When writing tests, it'd often be very useful to be able to intercept
calls to a function in the code being tested and replace it with a
test-specific stub. This has always been an obviously missing piece of
KUnit, and the solutions always involve some tradeoffs with cleanliness,
performance, or impact on non-test code. See the folowing document for
some of the challenges:
https://kunit.dev/mocking.html
This series consists of two prototype patches which add support for this
sort of redirection to KUnit tests:
1: static_stub: Any function which might want to be intercepted adds a
call to a macro which checks if a test has redirected calls to it, and
calls the corresponding replacement.
2: ftrace_stub: Functions are intercepted using ftrace and livepatch.
This doesn't require adding a new prologue to each function being
replaced, but does have more dependencies (which restricts it to a small
number of architectures, not including UML), and doesn't work well with
inline functions.
The API for both implementations is very similar, so it should be easy
to migrate from one to the other if necessary. Both of these
implementations restrict the redirection to the test context: it is
automatically undone after the KUnit test completes, and does not affect
calls in other threads. If CONFIG_KUNIT is not enabled, there should be
no overhead in either implementation.
Does either (or both) of these features sound useful, and is this
sort-of API the right model? (Personally, I think there's a reasonable
scope for both.) Is anything obviously missing or wrong? Do the names,
descriptions etc. make any sense?
Note that these patches are definitely still at the "prototype" level,
and things like error-handling, documentation, and testing are still
pretty sparse. There is also quite a bit of room for optimisation.
These'll all be improved for v1 if the concept seems good.
Cheers,
-- David
Daniel Latypov (1):
kunit: expose ftrace-based API for stubbing out functions during tests
David Gow (1):
kunit: Expose 'static stub' API to redirect functions
include/kunit/ftrace_stub.h | 84 +++++++++++++++++
include/kunit/static_stub.h | 106 +++++++++++++++++++++
lib/kunit/Kconfig | 11 +++
lib/kunit/Makefile | 5 +
lib/kunit/ftrace_stub.c | 138 ++++++++++++++++++++++++++++
lib/kunit/kunit-example-test.c | 64 +++++++++++++
lib/kunit/static_stub.c | 125 +++++++++++++++++++++++++
lib/kunit/stubs_example.kunitconfig | 11 +++
8 files changed, 544 insertions(+)
create mode 100644 include/kunit/ftrace_stub.h
create mode 100644 include/kunit/static_stub.h
create mode 100644 lib/kunit/ftrace_stub.c
create mode 100644 lib/kunit/static_stub.c
create mode 100644 lib/kunit/stubs_example.kunitconfig
--
2.35.1.894.gb6a874cedc-goog
On Fri, 18 Mar 2022 10:13:12 +0800
David Gow <[email protected]> wrote:
> Does either (or both) of these features sound useful, and is this
> sort-of API the right model? (Personally, I think there's a reasonable
> scope for both.) Is anything obviously missing or wrong? Do the names,
> descriptions etc. make any sense?
Obviously I'm biased toward the ftrace solution ;-)
-- Steve
+Steve Muckle - since I think this might affect things he is working on.
On Thu, Mar 17, 2022 at 10:13 PM David Gow <[email protected]> wrote:
>
> When writing tests, it'd often be very useful to be able to intercept
> calls to a function in the code being tested and replace it with a
> test-specific stub. This has always been an obviously missing piece of
> KUnit, and the solutions always involve some tradeoffs with cleanliness,
> performance, or impact on non-test code. See the folowing document for
> some of the challenges:
> https://kunit.dev/mocking.html
>
> This series consists of two prototype patches which add support for this
> sort of redirection to KUnit tests:
>
> 1: static_stub: Any function which might want to be intercepted adds a
> call to a macro which checks if a test has redirected calls to it, and
> calls the corresponding replacement.
>
> 2: ftrace_stub: Functions are intercepted using ftrace and livepatch.
> This doesn't require adding a new prologue to each function being
> replaced, but does have more dependencies (which restricts it to a small
> number of architectures, not including UML), and doesn't work well with
> inline functions.
>
> The API for both implementations is very similar, so it should be easy
> to migrate from one to the other if necessary. Both of these
> implementations restrict the redirection to the test context: it is
> automatically undone after the KUnit test completes, and does not affect
> calls in other threads. If CONFIG_KUNIT is not enabled, there should be
> no overhead in either implementation.
>
> Does either (or both) of these features sound useful, and is this
> sort-of API the right model? (Personally, I think there's a reasonable
> scope for both.) Is anything obviously missing or wrong? Do the names,
> descriptions etc. make any sense?
