Also change the discovery/nvmf.sh test to use it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I56bce9a84bd46f13b6d4f34da81abf23413f2598
The SCSI library already provides a callback when the task is released
(free_fn), so the user can update their own task counter.
Change-Id: I7fb13f6fff66dbba2315fd03fb06e49f793be123
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The task free callback function is required, so make its assignment part
of task construction.
Change-Id: I2f5fdf73b064653ee85b4e7961cb1653a0a4107d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Fix up a second copy of linux_iter_pci in the same manner as commit
6562e95092 (scripts/setup.sh: support
systems where more than one domain is used).
Change-Id: I3d9b842891d70c2960de8287e3b11c1a11b02d1f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Added 1 minute timeout for vhost to exit after receiving
INT signal. After timeout send KILL signal to vhost app
and exit with error.
Change-Id: Ib1659660667991b21103b3401bbd780290521433
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
The first time this test runs on a machine, it will not have the SSH key
in its known_hosts, so allow it to connect without user intervention.
Change-Id: I27f750f41e1819b526197af2d0b9ae15382d9aa4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Blockdevs already indicate support for unmap via
spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_UNMAP).
Change-Id: I634f27a281fd900bb3a6da2e4ff8a74e43579578
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
We plan to use these buffers for more than just reads.
Change-Id: I8fa6cb432a6cfe4406fbf240cd3aa2ae4ab5f3d5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The bdev layer manages the bdev_io caller_ctx internally.
Change-Id: Icb5a4f26b090d6f761af988fb26a319005e66a6d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The SCSI layer was not using the task ID for anything; the iSCSI layer
was using it to store the task tag, so move it there and rename it to
"tag" to make its purpose clear.
Change-Id: Ibda4f4e215056116b9be4a3a0264f98bc4c29535
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Move the scsi_nvme translation code from the SCSI library into bdev, and
provide a generic way to translate any bdev_io status into a SCSI
status.
Change-Id: Ib61a6209387c24543e31574e2b5ca249e2ac8b74
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
RocksDB's build system does not correctly generate dependencies - if
headers have been removed in the SPDK repository since the previous
build, the .d files will prevent the build from continuing.
Avoid this by cleaning up any left-over build outputs from the previous
run.
Change-Id: Ib8669273eb38c241f83e14714683693430469462
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Convert autotest.config to a sourceable file. This enables direct or
indirect usage of variables and a chance to remove error prone
statements like '$(cat | grep | awk')'
Change-Id: I3ae4152fd7548e4b2bfd01c00b614155871b2db3
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
The contents of struct spdk_scsi_lun don't need to be part of the public
API.
Change-Id: I101b77871054557380610fd901ab38bada463202
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The RocksDB SPDK environment no longer depends directly on DPDK.
Change-Id: Ifcabcd137292f86724a1e3fed93abb94fa5a2344
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Fix the incorrect iSCSI PID ($pid) and also terminate nvmf_tgt on
failure.
Change-Id: I92457591c98c12f101a1713763ed9dc1feb32cb9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Similar to commit 378fc7787c, remove
unnecessary process_core calls.
Change-Id: I2fb55e39770ca57274c177fce3161f46a94d4efe
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Setting max_sectors_kb to 4k was workaround and is no longer needed.
Vhost must work with arbitrary guest settings.
Change-Id: I9ff9d752916e200e79cdd2b4b3d64afdecae11bc
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Fix up all existing spacing errors in comments and add an automated
check for patterns like /*comment*/.
Change-Id: I28f61c93612dc0f8aed66bd509da78e91ea9737e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Also adds the automation test case for using remote
NVMe devices exported by NVMe-oF target.
Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
The new format is: domain.bus.device.function
For this format, since we use '.' as separator,
to avoid misusing, we only support the following:
1 domain.bus.device.function ( 4 values provided)
2 bus.device.function (3 values provoided with domain = 0)
3 bus.device (2 values provided with domain = 0, function = 0)
Change-Id: Ide03db38b4ac7802cf36f0e536e8b997101d6cd3
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
A 100us is so small that applying the quirk to the specific
SSDs that require the delay is more trouble than it is worth.
So remove the quirk and always wait 100us before re-enabling
the NVMe SSD during initialization.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id6a8cc6e35d103fffdf135580301fc3e5b27e722
The bdevperf reset test only executes in the nightly test run, so this
problem wasn't caught during review.
