPEL's|ELRepo's repos instead
of mixing it with unnecessary install step.
Change-Id: I3ba9c6f713ac7e28c7f1ba1e4482d2f2839eeb6a
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4018
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
There are 2 issues:
1. building of ninja fails with missed skbuild dependency.
This dependency can be installed with scikit-build module
2. The build of ninja wheel fails due to cmake version. E.g. centos 7.6
uses cmake 2.8 and cmake 3.x can be installed as another package.
It is needed to create a soft link to cmake3 with `cmake` name so
that cmake3 is used instead of cmake 2.8
Change-Id: I53910d8d73b233d3f3a9059ab10938477c5db57b
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4062
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Most distros ship these drivers as modules, however, some, like
clearlinux, have them compiled directly in the kernel (e.g. nvme).
In case modalias lookup fails, have a fallback prepared just in case.
Change-Id: Ib9da8ff69edbb0c4ce427aa089fa7d04b399a20b
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4047
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The previous approach didn't take into the account systems where root
partition is representend in a virtual form of /dev/root (see
clearlinux). In case the device was bound to virtio this would fool
setup.sh into thinking nothing is mounted and proceed with unbinding
the entire rootfs.
Fix this by checking the maj:min numbers of the device instead of its
name against the mountinfo list.
Change-Id: I0feb2584869f6bb72df3e9a4e619620240cfce3b
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4046
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Make sure that setup.sh waits for block devices during the tests.
This is to make sure that the underlying controllers are ready to
be talked to and minimize the amount of their flakiness.
Change-Id: Id09445de7ac7ccf4c9f679ee2b6c4bdd14c89733
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4029
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This script tries to be smart by mainly looking up all the kernel
events caught over the netlink. This is done in order to mitigate the
amount of flakiness test scripts may suffer from while waiting for
given device to pop up. This mainly concerns nvme devices when they
are moved back from user space - the time from when the ctrl is bound
to the driver until the block subsystem finally gets the device can be
substantial (even > 2s). Consider:
$ setup.sh reset; sleep 10s
vs
$ setup.sh reset; sync_dev_uevents.sh block/disk nvme0n1
Where the latter could recover some of the seconds back:
$ time { ./scripts/setup.sh reset; ./scripts/sync_dev_uevents.sh block/disk nvme0n1 ; }
0000:82:00.0 (8086 0953): uio_pci_generic -> nvme
0000:00:04.0 (8086 0e20): uio_pci_generic -> ioatdma
0000:80:04.0 (8086 0e20): uio_pci_generic -> ioatdma
0000:00:04.1 (8086 0e21): uio_pci_generic -> ioatdma
0000:80:04.1 (8086 0e21): uio_pci_generic -> ioatdma
0000:00:04.2 (8086 0e22): uio_pci_generic -> ioatdma
0000:80:04.2 (8086 0e22): uio_pci_generic -> ioatdma
0000:00:04.3 (8086 0e23): uio_pci_generic -> ioatdma
0000:80:04.3 (8086 0e23): uio_pci_generic -> ioatdma
0000:00:04.4 (8086 0e24): uio_pci_generic -> ioatdma
0000:80:04.4 (8086 0e24): uio_pci_generic -> ioatdma
0000:00:04.5 (8086 0e25): uio_pci_generic -> ioatdma
0000:80:04.5 (8086 0e25): uio_pci_generic -> ioatdma
0000:00:04.6 (8086 0e26): uio_pci_generic -> ioatdma
0000:80:04.6 (8086 0e26): uio_pci_generic -> ioatdma
0000:00:04.7 (8086 0e27): uio_pci_generic -> ioatdma
0000:80:04.7 (8086 0e27): uio_pci_generic -> ioatdma
* Found nvme0n1
real 0m2.427s
user 0m0.465s
sys 0m0.408s
Change-Id: Ib91efc73742fb00b058d4f54976d27390489ddbf
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3655
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ifeadc3fac3f5d09143dd13ef3d510de9cecffb8a
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4032
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
On its own these values don't uniquely identify a particular pci
device, hence there's no much point in storing them like so.
Change-Id: Icc5cfec28fcccae6cd935dff03fe3ef01826edc4
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3950
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
There are some devices for which nvme driver takes a long time to
finalize the unbind stage. With that in mind, each device would
add up a significant amount of time needed for setup.sh to complete.
