Search Linux Wireless

Re: [PATCH 08/18] wifi: mac80211: mesh: fix some kdoc warnings

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, 2023-09-29 at 08:48 +0200, Johannes Berg wrote:
> Hi,
> 
> > > +++ b/net/mac80211/mesh.c
> > > @@ -56,6 +56,8 @@ static void ieee80211_mesh_housekeeping_timer(struct timer_list *t)
> > >    *
> > >    * This function checks if the mesh configuration of a mesh point matches the
> > >    * local mesh configuration, i.e. if both nodes belong to the same mesh network.
> > > + *
> > > + * Returns: %true if both nodes belong to the same mesh
> > 
> > I thought kdoc used Return: not Returns:
> > 
> > <https://static.lwn.net/kerneldoc/doc-guide/kernel-doc.html#function-documentation>
> 
> Huh, indeed.
> 
> I'm sure this actually fixed it though because we did start running
> kernel-doc on this code internally (via W=1), the issue at hand was that
> we used to compile without mesh (not supported by iwlwifi) and now added
> it for hwsim testing.
> 
> So ... I looked, and indeed _both_ are accepted:

Hah, I suspect I used "Returns:" here because the file already used it
in other places. But, I am also used to "Returns:" from gtk-doc, so it
might also have felt natural to me because of that.

My hunch is, that it really doesn't matter and it can be left as-is for
now. But it is also obviously true that it may be surprising and the
extra consistency would help. So, a follow up patch to make everything
consistent seems reasonable to me.

Benjamin

> scripts/kernel-doc:
> 
>         } elsif ($newsection =~ m/^returns?$/i) {
>             $newsection = $section_return;
>         } elsif ($newsection =~ m/^\@return$/) {
>             # special: @return is a section, not a param description
>             $newsection = $section_return;
> 
> 
> which I guess means
> 
>  /**
>   * foo - ...
>   *
>   * Returns: bar
>   */
> 
> and
> 
>  /**
>   * foo - ...
>   *
>   * Return: bar
>   */
> 
> and
> 
>  /**
>   * foo - ...
>   *
>   * @return
>   * bar
>   */
> 
> are all accepted.
> 
> 
> Though "Return:" is more common, perhaps due to the docs:
> 
> $ git grep ' \* Returns:' -- | wc -l
> 3066
> $ git grep ' \* Return:' -- | wc -l
> 9189
> 
> 
> Personally, I find "Returns: %true if both nodes ..." to be nicer to
> read, since it's more like a sentence?
> 
> But then again it depends on whether you're reading it as a description
> or an instruction, I guess. "Returns: something" is a description, and
> "Return: something" is an instruction (for the function). Hmm.
> 
> I'm inclined to say I like the descriptive text better, but since it
> shows up in the generated documentation with "Return" anyway, maybe it's
> better to have it in the same way in the code?
> 
> Though of course it's (slightly) easier to just apply it as is :)
> 
> What do you (and others!) think? Does it matter?
> 
> johannes

Intel Deutschland GmbH
Registered Address: Am Campeon 10, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de <http://www.intel.de>
Managing Directors: Christin Eisenschmid, Sharon Heck, Tiffany Doon Silva  
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928




[Index of Archives]     [Linux Host AP]     [ATH6KL]     [Linux Wireless Personal Area Network]     [Linux Bluetooth]     [Wireless Regulations]     [Linux Netdev]     [Kernel Newbies]     [Linux Kernel]     [IDE]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite Hiking]     [MIPS Linux]     [ARM Linux]     [Linux RAID]

  Powered by Linux