It gets stuck in an infinite loop and produces lots of log output; turn
it off to avoid filling up the disk with logs until we determine what
broke it.
Change-Id: I7601ba324a0531de3733cf5784ccd46da5618082
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Queue aborts that would exceed the abort command limit
in software as a convenience for the user.
Change-Id: I8c1f0380984cc6c0cdb453db961939a7f571b336
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Core dumps stop working if the path is longer than 64 bytes.
Use readlink to shorten some of the very long relative paths.
Change-Id: I5e7eb6580ca581c5ac3a71afb7b62953836e2660
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
It has been discovered that some devices require
a very small delay before writing CC.EN to 1 after
CSTS.RDY goes to 0.
Change-Id: I73d31726d17ebf5bbec7ee528e2f98fcd05234dd
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This patch removes comments within configuration files for tests
that are not relevant to the test itself.
It reduces the need to update comments in test config files with each update.
As well it makes those same as other test configuration files.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I48116613435fadf0738b5a948d32f458c7ce06be
In the cache_append_no_cache test case, we need to
sync the file first, before calling cache_free_buffers().
Otherwise the buffers do not really get freed since they
contain dirty data.
This fixes a couple of memory leak bugs detected with
CONFIG_ASAN=y.
Reported-by: John Meneghini <john.meneghini@netapp.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If1ff6ed7d9e7a6dc67af0a2baf3b2555ace610ee
This parameter is not actually used by the unit
tests using the bdev, but setting it to a non-zero
value silences a couple of static analysis errors.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id53a17bcd1f4ec51f35a954ec0f6e6ea9803086e
This fixes multiple SCSI reset issue.
This patch does not remove sleep in iSCSI tests.
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: I5e9f3705e5dc34004b9d1b9e40fbdcb04a3bee4e
Some intermittent issues still observed with multiple
resets in quick succession. Reverting for now while the
issue is more fully root caused.
This reverts commit 7fa7f91ee3.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I493b564e8a110bbfb7a6cc47107d53d6eca40053
Automated test scripts now set up 4096 huge pages by default,
so we can now safely allocate the default 4GB for blobfs cache.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8aea322068a5df79ca75737c43a4e0e80e7d2be4
The bdev product_name field is used in the Inquiry command translation
and was previously uninitialized.
Change-Id: Ie9d056ddaddfbb60ce5c3444ed87f852c00a296d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Rather than re-running setup.sh during the vhost tests, just
pre-configure NRHUGE to the necessary value for all setup.sh invocations
during autotest.
Change-Id: I1a4d9a7f7ab52082a8f626c3c64807417a493b5f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Negative queue depths, I/O sizes or time durations do
not make sense, so exit() if user input contains any of
those cases.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9d0261d1151f41dfc11013c797bf949b736ebba3
For the automated tests, use this to directly write output
files to the output_dir.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I403f51a4f4d0cac0fca7ec610527bf616f97b20d
Change NO_PERF to USE_PERF which defaults to 1. This
will be overridden to 0 if set to 0 in the environment
or perf is not available on the system.
While here, disable perf in the automated test suite.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia182f6a3068d0da631973d4010dce24bd0f98eb5
- Renamed to integrity filesystem test
- Add detecting if qemu is installed
- Default vhost.conf using NVMe; ramdisk may be too small
for copying & building both dpdk and spdk
- Use kernel source (present on VM) for running test compilation
- rename to integrity test
Change-Id: Id369a5aac3008700cffbc51a5197b72f10645a1b
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
This is the initial commit for "blobfs", a lightweight
filesystem built on top of the SPDK blobstore.
Also included in this patch:
1) a shim for using SPDK bdevs as the backing store for
SPDK blobstore/blobfs
2) documentation for using blobfs as the storage engine
with RocksDB
3) scripts for running a set of workloads and collecting
profiling data with RocksDB and blobfs
See doc/blobfs/getting_started.md included in this commit
for more details on blobfs, including some of the current
limitations.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2a6d3d4b87236730051228ed62c0c04e04c42c73
A few instances of DPDK_INC were added after the env library abstraction
was introduced; replace them with the correct ENV_CFLAGS variable
defined by env.mk.
Change-Id: I54ab2b2360a72506049fee549491f6614601a148
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is the initial commit for the "blobstore", a lightweight,
highly parallel, persistent, power-fail safe block allocator.