To mitigate such a scenario, make sure the controllers are unbound
in a parallel fashion.
Examples taken from the system with 19 nvmes on board:
[root@supermicro4 spdk]# time ./scripts/setup.sh &>/dev/null
real 0m36.250s
user 0m1.024s
sys 0m1.990s
[root@supermicro4 spdk]# time ./scripts/setup.sh &>/dev/null
real 0m4.848s
user 0m0.867s
sys 0m17.605s
Also, take note that this is currently done only for the nvme
devices since other, i.e., ioatdma, seem to trigger a BUG in the
kernel when unbound in parallel. Some details here:
https://bugzilla.kernel.org/show_bug.cgi?id=209041
Change-Id: Icaeb2b2ecb306f149587bc5da73743b1519bc5d6
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3893
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This is a first patch from a series attempting to merge similar
routines used throughout setup.sh.
Change-Id: I4c71b88f7556b0e9e2f65a37b8175914022486a4
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3889
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Additionally, fall through potential yum failures to try and install
what's available in repos given system is using.
Change-Id: Icf7f5f3db1d72b67414c7d7e4badb8416a9d25b1
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3873
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
For some RHEL releases, especially 8.x, EPEL repo is crucial in order
to install most of the package dependencies, hence we need to make
sure it's in use.
Change-Id: Iaaa1f901e6fffddb65b98fa98b21b9e77e59760f
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3872
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
On rhel 8.x standard appstream and baseos repos don't provide python
package. Use python36 instead as it's available there.
Change-Id: I125aa3370bdb11f67f2774e00e70196df9680f13
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3871
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This limitation may simply fail if /usr is part of a separate
mount|fs so softlink it instead.
Change-Id: I1e0387c2792c0bbd20d83e9739b3882a7cbbbf50
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3870
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
When using --dif-type option --md-size should be
required as well.
Update & improve bdev_null_create rpc.py help
messages as well.
Change-Id: I6588a97aef6c8792bab7a41ece17c0461bb36844
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3949
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Allow toggling log timestamps on and off by adding new RPC call.
Change-Id: I34c84bf89fae352ade266fbf7fd20594ff67bced
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2024
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Add an abstract method in Initiator class to get
rid of pylint "no-member" error.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I83ac1a498d55f6390a5b36038a9532121a9e5d0a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3922
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Use --dif-type and --md-size option for bdev_null_create
rpc when creating null bdevs. Needed for DIF/DIX comparison
tests.
Change-Id: I15de25d498ff7c239cfca291e10aa16acb4f09eb
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3919
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
So far only single null block device was allowed in
the setup for the purpose of latency measurements.
Rename setting to "null_block_devices" and use an int,
rather than a bool, so that multiple null block
devices are created.
Change-Id: Id543cf3bfbc7d9c171f8605909e1971cabbdfcc7
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3900
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This is done to be more conscious of what net|rxe devices ar currently
in use to make sure we don't unnecessarily write to infiniband class.
Also, this makes use of net/infiniband class more consistent between
both the add and remove actions.
Change-Id: I0ba3bc3a14047e12a804bfdf3aca8c089d34d6b0
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3825
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
- reorganize the code to cleanup main loop
- other small syntax improvements
Signed-off-by: Pawel Piatek <pawelx.piatek@intel.com>
Change-Id: I0a2b9ac9e78db2f96ed49d3e7e4d938425c76fa9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3720
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
- rename to get_box_type
- use hash to map distro to box type
Signed-off-by: Pawel Piatek <pawelx.piatek@intel.com>
Change-Id: I5608b04d813ad1b37ae4dbbc418217a61bd96a52
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3480
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
To solve below issue:
https://github.com/spdk/spdk/issues/1550
The default one is Malloc0 which is not used.
Change-Id: I36b64457e8583c8243b8b9d6714c42e2d43ade7d
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3831
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This adds basic bash-completion support for apps commonly used within
the repo. It's meant to work with standard bash-completion package
available for most of the distros out there.
Change-Id: If2f92c85537aa7a56d32b46450f2afa0298e7fed
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3526
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I6c836114b71b1c5ee1dbbf7900c7314bb5619a48
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3551
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
link_up_rxes() is not used anywhere so remove it. That said, make sure
that link_up() is called while adding the rxe device(s).
Change-Id: Ic0f9573b911d83d58b42f62d104ef916aa010e4e
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3824
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
