From: Quentin Monnet <quentin.monnet(a)netronome.com>
[ Upstream commit c5fa5d602221362f8341ecd9e32d83194abf5bd9 ]
The return value for each test in test_libbpf.sh is compared with
if (( $? == 0 )) ; then ...
This works well with bash, but not with dash, that /bin/sh is aliased to
on some systems (such as Ubuntu).
Let's replace this comparison by something that works on both shells.
Signed-off-by: Quentin Monnet <quentin.monnet(a)netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski(a)netronome.com>
Signed-off-by: Alexei Starovoitov <ast(a)kernel.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/bpf/test_libbpf.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/bpf/test_libbpf.sh b/tools/testing/selftests/bpf/test_libbpf.sh
index 8b1bc96d8e0cc..2989b2e2d856d 100755
--- a/tools/testing/selftests/bpf/test_libbpf.sh
+++ b/tools/testing/selftests/bpf/test_libbpf.sh
@@ -6,7 +6,7 @@ export TESTNAME=test_libbpf
# Determine selftest success via shell exit code
exit_handler()
{
- if (( $? == 0 )); then
+ if [ $? -eq 0 ]; then
echo "selftests: $TESTNAME [PASS]";
else
echo "$TESTNAME: failed at file $LAST_LOADED" 1>&2
--
2.20.1
Add documentation for the Python script used to build, run, and collect
results from the kernel known as kunit_tool. kunit_tool
(tools/testing/kunit/kunit.py) was already added in previous commits.
Signed-off-by: Brendan Higgins <brendanhiggins(a)google.com>
---
Documentation/dev-tools/kunit/index.rst | 1 +
Documentation/dev-tools/kunit/kunit-tool.rst | 57 ++++++++++++++++++++
Documentation/dev-tools/kunit/start.rst | 5 +-
3 files changed, 62 insertions(+), 1 deletion(-)
create mode 100644 Documentation/dev-tools/kunit/kunit-tool.rst
diff --git a/Documentation/dev-tools/kunit/index.rst b/Documentation/dev-tools/kunit/index.rst
index 26ffb46bdf99d..c60d760a0eed1 100644
--- a/Documentation/dev-tools/kunit/index.rst
+++ b/Documentation/dev-tools/kunit/index.rst
@@ -9,6 +9,7 @@ KUnit - Unit Testing for the Linux Kernel
start
usage
+ kunit-tool
api/index
faq
diff --git a/Documentation/dev-tools/kunit/kunit-tool.rst b/Documentation/dev-tools/kunit/kunit-tool.rst
new file mode 100644
index 0000000000000..37509527c04e1
--- /dev/null
+++ b/Documentation/dev-tools/kunit/kunit-tool.rst
@@ -0,0 +1,57 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=================
+kunit_tool How-To
+=================
+
+What is kunit_tool?
+===================
+
+kunit_tool is a script (``tools/testing/kunit/kunit.py``) that aids in building
+the Linux kernel as UML (`User Mode Linux
+<http://user-mode-linux.sourceforge.net/>`_), running KUnit tests, parsing
+the test results and displaying them in a user friendly manner.
+
+What is a kunitconfig?
+======================
+
+It's just a defconfig that kunit_tool looks for in the base directory.
+kunit_tool uses it to generate a .config as you might expect. In addition, it
+verifies that the generated .config contains the CONFIG options in the
+kunitconfig; the reason it does this is so that it is easy to be sure that a
+CONFIG that enables a test actually ends up in the .config.
+
+How do I use kunit_tool?
+=================================
+
+If a kunitconfig is present at the root directory, all you have to do is:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run
+
+However, you most likely want to use it with the following options:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --timeout=30 --jobs=`nproc --all`
+
+- ``--timeout`` sets a maximum amount of time to allow tests to run.
+- ``--jobs`` sets the number of threads to use to build the kernel.
+
+If you just want to use the defconfig that ships with the kernel, you can
+append the ``--defconfig`` flag as well:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --timeout=30 --jobs=`nproc --all` --defconfig
+
+.. note::
+ This command is particularly helpful for getting started because it
+ just works. No kunitconfig needs to be present.