Documentation will be added in future patches.
Change-Id: I20a4daf899f1215d396f7931c3ec9a2e2bb269d0
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The user now must choose the name for each AIO bdev. This
provides consistency for names across restarts.
Change-Id: I13ced1d02bb28c51d314512d60f739499b0c7d8d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This patch changes following:
- device renamed to vhost-user-scsi-pci
- VM image set as first in boot order
Additionally cleaned up test paths.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
Change-Id: Ib89954ba083bf047e1a46bc350345e42272badd9
Simplify code that previously needed to check for subsystem type by
factoring out the discovery controller operations into a new ops
instance.
Change-Id: Id87b498e4623451993fe779ffb765be5a6743fd9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
No functional change, just rearranging code.
Change-Id: I28328dfefd7de269d326834c484f2c2fca4e6c1f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Names for the NVMe bdevs are now assigned by the user.
This means the same name will always be assigned to the
same device, even across restarts.
Change-Id: If9825ec9abcb5236b4671bc44a825e4f0d704fe3
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Option to run without fio job configs was unused - removed.
Removed related unnecessary code.
Change-Id: If5a8e39c446d575245518a90ee77283b999305cc
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
All devices must be specified by BDF. Add support for scripts
to use lspci to grab the available NVMe device BDFs for the
current machine.
Change-Id: I4a53b335e3d516629f050ae1b2ab7aff8dd7f568
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Move claimed flag to struct spdk_scsi_lun and remove RPC call that allow
SCSI LUN to be deleted by user.
Change-Id: I0fe57d33ab017816ab4799bce259807735e0c783
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
This patch creates a new test case: use rpc to
frequently add/delete a same subsystem to verify the
correctness of our code.
Change-Id: Ib582061f9fa909f271bea4865df29a3f432f807c
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change SCSI device configuration format from "DevX LUN0" to "Dev X LUN0"
This allow checking configuration against silly errors when device
number is out of range.
Also assert exactly only one LUN is given.
Change-Id: Idccd6878119282fc51947b092bdda7ae06aa94ad
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Make sure the automated test pool can catch any errors.
Change-Id: Icd1cb9f5501d5f18395b963453e6c3d5e58c48ee
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
SPDK_COUNTOF works like sizeof, except it returns the number of elements
in an array instead of the number of bytes.
Change-Id: I38ff4dd3485ed9b630cc5660ff84851d0031911f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
"meta" verification method in FIO is deprecated and is
suspected of producing false positives. MD5 checksums
are now used instead.
Change-Id: Id3a2b008837173f82e0eda48d96cd4a43e00531f
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
This patch adds a library, application and test scripts for extending
SPDK to present virtio-scsi controllers to QEMU-based VMs and
process I/O submitted to devices attached to those controllers.
This functionality is dependent on QEMU patches to enable
vhost-scsi in userspace - those patches are currently working their
way through the QEMU mailing list, but temporary patches to enable
this functionality in QEMU will be made available shortly through the
SPDK github repository.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
Signed-off-by: Michal Kosciowski <michal.kosciowski@intel.com>
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
Change-Id: I138e4021f0ac4b1cd9a6e4041783cdf06e6f0efb
Newer versions of nvme-cli return a non-zero status from the disconnect
command when it is successful. Catch this in the test scripts so that
the tests don't consider this a failure.
Change-Id: I3abf4042ec8b0bc8aed5a4c36128ae73230aa190
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The setup.sh parameters we use in autotest don't allow for enough RAM
for 12 subsystems; reduce it slightly while still testing a large
number.
Change-Id: I595589db6e5199ad87ad9511bb0ad63b7b60211d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
It is caused by this commit:
4163626c5c
nvmf_tgt_delete_subsystem registers the poller
with function subsystem_delete_event.
subsystem_delete_event is called asynchronously,
the deletion should happen in this function.
Otherwise, with the current code,
g_subsystems_shutdown = true
TAILQ_EMPTY(&g_subsystems) = true
when subsystem_delete_event is firstly called.
If there are multiple subsystems, the logic is wrong.
Thus other subsystem will never be delete. since
we already execute shutdown_complete().
Also add related test scripts.
Change-Id: I3823563fc9e8611c11a6d798685ff64e2939842e
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
This patch make the function spdk_scsi_task_process_null_lun() as public and
finish the task immediately once we get task in iscsi layer.
