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>
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>
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>
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>
It's now possible to specify a time to wait until a connection to the
discovery controller and the NVM controllers it exposes is made.
Whenever that time is exceeded, a callback is immediately executed.
However, depending on the stage of the discovery process, we might need
to wait a while before actually stopping it (e.g. because a controller
attach is in progress). That means that a discovery service might be
visible for a while after it timed out.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I2d01837b581e0fa24c8e777730d88d990c94b1d8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12684
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
By default, failback to the preferred I/O path is done automatically
if it is restored. Some users may want to keep using the backup I/O
path even if the preferred I/O path is restored. In this case,
bdev_nvme_set_preferred_path can be used to do manual failback.
We may be able to clear/fill I/O path cache more strictly but it will
be complicated and have bugs. This patch does the minimal change,
just skips an apparent case.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I78fe5faee6ff04e88ae3d7c6be6da1c20637c912
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12431
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: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Check that we're not already connected to a discovery service that has
the same address (or has a referall to) as the service we're trying to
start.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I84863fd959f62b30e9a348f69d10c7f1edffda7a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12627
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: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Otherwise, we'll try to use the same name to create the bdevs.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I38cfb9073f343a7bf3966754008065c7ab89e609
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12626
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: Aleksey Marchuk <alexeymar@nvidia.com>
Add a CLI param "-x" to specify which memory
domain operation must be executed during test
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I0dbee8d5d186da1e177e24beadc4e71909e4057f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12344
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The new blobstore ext API is used when the user
provides ext_io_opts in bdev layer.
To store blobstore ext_io_opts, vbdev_lvol reports
non-zero get_ctx_size in bdev module interface.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I64076b5369533be0c1d69ca48aef9d70a9abe488
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11373
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>
These function accept optional spdk_blob_ext_io_opts
structure. If this structure is provided by the user
then readv/writev_ext ops of base dev will be used
in data path
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I370dd43f8c56f5752f7a52d0780bcfe3e3ae2d9e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11371
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>
Introduce spdk_blob_ext_io_opts structure which
is used in the new *_ext functions.
Zeroes dev is updated with implementation of
readv_ext which uses memory domains memzero
or regular memset().
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Id94542196eff999827bf00591fd43804256fccb4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11369
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
This package is needed for testing SSL functionality.
We using both s_client and s_server mode from this package.
openssl-devel is already installed in scripts/pkgdep.
Here we adding the utility itself.
Change-Id: I389f4d273aefd82f2667aba620d00f17c3e557a9
Signed-off-by: Boris Glimcher <Boris.Glimcher@emc.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12645
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: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Michal Berger <michallinuxstuff@gmail.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
A recent change to the bdevio test changed one of the
tests from a 30x4K IO to a 30xLBA IO.
This has started giving us nightly test failures (see
issue #2499). It is because now the 30 * LBA (512B)
is 15KB which all fits within one 16KiB reduce chunk,
yet we only allocate REDUCE_MAX_IOVS
(17) iovs per reduce IO context. Previously when
they were 4KiB each, we would only need 4 iovecs
(16KiB / 4KiB).
We may need to think through this a bit more, to
make this more dynamic (based on the chunk size
and underlying block size), but for now let's at
least increase this enough to handle the most
common 16KiB chunk / 512B LBA case.
While here, run bdevio as part of per-patch testing,
that would have helped us catch this issue before the
a6e022463 patch got merged.
Fixes#2499.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2140c5b9d038b202edfdcbb2bfee059140f7703a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12672
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reduce the number of iterations by half. Currently this
test often leads to time-outs in CI nightly testing because
of how long it takes.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I3ccf9663706e5f79da609d6b9ffa08bb89dc31dd
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12658
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Increase waitforserial() function initial wait time.
When testing with 5.16.X kernel version 2 seconds were
not enough and issuing a "disconnect" immediatelly after
"connect" caused keepalive timeout.
Fixes#2467
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: Ib913be35b2383daf5f5037e258bbdd2477810921
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12587
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
In the following patches, nvme_ctrlr_process_init() will be used to
disable the controller when disconnecting the admin qpair for PCIe
transport. In this case, we will have to exit nvme_ctrlr_process_init()
after CSTS.RDY is 0. However, spdk_nvme_ctrlr_reset() and
spdk_nvme_ctrlr_reconnect_poll_async() have to continue
nvme_ctrlr_process_init() until the controller becomes ready.
To differentiate stop and continue clearly, add a new state
NVME_CTRLR_STATE_DISABLED to enum nvme_ctrlr_state.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ic0a5fb7114d4eeb1cefec28bc404184768fb0a96
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12613
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: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
The stub app runs during NVMe CI testing to minimize test time.
It runs as the primary process and all other NVMe tests
run as secondary processes and connect to the NVMe controller
via shared memory. The issue is when there is an Async Event
Notification, the secondary processes are not informed of the
event because the stub app does not process admin completions
and therefore they do not get added to other processes async
event list. This change adds a call to process admin
completions during the stub_sleep routine to cover this type
of AEN case. A configurable sleep time was also added to
allow the admin completion calls to occur more rapidly to help
minimize test time, if needed.