+
+For a list of all the flags supported by kunit_tool, you can run:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --help
diff --git a/Documentation/dev-tools/kunit/start.rst b/Documentation/dev-tools/kunit/start.rst
index aeeddfafeea20..4248a6f9038b8 100644
--- a/Documentation/dev-tools/kunit/start.rst
+++ b/Documentation/dev-tools/kunit/start.rst
@@ -19,7 +19,10 @@ The wrapper can be run with:
.. code-block:: bash
- ./tools/testing/kunit/kunit.py run
+ ./tools/testing/kunit/kunit.py run --timeout=30 --jobs=`nproc --all` --defconfig
+
+For more information on this wrapper (also called kunit_tool) checkout the
+:doc:`kunit-tool` page.
Creating a kunitconfig
======================
--
2.24.0.432.g9d3f5f5b63-goog
Currently the following command produces an error message:
linux# make kselftest TARGETS=bpf O=/mnt/linux-build
# selftests: bpf: test_libbpf.sh
# ./test_libbpf.sh: line 23: ./test_libbpf_open: No such file or directory
# test_libbpf: failed at file test_l4lb.o
# selftests: test_libbpf [FAILED]
The error message might not affect make return code, therefore one might
need to grep make output in order to detect it.
The current logic prepends $(OUTPUT) only to the first member of
$(TEST_PROGS). After that, run_one() does
cd `dirname $TEST`
For all tests except the first one, `dirname $TEST` is ., which means
they cannot access the files generated in $(OUTPUT).
Fix by using $(addprefix) to prepend $(OUTPUT)/ to each member of
$(TEST_PROGS).
Fixes: 1a940687e424 ("selftests: lib.mk: copy test scripts and test files for make O=dir run")
Signed-off-by: Ilya Leoshkevich <iii(a)linux.ibm.com>
---
v1->v2:
- Append / to $(OUTPUT).
- Use $(addprefix) instead of $(foreach).
v2->v3:
- Split the patch in two.
- Improve the commit message.
v3->v4:
- Drop the first patch.
- Add a note regarding make return code to the commit message.
tools/testing/selftests/lib.mk | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/lib.mk b/tools/testing/selftests/lib.mk
index 1c8a1963d03f..0cf510df1ee2 100644
--- a/tools/testing/selftests/lib.mk
+++ b/tools/testing/selftests/lib.mk
@@ -75,7 +75,8 @@ ifdef building_out_of_srctree
@rsync -aq $(TEST_PROGS) $(TEST_PROGS_EXTENDED) $(TEST_FILES) $(OUTPUT)
fi
@if [ "X$(TEST_PROGS)" != "X" ]; then
- $(call RUN_TESTS, $(TEST_GEN_PROGS) $(TEST_CUSTOM_PROGS) $(OUTPUT)/$(TEST_PROGS))
+ $(call RUN_TESTS, $(TEST_GEN_PROGS) $(TEST_CUSTOM_PROGS) \
+ $(addprefix $(OUTPUT)/,$(TEST_PROGS)))
else
$(call RUN_TESTS, $(TEST_GEN_PROGS) $(TEST_CUSTOM_PROGS))
fi
--
2.23.0
Greetings,
Find attached email very confidential. reply for more details
Thanks.
Peter Wong
----------------------------------------------------
This email was sent by the shareware version of Postman Professional.
Add documentation for the Python script used to build, run, and collect
results from the kernel known as kunit_tool. kunit_tool
(tools/testing/kunit/kunit.py) was already added in previous commits.
