test/check_so_deps: use VERSION to look for prior tags

The check_so_deps script compares the ABI of the current commit from the
ABI of the previous release. It used to find the previous release by

test/check_so_deps: use VERSION to look for prior tags

The check_so_deps script compares the ABI of the current commit from the
ABI of the previous release. It used to find the previous release by
looking at the first non-LTS, non-*pre tag and extracting the version
from that tag. This means that after any other tag is placed, it'll be
used to figure out the release to test against.

Instead, use VERSION to find the release tag to compare against. That
way, as long as VERSION points to a -pre tag, we can place any number of
-rc tags without breaking the ABI tests. And, once a release branch is
created, VERSION would get updated to the -rc tag that points to the
first commit on that branch, so the subsequent commits would be compared
to the ABI of that release.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I1b4f9118e6d3749e0c43da3fa917c08868796eb9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25524
Reviewed-by: Ben Walker <ben@nvidia.com>
Reviewed-by: Jim Harris <jim.harris@nvidia.com>
Community-CI: Community CI Samsung <spdk.community.ci.samsung@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/check_so_deps: remove unnecessary suppressions

These are no longer needed as the major version of the nvme and nvmf
libraries were bumped to 14 and 19, respectively.

Signed-off-by: Konrad Szty

test/check_so_deps: remove unnecessary suppressions

These are no longer needed as the major version of the nvme and nvmf
libraries were bumped to 14 and 19, respectively.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I64146dfeb8f0fab0d358740e031d9b45c6e8a12d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25163
Community-CI: Community CI Samsung <spdk.community.ci.samsung@gmail.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
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>

show more ...

nvme: drop support for submit_accel_crc32

It was deprecated, and the TCP transport no longer uses it for data
digest calculations.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id

nvme: drop support for submit_accel_crc32

It was deprecated, and the TCP transport no longer uses it for data
digest calculations.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ic9d5be8972b73325a35bcaff7ed4998fd34f7b8b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25160
Community-CI: Community CI Samsung <spdk.community.ci.samsung@gmail.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot

show more ...

nvme: require TLS PSKs to be specified via keyring

It's no longer possible to pass the key directly in spdk_nvme_ctrl_opts.
This method was deprecated and was supposed to be removed in the
upcoming

nvme: require TLS PSKs to be specified via keyring

It's no longer possible to pass the key directly in spdk_nvme_ctrl_opts.
This method was deprecated and was supposed to be removed in the
upcoming release.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: If06e087abb83da6b2f22c4a9f7129720f26e6f0d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24810
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>

show more ...

test: Stop grep'ing through config.h, *.mk, etc. for build config

All the parameters are already available in the environment through
build_config.sh.

Change-Id: Icd6f536c1a31d6bb39f289f84cb0ac7cd8

test: Stop grep'ing through config.h, *.mk, etc. for build config

All the parameters are already available in the environment through
build_config.sh.

Change-Id: Icd6f536c1a31d6bb39f289f84cb0ac7cd87ca57d
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24770
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

nvme/pcie: add disable_pcie_sgl_merge option

Currently SPDK always merges physically contiguous SGL elements for the
PCIe transport. Add an option to disable this. This is useful for
testing.

Chang

nvme/pcie: add disable_pcie_sgl_merge option

Currently SPDK always merges physically contiguous SGL elements for the
PCIe transport. Add an option to disable this. This is useful for
testing.

Change-Id: I2ba7949e8175681a846af6718372269439344243
Signed-off-by: Vincent Fu <vincent.fu@samsung.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24194
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/check_so_deps: add option to specify release

This option allows user to run ABI tests against a specific release.
As it doesn't rely on tags, it'll be especially useful for testing
patches to t

test/check_so_deps: add option to specify release

This option allows user to run ABI tests against a specific release.
As it doesn't rely on tags, it'll be especially useful for testing
patches to the spdk-abi repo before a release.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I23566dde8a858cc36bf38c722cbb2f6a8682543c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24140
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

nvmf: add option to disable command passthru

This option allows user to control whether the target will submit
commands that are not known through NVME_PASSTHRU.