Signed-off-by: Curt Bruns <curt.e.bruns@gmail.com>
Change-Id: I72d5afc511c4409ec4a03cd969dbbc0d7dd4f256
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12555
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: Jim Harris <james.r.harris@intel.com>
Allow DAOS modules to be not built during scanbuild test.
The CI systems which do the autobuild & scanbuild tests are
DAOS development packages necessary to build "--with-daos"
are not available for Fedora systems which are used to run
build tests.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I3e01413a64d55dae2a0d77c3bf5fa9ed23c7ad87
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12408
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>
Changed 'git -C' to 'git --git-dir' when adding
submodules to cloned spdk repo. Both commands are not
equivalent. 'git -C' does not override the GIT_DIR
environment variable and some systems may have set
the spdk repository location incorrectly (not visible
by OS). Using 'git --git-dir' allows to set repo
options remotely (without being inside).
Changed the ownership of the spdk repo (Linux only)
to the actual $USER while checking if repo exists.
That prevents from unsafe directory detection run
by GIT.
Such issues happend on Ubuntu 18.04 and 20.04
Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I61dceddebde7e3c66e4b37990e205c513fe38414
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12551
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Michal Berger <michallinuxstuff@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
These were deprecated in 2019, it's time to remove
support for them now.
Fix ocf test script that was still using the
deprecated get_bdevs RPC name - change it to
bdev_get_bdevs.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7f8caedc250b80503671a0236694181613f63860
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12553
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
We have separated the process_shm function in order to avoid its call-out in "nvmftestinit" when there are no rdma cards
Signed-off-by: Jaroslaw Chachulski <jaroslawx.chachulski@intel.com>
Change-Id: I42d2d3426b71199a248bbfb238ab3811457c68bd
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12394
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>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
We enable multiple engines by:
* getting rid of the globals that point to the one available HW
and one available SW engine
* adding a submit_tasks() entry point for the SW engine so that
it is treated like any other engine allowing us to just call
submit_tasks() to the assigned engine for the opcode instead of
checking what is supported
* changing the definition of engine capabilities from
"HW accelerated" to simply "supported"
* during init, use a global (g_engines_opc) that contains engines
and is indexed by opcode so we know what the best engine is for each
op code
* future patches will add RPC's to override engine priorities or
specifically assign an opcode(s) to an engine.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I9b9f3d5a2e499124aa7ccf71f0da83c8ee3dd9f9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11870
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
The NVMe bdev module supported active-passive policy for multipath mode
first. By this patch, the NVMe bdev module supports active-active policy
for multipath node next. Following the Linux kernel native NVMe multipath,
the NVMe bdev module supports round robin algorithm for active-active
policy.
The multipath policy, active-passive or active-active, is managed per
nvme_bdev. The multipath policy is copied to all corresponding
nvme_bdev_channels.
Different from active-passive, active-active caches even non_optimized
path to provide load balance across multiple paths.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ie18b24db60d3da1ce2f83725b6cd3079f628f95b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12001
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
If we specify a preferred path manually for each NVMe bdev, we will
be able to realize a simple static load balancing and make the failover
more controllable in the multipath mode.
The idea is to move I/O path to the NVMe-oF controller to the head of
the list and then clear the I/O path cache for each NVMe bdev channel.
We can set the I/O path to the I/O path cache directly but it must be
conditional and make the code very complex. Hence, let find_io_path() do
that.
However, a NVMe bdev channel may be acquired after setting the preferred
path. To cover such case, sort the nvme_ns list of the NVMe bdev too.
This feature supports only multipath mode. The NVMe bdev module supports
failover mode too. However, to support the latter, the new RPC needs to
have trid as parameters and the code and the usage will be come very
complex. Add a note for such limitation.
To verify one by one exactly, add unit test.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ia51c74f530d6d7dc1f73d5b65f854967363e76b0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12262
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: <tanl12@chinatelecom.cn>
Reviewed-by: GangCao <gang.cao@intel.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>
If the fuzzer crashes or hangs, LLVM spits out a file
containing the raw data input for the iteration that
caused the crash or hang. Add a -N option to
llvm_nvme_fuzz to allow the user to specify one of these
files. When specified, the fuzzer will only run one
iteration with that specific input data.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0d86cca7b2a0b6eaee263665478c31ee4060a8b8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12451
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: Changpeng Liu <changpeng.liu@intel.com>
We do not want the fuzzing thread to compete with
one of the nvmf target cores - otherwise they will
keep preempting each other and drastically reducing
the operation rate.
This sped up the operation rate approximately 20x
in my test VM.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I75d8c59d75e49cafaf7ee94f4796f9184b690647
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12403
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
If an input causes a hang, the fuzzing thread won't
terminate itself, since it is waiting for all
outstanding commands to complete. So raise a SIGSEGV
in the SPDK shutdown handler instead, which will
cause the fuzzer thread to exit as well as generating
an input file of the hung input.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I5753977740e27ca7827222b9e3cee1e939ef31a1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12407
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <yifan.bian@intel.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
When the LLVM fuzzer is done, it calls exit(0)
explicitly. This triggers the DPDK exit handler
to run which starts unmapping huge pages while
our reactor thread is still running.
