Re: [PATCH 24/26] locking/atomic: scripts: generate kerneldoc comments

[Akira Yokosawa <akiyks@gmail.com>]

Hi Randy,

On 2023/05/26 13:51, Randy Dunlap wrote:
> Hi Akira,
> 
> On 5/25/23 20:17, Akira Yokosawa wrote:
>> On Wed, 24 May 2023 16:11:52 +0200, Peter Zijlstra wrote:
>>> On Wed, May 24, 2023 at 11:03:58PM +0900, Akira Yokosawa wrote:
>>>
>>>>> * All ops are described as an expression using their usual C operator.
>>>>>   For example:
>>>>>
>>>>>   andnot: "Atomically updates @v to (@v & ~@i)"
>>>>
>>>> The kernel-doc script converts "~@i" into reST source of "~**i**",
>>>> where the emphasis of i is not recognized by Sphinx.
>>>>
>>>> For the "@" to work as expected, please say "~(@i)" or "~ @i".
>>>> My preference is the former.
>>>
>>> And here we start :-/ making the actual comment less readable because
>>> retarded tooling.
>>>
>>>>>   inc:    "Atomically updates @v to (@v + 1)"
>>>>>
>>>>>   Which may be clearer to non-naative English speakers, and allows all
>>>>                             non-native
>>>>
>>>>>   the operations to be described in the same style.
>>>>>
>>>>> * All conditional ops have their condition described as an expression
>>>>>   using the usual C operators. For example:
>>>>>
>>>>>   add_unless: "If (@v != @u), atomically updates @v to (@v + @i)"
>>>>>   cmpxchg:    "If (@v == @old), atomically updates @v to @new"
>>>>>
>>>>>   Which may be clearer to non-naative English speakers, and allows all
>>>>
>>>> Ditto.
>>>
>>> How about we just keep it as is, and all the rst and html weenies learn
>>> to use a text editor to read code comments?
>>
>> :-) :-) :-)
>>
>> It turns out that kernel-doc is aware of !@var [1].
>> Similar tricks can be added for ~@var.
>> So let's keep it as is!
>>
>> I'll ask documentation forks for updating kernel-doc when this change
>> is merged eventually.
> 
> What do you mean by that?
> What needs to be updated and how?
 
I mean, scripts/kernel-doc needs to be updated so that "~@var"
is converted into "**~var**".

I think adding "~" to the substitution pattern added in [1] as follows
should do the trick (not well tested):

diff --git a/scripts/kernel-doc b/scripts/kernel-doc
index 2486689ffc7b..eb70c1fd4e86 100755
--- a/scripts/kernel-doc
+++ b/scripts/kernel-doc
@@ -64,7 +64,7 @@ my $type_constant = '\b``([^\`]+)``\b';
 my $type_constant2 = '\%([-_\w]+)';
 my $type_func = '(\w+)\(\)';
 my $type_param = '\@(\w*((\.\w+)|(->\w+))*(\.\.\.)?)';
-my $type_param_ref = '([\!]?)\@(\w*((\.\w+)|(->\w+))*(\.\.\.)?)';
+my $type_param_ref = '([\!~]?)\@(\w*((\.\w+)|(->\w+))*(\.\.\.)?)';
 my $type_fp_param = '\@(\w+)\(\)';  # Special RST handling for func ptr params
 my $type_fp_param2 = '\@(\w+->\S+)\(\)';  # Special RST handling for structs with func ptr params
 my $type_env = '(\$\w+)';

Thoughts?

        Thanks, Akira

> 
> 
>> [1]: ee2aa7590398 ("scripts: kernel-doc: accept negation like !@var")
> 
> thanks.