Signed-off-by: Konrad Sztyber <konr

nvmf: add option to disable command passthru

This option allows user to control whether the target will submit
commands that are not known through NVME_PASSTHRU.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: If1707edec76db97be004db5f09395a8105f26d7c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23373
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

check_so_deps: remove unnecessary suppress entries

None of these are necessary, as the major versions of the associated
libraries have been bumped since they were added.

Signed-off-by: Konrad Sztyb

check_so_deps: remove unnecessary suppress entries

None of these are necessary, as the major versions of the associated
libraries have been bumped since they were added.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I623e5ae541f5347f18d59f6384ea9a44954022da
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23424
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>

show more ...

env_dpdk: Add an option to run without huge pages.

This change adds a new SPDK commandline parameter "--no-huge", which
allows to run SPDK without huge pages. This is useful to run NVMf
with tcp/ip

env_dpdk: Add an option to run without huge pages.

This change adds a new SPDK commandline parameter "--no-huge", which
allows to run SPDK without huge pages. This is useful to run NVMf
with tcp/ip configuration.
Tested this by running a NVMf test over TCP between target and host.
Configured the NVMf target to run SPDK without hugepages by passing
the "no-huge" parameter. Then configured the NVMf host and connected
it to the target over TCP transport and verified that the subsystem
is getting exposed.
Modified existing NVMf tests to accept "--no-hugepages" parameter and
verified functionality by running those tests.

Fixes #2973

Change-Id: I68773bc6c818e5fc5173142213c3928fcbca96c8
Signed-off-by: Sarvesh Lanke <sarvesh.lanke@nutanix.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20171
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Abhineet Pandey <abhineet.pandey@nutanix.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

check_so_deps: remove the SPDK_NO_LIB_DEPS variable

This variable used to control whether the dependencies of a .so library
were listed during linking. But the way confirm_makefile_deps() works
is

check_so_deps: remove the SPDK_NO_LIB_DEPS variable

This variable used to control whether the dependencies of a .so library
were listed during linking. But the way confirm_makefile_deps() works
is it builds a list of dependencies by going through each undefined
symbol in a library, finding the library that defines that symbol, and
then comparing that list to the list from mk/spdk.lib_deps.mk. It
doesn't matter whether any of those libraries were linked in or not.

The only difference, in terms of confirm_makefile_deps(), setting this
variable makes is that the undefined symbols have their type filled in
properly:

w/o SPDK_NO_LIB_DEPS:
Num: Value Size Type Bind Vis Ndx Name
0: 0000000000000000 0 NOTYPE LOCAL DEFAULT UND
1: 0000000000000000 0 FUNC GLOBAL DEFAULT UND spdk_dif_ctx_set_data_offset
2: 0000000000000000 0 FUNC GLOBAL DEFAULT UND spdk_bdev_comparev_and_writev_blocks
3: 0000000000000000 0 FUNC GLOBAL DEFAULT UND spdk_bdev_queue_io_wait
4: 0000000000000000 0 FUNC GLOBAL DEFAULT UND free@GLIBC_2.2.5 (2)
...

w/ SPDK_NO_LIB_DEPS=1:
Num: Value Size Type Bind Vis Ndx Name
0: 0000000000000000 0 NOTYPE LOCAL DEFAULT UND
1: 0000000000000000 0 NOTYPE GLOBAL DEFAULT UND spdk_dif_ctx_set_data_offset
2: 0000000000000000 0 NOTYPE GLOBAL DEFAULT UND spdk_bdev_comparev_and_writev_blocks
3: 0000000000000000 0 NOTYPE GLOBAL DEFAULT UND spdk_bdev_queue_io_wait
4: 0000000000000000 0 FUNC GLOBAL DEFAULT UND free@GLIBC_2.2.5 (2)
...

It required a slight modification to the way we search for the undefined
symbols, but it actually ended up making it a bit simpler.

Another advantage of this change is that it's now possible to run this
test without having to rebuild everything with SPDK_NO_LIB_DEPS=1.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I62c13c7328f9593480c09b858cf575bf621b62eb
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20846
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Ben Walker <ben@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

check_so_deps: use make to get a list of deps from lib_deps.mk

Not only is this cleaner, but it also takes into account the
configuration, i.e. if a dependency is conditionally added based on one
of

check_so_deps: use make to get a list of deps from lib_deps.mk

Not only is this cleaner, but it also takes into account the
configuration, i.e. if a dependency is conditionally added based on one
of the CONFIG_* variables, it'll only be included if that variable is
set.