Signed-off-by: Brendan Higgins <brendanhiggins(a)google.com>
---
Documentation/dev-tools/kunit/index.rst | 1 +
Documentation/dev-tools/kunit/kunit-tool.rst | 57 ++++++++++++++++++++
Documentation/dev-tools/kunit/start.rst | 3 ++
3 files changed, 61 insertions(+)
create mode 100644 Documentation/dev-tools/kunit/kunit-tool.rst
diff --git a/Documentation/dev-tools/kunit/index.rst b/Documentation/dev-tools/kunit/index.rst
index 26ffb46bdf99d..c60d760a0eed1 100644
--- a/Documentation/dev-tools/kunit/index.rst
+++ b/Documentation/dev-tools/kunit/index.rst
@@ -9,6 +9,7 @@ KUnit - Unit Testing for the Linux Kernel
start
usage
+ kunit-tool
api/index
faq
diff --git a/Documentation/dev-tools/kunit/kunit-tool.rst b/Documentation/dev-tools/kunit/kunit-tool.rst
new file mode 100644
index 0000000000000..aa1a93649a45a
--- /dev/null
+++ b/Documentation/dev-tools/kunit/kunit-tool.rst
@@ -0,0 +1,57 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=================
+kunit_tool How-To
+=================
+
+What is kunit_tool?
+===================
+
+kunit_tool is a set of scripts that aid in building the Linux kernel as UML
+(`User Mode Linux <http://user-mode-linux.sourceforge.net/old/>`_), running
+KUnit tests, parsing the test results and displaying them in a user friendly
+manner.
+
+What is a kunitconfig?
+======================
+
+It's just a defconfig that kunit_tool looks for in the base directory.
+kunit_tool uses it to generate a .config as you might expect. In addition, it
+verifies that the generated .config contains the CONFIG options in the
+kunitconfig; the reason it does this is so that it is easy to be sure that a
+CONFIG that enables a test actually ends up in the .config.
+
+How do I use kunit_tool?
+=================================
+
+If a kunitconfig is present at the root directory, all you have to do is:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run
+
+However, you most likely want to use it with the following options:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --timeout=30 --jobs=8
+
+- ``--timeout`` sets a maximum amount of time to allow tests to run.
+- ``--jobs`` sets the number of threads to use to build the kernel.
+
+If you just want to use the defconfig that ships with the kernel, you can
+append the ``--defconfig`` flag as well:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --timeout=30 --jobs=8 --defconfig
+
+.. note::
+ This command is particularly helpful for getting started because it
+ just works. No kunitconfig needs to be present.
+
+For a list of all the flags supported by kunit_tool, you can run:
+
+.. code-block:: bash
+
+ ./tools/testing/kunit/kunit.py run --help
diff --git a/Documentation/dev-tools/kunit/start.rst b/Documentation/dev-tools/kunit/start.rst
index aeeddfafeea20..1535c4394cfa2 100644
--- a/Documentation/dev-tools/kunit/start.rst
+++ b/Documentation/dev-tools/kunit/start.rst
@@ -21,6 +21,9 @@ The wrapper can be run with:
./tools/testing/kunit/kunit.py run
+For more information on this wrapper (also called kunit_tool) checkout the
+:doc:`kunit-tool` page.
+
Creating a kunitconfig
======================
The Python script is a thin wrapper around Kbuild as such, it needs to be
--
2.24.0.432.g9d3f5f5b63-goog
On Thu, Nov 14, 2019 at 1:31 PM Brendan Higgins
<brendanhiggins(a)google.com> wrote:
>
> +kselftest and kunit lists to document this decision.
Sorry for the spam. I accidentally CC'ed the doc list instead of the
kselftest list in my previous email.
> On Wed, Nov 13, 2019 at 11:54 PM Alan Maguire <alan.maguire(a)oracle.com> wrote:
> >
> > On Wed, 13 Nov 2019, Stephen Boyd wrote:
> >
> > > Quoting Brendan Higgins (2019-11-11 13:41:38)
> > > > +Stephen Boyd - since he is more of an expert on the hung task timer than I am.
> > > >
> > > > On Fri, Nov 8, 2019 at 7:30 AM Alan Maguire <alan.maguire(a)oracle.com> wrote:
> > > > >
> > > > > On Thu, 7 Nov 2019, Brendan Higgins wrote:
> > > > >
> > > > > > On Thu, Oct 17, 2019 at 11:09 AM Alan Maguire <alan.maguire(a)oracle.com> wrote:
> > > > > > > +MODULE_LICENSE("GPL");
> > > > > > > diff --git a/lib/kunit/try-catch.c b/lib/kunit/try-catch.c
> > > > > > > index 1c1e9af..72fc8ed 100644
> > > > > > > --- a/lib/kunit/try-catch.c
> > > > > > > +++ b/lib/kunit/try-catch.c
> > > > > > > @@ -31,6 +31,8 @@ static int kunit_generic_run_threadfn_adapter(void *data)
> > > > > > > complete_and_exit(try_catch->try_completion, 0);
> > > > > > > }
> > > > > > >
> > > > > > > +KUNIT_VAR_SYMBOL(sysctl_hung_task_timeout_secs, unsigned long);
> > > > > >
> > > > > > Can you just export sysctl_hung_task_timeout_secs?
