[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YY+/t75dqrkDA9CM@sashalap>
Date: Sat, 13 Nov 2021 08:37:59 -0500
From: Sasha Levin <sashal@...nel.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Ingo Molnar <mingo@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] tools/lib/lockdep: drop liblockdep
On Fri, Nov 12, 2021 at 11:16:14AM -0800, Linus Torvalds wrote:
>On Fri, Nov 12, 2021 at 7:16 AM Sasha Levin <sashal@...nel.org> wrote:
>>
>> TL;DR: While a tool like liblockdep is useful, it probably doesn't
>> belong within the kernel tree.
>
>I have eagerly applied this patch, because I wasn't too happy with
>that thing in the first place.
Thank you for providing the room for this experiment nonetheless.
>However - I do note that there's signs of liblockdep in the
>tools/include/ directory still.
>
>Some of those headers may now be dead, and should be removed, but I
>didn't end up trying to figure it out, so I left them alone.
>
>This stupid shell command
>
> git grep -il liblockdep tools/include/ |
> sed 's:tools/include/::' |
> while read i; do echo; echo $i:; echo; git grep $i tools/ ; done
>
>seems to say that yes, the <linux/module.h> header in the tools
>directory is still used, but the rest of them seem to only reference
>other dead headers.
>
>Hmm? It would be lovely to clean those headers up too and get rid of
>them - because those headers are very much part of why I hated the
>whole liblockdep mess, and why I was so happy to immediately apply
>this patch..
Yup, there are remenants in the common header directory which need more
careful auditing before removing. I'll send a follow-up patch during one
of the early -rc cycles.
Thanks!
--
Thanks,
Sasha
Powered by blists - more mailing lists