Re: [PATCH] tick/nohz: Update comments some more

[Frederic Weisbecker]

On Thu, Sep 28, 2023 at 11:07:01AM +0200, Ingo Molnar wrote:
> > + * infrastructure actually relies on the tick itself as a backend in
> > + * low-resolution mode (see hrtimer_run_queues()).
> > + *
> > + * This low-resolution handler still makes use of some hrtimer APIs meanwhile
> > + * for commodity with expiration calculation and forwarding.
> 
> commodity?

I meant 'convenience', my usual frenglish issues...

> 
> >   */
> >  static void tick_nohz_lowres_handler(struct clock_event_device *dev)
> >  {
> 
> As well-deserved penace for my nitpicking, I've fixed these on top of
> tip:timers/core, and have also done a full scan of kernel/time/tick-sched.c
> for spelling, consistency of style and readability of comments - see
> the patch below.
> 
> Thanks,
> 
> 	Ingo
> 
> ===========================>
> From: Ingo Molnar <mingo@xxxxxxxxxx>
> Date: Thu, 28 Sep 2023 10:45:54 +0200
> Subject: [PATCH] tick/nohz: Update comments some more
> 
> Inspired by recent enhancements to comments in kernel/time/tick-sched.c,
> go through the entire file and fix/unify its comments:
> 
>  - Fix over a dozen typos, spelling mistakes & cases of bad grammar.
> 
>  - Re-phrase sentences that I needed to read three times to understand.
> 
>     [ I used the following arbitrary rule-of-thumb:
>        - if I had to read a comment twice, it was usually my fault,
>        - if I had to read it a third time, it was the comment's fault. ]
> 
>  - Comma updates:
> 
>     - Add commas where needed
> 
>     - Remove commas where not needed
> 
>     - In cases where a comma is optional, choose one variant and try to
>       standardize it over similar sentences in the file.   
> 
>  - Standardize on standalone 'NOHZ' spelling in free-flowing comments:
> 
>       s/nohz/NOHZ
>       s/no idle tick/NOHZ
> 
>    Still keep 'dynticks' as a popular synonym.
> 
>  - Standardize on referring to variable names within free-flowing
>    comments with the "'var'" nomenclature, and function names as
>    "function_name()".
> 
>  - Standardize on '64-bit' and '32-bit':
>      s/32bit/32-bit
>      s/64bit/64-bit
> 
>  - Standardize on 'IRQ work':
>      s/irq work/IRQ work
> 
>  - A few other tidyups I probably missed to list.
> 
> No change in functionality intended - other than one small change to
> a syslog output string.
> 
> Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>

Looks good, thanks!