linux-kernel - Re: linux-next: build warnings after merge of the tomoyo tree

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

Message-ID: <9353ac77-f473-3f64-7e61-566807cd47e2@I-love.SAKURA.ne.jp>
Date:   Tue, 31 May 2022 13:31:46 +0900
From:   Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>
To:     Stephen Rothwell <sfr@...b.auug.org.au>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Linux Next Mailing List <linux-next@...r.kernel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: linux-next: build warnings after merge of the tomoyo tree

On 2022/05/31 9:44, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the tomoyo tree, today's linux-next build (powerpc
> ppc64_defconfig) produced these warnings:
> 
> drivers/scsi/mpt3sas/mpt3sas_scsih.c:12412:9: note: '#pragma message: Please avoid flushing system-wide workqueues.'

This is neither "warning:" nor "error:". This is a "note:" message which also appears with "make W=1" case.

> Also, my x86_64 allmodconfig build gained these:

Yes, currently there are many matches when building all modules.
But do developers build all modules?

> Linus will not be happy ...

Hmm, an important part which was missing in commit 771c035372a036f8 ("deprecate
the '__deprecated' attribute warnings entirely and for good") is that Linus
builds all modules and reads all messages during the build?

If that is the case, reporting "Please avoid flushing system-wide workqueues."
every time sounds noisy for him.

But since __deprecated became no-op in v4.19, I guess that developers abandoned
efforts for removing __deprecated usage.

  v5.18: 13 references
  v5.10: 12 references
  v4.19: 12 references
  v4.18: 20 references
  v4.10: 25 references
  v4.0:  25 references
  v3.10: 37 references

What I'm trying to do is to teach developers about possibility of problems
at compile time. __deprecated suits here but is no-op.

Something like console_loglevel which can be controlled on per directory
basis (so that developers can deeply listen to messages from only modules
they maintain) is needed for kernel build?