Virtio-blk supports one to one mapping between devices and volumes.
Based on that, we can report underlying bdev capabilities as device
level QoS capabilities.
Signed-off-by: Artsiom Koltun <artsiom.koltun@intel.com>
Change-Id: If7c28c75d087fecec0105002994581fd7ac57f51
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15260
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
(84ab68c) test/scheduler: add a for_each_reactor shutdown test
Patch above added regression test for handling spdk_for_each_reactor
during shutdown, by adding constant repeat of this operation
while application is running.
Reactor event processing (especially constant) has impact on the
reactor_interrupt_run(). spdk_fd_group_wait() will almost always
execute an event, skewing the results of scheduler test.
Reactor that should have been idle, will show active usage via
/proc/stat.
Fixes#1950
This patch makes this regression test optional, and enables it
only in test that does not measure CPU utilization from the system.
The ./test/event/scheduler/scheduler.sh is the only one where it is
enabled, as it's purpose is to verify the test scheduler application.
Remaining ./test/scheduler/*.sh tests do verify CPU utilization,
so the regression test is disabled in those.
Modified the for_each_done, to for_each_reactor_start, to better
reflect the intention.
On my system enabling spdk_for_each_reactor test flag on the
scheduler application with no threads (except app thread),
consumes ~20-25% CPU from every core in CPU mask.
Meanwhile disabling it, idle cores are 100% idle
and active cores spend 100% of CPU time in usr.
Change-Id: I40eda15a748e76b95dc5441144cd8931e46edee5
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15210
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
The bdev*_with_md APIs now allow to pass NULL md
pointer, so calling this function without checking
for metadata simplifies code
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I32b2f85e375b858c25beb7e2128132dd604ca040
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15092
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>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Add a new parameter "-c" to display the per channel IO statistics
for required Bdev
./scripts/rpc.py bdev_get_iostat -b Malloc0 -h
usage: rpc.py [options] bdev_get_iostat [-h] [-b NAME] [-c]
optional arguments:
-h, --help show this help message and exit
-b NAME, --name NAME Name of the Blockdev. Example: Nvme0n1
-c, --per-channel Display per channel IO stats for specified device
This could give more intuitive information on each channel's processing
of the IOs with the associated thread on the same Bdev.
Please also be aware that the IO statistics are collected from SPDK
thread's related channel's information. So that it is more relating
to the SPDK thread. And in the dynamic scheduling case, different
SPDK thread could be running on the same Core.
In this case, any seperate channel's IO statistics are returned to
the RPC call and if needed, further parse of the data is needed to
get the per Core information although usually there is one thread
per Core.
On the other hand, user could run the framework_get_reactors RPC
method to get the relationship of the thread and CPU Cores so as
to get the precise information of IO runnings on each thread and
each Core for the same Bdev.
Change-Id: I39d6a2c9faa868e3c1d7fd0fb6e7c020df982585
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/13011
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
If verify is enabled, both data and metadata are checked. However,
if verify is disabled, read data is not checked even if dif_check_flags
is not zero. Add DIF/DIX verification for read I/O at completion.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ibde44bc244f84e40cef68653978191363acca5ce
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15074
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
NVMe-RDMA target has a helper function get_rdma_qpair_from_wc() and
uses it to identify a qpair from a WC.
NVMe-RDMA initiator has a similar function
nvme_rdma_poll_group_get_qpair_by_id().
NVMe-RDMA initiator will support SRQ in the following patches, and
it will want to identify a qpair from a WC.
get_rdma_qpair_from_wc() of NVMe-RDMA target uses wc->qp_num internally
anyway.
However, the upcoming custom transport for RDMA will have to use other
variables of WC.
Hence, it will be convenient to pass WC instead of qp_num if we consider
future enhancements.
Based on these thoughts, for NVMe-RDMA initiator rename
nvme_rdma_poll_group_get_qpair_by_id() by get_rdma_qpair_from_wc().
remove unnecessary declaration, and pass WC instead of qp_num.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: Denis Nagorny <denisn@nvidia.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@nvidia.com>
Change-Id: I01ead4730207e2c6ac53b83f151bd5f977a11465
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14279
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
With RDMA, the admin poller can experience a remote disconnect when
processing completions. The admin qpair will be disconnected to handle
this. The disconnect code path will manually complete queued aborts.
However, the completion callback for the abort will attempt to resubmit
other queued aborts from the queue, which will result in a very large
stack and can eventually cause a segfault.
The fix is to not resubmit queued aborts if the admin qpair is in any
kind of failed state.
Change-Id: I4a6f959232c8a1bd30c87ca50459014e556cbaa0
Signed-off-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15114
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: Michael Haeuptle <michaelhaeuptle@gmail.com>
This function was a bit cluttered so simplify it. Also, it was
assuming nbd module was loaded into the kernel prior running it -
this could silently fail the nbd_all[@] setup in case it wasn't.
Always attempt to load nbd driver before the setup and fail hard
if the driver is not in place.
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I5884973944eae3e827eaafec16ba72e2cb4f70e7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14994
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: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
zram doesn't support some of the uring-specific setup that xnvme may
be using, in particular the IORING_SETUP_IOPOLL, so switch to something
more common like null_blk module. For details see:
https://github.com/spdk/spdk/issues/2708
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I48b91ba356b4054433eb1835fa3e2708c8d2628c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14920
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Kamil Godzwon <kamilx.godzwon@intel.com>
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>
To fix issue: #2726
And also fix the examples/nvme/perf tool.
