On Sat, Feb 15, 2025 at 4:03 AM David Gow <david...@google.com> wrote:
>
> On Fri, 14 Feb 2025 at 22:41, Tamir Duberstein <tam...@gmail.com> wrote:
> >
> > On Fri, Feb 14, 2025 at 2:42 AM David Gow <david...@google.com> wrote:
> > >
> > > From: José Expósito <jose.exposit...@gmail.com>
> > >
> > > In some cases, we need to call test-only code from outside the test
> > > case, for example, to mock a function or a module.
> > >
> > > In order to check whether we are in a test or not, we need to test if
> > > `CONFIG_KUNIT` is set.
> > > Unfortunately, we cannot rely only on this condition because:
> > > - a test could be running in another thread,
> > > - some distros compile KUnit in production kernels, so checking at runtime
> > > that `current->kunit_test != NULL` is required.
> > >
> > > Forturately, KUnit provides an optimised check in
> > > `kunit_get_current_test()`, which checks CONFIG_KUNIT, a global static
> > > key, and then the current thread's running KUnit test.
> > >
> > > Add a safe wrapper function around this to know whether or not we are in
> > > a KUnit test and examples showing how to mock a function and a module.
> > >
> > > Signed-off-by: José Expósito <jose.exposit...@gmail.com>
> > > Co-developed-by: David Gow <david...@google.com>
> > > Signed-off-by: David Gow <david...@google.com>
> > > Co-developed-by: Miguel Ojeda <oj...@kernel.org>
> > > Signed-off-by: Miguel Ojeda <oj...@kernel.org>
> > > ---
> > >
> > > Changes since v5:
> > > https://lore.kernel.org/all/20241213081035.2069066-4-david...@google.com/
> > > - Greatly improved documentation, which is both clearer and better
> > > matches the rustdoc norm. (Thanks, Miguel)
> > > - The examples and safety comments are also both more idiomatic an
> > > cleaner. (Thanks, Miguel)
> > > - More things sit appropriately behind CONFIG_KUNIT (Thanks, Miguel)
> > >
> > > Changes since v4:
> > > https://lore.kernel.org/linux-kselftest/20241101064505.3820737-4-david...@google.com/
> > > - Rebased against 6.13-rc1
> > > - Fix some missing safety comments, and remove some unneeded 'unsafe'
> > > blocks. (Thanks Boqun)
> > >
> > > Changes since v3:
> > > https://lore.kernel.org/linux-kselftest/20241030045719.3085147-8-david...@google.com/
> > > - The example test has been updated to no longer use assert_eq!() with
> > > a constant bool argument (fixes a clippy warning).
> > >
> > > No changes since v2:
> > > https://lore.kernel.org/linux-kselftest/20241029092422.2884505-4-david...@google.com/
> > >
> > > Changes since v1:
> > > https://lore.kernel.org/lkml/20230720-rustbind-v1-3-c80db349e...@google.com/
> > > - Rebased on top of rust-next.
> > > - Use the `kunit_get_current_test()` C function, which wasn't previously
> > > available, instead of rolling our own.
> > > - (Thanks also to Boqun for suggesting a nicer way of implementing this,
> > > which I tried, but the `kunit_get_current_test()` version obsoleted.)
> > > ---
> > > rust/kernel/kunit.rs | 66 ++++++++++++++++++++++++++++++++++++++++++++
> > > 1 file changed, 66 insertions(+)
> > >
> > > diff --git a/rust/kernel/kunit.rs b/rust/kernel/kunit.rs
> > > index 9e27b74a605b..3aad7a281b6d 100644
> > > --- a/rust/kernel/kunit.rs
> > > +++ b/rust/kernel/kunit.rs
> > > @@ -286,11 +286,77 @@ macro_rules! kunit_unsafe_test_suite {
> > > };
> > > }
> > >
> > > +/// Returns whether we are currently running a KUnit test.
> > > +///
> > > +/// In some cases, you need to call test-only code from outside the test
> > > case, for example, to
> > > +/// create a function mock. This function allows to change behavior
> > > depending on whether we are
> > > +/// currently running a KUnit test or not.
