Re: [PATCH v2 0/3] decoupling the ext4 test module

["yebin (H)" <yebin10@huawei.com>]

On 2026/3/15 14:44, Ojaswin Mujoo wrote:
> On Sat, Mar 14, 2026 at 03:52:55PM +0800, Ye Bin wrote:
>> From: Ye Bin <yebin10@huawei.com>
>>
>> This patchset is split out from the "Fix some issues about ext4-test"
>> patchset. It decouples mballoc-test.c and extents-test.c from the
>> ext4 module, resolving the issue where these two tests could not be run
>> when EXT4_KUNIT_TESTS is set to M.
>>
>> Ye Bin (3):
>>    ext4: introduce EXPORT_SYMBOL_FOR_EXT4_TEST() helper
>>    ext4: fix mballoc-test.c is not compiled when EXT4_KUNIT_TESTS=M
>>    ext4: fix extents-test.c is not compiled when EXT4_KUNIT_TESTS=M
>
> Hi Ye,
>
> So thinking about this a bit more, is there any reason we are not using
> the VISIBLE_IF_KUNIT and EXPORT_SYMBOL_IF_KUNIT macros directly?  Is
> there any case that they don't handle?
>
A limitation of directly using the VISIBLE_IF_KUNIT and
EXPORT_SYMBOL_IF_KUNIT macros is that they rely entirely on the
CONFIG_KUNIT configuration being enabled. However, enabling
CONFIG_KUNIT does not necessarily mean that EXT4_KUNIT_TESTS is
enabled. If EXT4_KUNIT_TESTS is not enabled, there is no need to export
the symbols.
> [1] https://docs.kernel.org/dev-tools/kunit/usage.html#testing-static-functions
>>
>>   fs/ext4/Makefile       |   5 +-
>>   fs/ext4/ext4.h         |   5 ++
>>   fs/ext4/ext4_extents.h |  12 +++++
>>   fs/ext4/extents-test.c |   8 ++--
>>   fs/ext4/extents.c      |  39 +++++++++++++---
>>   fs/ext4/mballoc-test.c |  81 ++++++++++++++++----------------
>>   fs/ext4/mballoc.c      | 102 +++++++++++++++++++++++++++++++++++++++--
>>   fs/ext4/mballoc.h      |  30 ++++++++++++
>>   8 files changed, 227 insertions(+), 55 deletions(-)
>>
>> --
>> 2.34.1
>>
>
>
> .
>