The srand() only needs to be called once to set the seed of
futher more calls of rand().
Change-Id: I41ab3a46593513516ad11ea7a5b8960b449e9867
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15108
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Do not run test with message send using hello_sock client with
KTLS enabled. CI does not have systems that run with openssl-3
and KTLS and it fails under Fedora36.
This test should be brought back after systems do run openssl-3
on board.
Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: Id7c46ced745fb4564b7428202a902cb332d30b33
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14830
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Also, clear generated fuzz logs upon successful test run - they may
get quite big (> 250MB) in size which doesn't play well with the
artifacts limit CI enforces. If the test succeeded then there's no
point in keeping them around for any potential debugging.
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I240e8217cd50dba258310604c6985d8f6a832756
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15030
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: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: <sebastian.brzezinka@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Don't touch existing hp allocations when they already meet the NRHUGE
requirement. Introduce new environment switch, SHRINK_HUGE, to
override that new behavior.
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I0cd124b98c3deb7c21b4fbd57529d0995978c2e5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14912
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>
The SMA tests are now tested on CI, so they no longer need to be on the
skipped_tests list.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I5746da06540efb1d3451a688fec4727d253bfc89
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14151
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: Michal Berger <michal.berger@intel.com>
Allow to run AUTOBUILD test in two versions:
* full - original one
* tiny - minimalistic, currently only check format and
check so deps
We want to reduce first phase of per-patch (called Pre-Tests)
by running only basic tests. To accomplish that we will run
autobuild test in tiny version. Full version will be executed
in second phase of per-patch (called Tests).
Signed-off-by: Pawel Piatek <pawelx.piatek@intel.com>
Change-Id: I87d503ec272080ab238cf5129c44f92d8bb690ed
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14481
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Move the RPM builds away from autopackage.sh into a dedicated test
suite - this will allow the actual packaging tests to be run in a
separate instance.
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I23a824db7aa3a5e469bea9b2536a83ae0a5c8e04
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14937
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>
See:
https://review.spdk.io/gerrit/c/spdk/spdk/+/12893
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I9dfd22e97645411bc24fdc0e62bb0dff9eed9abd
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14841
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I82d0cb1116b822f1a1111b19d2dd1286d8a399c4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14829
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I2c7b0c450f76057af31c9501c1b9ac843e0e1ee8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14828
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Without this, the second run of grep adds output directly behind the
last line of the first one. This results in the last test (from the
first run) always being filtered out from the list by the next sed
call.
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: I1aa5d2acfc28cccdee1464462b0350f0cbc346c7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14756
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
To fix issue: 2719
In this case, if there is successfully configured job and continue_on_failure (-f)
parameter is set, it will continue to run for these jobs. Otherwise, it will just
stop.
Change-Id: Ia2940a9e1b5da63e50857da11ff7f811c7576e3e
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14747
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Modified the code to fix the build error
"error: redefinition of ‘io_uring_get_sqe’" when building SPDK
with the latest liburing code.
Signed-off-by: Krishna Kanth Reddy <krish.reddy@samsung.com>
Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: Ie331ef0bbce143c92e2d7192465af995c8f451e9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12893
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: Aleksey Marchuk <alexeymar@nvidia.com>
If the module stop handler is asynchronous we must wait until it
finishes before unregistering the io_device.
Change-Id: I149b716d9f4b0c1680b3e43b395fc9ec5b90d70c
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14717
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Make writes larger when necessary to comply with the bdev's write unit
size requirement.
Change-Id: Ief60d369bb3efb5de083c5951672e20f2dd9e0af
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14715
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>
Refactor blockdev_write_read_offset_plus_nbytes_equals_bdev_size() and
blockdev_write_read_offset_plus_nbytes_gt_bdev_size() to use
blockdev_write_read() similar to how it's done in other tests.
Change-Id: Iba853a0df1facd88ce7c206163386f6c8a9c7c1b
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14714
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>
Implement support for full stripe writes without parity calculation.
The size and alignment of write IOs must be a multiple of full stripe
size. To reflect this, the raid bdev's write_unit_size is set
accordingly. We rely on the bdev layer to split larger IOs based on
that.
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: I6940280ad870f3bd678fd19346b06ba4bdadd52e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7702
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add new bdev property split_on_write_unit which, if set to true, causes
writes to be split to match write_unit_size and fail if not aligned to
or not multiple of write_unit_size.
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: Id49f58a3288ddf5cfe4921ce4020ae4bcdd67298
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11390
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Move device cleanup to spdk_io_device_unregister() callback. This fixes
a case when the device would be freed before its last io channel was
closed, leading to use after free condition.
Repurpose raid_bdev_free() to actually free the bdev.
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: Ib667b4d5ac1b34a0f2dda69f6b0775d9363dbfee
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11398
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Depend on bdev layer to send down chunk sized IOs, based on
optimal_io_boundary.
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: Iec45f4917117d35c3a9e807940e49091dfcba870
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7699
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: If5369e76d5be42005a2c26d10c04dc0b03cb9092
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14665
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
