Maximum Data Transfer Size indicates the maximum data transfer size
for a command that transfers data between memory and the controller.
SPDK NVMe driver will split IO request based on MDTS and stripe
boundary, however, if user submited a big enough IO size, SPDK
NVMe driver will run out of requests data structure and return
-EINVAL as the error code, it's OK for application who call SPDK
NVMe driver directly.
SPDK bdev module also will do similar split based on `max_segment_size`
and `max_num_segments`, but we don't set them for bdev/nvme driver,
once there was a big enough IO request, the NVMe driver will return
-EINVAL to bdev module, here we set `max_segment_size` based on MDTS
and set `max_num_segments` based on number of requests of NVMe
controller.
Fix#2403.
Change-Id: Ic6e14a0b12413783597122285ac648b87e1f1e16
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12186
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 <smatsumoto@nvidia.com>
Only 4 bytes or 8 bytes are valid numbers when to access NVMe
registers, add the check here.
Fix issue #2495.
Change-Id: I63b6e16a156f6eba17f397ec9d1a447e6a80b4da
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12643
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Reviewed-by: John Levon <levon@movementarian.org>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
During ctrlr reconnection, spdk_nvme_ctrlr_reconnect_poll_async()
is executed by a non-timed poller.
We should poll adminq more often during ctrlr disconnection too.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ib1f5b41015aed20deda8df6f2c837981ac233c04
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12615
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
As described in the NVMe specification, a controller level reset
includes the following actions:
- the controller stops processing any outstanding admin or I/O
commands;
- all I/O SQs and CQs are deleted.
In a full controller reset sequence for a PCIe controller, if we do
a controller level reset first, we can abort outstanding commands
after the hardware has actually been stopped.
For NVMe-oF controller, each I/O qpair is an independent network
connection and is disconnected safely. We do not want to change
NVMe-oF controller.
Fixes the issue #2360
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: If05febac74705bfd3df5abd15064c1203126e027
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12447
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
For PCIe transport, we need to stop any activity of the controller
before deleting I/O qpair resource in a controller reset sequence.
However, we set I/O qpairs to failed before disabling a controller.
In the NVMe bdev module, this caused disconnected qpair callback to
delete I/O qpairs before disabling the controller.
Hence, change the code slightly to set I/O qpairs to failed only if
reset is synchronous to keep backward compatibility.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ica71aad0a1dabce45616dfdfff5f11b07131bbd1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12736
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
The test uses `scripts/sma-client.py` to send a series of gRPC methods
that are serviced by the SMA then verify that new vhost device are
available on qemu guest os.
Change-Id: Ic4a1d56da59a636556fd6dfd4787fc613a4eb325
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12412
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This patch utilize generic sma implementation by adding vhost-user
devices manager. It's allow to expose virtualized block devices to
QEMU instances or other arbitrary processes.
Max device capacity depend on available `pci-bridge`
```
devices:
- name: 'vhost_blk'
params:
buses:
- name: 'pci.spdk.0'
count: 32
- name: 'pci.spdk.1'
count: 32
qmp_addr: 127.0.0.1
qmp_port: 9090
```
To attach `pci-bridge` at boot time we need to run qemu with this option
```
device pci-bridge,chassis_nr=1,id=pci.spdk.0
device pci-bridge,chassis_nr=2,id=pci.spdk.1
``
Change-Id: Idbe841b12cf59975ff0e6717f8dc905d38379357
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12207
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Artek Koltun <artsiom.koltun@intel.com>
This patch adds python scripts for basic handling of QEMU Machine
Protocol connection for use with vfio_user. Basic low-level connection
and error handling added as extended socket class.
QMPClient: Wrapper class for handling connection with QMP was added.
Protected class method _receive relies on the fact, that QMP always
sends command 'in one line' - each one is always terminated with CRLF.
https://github.com/qemu/qemu/blob/master/docs/interop/qmp-spec.txt
Usage example, command line calls:
1) Without event parameter:
```json
{
"request": {
"execute": "device-list-properties",
"arguments": {
"typename": "vfiouser-0-1-1"
}
}
}
```
2) With event parameter specified. Specifying 'event'
parameter will set script to block wait for occurrence
of such one after a valid execution of specified request:
```json
{
"event": {
"event": "DEVICE_DELETED",
"data": {
"device": "vfiouser-0-1-1"
}
},
"request": {
"execute": "device_del",
"arguments": {
"id": "vfiouser-0-1-1"
}
}
}
```
Signed-off-by: mionsz <milosz.linkiewicz@intel.com>
Change-Id: I7a548f61d31097ad425994b2ea4bac8bc6800b5b
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11106
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Since SMA keeps track of attached volumes and a volume might get
disconnected at any time (in which case they're also removed from the
host side), a mechanism is needed to monitor if any of the volumes are
no longer accessible.