Change-Id: I4ada027d3a324dce8ef0d0f7706dbc14184ead96
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
After checking the code, aerl in our session is 0,
so there will be only 1 AER. So currently,
we will only handle 1 AER case.
When the AER event is triggered by real NVMe device owned
by the subsystem, it notifies all sessions belonging to
the subsystem.
Change-Id: Ia80fb0f03e893c20d8dd14afbed8db10db38301c
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
vtophys_get_paddr() and vtophys_get_dpdk_paddr() are doing similar
things; combine them into one function that works for all DPDK
memory addresses.
Part of the vtophys test is temporarily disabled until the next commit,
which will register all DPDK memory at startup and stop lookiing up
addresses at runtime.
Change-Id: I91312837aa1e6170bacaf3b0d2adbdc4391d3afa
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
We already know the PID of the iscsi_tgt process since we launched it
ourselves; drop the use of the pidfile.
Change-Id: Ib1569c5b507ac85f26020937a007f5d7df090d99
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch makes spdk_scsi_lun_construct behave as documented.
spdk_scsi_lun_construct will return only newly created LUN.
If LUN with that name already exists, NULL will be returned.
Unit test relevant to this behaviour is now changed to show
this functionality is now working.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I002903d6f96555c638aba3fa99cc2c2504ced603
This is necessary to prevent claiming the same LUN twice
and properly cleanup in case of an error during spdk_scsi_dev_construct.
This patch addresses three issues:
- spdk_scsi_lun_claim error is correctly handled in spdk_scsi_dev_add_lun
- on error when constructing scsi dev, it is now correctly removed along with attached luns
- spdk_scsi_dev_destruct not only unclaims, but calls spdk_scsi_lun_destruct on each lun in dev
Unit tests relevant to this behaviour are changed to show this functionality is now working.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I111c320f875e5003e3f1f7748a2630097301ce1b
This patch adds two new unit tests for scsi device:
- creating two different devices, each containing the same lun
- creating one device, with the same lun twice
As noted in code, three asserts are incorrectly set to show functionality
that is not working currently.
Next patch in series implements that functionality and changes asserts
in the unit tests.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I2645401fee4f2cd986458e0a4db108ce4e1bf9db
This patch adds new unit test for scsi lun:
- creates the same lun twice
As noted in code, one assert is incorrectly set to show functionality
that is not working currently.
Next patch in series implements that functionality and changes that assert
in the unit test.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ie1ea6034ba295bc1c24d21529c6adcf20ac10c53
By default, all SPDK applications will not share memory.
To share memory, start the applications with the same
shared memory id.
Change-Id: Ib6180369ef0ed12d05983a21d7943e467402b21a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Make sure that the output strings are not NULL before comparing them
with strcmp(). This silences some scan-build warnings.
Change-Id: I3bfb7e44abe0451ae8afb4647c3b2d3f787b4faa
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The direct and virtual mode code is identical; move it to session.c like
the other virtualized get/set features.
Change-Id: I0a0e2dd795197c142ad5d9d0e4ddedb2aa5c8c2a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Instance ID is too overloaded and the uses are beginning
to conflict. Separate the RPC configuration out.
Change-Id: I712731130339fee4fc8de4dc2d0fea7040773c58
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The host and port output parameters point into the (non-const) char *ip,
so it makes more sense for them to be non-const as well.
This allows the flexibility to pass non-const char pointers as the
output parameters, which will be used in the nvmf_tgt/conf.c parsing
code.
Change-Id: I1d5b102fc389c06d36432904e4fda944437b659e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
For namespaces with end-to-end protection information, metadata size
of exactly 8 bytes, and extended LBA configured, the NVMe driver would
calculate the size of the data block incorrectly. The NVMe spec has a
special provision for this specific case (8-byte metadata only) and
PRACT = 1 that requires that the host does not send the metadata as part
of the host memory buffer.
To fix this, clean up the calculation of the per-block data transfer
size by adding a new extended_lba_size field in the namespace, which
represents the total size of data to be transferred per block based on
the namespace's configured metadata size and whether it transfers
metadata as part of the data buffer. Then add the special case for
PRACT = 1 and PI configured and extended LBA in the R/W helper
functions.
Change-Id: I0b383a58c773cac06e6c018858b57129064c6059
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