It makes it a lot easier to run this test locally, as it's no longer
necessary to enable all dependencies in configure.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ib5a809576dd1fe3d3a251a0dd021377b0f576cdc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20843
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <ben@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

check_so_deps: sort libraries in confirm_makefile_deps

While here, mark all local variables and change the way the error
message is printed (it helps to keep the output mangled due to the tests
bein

check_so_deps: sort libraries in confirm_makefile_deps

While here, mark all local variables and change the way the error
message is printed (it helps to keep the output mangled due to the tests
being run in parallel).

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: If26184c59199909c7fed1b7ee3a0757ff25dcc6c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20842
Reviewed-by: Ben Walker <ben@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>

show more ...

check_so_deps: suppress types only for single SO_VER

Suppressing types with libabigail is a manual process,
that should be used with care. Otherwise ABI changes
might be missed or ignored.

Main req

check_so_deps: suppress types only for single SO_VER

Suppressing types with libabigail is a manual process,
that should be used with care. Otherwise ABI changes
might be missed or ignored.

Main requirement going forward should be to specify
name of the type and specific SO_VER of the library,
when adding new suppression.
Just with this change, the suppress type entry can be
removed at any later time. Rather than keep up with
SO_VER changes.

This leaves a gap during that specific SO_VER.
Where ABI breaking change could be introduced, but would
be ignored due to the suppression.
There are more specific ways to limit the scope of the
suppressions. Most useful ones are has_data_member_*,
where a name of removed field or offset of the field
can be specified. The more specific the better chances
of catching just that change.
Please refer to libabigail manual for [suppress_type].

Three examples in this patch are:
- spdk_app_opts, where reserved185 field name was removed
and new member inserted between 1480-1488 bits
- spdk_nvme_cap_register, where due to limitations of
libabigail it was not possible to refer to change in
bits part of the union and narrow down this suppression
- spdk_bdev_opts, where reserved field name was removed
and new members inserted at 192-224 bits.

As the second example shows, this should be best effort.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I8d9bb0fcc4d305426ee1b03d9c43ad20d299ae51
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20544
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

bdev: make iobuf caches configurable

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I5bf79dd434811cdd636880e4d26f23a0b1239f76
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/s

bdev: make iobuf caches configurable

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I5bf79dd434811cdd636880e4d26f23a0b1239f76
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20543
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

check_so_deps: remove [suppress_type] entries

These should always be removed when the associated
library's SO_MAJOR gets bumped.

spdk_nvme_cap_registers was just added, so keep
that one. All of the

check_so_deps: remove [suppress_type] entries

These should always be removed when the associated
library's SO_MAJOR gets bumped.

spdk_nvme_cap_registers was just added, so keep
that one. All of the rest were committed prior to
22.09 release, but spdk_app_opts has since made
another change that we want to suppress. So keep
spdk_app_opts too.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I4cf3387954f392eddbd43209ac8c0e19c0e53817
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20515
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>

show more ...

examples/identify: don't read PMPCAP and CMBSZ if not support in CAP

I got error logs when running `identify` utility with NVMe-oF for above
2 registers access, so add a check before reading them in

examples/identify: don't read PMPCAP and CMBSZ if not support in CAP

I got error logs when running `identify` utility with NVMe-oF for above
2 registers access, so add a check before reading them in `identify`.

Change-Id: Ifb31c3d2ab2b0426c6af7b69f189ea2f1cdde846
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20471
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/abi: fix excluding "-pre" tags from ABI check

Commit below introduced nullglob to autotest:
(4641aa5)test/common: Provide {ext,null}glob by default in autotest suite

Without change in this com

test/abi: fix excluding "-pre" tags from ABI check

Commit below introduced nullglob to autotest:
(4641aa5)test/common: Provide {ext,null}glob by default in autotest suite

Without change in this commit, this resulted in not properly excluding tags
that end with "-pre". This resulted in comparing against pre-release SPDK version.
Since that version didn't yet have spdk-abi XML files for comparison,
and sanity check were missing in confirm_abi_deps() were missing.
All shared files comparison was skipped.