This patch implements that mechanism by adding a new thread running in
the background that will periodically (60s by default) send a
bdev_get_bdevs RPC and check that all previously attached volumes are
available. If any of them are not, it'll remove it and stop the
associated discovery services (if their refcount goes down to zero).
The period can be changed through the `volume_cleanup_period` variable
in the config file.
One important thing to note is that we assume that any intermittent
connection issues are handled internally by the SPDK application and a
bdev is only removed after all reconnection attempts are performed.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I5b9e63698879527d9f79156a0eda1c8bc5e66def
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12699
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
The test checks that it's possible to attach remote volumes through
discovery service. The tests starts up three SPDK instances: one is being
managed by SMA, while the other two are acting as remote storage nodes.
The test verifies that volumes can be successfully discvered and
attached to a device and that the connections to the discovery
subsystems are refcounted correctly.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I3488b822880b99ef9cfd70e03de3ed9054c13901
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12414
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 patch adds support for connecting volumes via discovery service.
The user specifies a volume UUID/GUID and a list of discovery endpoints,
which are then used to start the discovery service on and attach the
volume to a device.
SMA will keep track of the attached volumes and will also refcount the
connections to discovery services. So if two volumes are attached using
the same discovery endpoint, it'll be disconnected only after both of
them are detached.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ie8ea50a2a784cf0db8a5953234c6bb2b68685d7c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12413
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 will make it easier to control various options through config file.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I19ab9c020d536521f5985227bc43965bd04d2707
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12724
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ibe602571413531f151136a68ad4bdcd0a6be32e9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12117
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 of the devices need to send RPCs during initialization (e.g. create
a transport), so we need to take care that they're initialized only
after we are certain that the SPDK process is listening.
The mechanisim is similar to the `waitforlisten` function used in our
test scripts - it sends a series of `rpc_get_methods` calls and times
out after a period of time if no response is received. For now, the
timeout is hardcoded to 60s.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Iddadc04ad4c486d2894bc40e1a899a9d204400fc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11802
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
The test verifies that it's possible to register device managers from
out-of-tree plugins. The test defines two plugins, each defining two
device managers implementing the same two protocols and verifies that
it's possible to register different combinations.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I2144b40db603fea95bf8b571777e6662b4de9bc7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11729
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
The test uses `scripts/sma-client.py` to send a series of gRPC methods
that are serviced by the SMA and then uses the regular SPDK RPC
interface to verify its effects on the application.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I2b61c61cb475ec906bb6a594cda3082c0bd8ab44
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11728
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
If the app is killed with either one of these signals, it'll shutdown
gracefully and exit with zero.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I630f9e34c3ccb382c8e1b53d0f589f3ec4d1483b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11727
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 allows a user to specify the types of devices to support. Those
not specified in the config, won't get initialized and will not service
user's requests.
Additionally, each device manager can now receive its own configuration
in its init() method.
The device configuration is structured as a list of objects with two
properties: "name" and "params". The former identifies a device to
enable, while the latter contains a set of options (if any) specific to
that device manager. For instance:
```
devices:
- name: 'nvmf-tcp'
params:
max_queue_depth: 256
io_unit_size: 8192
- name: 'nvmf-vfiouser'
```
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I967016502ad93c243b3a7af58992bde14c44953c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11713
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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 patch adds support for loading configuration from a yaml-formatted
file. The command-line options takes precedence over those from the
config file.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Iafb5b4aec0a64afe358fce2d6c55361fc4093223
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11712
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Monica Kenguva <monica.kenguva@intel.com>
The CSTS.SHN is changed only in shutting down the controller,
nvmf library already ensure that all the outstanding IOs will
be flushed before that, so we can remove this check here.
Change-Id: Ib93a256e986b7b2ec1da0fc7992feb3a02c1d657
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11674
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
After finishing migration in source VM, the subsystem is in
PAUSED state, the controller is dead for the source VM, we will
destroy the controller when disconnecting socket, but after that,
we should RESUME the subsystem so that it can be ready for the
next new client.
Fix issue #2363.
Change-Id: Icf0999b9085cebe8be4c8783e1a43bb13d4f7987
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11422
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This OCF update is mainly focused on:
- Fix OCF crash when handling flush I/O with large size
Previous implementation attempts to use cache data I/O interface
(ocf_submit_cache_reqs function) instead of submitting empty flush to
the underlying cache device. This function is designed to read/write
from mapped cachelines while there is no traversation/mapping
performed on flush I/O.
