Re: [PATCH] Documentation: workqueue: Inline WQ_* flag pattern

[Akira Yokosawa]

[+Cc: Jon]

On Thu, 18 Apr 2024 14:21:01 +0700, Bagas Sanjaya wrote:
> Sphinx reports htmldocs warning:
> 
> Documentation/core-api/workqueue:776: ./include/linux/workqueue.h:493: WARNING: Inline literal start-string without end-string.
> 
> The warning is caused by unescaped wildcard in the pattern that matches
> all workqueue flags (WQ_*). The pattern is rendered as normal text
> (with surrounding double backticks due to @argument syntax usage)
> instead of inline text like in individual flag names as a result.
> 
> Inline the pattern to fix the warning.
> 
> Fixes: 5797b1c18919 ("workqueue: Implement system-wide nr_active enforcement for unbound workqueues")
> Signed-off-by: Bagas Sanjaya <bagasdotme@xxxxxxxxx>
> ---
>  include/linux/workqueue.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
> index 158784dd189ab2..c9fec74e130459 100644
> --- a/include/linux/workqueue.h
> +++ b/include/linux/workqueue.h
> @@ -463,7 +463,7 @@ void workqueue_softirq_dead(unsigned int cpu);
>  /**
>   * alloc_workqueue - allocate a workqueue
>   * @fmt: printf format for the name of the workqueue
> - * @flags: WQ_* flags
> + * @flags: ``WQ_*`` flags
>   * @max_active: max in-flight work items, 0 for default
>   * remaining args: args for @fmt
>   *
> @@ -490,7 +490,7 @@ void workqueue_softirq_dead(unsigned int cpu);
>   * min_active which is set to min(@max_active, %WQ_DFL_MIN_ACTIVE). This means
>   * that the sum of per-node max_active's may be larger than @max_active.
>   *
> - * For detailed information on %WQ_* flags, please refer to
> + * For detailed information on ``WQ_*`` flags, please refer to
>   * Documentation/core-api/workqueue.rst.
>   *
>   * RETURNS:

In my opinion, reST-specific clutters like these should be avoided in
kernel-doc comments as far as possible.

Instead, I think you can educate kernel-doc (script) so that "*" is
allowed in the %CONSTANT pattern, meaning %WQ_* can be converted
to ``WQ_*`` in reST.

For similar changes made against the @param pattern, see commits
69fc23efc7e5 ("kernel-doc: Add unary operator * to $type_param_ref")
and 8aaf297a0dd6 ("docs: scripts: kernel-doc: accept bitwise negation
like ~@var").

I guess it is $type_constant2 that needs a tweak in this case.

HTH,
Akira