linux-kernel - Re: [GIT PULL] trivial for 4.9

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: <1475870688.1945.13.camel@perches.com>
Date:   Fri, 07 Oct 2016 13:04:48 -0700
From:   Joe Perches <joe@...ches.com>
To:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Jiri Kosina <jikos@...nel.org>,
        Colin Ian King <colin.king@...onical.com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] trivial for 4.9

On Fri, 2016-10-07 at 12:22 -0700, Linus Torvalds wrote:
> On Fri, Oct 7, 2016 at 1:51 AM, Jiri Kosina <jikos@...nel.org> wrote:
> > Colin Ian King (5):
> >       netfilter: Add missing \n to pr_err() message
> >       agp/intel: add missing \n to end of dev_emerg message
> >       lightnvm: add missing \n to end of dev_err message
> >       nvme: add missing \n to end of dev_warn message
> 
> 
> We really shouldn't be needing these final '\n' characters any more, afaik.
> 
> If the next printk isn't done by the same process, and doesn't have a
> KERN_CONT, the printk machinery should add the newline on its own.
> 
> I realize that people have been adding these '\n' characters for a
> while, but is there actually any point to it? They make the code less
> legible, imho. And we actually have a number of logging functions that
> explicitly don't want the newline (eg ext4_warning/error()), so it's
> actually more consistent to *not* have a newline than it is to have
> one.

Not remotely true.

It's _way_ more consistent to use a newline termination.
Macros without the newline are _far_ less common than those
with newlines.

Any printk without a KERN_<level> prefix, and there
are still many of those, can cause random interleaving.

> And if those '\n' characters actually make a difference, that should
> be noted. Because that would imply that the printk logic isn't working
> right.

Not at all.  Until printk KERN_<level> uses are mandated,
then these newlines are still useful.