Re: Re: What is the best way to compare an unsigned and a constant?

[Brendan Higgins <brendanhiggins@google.com>]

Sorry for the delay, I was on vacation. (I still am, but I was too ;-).)

On Tue, Jan 7, 2020 at 3:52 AM SeongJae Park <sjpark@amazon.com> wrote:
>
> On   Fri, 27 Dec 2019 13:52:27 +0100   Bernd Petrovitsch <bernd@petrovitsch.priv.at> wrote:
>
> > This is a multi-part message in MIME format.
> > --------------D98A0A31D62B0BC2939BAEE9
> > Content-Type: text/plain; charset=utf-8
> > Content-Transfer-Encoding: quoted-printable
> >
> > Hi all!
> >
> > On 27/12/2019 13:39, SeongJae Park wrote:
> > [...]
> > > I have a function returning 'unsigned long', and would like to write a =
> > kunit
> > > test for the function, as below.
> > >=20
> > >     unsigned long foo(void)
> > >     {
> > >             return 42;
> > >     }
> > >=20
> > >     static void foo_test(struct kunit *test)
> > >     {
> > >         KUNIT_EXPECT_EQ(test, 42, foo());
> > >     }
> >
> > For this case: shouldn't=20
> > ----  snip  ----
> > static void foo_test(struct kunit *test)
> > {
> >      KUNIT_EXPECT_EQ(test, 42ul, foo());
> > }
> > ----  snip  ----
> > do the trick?
>
> Unfortunately, it doesn't works.
>
>     [13:04:58] Building KUnit Kernel ...
>     In file included from /.../linux/include/linux/list.h:9:0,
>                      from /.../linux/include/linux/wait.h:7,
>                      from /.../linux/include/linux/wait_bit.h:8,
>                      from /.../linux/include/linux/fs.h:6,
>                      from /.../linux/include/linux/debugfs.h:15,
>                      from /.../linux/mm/damon.c:12:
>     /.../linux/mm/damon-test.h: In function ‘damon_test_foo’:
>     /.../linux/include/linux/kernel.h:842:29: warning: comparison of distinct pointer types lacks a cast
>        (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
>                                  ^
>     /.../linux/include/kunit/test.h:493:9: note: in expansion of macro ‘__typecheck’
>       ((void)__typecheck(__left, __right));           \
>              ^~~~~~~~~~~
>     /.../linux/include/kunit/test.h:517:2: note: in expansion of macro ‘KUNIT_BASE_BINARY_ASSERTION’
>       KUNIT_BASE_BINARY_ASSERTION(test,           \
>       ^~~~~~~~~~~~~~~~~~~~~~~~~~~
>     /.../linux/include/kunit/test.h:606:2: note: in expansion of macro ‘KUNIT_BASE_EQ_MSG_ASSERTION’
>       KUNIT_BASE_EQ_MSG_ASSERTION(test,           \
>       ^~~~~~~~~~~~~~~~~~~~~~~~~~~
>     /.../linux/include/kunit/test.h:616:2: note: in expansion of macro ‘KUNIT_BINARY_EQ_MSG_ASSERTION’
>       KUNIT_BINARY_EQ_MSG_ASSERTION(test,           \
>       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>     /.../linux/include/kunit/test.h:979:2: note: in expansion of macro ‘KUNIT_BINARY_EQ_ASSERTION’
>       KUNIT_BINARY_EQ_ASSERTION(test, KUNIT_EXPECTATION, left, right)
>       ^~~~~~~~~~~~~~~~~~~~~~~~~
>     /.../linux/mm/damon-test.h:565:2: note: in expansion of macro ‘KUNIT_EXPECT_EQ’
>       KUNIT_EXPECT_EQ(test, 42ul, (int)foo());
>       ^~~~~~~~~~~~~~~

Isn't the issue here that you fixed the 42, but are now casting the
result of foo() to an int?

Or have you fixed that now too?

Worst case (gross) scenario, you could just cast 42 to whatever type
foo() returns.

> Some other thoughts?
>
>
> Thanks,
> SeongJae Park
>
>
> >
> > MfG,
> >       Bernd
> > --=20
> > "I dislike type abstraction if it has no real reason. And saving
> > on typing is not a good reason - if your typing speed is the main
> > issue when you're coding, you're doing something seriously wrong."
> >     - Linus Torvalds