linux-kernel - Re: linux.git: printk() problem

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: <1476288538.16823.9.camel@perches.com>
Date:   Wed, 12 Oct 2016 09:08:58 -0700
From:   Joe Perches <joe@...ches.com>
To:     Michal Hocko <mhocko@...nel.org>,
        Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>
Cc:     torvalds@...ux-foundation.org, linux-kernel@...r.kernel.org
Subject: Re: linux.git: printk() problem

On Wed, 2016-10-12 at 16:35 +0200, Michal Hocko wrote:
> On Wed 12-10-16 22:30:03, Tetsuo Handa wrote:
> > Hello.
> > 
> > I noticed that current linux.git generates hardly readable console output
> > due to KERN_CONT changes. Are you suggesting developers that output like
> > this be fixed?
> 
> 
> Joe has already posted a patch
> http://lkml.kernel.org/r/c7df37c8665134654a17aaeb8b9f6ace1d6db58b.1476239034.git.joe@perches.com
> 
> And in general I think that adding those KERN_CONT is a good thing to
> do.

As do I, but there are about a quarter _million_ uses of
printk/logging messages in the kernel tree with newlines.

Maybe 100 or so of the existing messages lack terminating
newlines.

There are about 1800 uses of KERN_CONT/pr_cont today.

There are multiple thousands of uses of bare printks used as
printk line continuations that would need updating.

Most of those multiple thousands are in drivers of ancient and
out of manufacture devices or are subsystems that are old, crufty
and effectively unmaintained.  These might not need to be updated.

Still, there are probably hundreds to low thousands of actively
used device drivers that will need KERN?CONT or pr_cont update to
avoid logging output defects.