Re: [PATCH] selftests/mount_setattr: fix idmap_mount_tree_invalid failed to run

[Shuah Khan <skhan@linuxfoundation.org>]

On 10/25/24 02:08, zhouyuhang wrote:
> 
> 
> 在 2024/10/24 22:26, Shuah Khan 写道:
>> On 10/24/24 03:50, zhouyuhang wrote:
>>> From: zhouyuhang <zhouyuhang@kylinos.cn>
>>>
>>> Test case idmap_mount_tree_invalid failed to run on the newer kernel
>>> with the following output:
>>>
>>>   #  RUN mount_setattr_idmapped.idmap_mount_tree_invalid ...
>>>   # mount_setattr_test.c:1428:idmap_mount_tree_invalid:Expected sys_mount_setattr(open_tree_fd, "", AT_EMPTY_PATH, &attr, sizeof(attr)) (0) ! = 0 (0)
>>>   # idmap_mount_tree_invalid: Test terminated by assertion
>>>
>>> This is because tmpfs is mounted at "/mnt/A", and tmpfs already
>>> contains the flag FS_ALLOW_IDMAP after the commit 7a80e5b8c6fa ("shmem:
>>> support idmapped mounts for tmpfs"). So calling sys_mount_setattr here
>>> returns 0 instead of -EINVAL as expected.
>>>
>>> Ramfs is mounted at "/mnt/B" and does not support idmap mounts.
>>> So we can use "/mnt/B" instead of "/mnt/A" to make the test run
>>> successfully with the following output:
>>>
>>>   # Starting 1 tests from 1 test cases.
>>>   #  RUN mount_setattr_idmapped.idmap_mount_tree_invalid ...
>>>   #            OK mount_setattr_idmapped.idmap_mount_tree_invalid
>>>   ok 1 mount_setattr_idmapped.idmap_mount_tree_invalid
>>>   # PASSED: 1 / 1 tests passed.
>>>
>>
>> Sounds like this code is testing this very condition passing
>> in invalid mount to see what happens. If that is the intent
>> this patch is incorrect.
>>
> 
> I think I probably understand what you mean, what you're saying is that the output of this line of errors is the condition,
> and the main purpose of the test case is to see what happens when it invalid mount. But it's valid now, isn't it?
> So we need to fix it. I don't think that constructing this error with ramfs will have any impact on the code that follows.
> If you feel that using "/mnt/B" is unreliable, I think we can temporarily mount ramfs to "/mnt/A" here and continue using "/mnt/A".
> Do you think this is feasible? Looking forward to your reply, thank you.
> 

What I am saying is if this test is intended to test invalid mounts, passing
"/mnt/A" makes perfect sense.

>>> Signed-off-by: zhouyuhang <zhouyuhang@kylinos.cn>
>>> ---
>>>   tools/testing/selftests/mount_setattr/mount_setattr_test.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/tools/testing/selftests/mount_setattr/mount_setattr_test.c b/tools/testing/selftests/mount_setattr/mount_setattr_test.c
>>> index c6a8c732b802..54552c19bc24 100644
>>> --- a/tools/testing/selftests/mount_setattr/mount_setattr_test.c
>>> +++ b/tools/testing/selftests/mount_setattr/mount_setattr_test.c
>>> @@ -1414,7 +1414,7 @@ TEST_F(mount_setattr_idmapped, idmap_mount_tree_invalid)
>>>       ASSERT_EQ(expected_uid_gid(-EBADF, "/tmp/B/b", 0, 0, 0), 0);
>>>       ASSERT_EQ(expected_uid_gid(-EBADF, "/tmp/B/BB/b", 0, 0, 0), 0);
>>>   -    open_tree_fd = sys_open_tree(-EBADF, "/mnt/A",
>>> +    open_tree_fd = sys_open_tree(-EBADF, "/mnt/B",
>>>                        AT_RECURSIVE |
>>>                        AT_EMPTY_PATH |
>>>                        AT_NO_AUTOMOUNT |
>>
>> thanks,
>> -- Shuah
> 

thanks,
-- Shuah