Re: [PATCH v2] Docs: usb: update struct usb_driver, __init and __exit

[Philipp Hortmann <philipp.g.hortmann@gmail.com>]

On 10/21/21 10:00 AM, Greg KH wrote:
> On Wed, Oct 20, 2021 at 10:14:46PM +0200, Philipp Hortmann wrote:
>> update struct usb_driver from usb-skeleton.c.
>> update __init and __exit functions that are moved from
>> usb-skeleton.c to common used multi-stage macros.
>>
>> Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
>> ---
>> V1 -> V2: changed :c:func:`usb_register` to usb_register()
>>            changed the :c:func:`usb_deregister` to usb_deregister()
>>            used literal blocks for makro module_usb_driver and added one more
>>            stage of multi-stage macros.
>>
>>   .../driver-api/usb/writing_usb_driver.rst     | 70 ++++++++++---------
>>   1 file changed, 36 insertions(+), 34 deletions(-)
>>
>> diff --git a/Documentation/driver-api/usb/writing_usb_driver.rst b/Documentation/driver-api/usb/writing_usb_driver.rst
>> index 2176297e5765..12e0481cceae 100644
>> --- a/Documentation/driver-api/usb/writing_usb_driver.rst
>> +++ b/Documentation/driver-api/usb/writing_usb_driver.rst
>> @@ -54,12 +54,15 @@ information is passed to the USB subsystem in the :c:type:`usb_driver`
>>   structure. The skeleton driver declares a :c:type:`usb_driver` as::
>>   
>>       static struct usb_driver skel_driver = {
>> -	    .name        = "skeleton",
>> -	    .probe       = skel_probe,
>> -	    .disconnect  = skel_disconnect,
>> -	    .fops        = &skel_fops,
>> -	    .minor       = USB_SKEL_MINOR_BASE,
>> -	    .id_table    = skel_table,
>> +           .name        = "skeleton",
>> +           .probe       = skel_probe,
>> +           .disconnect  = skel_disconnect,
>> +           .suspend     = skel_suspend,
>> +           .resume      = skel_resume,
>> +           .pre_reset   = skel_pre_reset,
>> +           .post_reset  = skel_post_reset,
>> +           .id_table    = skel_table,
>> +           .supports_autosuspend = 1,
> 
> Why remove the tabs?  Is that needed here?
You are right will be changed.
> 
>>       };
>>   
>>   
>> @@ -81,36 +84,35 @@ this user-space interaction. The skeleton driver needs this kind of
>>   interface, so it provides a minor starting number and a pointer to its
>>   :c:type:`file_operations` functions.
>>   
>> -The USB driver is then registered with a call to :c:func:`usb_register`,
>> -usually in the driver's init function, as shown here::
>> -
>> -    static int __init usb_skel_init(void)
>> -    {
>> -	    int result;
>> -
>> -	    /* register this driver with the USB subsystem */
>> -	    result = usb_register(&skel_driver);
>> -	    if (result < 0) {
>> -		    err("usb_register failed for the "__FILE__ "driver."
>> -			"Error number %d", result);
>> -		    return -1;
>> -	    }
>> -
>> -	    return 0;
>> -    }
>> -    module_init(usb_skel_init);
>> -
>> +The USB driver is then registered with a call to usb_register()
>> +which is usually in the driver's init function. Since this functionality
>> +is usable with many USB drivers, it is hidden behind multi-stage macros.
> 
> I don't understand the need for the "multi-stage macros" term here.
I am not a native English speaker so "multi-stage macros" is just not a 
fitting wording. May be “staged macros” is better or something else…
> 
> And what functionality is referred to here by "this"?
The “this” is replacing the “init function” but when this is unclear I 
will change in a later proposal…
> 
> 
>> +While the first macros are USB specific the later macros are used in different
>> +subsystems. This removes a lot of boilerplate code.
> 
> What later macros?  Is that really needed to describe here?
I will improve wording...

> I think the above code example should remain, as it is good for learning
> and understanding, and maybe just add something that says "Or you can
> use the following macro to replace all of the above common code."
I understand the need for keeping the code examples. But I would like to 
inform the reader about the macros first.
> 
> 
>>   
>>   When the driver is unloaded from the system, it needs to deregister
>> -itself with the USB subsystem. This is done with the :c:func:`usb_deregister`
>> -function::
>> -
>> -    static void __exit usb_skel_exit(void)
>> -    {
>> -	    /* deregister this driver with the USB subsystem */
>> -	    usb_deregister(&skel_driver);
>> -    }
>> -    module_exit(usb_skel_exit);
>> +itself with the USB subsystem. This is done with usb_deregister()
>> +which is also hidden behind multi-stage macros.
>> +
>> +The init and exit functions are included in the macro module_usb_driver.
>> +Find the first three stages of macros below::
>> +
>> +    module_usb_driver(skel_driver);
>> +                         |
>> +                         V
>> +    module_driver(__usb_driver, usb_register, usb_deregister)
>> +                         |               \               \
>> +                         V                ----------      ----------
>> +    static int __init __driver##_init(void) \      |               |
>> +    { \                 v---------------------------               |
>> +            return __register(&(__driver) , ##__VA_ARGS__); \      |
>> +    } \                                                            |
>> +    module_init(__driver##_init); \                                |
>> +    static void __exit __driver##_exit(void) \                     |
>> +    { \            v------------------------------------------------
>> +            __unregister(&(__driver) , ##__VA_ARGS__); \
>> +    } \
>> +    module_exit(__driver##_exit);
> 
> As the one who wrote these macros, I can't really understand the
> ascii-art here, so I worry about anyone else :)
Code is just better readable, even when code uses more lines. Will be 
changed in next proposal.
> 
> Again, do not think trying to show an implementation detail like this is
> needed.
The big question for me is for whom is this document written? For the 
USB subsystem maintainer that has even written the code by himself? I 
guess not, but may be I am wrong. Or for the kernel newbies like me? 
Please consider that the changed lines are may be not so much of use for 
me anymore as I am in the details.

When I saw the __init and __exit code example first, I was very happy to 
see it and then I was searching in the code for it. I did not find 
“init” and “exit” and was very frustrated. I want to help others to get 
into this example more smoothly.
> 
> thanks,
> 
> greg k-h
> 
Thanks that you replied at all.
Thanks for your very fast reply.

Kernelnewbie Philipp G. Hortmann