> > > > > >
> > > > > > I don't mean to make you redo all this work for one symbol twice, but
> > > > > > I thought we agreed on just exposing this symbol, but in a namespace.
> > > > > > It seemed like a good use case for that namespaced exporting thing
> > > > > > that Luis was talking about. As I understood it, you would have to
> > > > > > export it in the module that defines it, and then use the new
> > > > > > MODULE_IMPORT_NS() macro here.
> > > > > >
> > > > >
> > > > > Sure, I can certainly look into that, though I wonder if we should
> > > > > consider another possibility - should kunit have its own sysctl table for
> > > > > things like configuring timeouts? I can look at adding a patch for that
> > > >
> > > > So on the one hand, yes, I would like to have configurable test
> > > > timeouts for KUnit, but that is not what the parameter check is for
> > > > here. This is to make sure KUnit times a test case out before the hung
> > > > task timer does.
> > > >
> > > > > prior to the module patch so the issues with exporting the hung task
> > > > > timeout would go away. Now the reason I suggest this isn't as much a hack
> > > > > to solve this specific problem, rather it seems to fit better with the
> > > > > longer-term intent expressed by the comment around use of the field (at
> > > > > least as I read it, I may be wrong).
> > > >
> > > > Not really. Although I do agree that adding configurability here might
> > > > be a good idea, I believe we would need to clamp such a value by
> > > > sysctl_hung_task_timeout_secs regardless since we don't want to be
> > > > killed by the hung task timer; thus, we still need access to
> > > > sysctl_hung_task_timeout_secs either way, and so doing what you are
> > > > proposing would be off topic.
> > > >
> > > > > Exporting the symbol does allow us to piggy-back on an existing value, but
> > > > > maybe we should support out our own tunable "kunit_timeout_secs" here?
> > > > > Doing so would also lay the groundwork for supporting other kunit
> > > > > tunables in the future if needed. What do you think?
> > > >
> > > > The goal is not to piggy back on the value as I mentioned above.
> > > > Stephen, do you have any thoughts on this? Do you see any other
> > > > preferable solution to what Alan is trying to do?
> > >
> > > One idea would be to make some sort of process flag that says "this is a
> > > kunit task, ignore me with regards to the hung task timeout". Then we
> > > can hardcode the 5 minute kunit timeout. I'm not sure we have any more
> > > flags though.
> > >
> > > Or drop the whole timeout clamping logic, let the hung task timeout kick
> > > in and potentially oops the kernel, but then continue to let the test
> > > run and maybe sometimes get the kunit timeout here. This last option
> > > doesn't sound so bad to me given that this is all a corner case anyway
> > > where we don't expect to actually ever hit this problem so letting the
> > > hung task detector do its job is probably fine. This nicely avoids
> > > having to export this symbol to modules too.
> > >
> >
> > Thanks for suggestions! This latter approach seems fine to me; presumably
> > something has gone wrong if we are tripping the hung task timeout anyway,
> > so having an oops to document that seems fine. Brendan, what do you think?
>
> If Stephen thinks it is fine to drop the clamping logic, I think it is
> fine too. I think it would probably be good to replace it with a
> comment under the TODO that explains that a hung test *can* cause an
> oops if the hung task timeout is less than the kunit timeout value. It
> would probably be good to also select a timeout value that is less
> than the default hung task timeout. We might also want to link to this
> discussion. I fully expect that the timeout logic will get more
> attention at some point in the future.
>
> One more thing: Alan, can you submit the commit that drops the
> clamping logic in its own commit? I would prefer to make sure that it
> is easy to spot in the commit history.
>
> Cheers!