> > > +///
> > > +/// # Examples
> > > +///
> > > +/// This example shows how a function can be mocked to return a
> > > well-known value while testing:
> > > +///
> > > +/// ```
> > > +/// # use kernel::kunit::in_kunit_test;
> > > +/// fn fn_mock_example(n: i32) -> i32 {
> > > +/// if in_kunit_test() {
> > > +/// return 100;
> > > +/// }
> > > +///
> > > +/// n + 1
> > > +/// }
> > > +///
> > > +/// let mock_res = fn_mock_example(5);
> > > +/// assert_eq!(mock_res, 100);
> > > +/// ```
> > > +///
> > > +/// Sometimes, you don't control the code that needs to be mocked. This
> > > example shows how the
> > > +/// `bindings` module can be mocked:
> >
> > [`bindings`] here, please. There are two more instances below but
> > those aren't doc comments, so I don't think bracketing them will do
> > anything.
> >
>
> Done in v7. Alas, I'll have to keep getting used to the differences
> between kerneldoc and rustdoc...
>
> > > +///
> > > +/// ```
> > > +/// // Import our mock naming it as the real module.
> > > +/// #[cfg(CONFIG_KUNIT)]
> > > +/// use bindings_mock_example as bindings;
> > > +/// #[cfg(not(CONFIG_KUNIT))]
> > > +/// use kernel::bindings;
> > > +///
> > > +/// // This module mocks `bindings`.
> > > +/// #[cfg(CONFIG_KUNIT)]
> > > +/// mod bindings_mock_example {
> > > +/// /// Mock `ktime_get_boot_fast_ns` to return a well-known value
> > > when running a KUnit test.
> > > +/// pub(crate) fn ktime_get_boot_fast_ns() -> u64 {
> > > +/// 1234
> > > +/// }
> > > +/// }
> > > +///
> > > +/// // This is the function we want to test. Since `bindings` has been
> > > mocked, we can use its
> > > +/// // functions seamlessly.
> > > +/// fn get_boot_ns() -> u64 {
> > > +/// // SAFETY: `ktime_get_boot_fast_ns()` is always safe to call.
> > > +/// unsafe { bindings::ktime_get_boot_fast_ns() }
> > > +/// }
> > > +///
> > > +/// let time = get_boot_ns();
> > > +/// assert_eq!(time, 1234);
> > > +/// ```
> >
> > Isn't this swapping out the bindings module at compile time, and for
> > the whole build? In other words cfg(CONFIG_KUNIT) will apply to all
> > code, both test and non-test.
> >
>
> I believe so, so this is probably something best done only in test files.
Why would you need conditional compilation of this kind in test files?
What I was getting at with this comment is that this example might
mislead a user to think that this is how they should imbue their code
with test-specific behavior, which is not what this will do. Instead
this would break kernels built with CONFIG_KUNIT, which I think is not
where we want to be going. Right?
>
> Ideally, we'd have support for something like the KUnit function
> mocking features here, but that's horribly C-specific at the moment.
>
> > > +pub fn in_kunit_test() -> bool {
> > > + // SAFETY: `kunit_get_current_test()` is always safe to call (it has
> > > fallbacks for
> > > + // when KUnit is not enabled).
> > > + unsafe { !bindings::kunit_get_current_test().is_null() }
> >
> > Nit if you care about reducing unsafe blocks:
> >
> > !unsafe { bindings::kunit_get_current_test() }.is_null()
> >
> >
>
> Huh, I thought this wouldn't work, but it's working fine for me here,
> so I've made the change.
>
> Thanks!
>
> > > +}
> > > +
> > > #[kunit_tests(rust_kernel_kunit)]
> > > mod tests {
> > > + use super::*;
> > > +
> > > #[test]
> > > fn rust_test_kunit_example_test() {
> > > #![expect(clippy::eq_op)]
> > > assert_eq!(1 + 1, 2);
> > > }
> > > +
> > > + #[test]
> > > + fn rust_test_kunit_in_kunit_test() {
> > > + assert!(in_kunit_test());
> > > + }
> > > }
> > > --
> > > 2.48.1.601.g30ceb7b040-goog
> > >
> > >
>
> Thanks a lot, these should be fixed in v7.
>
> Cheers,
> -- David
