linux-kernel - Re: [PATCH v3 00/30] staging: rtl8723bs: remove RT

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 for Android: free password hash cracker in your pocket

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

Date:   Sat, 3 Apr 2021 17:21:44 +0200
From:   Fabio Aiuto <fabioaiuto83@...il.com>
To:     Joe Perches <joe@...ches.com>, gregkh@...uxfoundation.org
Cc:     dan.carpenter@...cle.com, linux-staging@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 00/30] staging: rtl8723bs: remove RT_TRACE logs in
 core/*

On Sat, Apr 03, 2021 at 08:02:25AM -0700, Joe Perches wrote:
> On Sat, 2021-04-03 at 11:13 +0200, Fabio Aiuto wrote:
> > This patchset removes all RT_TRACE usages in core/ files.
> 
> and hal and include and os_dep

Hi, 

I was just about to send the second patchset relative to hal/ files.
The whole has been split up in directories in order to reduce the
number of patch per patchset

> 
> > 
> > This is the first of a series aimed at removing RT_TRACE macro.
> > 
> > The whole private tracing system is not tied to a configuration
> > symbol and the default behaviour is _trace nothing_. It's verbose
> > and relies on a private log level tracing doomed to be
> > removed.
> 
> It's nice, but individual patches per file done by hand are difficult
> to review because you are interleaving removal patches with cleanup
> patches.
> 
> I believe this should be a patch series with a single patch to remove
> all RT_TRACE macro uses using coccinelle and then use separate patches
> to do whatever cleanups around these removals you want to do.

It's a good idea, but the patches relative to RT_TRACE removal
could be huge

> 
> All of these below should be done for all files in drivers/staging/rtl8723bs
> at once instead of submitting per-file patches.
> 
> IMO something like:
> 
> Cover-letter: Explain why you are doing this
> Patch 1 of N: Remove all RT_TRACE macro uses using a coccinelle script
>               and include the coccinelle script in the commit message
> Patch 2 of N: Remove commented out RT_TRACE uses
> Patch 3 of N: Remove RT_TRACE macro definition
> Patch 4 of N: Cleanup coccinelle generated {} uses, if/else braces and
>               the now unnecessary if tests around the RT_TRACE removals
> Patch 5 of N: Cleanup whitespace
> Patcn x of N: Whatever else related to these RT_TRACE sites...
> 
> https://lore.kernel.org/lkml/c845d8ea7d0d8e7a613471edb53d780d660142a9.camel@perches.com/
> 
> Using a sequence like the above would be much easier to review and
> would be a significant shorter patch set.
> 

moreover every non RT_TRACE deletion patch (clean up patch) is near 
to the contextual deletion patch (parent patch or grand-parent)

but I do not have experience in code reviewing, so I will do like you
say. Maybe I wait for other opinions, but what you say is good and
elegant.

thank you,

fabio