linux-toradex.git/lib/kunit, branch v6.11-rc3

kunit: executor: Simplify string allocation handling

2024-07-12T16:11:48+00:00

The alloc/copy code pattern is better consolidated to single kstrdup (and
kstrndup) calls instead. This gets rid of deprecated[1] strncpy() uses as
well. Replace one other strncpy() use with the more idiomatic strscpy().

Link: https://github.com/KSPP/linux/issues/90 [1]
Reviewed-by: David Gow 
Signed-off-by: Kees Cook 
Signed-off-by: Shuah Khan

kunit/usercopy: Disable testing on !CONFIG_MMU

2024-07-02T16:11:40+00:00

Since arch_pick_mmap_layout() is an inline for non-MMU systems, disable
this test there.

Reported-by: kernel test robot 
Closes: https://lore.kernel.org/oe-kbuild-all/202406160505.uBge6TMY-lkp@intel.com/
Signed-off-by: Kees Cook 
Reviewed-by: Rae Moar 
Reviewed-by: David Gow 
Signed-off-by: Shuah Khan

kunit: test: Add vm_mmap() allocation resource manager

2024-06-15T01:31:33+00:00

For tests that need to allocate using vm_mmap() (e.g. usercopy and
execve), provide the interface to have the allocation tracked by KUnit
itself. This requires bringing up a placeholder userspace mm.

This combines my earlier attempt at this with Mark Rutland's version[1].

Normally alloc_mm() and arch_pick_mmap_layout() aren't exported for
modules, so export these only for KUnit testing.

Link: https://lore.kernel.org/lkml/20230321122514.1743889-2-mark.rutland@arm.com/ [1]
Co-developed-by: Mark Rutland 
Signed-off-by: Mark Rutland 
Reviewed-by: David Gow 
Signed-off-by: Kees Cook 
Signed-off-by: Shuah Khan

kunit: add missing MODULE_DESCRIPTION() macros to core modules

2024-06-07T21:59:05+00:00

make allmodconfig && make W=1 C=1 reports in lib/kunit:
WARNING: modpost: missing MODULE_DESCRIPTION() in lib/kunit/kunit.o
WARNING: modpost: missing MODULE_DESCRIPTION() in lib/kunit/kunit-test.o
WARNING: modpost: missing MODULE_DESCRIPTION() in lib/kunit/kunit-example-test.o

Add the missing invocations of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson 
Reviewed-by: Rae Moar 
Reviewed-by: David Gow 
Signed-off-by: Shuah Khan

kunit: Cover 'assert.c' with tests

2024-05-30T18:53:47+00:00

There are multiple assertion formatting functions in the `assert.c`
file, which are not covered with tests yet. Implement the KUnit test
for these functions.

The test consists of 11 test cases for the following functions:

1) 'is_literal'
2) 'is_str_literal'
3) 'kunit_assert_prologue', test case for multiple assert types
4) 'kunit_assert_print_msg'
5) 'kunit_unary_assert_format'
6) 'kunit_ptr_not_err_assert_format'
7) 'kunit_binary_assert_format'
8) 'kunit_binary_ptr_assert_format'
9) 'kunit_binary_str_assert_format'
10) 'kunit_assert_hexdump'
11) 'kunit_mem_assert_format'

The test aims at maximizing the branch coverage for the assertion
formatting functions.

As you can see, it covers some of the static helper functions as
well, so mark the static functions in `assert.c` as 'VISIBLE_IF_KUNIT'
and conditionally export them with EXPORT_SYMBOL_IF_KUNIT. Add the
corresponding definitions to `assert.h`.

Build the assert test when CONFIG_KUNIT_TEST is enabled, similar to
how it is done for the string stream test.

Signed-off-by: Ivan Orlov 
Reviewed-by: Rae Moar 
Acked-by: David Gow 
Signed-off-by: Shuah Khan

kunit: bail out early in __kunit_test_suites_init() if there are no suites to test

2024-05-06T20:22:02+00:00

Commit c72a870926c2 added a mutex to prevent kunit tests from running
concurrently.  Unfortunately that mutex gets locked during module load
regardless of whether the module actually has any kunit tests.  This
causes a problem for kunit tests that might need to load other kernel
modules (e.g. gss_krb5_test loading the camellia module).

So check to see if there are actually any tests to run before locking
the kunit_run_lock mutex.

Fixes: c72a870926c2 ("kunit: add ability to run tests after boot using debugfs")
Reported-by: Nico Pache 
Signed-off-by: Scott Mayhew 
Reviewed-by: Rae Moar 
Reviewed-by: David Gow 
Signed-off-by: Shuah Khan

kunit: string-stream-test: use KUNIT_DEFINE_ACTION_WRAPPER

2024-05-06T20:22:02+00:00

Use KUNIT_DEFINE_ACTION_WRAPPER macro to define the 'kfree' and
'string_stream_destroy' wrappers for kunit_add_action.

Signed-off-by: Ivan Orlov 
Reviewed-by: Rae Moar 
Acked-by: David Gow 
Signed-off-by: Shuah Khan

kunit: test: Move fault tests behind KUNIT_FAULT_TEST Kconfig option

2024-05-06T20:22:02+00:00

The NULL dereference tests in kunit_fault deliberately trigger a kernel
BUG(), and therefore print the associated stack trace, even when the
test passes. This is both annoying (as it bloats the test output), and
can confuse some test harnesses, which assume any BUG() is a failure.

Allow these tests to be specifically disabled (without disabling all
of KUnit's other tests), by placing them behind the
CONFIG_KUNIT_FAULT_TEST Kconfig option. This is enabled by default, but
can be set to 'n' to disable the test. An empty 'kunit_fault' suite is
left behind, which will automatically be marked 'skipped'.

As the fault tests already were disabled under UML (as they weren't
compatible with its fault handling), we can simply adapt those
conditions, and add a dependency on !UML for our new option.

Suggested-by: Guenter Roeck 
Link: https://lore.kernel.org/all/928249cc-e027-4f7f-b43f-502f99a1ea63@roeck-us.net/
Fixes: 82b0beff3497 ("kunit: Add tests for fault")
Signed-off-by: David Gow 
Reviewed-by: Mickaël Salaün 
Reviewed-by: Rae Moar 
Signed-off-by: Shuah Khan

kunit: unregister the device on error

2024-05-06T20:22:02+00:00

kunit_init_device() should unregister the device on bus register error,
but mistakenly it tries to unregister the bus.

Unregister the device instead of the bus.

Signed-off-by: Wander Lairson Costa 
Fixes: d03c720e03bd ("kunit: Add APIs for managing devices")
Reviewed-by: Greg Kroah-Hartman 
Signed-off-by: Shuah Khan

kunit: Fix race condition in try-catch completion

2024-05-06T20:22:02+00:00

KUnit's try-catch infrastructure now uses vfork_done, which is always
set to a valid completion when a kthread is created, but which is set to
NULL once the thread terminates. This creates a race condition, where
the kthread exits before we can wait on it.

Keep a copy of vfork_done, which is taken before we wake_up_process()
and so valid, and wait on that instead.

Fixes: 93533996100c ("kunit: Handle test faults")
Reported-by: Linux Kernel Functional Testing 
Closes: https://lore.kernel.org/lkml/20240410102710.35911-1-naresh.kamboju@linaro.org/
Tested-by: Linux Kernel Functional Testing 
Acked-by: Mickaël Salaün 
Signed-off-by: David Gow 
Reviewed-by: Rae Moar 
Tested-by: Miguel Ojeda 
Signed-off-by: Shuah Khan