Currently, this doesn't cause any problems since
the fuzzing thread and reactor thread are running
on the same core. But the next patch will
unaffinitize the fuzzing thread, meaning that the
reactor thread will be actively trying to read
hugepage memory while the fuzzing thread is in
DPDK exit handlers unmapping that same memory.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie304ebbb1962855796dac699849a0726cfdcd0d4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12406
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: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Similar to examples/nvme/perf/perf.c
Usage:
-z disable zcopy
-Z enable zcopy
Signed-off-by: Boris Glimcher <Boris.Glimcher@emc.com>
Change-Id: I56f1b0d9c79cbfdba57d0e132a18cb5dce0b7904
Signed-off-by: Boris Glimcher <Boris.Glimcher@emc.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12435
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.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>
When we use Fuzzer 0 to do random testing of admin commands,
it's should be a normal one, instead of Fused operation.
Change-Id: I652a725798e79842866be01119be817c965fcee7
Signed-off-by: Yifan Bian <yifan.bian@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12421
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Added a check before any of the parted calls to
confirm the size is what is expected.
Fixes#2388
Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I18720c66d70da82b8943d3231348a4015108a598
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11820
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Michal Berger <michallinuxstuff@gmail.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>
blockdev_compare_and_write and blockdev_write_read check
if data_length is multiple of block size and silently
succeed if it is not. There are multiple test that test
e.g. invalid size but the actual bdev API is
never called and tested because of the early return.
The intend of this behavior presumably was to allow
to run test like e.g. the 512bytes rw test on a 4K
formatted namespace without failing. Instead of silently
succeeding which introduces the issues described this
patch changes all sizes to be multiple of block size and
removes the test for valid block size to let the
return value of the bdev API functions indicate failure like
intended. All previous 4k tests are now performing io of
1 block size. Previous 512byte tests now perform
8 * block size (to have 4k test size for 512byte formatted
namespaces). Write zeroes test are now using the
write zeroes buffer size instead of a hardcoded value
and align accordingly to block size io.
Furthermore, no cmp&write test is ever executed because
data_length is set to 1 trying to indicate 1 block but
it is tested against valid block bytes and silently
succeeded. Also write and read helpers in cmp&write test
expect bytes not blocks. For consistency use bytes in the
test and only convert to blocks when calling the cmp&write
blocks API.
Signed-off-by: Jonas Pfefferle <pepperjo@japf.ch>
Change-Id: I662094c2bcd08c0e9dc5573177a2e7a0edd275ce
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12382
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: Ben Walker <benjamin.walker@intel.com>
Can only 'return' from a function or sourced script.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: Ie6104adeec0f07557989d9ba25ca9aa2f563e1d5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12422
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Dong Yi <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@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>
Do not perform strict host checking and mask
known hosts file. The test might run just after
IP address is set up and we don't want to get
stuck at ssh interactive prompt for adding remote
host to known hosts list.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I10e5e473520994e09562e667b27ee655fceb2dd5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12348
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Put the SSH routine into a function and allow to
use custom SSH key by using an optionally set
environment variable.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I6b1f552af9e28ab0b94d027c8848ef0dd5480791
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12339
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Branch "vfio-user-v0.93" has been silently removed.
To fix pkgdep script, branch "vfio-user-dbfix" been
chosen, but the reason behind this decision is
unknown to me.
Signed-off-by: Pawel Piatek <pawelx.piatek@intel.com>
Change-Id: I56dece900efcddfc020e13f2f8ef1019cc2185ec
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12358
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Monica Kenguva <monica.kenguva@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
spdk_bdev_get_acwu() is a 1-based number, so we need
to subtract 1 from it before assigning the value to
nsdata->nacwu.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I32708b28a35670cba6013a48b79389fa48226285
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12399
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Fixing grep because in freebsd grep interprets --exclude,include differently
which causes match on run_test()
Fixes: #2479
Signed-off-by: Jaroslaw Chachulski <jaroslawx.chachulski@intel.com>
Change-Id: I3ab5daa7cd49d2bc8f1af6f7517f9b74b2dd88de
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12392
Reviewed-by: Michal Berger <michallinuxstuff@gmail.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
The bdevio test app has some test cases verifying
that write zeroes commands are handled correctly,
but using knowledge of the ZERO_BUFFER_SIZE that
the bdev library uses for splitting larger write
zeroes commands. Instead of hardcoding that 1MB
value in bdevio.c, have bdevio.c use ZERO_BUFFER_SIZE
directly instead. But this requires moving
ZERO_BUFFER_SIZE into bdev_internal.h and having
bdevio.c include that file.
We do this instead of putting ZERO_BUFFER_SIZE in
the public API because we don't want users to
make any kind of dependencies on this value.
While here, also rename the tests that are using this
value, so that the test names don't include any reference
to the specific size of this bdev-internal zero buffer
size.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia29d92a706cb1f86b4c29374dc2a9beccf679208
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12383
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: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