This commit accomplishes the following:
- Correctly excludes SPDK tags with "-pre"
- Checks if spdk-abi contains the directory for relevant SPDK version
- Checks if any shared objects were processed

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reported-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: I43036662e23743ceba4b340386cd0195c7143ffd
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20074
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/check_so_deps: parametrize the script options

No longer rely on positional arguments when calling the
test script.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Icc4726

test/check_so_deps: parametrize the script options

No longer rely on positional arguments when calling the
test script.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Icc4726b50f0b34d99fdb6828b4fd03e83788e008
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19281
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/check_so_deps: only rebuild SPDK if config file is used

Prior to this patch SPDK was always rebuilt and it
was not possible to run the test script without config file.

Now the check_so_deps.sh

test/check_so_deps: only rebuild SPDK if config file is used

Prior to this patch SPDK was always rebuilt and it
was not possible to run the test script without config file.

Now the check_so_deps.sh can be re-run multiple times by
the developer. SPDK has to either be built manually,
or once using a config file.

Change-Id: I8a556d0cb5d99be43300e151bc00176ada1784c4
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19280
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/check_so_deps: fail each test case independently

Rather than produce a fail_file and exit at the end of
whole script, do so within each run_test.

confirm_makefile_deps failure is still done vi

test/check_so_deps: fail each test case independently

Rather than produce a fail_file and exit at the end of
whole script, do so within each run_test.

confirm_makefile_deps failure is still done via file,
due to parrarell execution of confirm_deps.

Meanwhile confirm_abi_deps use a local variable for the
same purpose.

Each run_test will now fail independently.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I40a985a94c49d5d48b6bf908c39415f0362b6e2d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19279
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/check_so_deps: reorder functions closer to their use

This patch makes the script more readable by moving functions
closer to their use. Test are now executed one after another
with building SPD

test/check_so_deps: reorder functions closer to their use

This patch makes the script more readable by moving functions
closer to their use. Test are now executed one after another
with building SPDK preceeding.

No need to disable/restore extraces in check_header_filenames,
as the tests were already without them.

Change-Id: If85343e37753eb934a1ac7aa6348ab0367eb3300
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19278
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>

show more ...

test/check_so_deps: use current test dir for temp files

During the check_so_deps.sh testing some files are generated
temporarily, but are not actual test result and artifact
worth keeping.

spdk-abi

test/check_so_deps: use current test dir for temp files

During the check_so_deps.sh testing some files are generated
temporarily, but are not actual test result and artifact
worth keeping.

spdk-abi repo, abigail suppression file and fail file are
now kept within test directory for duration of the test.

Change-Id: Icc5b75be5c591639dc6ba11d1ec79e8a45c49f5f
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19277
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/check_so_deps: do not ignore exit on FreeBSD

This wasn't executed within CI on FreeBSD right now,
but there is no reason to not produce an error when trying.

Signed-off-by: Tomasz Zawadzki <to

test/check_so_deps: do not ignore exit on FreeBSD

This wasn't executed within CI on FreeBSD right now,
but there is no reason to not produce an error when trying.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I473e3b17344fe47cdb9d1030b3435a3ebc7cfece
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19276
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>

show more ...

test/check_so_deps: refactor pulling spdk-abi repo

This patch refactors pulling spdk-abi repository for
confirm_abi_deps test.

Main goal is to make it easier to developers to execute
check_so_deps.

test/check_so_deps: refactor pulling spdk-abi repo

This patch refactors pulling spdk-abi repository for
confirm_abi_deps test.

Main goal is to make it easier to developers to execute
check_so_deps.sh, by allowing to have a pre-existing
spdk-abi repository on their system and point to it.
Rather than hide it in SPDK_ABI_DIR variable passed in
autotest config file.

Now if second argument is provided it will be used as
path to spdk-abi. This repository will not be removed
at the end of the test.

If not provided, the repo will be cloned and then deleted
at the end of the test.

CI used to pass the path via SPDK_ABI_DIR, so it is now
included as second parameter when called from autobuild.

Change-Id: I8110c2bf4b25658d7f7f6758d00ea57e3ea6e60a
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19275
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...