[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-id: <53AB31C3.1080309@samsung.com>
Date: Wed, 25 Jun 2014 14:32:03 -0600
From: Shuah Khan <shuah.kh@...sung.com>
To: Davidlohr Bueso <davidlohr@...com>
Cc: gregkh@...uxfoundation.org, akpm@...ux-foundation.org,
colin.king@...onical.com, linux-kernel@...r.kernel.org,
Shuah Khan <shuah.kh@...sung.com>
Subject: Re: [PATCH] tools: msgque improve error handling when not running as
root
On 06/25/2014 02:22 PM, Davidlohr Bueso wrote:
> On Wed, 2014-06-25 at 14:15 -0600, Shuah Khan wrote:
>> On 06/25/2014 02:05 PM, Davidlohr Bueso wrote:
>>> On Wed, 2014-06-25 at 13:40 -0600, Shuah Khan wrote:
>>>> The test fails in the middle when it is not run as root while
>>>> accessing /proc/sys/kernel/msg_next_id. Changed it to check
>>>> for root at the beginning of the test. Also added messages
>>>> to indicate start and test pass status.
>>>>
>>>> Signed-off-by: Shuah Khan <shuah.kh@...sung.com>
>>>> ---
>>>> tools/testing/selftests/ipc/msgque.c | 7 +++++++
>>>> 1 file changed, 7 insertions(+)
>>>>
>>>> diff --git a/tools/testing/selftests/ipc/msgque.c b/tools/testing/selftests/ipc/msgque.c
>>>> index aa290c0..30c627a 100644
>>>> --- a/tools/testing/selftests/ipc/msgque.c
>>>> +++ b/tools/testing/selftests/ipc/msgque.c
>>>> @@ -193,6 +193,12 @@ int main(int argc, char **argv)
>>>> int msg, pid, err;
>>>> struct msgque_data msgque;
>>>>
>>>> + if (getuid() != 0) {
>>>> + fprintf(stderr, "Please run the test as root - Exiting.\n");
>>>> + exit(1);
>>>> + }
>>>
>>> Please use err(3).
>>
>> I will make this printf to be consistent with the rest of the
>> printfs in this file.
>
> Can you please update the file then. err(EXIT_FAILURE, ...) is a lot
> nicer than fprintf+exit combos all over the place.
That would make sense as a separate patch and not as part of fix for
this problem.
>
>>
>>>
>>>> + printf("Starting ipc tests:\n");
>>>
>>> This is unnecessary and adds nothing to the program.
>>
>> The reason for doing this is it becomes easier to parse when this test
>> is run as part of run_tests at the top level:
>
> Parsing reasons should not determine such things. What if I add shm, or
> semaphore tests to the ipc dir?? Then the text would be bogus. What if
> not all tests pass? There is no way of determining a percentage of
> completeness. Please do not add such things for a simple msg queue
> program as it doesn't necessarily represent ipc selftests.
>
I can drop these messages. However, I do think it would make sense
to add a summary of tests passed. Similar to what mqueue does for
example.
-- Shuah
--
Shuah Khan
Senior Linux Kernel Developer - Open Source Group
Samsung Research America(Silicon Valley)
shuah.kh@...sung.com | (970) 672-0658
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists