On Sat, May 8, 2021 at 5:31 AM Brendan Higgins wrote: > > Document QEMU support, what it does, and how to use it in kunit_tool. > > Signed-off-by: Brendan Higgins > --- This is a good start, and probably meets the minimum requirements, but I do have a number of comments and suggestions below. Cheers, -- David > Documentation/dev-tools/kunit/usage.rst | 37 +++++++++++++++++++++---- > 1 file changed, 31 insertions(+), 6 deletions(-) > > diff --git a/Documentation/dev-tools/kunit/usage.rst b/Documentation/dev-tools/kunit/usage.rst > index 650f99590df57..b74bd7c87cc20 100644 > --- a/Documentation/dev-tools/kunit/usage.rst > +++ b/Documentation/dev-tools/kunit/usage.rst > @@ -612,14 +612,39 @@ only things to be aware of when doing so. > The biggest impediment will likely be that certain KUnit features and > infrastructure may not support your target environment. For example, at this > time the KUnit Wrapper (``tools/testing/kunit/kunit.py``) does not work outside > -of UML. Unfortunately, there is no way around this. Using UML (or even just a > -particular architecture) allows us to make a lot of assumptions that make it > -possible to do things which might otherwise be impossible. > +of UML and QEMU. Unfortunately, there is no way around this. Using UML and QEMU > +(or even just a particular architecture) allows us to make a lot of assumptions > +that make it possible to do things which might otherwise be impossible. This is a bit more awkward now, and I don't think gives quite the right impression. Particularly the "Unfortunately, there is no way around this." bit: there's no fundamental reason that someone couldn't implement support for some other emulator (or even a setup which pushed to real hardware and read results over serial), it'd just take a bit of work to implement (like this patch series has done for qemu). Personally, I think it'd be easiest to simplify this section and say that kunit_tool currently only fully supports some architectures, via UML and QEMU. > > Nevertheless, all core KUnit framework features are fully supported on all > -architectures, and using them is straightforward: all you need to do is to take > -your kunitconfig, your Kconfig options for the tests you would like to run, and > -merge them into whatever config your are using for your platform. That's it! > +architectures, and using them is straightforward: Most popular architectures > +are supported directly in the KUnit Wrapper via QEMU. Currently, supported > +architectures on QEMU include: > + > +* i386 > +* x86_64 > +* arm > +* arm64 > +* alpha > +* powerpc > +* riscv > +* s390 > +* sparc > + > +In order to run KUnit tests on one of these architectures via QEMU with the > +KUnit wrapper, all you need to do is specify the flags ``--arch`` and > +``--cross_compile`` when invoking the KUnit Wrapper. For example, we could run > +the default KUnit tests on ARM in the following manner (assuming we have an ARM > +toolchain installed): > + > +.. code-block:: bash > + > + tools/testing/kunit/kunit.py run --timeout=60 --jobs=12 --arch=arm --cross_compile=arm-linux-gnueabihf- > + Is it worth also documenting the --qemu_config option here? (Particularly given the restriction on its path?) Or is that something best added to the kunit_tool page? That being said, changes to the kunit_tool page are probably worth adding as a section in the updated page: https://patchwork.kernel.org/project/linux-kselftest/patch/20210417034553.1048895-1-davidgow@google.com/ At the very least, it'd be nice to have the new QEMU-related options documented there. > +Alternatively, if you want to run your tests on real hardware or in some other > +emulation environment, all you need to do is to take your kunitconfig, your > +Kconfig options for the tests you would like to run, and merge them into > +whatever config your are using for your platform. That's it! > > For example, let's say you have the following kunitconfig: > > -- > 2.31.1.607.g51e8a6a459-goog >