linux-kernel - Re: [PATCH 0/2] sched: Refine scheduler naming for clarity and specificity

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <EBFD5AF7-8B9D-480E-9A01-74898E4EBC60@gmail.com>
Date: Thu, 20 Feb 2025 14:48:41 +0800
From: Jemmy Wong <jemmywong512@...il.com>
To: Steven Rostedt <rostedt@...dmis.org>
Cc: Jemmy <jemmywong512@...il.com>,
 Masami Hiramatsu <mhiramat@...nel.org>,
 Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
 Mark Rutland <mark.rutland@....com>,
 Jonathan Corbet <corbet@....net>,
 Ingo Molnar <mingo@...hat.com>,
 Peter Zijlstra <peterz@...radead.org>,
 Juri Lelli <juri.lelli@...hat.com>,
 Vincent Guittot <vincent.guittot@...aro.org>,
 Dietmar Eggemann <dietmar.eggemann@....com>,
 Ben Segall <bsegall@...gle.com>,
 Mel Gorman <mgorman@...e.de>,
 Valentin Schneider <vschneid@...hat.com>,
 linux-kernel@...r.kernel.org,
 linux-trace-kernel@...r.kernel.org,
 linux-doc@...r.kernel.org
Subject: Re: [PATCH 0/2] sched: Refine scheduler naming for clarity and
 specificity

Thanks for filling me in on the background of the code status
and the potential consequences of these changes. 
I really appreciate the context! 

As Phil Karlton once said:
There are only two hard things in Computer Science: 
cache invalidation and naming things. 

A good name makes code easier to read and understand.
I personally value code readability and maintainability. 
However, based on your input and the community’s opinion,
I agree that these changes could introduce indirection 
for git blame and make maintenance even harder. 

So, I’ll discard the changes.

BR,
Jemmy  

> On Feb 20, 2025, at 6:24 AM, Steven Rostedt <rostedt@...dmis.org> wrote:
> 
> On Thu, 20 Feb 2025 02:20:18 +0800
> Jemmy Wong <jemmywong512@...il.com> wrote:
> 
>> Hi, I'm relatively new to Linux and eager to contribute to the community
>> with some foundational work.
> 
> Welcome to the community.
> 
>> 
>> I aim to help improve code readability and maintainability.
>> While reading the scheduler code, I found some naming inconsistencies
>> that made understanding the code more difficult.
> 
> I do plan on updating the scheduler comments soon, so that every function
> has a purpose. But there's a lot of history to the scheduler code that you
> really can't just go and rename items. It causes a lot of churn which also
> causes noise in the git history, where git blame is used a lot to find why
> things were done. Adding renames causes one more level of indirection that
> makes it harder on the maintainers to do git forensics.
> 
>> 
>> Specifically:
>> 1. Some function names do not accurately reflect their purpose.
>> 2. Certain names are too general, making it unclear what they represent.
>> 3. Some concepts are ambiguous, leading to potential confusion.
>> 
>> - Rename ttwu_do_wakeup to ttwu_set_running:
>>    - This function only sets task state to TASK_RUNNING,
>>        not performing the actual wakeup.
>> 
>> - Rename update_cfs_group to update_cfs_group_shares:
>>    - The name was too generic; specifying "shares" clarifies its purpose.
>> 
>> - Rename place_entity to update_entity_sched:
>>    - The function does not handle entity placement but updates
>>        sched info: vruntime, slice, and deadline.
>> 
>> - Rename update_load_<set, add, sub> to load_weight_<set, add, sub>:
>>    - "load" can refer to either PELT load or load weight, causing ambiguity;
>>        "load_weight" specifies it's dealing with weight.
>> 
>> - Rename struct sched_avg to struct sched_pelt:
>>    - This structure includes not just average statistics
>>        but also sums like <load, runnable, util>_sum, last_updae_time,
>>        all of which are PELT (Per-Entity Load Tracking) metrics.
>> 
>> - Rename init_entity_runnable_average to init_entity_pelt
>> 
>> This patch aims to improve code readability and reduce confusion by
>> ensuring names are more descriptive of their actual functionality or purpose.
> 
> I personally am not against these updates. But as I mentioned, there's a
> lot of history here and it's really Peter Zijlstra's call (and he's been
> against changes like this in the past).
> 
> So please do not be discourage if this doesn't get much traction.
> 
> -- Steve