From: Zebediah Figura <z.figura12@gmail.com>
Subject: Re: [PATCH] ntoskrnl.exe/tests: Use NULL instead of casting 0
Message-Id: <e6f39a14-7309-862c-be61-afec6a076dce@gmail.com>
Date: Fri, 7 Dec 2018 15:43:08 -0600
In-Reply-To: <003c5bc7-364a-41e2-91e6-31e1c52ef6d0@winehq.org>
References: <20181207205717.11320-2-mstefani@winehq.org> <95aed620-34c9-6ac3-d594-6c9d889f3601@gmail.com> <003c5bc7-364a-41e2-91e6-31e1c52ef6d0@winehq.org>

On 12/07/2018 03:40 PM, Michael Stefaniuc wrote:
> On 12/7/18 10:03 PM, Zebediah Figura wrote:
>> On 12/07/2018 02:57 PM, Michael Stefaniuc wrote:
>>> Signed-off-by: Michael Stefaniuc <mstefani@winehq.org>
>>> ---
>>>    dlls/ntoskrnl.exe/tests/driver.c | 4 ++--
>>>    1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/dlls/ntoskrnl.exe/tests/driver.c
>>> b/dlls/ntoskrnl.exe/tests/driver.c
>>> index 17e54cd508..8f41d39747 100644
>>> --- a/dlls/ntoskrnl.exe/tests/driver.c
>>> +++ b/dlls/ntoskrnl.exe/tests/driver.c
>>> @@ -435,7 +435,7 @@ static void test_sync(void)
>>>            ok(ret == i - 9, "expected %d, got %d\n", i - 9, ret);
>>>        }
>>>    -    run_thread(mutex_thread, (void *)0);
>>> +    run_thread(mutex_thread, NULL);
>>>          ret = wait_single(&test_mutex, 0);
>>>        ok(ret == 0, "got %#x\n", ret);
>>> @@ -445,7 +445,7 @@ static void test_sync(void)
>>>        ret = KeReleaseMutex(&test_mutex, 0);
>>>        ok(ret == 0, "got %#x\n", ret);
>>>    -    run_thread(mutex_thread, (void *)0);
>>> +    run_thread(mutex_thread, NULL);
>>>          /* test timers */
>>>        KeInitializeTimerEx(&timer, NotificationTimer);
>>>
>>
>> I'd argue this should be left as it is; the test is casting an NTSTATUS
>> value to the (void *) used as the thread parameter.
> If it is a NTSTATUS shouldn't you use STATUS_SUCCESS instead of 0 then?
> 

Well, maybe, but given that KeWaitForMultipleObjects returns an index, 
this isn't clearly better.