test/fuzz/llvm: Make llvo-gcov.sh more portable

For FreeBSD's sake. That said, it's not really needed to use such
wrapper under FreeBSD since gcov by the very default executes there
in llvm context.

test/fuzz/llvm: Make llvo-gcov.sh more portable

For FreeBSD's sake. That said, it's not really needed to use such
wrapper under FreeBSD since gcov by the very default executes there
in llvm context. But this aligns it more with what we do for Linux
and doesn't require any extra distro checks.

Change-Id: Ib22d80ee0a28f3b611193be49f327594465b953f
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24775
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Bypass-Merge-Requirements: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

autotest: Stop rejecting coverage collection under clang

78d5bfe575 claimed that collecting coverage from a clang build
takes signficant amount of time, hence not being collected by
the autotest nor

autotest: Stop rejecting coverage collection under clang

78d5bfe575 claimed that collecting coverage from a clang build
takes signficant amount of time, hence not being collected by
the autotest nor unittest.

However, currently this doesn't seem to be the case anymore.
To align coverage and make sure there are no any odd exceptions
let's have it always collected.

This also merges llvm coverage into the one collected by the
autotest instance as there's no point now to keep it separately.

Change-Id: Ic46408f313fff0273a226adbf3fb5f56c916aedf
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24722
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/fuzz: create output dir only when coverage is collected

The folder is only created when needed, avoiding unnecessary
creation of empty artifacts

Resolves #3491

Change-Id: Ib7b1e4fd94ea02bead6

test/fuzz: create output dir only when coverage is collected

The folder is only created when needed, avoiding unnecessary
creation of empty artifacts

Resolves #3491

Change-Id: Ib7b1e4fd94ea02bead6ceaf12e6a59661194311c
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24591
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: Tomasz Zawadzki <tomasz@tzawadzki.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>

show more ...

python/rpc: remove deprecated retry_count param

Remove deprecated retry_count parameter from
bdev_nvme_set_options rpc function

Change-Id: I1b415cfcfd7b3226d6331960ad42c500ba77f0aa
Signed-off-by: E

python/rpc: remove deprecated retry_count param

Remove deprecated retry_count parameter from
bdev_nvme_set_options rpc function

Change-Id: I1b415cfcfd7b3226d6331960ad42c500ba77f0aa
Signed-off-by: Eugene Kobyak <eugene.kobyak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23498
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot

show more ...

test/fuzz: Rename llvm fuzzing tests

Make it clear they are llvm-related - the nvmf_fuzz is already used
by the fabrics_fuzz.sh tests from the nvmf suite.

Change-Id: I23064b3166b7d0c80cd8d082aa73d1

test/fuzz: Rename llvm fuzzing tests

Make it clear they are llvm-related - the nvmf_fuzz is already used
by the fabrics_fuzz.sh tests from the nvmf suite.

Change-Id: I23064b3166b7d0c80cd8d082aa73d1e1624f8b20
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23467
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/iscsi: remove duplicate fuzzing test

Both autofuzz_iscsi.sh and fuzz.sh are the same.
The difference is passing the time to fuzz and
minor difference in how the application was started.

Remove

test/iscsi: remove duplicate fuzzing test

Both autofuzz_iscsi.sh and fuzz.sh are the same.
The difference is passing the time to fuzz and
minor difference in how the application was started.

Removed fuzz.sh which was called directly in per-patch
autotest for short period of 30 sec. Replaced it with call
to autofuzz_iscsi.sh with the same timer.

Adjust ISCSI_APP to start in the same way, most importantly
per-patch now won't generate megabytes of logs due to
lack of output redirection in the removed fuzz.sh.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I79927a0634d9aa96014f9b589fd2883268ee18f5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22518
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

llvm/nvmf: supress false memory leak of `qpair_ctx` and `ctrlr`

This behavior is only reproducible during fuzzing test, probably due
to asynchronous nature of `spdk_nvmf_qpair_disconnect` and
`nvmf_

llvm/nvmf: supress false memory leak of `qpair_ctx` and `ctrlr`

This behavior is only reproducible during fuzzing test, probably due
to asynchronous nature of `spdk_nvmf_qpair_disconnect` and
`nvmf_ctrlr_create`. It was verified that there is no real memory leak.

Fixes issue #3285.

Change-Id: I94129bb0cbb63e4d5f7c96ae856d0a398f812f33
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22713
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

llvm/vfio: supress false memory leak of `qpair_ctx` and `ctrlr`

This behavior is only reproducible during fuzzing test, probably due
to asynchronous nature of `spdk_nvmf_qpair_disconnect` and
`nvmf_

llvm/vfio: supress false memory leak of `qpair_ctx` and `ctrlr`

This behavior is only reproducible during fuzzing test, probably due
to asynchronous nature of `spdk_nvmf_qpair_disconnect` and
`nvmf_ctrlr_create`. It was verified that there is no real memory leak.

Fixes issue #3285.

Change-Id: I97ebb4d2bb7ff8d0dab9b4cb2c9eee1b46d8b3b7
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22226
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

test: remove unnecessary RPC address setting

Since most of the apps no longer attempt to start RPC server by default,
we no longer need to assign unique addresses.

Change-Id: I714e30268141efef6f911

test: remove unnecessary RPC address setting

Since most of the apps no longer attempt to start RPC server by default,
we no longer need to assign unique addresses.

Change-Id: I714e30268141efef6f9117d3d4e6dab04d4fdc50
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22635
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

test/fuzz/iscsi: add --wait-for-rpc for ISCSI_APP parameters

After running fuzzer it turned out that we have to add this
parameter as it was done for test/iscsi_tgt/fuzz/fuzz.sh

Signed-off-by: Kami

test/fuzz/iscsi: add --wait-for-rpc for ISCSI_APP parameters

After running fuzzer it turned out that we have to add this
parameter as it was done for test/iscsi_tgt/fuzz/fuzz.sh

Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I382a3ddfcf3718d23f89868fc35a014db2b41dfd
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22412
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
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: Michal Berger <michal.berger@intel.com>

show more ...

test/fuzz/autofuzz: code refactor

Avoid duplicated code by creating function for:
* config creation (and parameters sanity check)
* running specified fuzzer with run_fuzzer

Signed-off-by: Kamil God

test/fuzz/autofuzz: code refactor

Avoid duplicated code by creating function for:
* config creation (and parameters sanity check)
* running specified fuzzer with run_fuzzer

Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I1cc04d0954c3cfaa877da9302481f3e36a12b25b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22414
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

test/fuzz/autofuzz: add missing iscsi autofuzz tests

iSCSI autofuzzer was never added for some reason.

Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I032faa7b56f509e700d3eb2534

test/fuzz/autofuzz: add missing iscsi autofuzz tests

iSCSI autofuzzer was never added for some reason.

Signed-off-by: Kamil Godzwon <kamilx.godzwon@intel.com>
Change-Id: I032faa7b56f509e700d3eb25347fb384b170a659
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22374
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
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>

show more ...

fuzzer: enable reporting address of leaked memory

This patch turns on reporting of the address of leaked
memory object, when a leak sanitizer is detecting a memory
leak during fuzz tests. This is do

fuzzer: enable reporting address of leaked memory

This patch turns on reporting of the address of leaked
memory object, when a leak sanitizer is detecting a memory
leak during fuzz tests. This is done by setting
LSAN_OPTIONS vaiable to 'report_objects=1'

Can be used to verify if the leaked object is really
a leak, or during memory leak debugging.
Example output:

Objects leaked above:
0x60700000a820 (72 bytes)

SUMMARY: AddressSanitizer: 72 byte(s) leaked in 1 allocation(s).

If 'report_objects' is not enabled, the address of leaked object
is not provided (only the size is reported)

Change-Id: I0d59ff82096b64ed17fb89e369b439c9f565f500
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21896
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>

show more ...

test/fuzz/llvm: Add option to compress fuzzer's output

Change-Id: I15865076f1205f86addaae64f27926f399d97ded
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io

test/fuzz/llvm: Add option to compress fuzzer's output

Change-Id: I15865076f1205f86addaae64f27926f399d97ded
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21111
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>

show more ...

test/fuzz/llvm: Add support for sending fuzzer output to syslog

The long variant of the llvm fuzzer test is producing ~30GB of output
data. Since this may put the underlying system under a severe st

test/fuzz/llvm: Add support for sending fuzzer output to syslog

The long variant of the llvm fuzzer test is producing ~30GB of output
data. Since this may put the underlying system under a severe stress,
depending on how its storage is set up, add an alternative for the
data destination.

Change-Id: I3a4c0efa8df3919642a531bcbe73c5a3daaa97e0
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21110
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/fuzz/llvm: Simplify parallel execution logic

Change-Id: I2b02b4efebc44975f067f71a7ff6f16bd1e430b4
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerr

test/fuzz/llvm: Simplify parallel execution logic

Change-Id: I2b02b4efebc44975f067f71a7ff6f16bd1e430b4
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21109
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>

show more ...

Revert "llvm/vfio: increase memory in short fuzz test"

Enabling legacy mode `-s` option cause `failed: Cannot allocate memory`,
there is not enough memory left.

This reverts commit c8a7742595d94313

Revert "llvm/vfio: increase memory in short fuzz test"

Enabling legacy mode `-s` option cause `failed: Cannot allocate memory`,
there is not enough memory left.

This reverts commit c8a7742595d94313e98431a9496c4a4f02b8c228.

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I92968783dc82b9d66826c482ac584cda01e36738
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20010
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

llvm/vfio: increase hugepages to 2048 pre test

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: Id700d20d9700c2d7823e679304e9d6c30cbb1dc8
Reviewed-on: https://review.spd

llvm/vfio: increase hugepages to 2048 pre test

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: Id700d20d9700c2d7823e679304e9d6c30cbb1dc8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19136
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

llvm/vfio: increase memory in short fuzz test

When spdk is built with `--with-ocf` support, processes have more memory
requirements.

Also, make sure that short fuzz allocated enough hugepages(5GiB)

llvm/vfio: increase memory in short fuzz test

When spdk is built with `--with-ocf` support, processes have more memory
requirements.

Also, make sure that short fuzz allocated enough hugepages(5GiB).

Fixes: #2992

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I0150c61ab5ea3ea84715b7ee7897ebf2b04d59a9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/18835
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

autotest: Stop running extra git clean against .gcda files

These files are part of .gitignore hence this command is foobar.

Change-Id: Ied1bc156cf9b2c428e5f3fdb3b31cf4715c7d07f
Signed-off-by: Micha

autotest: Stop running extra git clean against .gcda files

These files are part of .gitignore hence this command is foobar.

Change-Id: Ied1bc156cf9b2c428e5f3fdb3b31cf4715c7d07f
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/18120
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>

show more ...

fuzz/llvm: move coverage data to `llvm/coverage`

There is no access to fuzzer logs if `index.html` is in the same dir,
move covrage to `$output_dir/llvm/coverage`.

Signed-off-by: Sebastian Brzezink

fuzz/llvm: move coverage data to `llvm/coverage`

There is no access to fuzzer logs if `index.html` is in the same dir,
move covrage to `$output_dir/llvm/coverage`.

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I139a6d780754aaf5b1333a2e5b0183bd24488bfa
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16341
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>

show more ...

fuzz/llvm: provide a prefix to use when saving artifacts

Save crash files and other artifacts in `$output_dir/llvm/`

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I2

fuzz/llvm: provide a prefix to use when saving artifacts

Save crash files and other artifacts in `$output_dir/llvm/`

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I2ff82b414592cc492b79c9178b7257b2e87440b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15827
Reviewed-by: Michal Berger <michal.berger@intel.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>

show more ...

llvm_nvme_fuzz: enable running llvm nvmf test in parallel

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: Iad129c1bc62116a93701a5f68c78351f01a4c878
Reviewed-on: https:/

llvm_nvme_fuzz: enable running llvm nvmf test in parallel

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: Iad129c1bc62116a93701a5f68c78351f01a4c878
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16249
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

llvm_vfio_fuzz: start fuzzer tests in parallel

With corpus files persistent between fuzzer weekend run it may be
better to start all test for fraction of time instead of different
test every week.

llvm_vfio_fuzz: start fuzzer tests in parallel

With corpus files persistent between fuzzer weekend run it may be
better to start all test for fraction of time instead of different
test every week.

Remove `poll_groups_mask` from config, this patch run every test on
single core and since then there is no need to specify another mask.

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I4448724801bdf1a3c496f829fd168b840c2efa67
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15384
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: Michal Berger <michal.berger@intel.com>

show more ...

fuzz/llvm: add `common.sh` for llvm fuzzers

`common.sh` - add common function to start fuzzers in
parallel and quick sequential run

add `get_testn` - get number of test to run in parallel

Signed-o

fuzz/llvm: add `common.sh` for llvm fuzzers

`common.sh` - add common function to start fuzzers in
parallel and quick sequential run

add `get_testn` - get number of test to run in parallel

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I7c70b5221887c29b495a1632545877ca7cca0945
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16323
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>

show more ...