>
> Note that these patches are definitely still at the "prototype" level,
> and things like error-handling, documentation, and testing are still
> pretty sparse. There is also quite a bit of room for optimisation.
> These'll all be improved for v1 if the concept seems good.
>
> Cheers,
> -- David
>
> Daniel Latypov (1):
> kunit: expose ftrace-based API for stubbing out functions during tests
>
> David Gow (1):
> kunit: Expose 'static stub' API to redirect functions
>
> include/kunit/ftrace_stub.h | 84 +++++++++++++++++
> include/kunit/static_stub.h | 106 +++++++++++++++++++++
> lib/kunit/Kconfig | 11 +++
> lib/kunit/Makefile | 5 +
> lib/kunit/ftrace_stub.c | 138 ++++++++++++++++++++++++++++
> lib/kunit/kunit-example-test.c | 64 +++++++++++++
> lib/kunit/static_stub.c | 125 +++++++++++++++++++++++++
> lib/kunit/stubs_example.kunitconfig | 11 +++
> 8 files changed, 544 insertions(+)
> create mode 100644 include/kunit/ftrace_stub.h
> create mode 100644 include/kunit/static_stub.h
> create mode 100644 lib/kunit/ftrace_stub.c
> create mode 100644 lib/kunit/static_stub.c
> create mode 100644 lib/kunit/stubs_example.kunitconfig
>
> --
> 2.35.1.894.gb6a874cedc-goog
>
On Fri, Mar 18, 2022 at 9:22 AM Steven Rostedt <[email protected]> wrote:
>
> On Fri, 18 Mar 2022 10:13:12 +0800
> David Gow <[email protected]> wrote:
>
> > Does either (or both) of these features sound useful, and is this
> > sort-of API the right model? (Personally, I think there's a reasonable
> > scope for both.) Is anything obviously missing or wrong? Do the names,
> > descriptions etc. make any sense?
>
> Obviously I'm biased toward the ftrace solution ;-)
Personally, I like providing both - as long as we can keep the
interface the same.
Ftrace is less visually invasive, but it is also less flexible in
capabilities, and requires substantial work to support on new
architectures.
+Joe Fradley who is also looking at KUnit with Android.
On 4/15/22 14:43, Steve Muckle wrote:
> On 4/4/22 13:13, Brendan Higgins wrote:
>> On Fri, Mar 18, 2022 at 9:22 AM Steven Rostedt <[email protected]>
>> wrote:
>>>
>>> On Fri, 18 Mar 2022 10:13:12 +0800
>>> David Gow <[email protected]> wrote:
>>>
>>>> Does either (or both) of these features sound useful, and is this
>>>> sort-of API the right model? (Personally, I think there's a reasonable
>>>> scope for both.) Is anything obviously missing or wrong? Do the names,
>>>> descriptions etc. make any sense?
>>>
>>> Obviously I'm biased toward the ftrace solution ;-)
>>
>> Personally, I like providing both - as long as we can keep the
>> interface the same.
>>
>> Ftrace is less visually invasive, but it is also less flexible in
>> capabilities, and requires substantial work to support on new
>> architectures.
>
> The general feature looks useful to me. I'm not sure the ftrace based
> API is worth it given it is only offering a visual improvement and has
> some drawbacks compared to the other implementation (won't work with
> inline functions, dependencies on other features). Livepatch is absent
> on arm64 which mostly rules it out for my purposes (Android Generic
> Kernel Image testing).
>
> cheers,
> Steve
On 4/4/22 13:13, Brendan Higgins wrote:
> On Fri, Mar 18, 2022 at 9:22 AM Steven Rostedt <[email protected]> wrote:
>>
>> On Fri, 18 Mar 2022 10:13:12 +0800
>> David Gow <[email protected]> wrote:
>>
>>> Does either (or both) of these features sound useful, and is this
>>> sort-of API the right model? (Personally, I think there's a reasonable
>>> scope for both.) Is anything obviously missing or wrong? Do the names,
>>> descriptions etc. make any sense?
>>
>> Obviously I'm biased toward the ftrace solution ;-)
>
> Personally, I like providing both - as long as we can keep the
> interface the same.
>
> Ftrace is less visually invasive, but it is also less flexible in
> capabilities, and requires substantial work to support on new
> architectures.
The general feature looks useful to me. I'm not sure the ftrace based
API is worth it given it is only offering a visual improvement and has
some drawbacks compared to the other implementation (won't work with
inline functions, dependencies on other features). Livepatch is absent
on arm64 which mostly rules it out for my purposes (Android Generic
Kernel Image testing).
cheers,
Steve