If request map allocation succeeds, this results in sending I/O to
addres 0 with size and flags inherited from the top adapter I/O.
This doesn't make any sense, and can even result in invalid I/O if the
size is greater than cache device size.
Even worse, if flush request map allocation fails (which happens
always in case of large flush requests) then the erroneous call to
ocf_submit_cache_reqs results in NULL pointer dereference.
Other fixes included in this OCF update:
- Fix race condition in fast path
- Fix use after free during metadata load
- Fix cleaner hung after cleaning policy switch
- Fix use after free during volume close
- Fix stack overflow when trying to insert to disabled io class
and the sequential cutoff is on
Signed-off-by: Rafal Stefanowski <rafal.stefanowski@intel.com>
Change-Id: Id38f54df7b6c58c84f7f15f4354530755eb9171f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12744
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
These were deprecated in 2019, it's time to remove
support for them now.
Change-Id: Ic80ce74344b24814dad792cfff6a4791d0430527
Signed-off-by: wanghailiangx <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12741
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
We try to modprobe msr so that DPDK can more easily figure
out TSC on x86 systems. But that just results in a
"Module msr not found" message on non-x86 systems. setup.sh
still works, but the error message makes it appear
otherwise. So just redirect the error message to reduce
this confusion.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I71074c7836519b003933551ce51a193070b1bfe6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12671
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Some environments (such as a VM on my M1-based
MacBook Pro) report CPU with an Implementer+Part Number
that DPDK doesn't explicitly support. In my case, it
is Implementer ARM (0x41) but no part number (0x0).
DPDK doesn't just default to a generic build in this
case, it requires the user to specify
-Dplatform=generic.
We could try to plumb this into a configure option,
but for now I'm fine in my dev environment to just
specify this DPDKBUILD_FLAGS in my environment to
work around it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0465e3747de9b1ebacaf4ca53c2e6ccf36db1091
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12670
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
DPDK reports that -Dmachine is deprecated, we should
use -Dcpu_instruction_set instead.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1dbd81fe9f34d875bf00c0fa3a7868a8fc30cac3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12669
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Spread out used block devices evenly between
used initiators and then between initiators NIC
interfaces, so that in result each initiator
connects to the same number of subsystems.
Previous version had problems spreading block
devices / subsystems between interfaces in case
division remainder was not 0.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I045dc9313b0613685fc997f821fe66238026a645
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12732
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Save Initiator addressing information in Target object.
This will be needed later to properly spread created
subsystems between initiators and NICs.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I9e5545b7847bb16f3ce3d0ee56038003fda74e6d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12731
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
The completion callback of `spdk_nvmf_subsystem_resume`
and `spdk_nvmf_subsystem_pause` can run in different
core other than the `vfu_ctx` core, this may lead to
race condition when changing controller's state. Here
we use a thread message to change it in the same thread
context.
Change-Id: I53d139adcca6ff72a3b91a2a931f1239f3271fa9
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12558
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: John Levon <levon@movementarian.org>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
With high number of NVMe drives being attached to
SPDK it might take some time to fully initialize all
controllers. Wait a bit longer before trying to
initialize performance tests.
Change-Id: I5104f962e667ee39db6054750afa0fe7bcb6f26a
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11424
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
.driver_specific.nvme is an array, so return all elements
of this array before trying to access .pci_address field.
We only expect a single PCIe address in this test, so it's
safe.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I0697ab2ffb1adff2c6f288bba460858bf594b219
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11406
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The following patches swaps the ordering of destrloying I/O qpairs
and disconnecting a controller for PCIe transport.
prepare_for_reset is a flag for PCIe transport.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I3009de9fea089fc93ecf87adba42e85c9a77e715
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12582
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
As described in the previous patches, we need to delete all I/O
SQ/CQs before aborting trackers when disconnecting a controller.
The following patches reorder the operations. This patch changes
adminq disconnection to initiate a Controller Level Reset and
adminq completion processes it if ctrlr->is_disconnecting is true.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I64f06bae2ce8a9127124029fd042db0028198e3c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12560
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Make this a transport-level decision instead. TCP and RDMA do want to
abort, but PCIe cannot because these commands may still be receiving DMA
operations from the device.
Change-Id: I305acddc3819c903eb3217e8f710d4216d0b3931
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11509
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Previously, we did not do any Controller Level Reset when disconnecting
the admin qpair.
However, for PCIe transport, we need to stop any activity of the
controller, i.e., delete all I/O SQ and CQs before
nvme_transport_ctrlr_disconnect_qpair_done() calls
nvme_transport_qpair_abort_reqs() (i.e., nvme_pcie_qpair_abort_trackers()).
