linux-kernel - Re: [PATCH] workqueue: combine judgments in the loop ofmaybe_create

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>] [day] [month] [year] [list]

Date:   Thu, 13 Sep 2018 10:32:15 +0800
From:   Lai Jiangshan <jiangshanlai@...il.com>
To:     liu.song11@....com.cn
Cc:     Tejun Heo <tj@...nel.org>, liusong@....com,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] workqueue: combine judgments in the loop ofmaybe_create_worker

On Thu, Sep 13, 2018 at 9:51 AM <liu.song11@....com.cn> wrote:
>
> >> From: Liu Song <liu.song11@....com.cn>
> >>
> >> Although the 'need_to_create_worker' has been determined to be
> >> true before entering the function. However, adjusting the order
> >> of judgment can combine two judgments in the loop. Also improve
> >> the matching between logical operations and function naming.
> >>
> > >Signed-off-by: Liu Song <liu.song11@....com.cn>
> >
> >idk, this doesn't really improve anything.
> >
> >Thanks.
> >
> >--
> >tejun
>
> Hi, tejun
> Indeed, the logic of the code has not changed.

The logic of the code is changed in the patch when create_worker()
happened to be failed.

Failed create_worker() doesn't always result to COOLDOWN due to
the possible condition changes. if (!need_to_create_worker(pool)),
it dosn't need to COOLDOWN.



> The idea that I made this change is
> that when reading the workqueue code, from 'maybe_create_worker' naming, it is
> possible to create a worker, but the implementation will definitely create a worker,
> which makes me somewhat confused. After analysis, I found the determine whether
> to create a worker before the function call. So this patch seems to be able to reduce
> the confusion between the function naming and implementation.
>
> Thanks
> Best regards