Otherwise, some corruption may occur because completed I/Os may still be
in progress on the NVMe device.
Not to change any public API, nvme_pcie_ctrlr_disconnect_qpair() is a
convenient place to initiate a Controller Level Reset because it is
called from spdk_nvme_ctrlr_disconnect(). Then
nvme_pcie_qpair_process_completions() can process it until completion.
However, necessary functions are not accessible from PCIe transport.
This patch adds two helper functions and guards us from some undesirable
behaviors because it was not assumed that nvme_ctrlr_process_init() is
called from the completion context and ends in the middle of transition.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I3d986e94ba71b83beeff7e75cf92033b5fa6f075
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12559
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
When a new cluster is added to a thin provisioned blob,
md_page is allocated to update extents in base dev
This memory allocation reduces perfromance, it can
take 250usec - 1 msec on ARM platform.
Since we may have only 1 outstainding cluster
allocation per io_channel, we can preallcoate md_page
on each channel and remove dynamic memory allocation.
With this change blob_write_extent_page() expects
that md_page is given by the caller. Sicne this function
is also used during snapshot deletion, this patch also
updates this process. Now we allocate a single page
and reuse it for each extent in the snapshot.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I815a4c8c69bd38d8eff4f45c088e5d05215b9e57
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12129
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Compilation with newer openssl fails, as MD5_*() API is now marked
deprecated. Update submodule to fix this.
Fixes#2459
Signed-off-by: John Levon <john.levon@nutanix.com>
Change-Id: I64f5e1946a956ed5252e4f4213399ee2808a2e2a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12639
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
To fix issue: #2484
When unregistering the bdev, will send out the message
to each thread to abort all the IOs including IOs from
nomem_io queue, need_buf_small queue and need_buf_large queue.
The new SPDK_BDEV_STATUS_UNREGISTERING state is newly
added to indicate this unregister operation.
In this case, the bdev unregister operation becomes the
async operation as each thread will be sent the message
to abort the IOs and as the last step, it will unregister
the required bdev and associted io device.
On the other hand, the queued_resets will be handled
separately and not aborted in the bdev unregister.
New unit test cases are also added:
enomem_multi_bdev_unregister: to abort the IO from
nomem_io queue during the unregister operation
bdev_open_ext_unregister: to handle the events and
async operations from the unregister operation
Change-Id: Ib1663c0f71ffe87144869cb3a684e18eb956046b
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12573
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Instead of releasing the batch memory when the batch generates a
completion, instead do it via refcnt. This will allow us to later hold
onto batch memory longer if vectored transactions end up spanning a
batch.
Change-Id: I942d6aa5052029eb0951e51a046dd98943108b94
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12259
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
If nbytes is not set, then the desination iovec sent to the underlying
driver has a length of 0.
Change-Id: Ia55f5ece942bd70f32bfdb3bcf02134ba98fca96
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12612
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Also, fix the way how fio's src dir is looked up and make sure
check_dir() is called before resolving the path.
Fixes issue #2487
Signed-off-by: Michal Berger <michallinuxstuff@gmail.com>
Change-Id: Ifb418abf8c51fdb469a3ba52c77c71c56d708142
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12388
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
When cross-compiling DPDK and changing the PKG_CONFIG_SYSROOT_DIR,
the resulting .pc files do not include it in libdir and includedir
variables.
To get the actual paths, those two variables need to be appended
with pc_sysrootdir.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ic45f5d2d25c24284561f4f022ca8631f3bfbae07
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12397
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
A device can now be identified by a protocol and a name. This allows to
have multiple types of devices providing the same protocol. The
selection of which device manager to use for specific protocol will be
added in following patches.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ib4a6c323ba6a2a5d36ff958ddc40631fd9329cb1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11683
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
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>
Change-Id: I7813a31060af7b2f9aea1b0a0d99fb699670509a
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11667
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
It's now possible to register SMA device managers defined in a module
outside of the regular directory. To do that, a global variable called
`devices` containing a list of device manager classes needs to be defined
in a module that's added to SMA_PLUGINS environment variable.
For example:
```
$ cat /path/to/plugins/external_devices/__init__.py
import spdk.sma as sma
class MyDeviceManager1(sma.DeviceManager):
pass
class MyDeviceManager2(sma.DeviceManager):
pass
devices = [MyDeviceManager1, MyDeviceManager2]
$ SMA_PLUGINS=external_devices scripts/sma.py
```
Multiple plugins can be specified by separating them with a colon, e.g.:
```
SMA_PLUGINS=plugin1:plugin2
```
Of course, the location at which these modules reside must be in
PYTHONPATH.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I2bb4cfea5191710f97d66abe3a21e4deacc6338a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11412
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
