linux-kernel - Re: [GIT PATCH] driver core patches for 3.5-rc1

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

Date:	Tue, 22 May 2012 15:12:16 -0700
From:	Joe Perches <joe@...ches.com>
To:	Greg KH <gregkh@...uxfoundation.org>
Cc:	Kay Sievers <kay@...y.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [GIT PATCH] driver core patches for 3.5-rc1 - try 2

(adding lkml and not editing Greg's reply)

On Tue, 2012-05-22 at 15:05 -0700, Greg KH wrote:
> On Tue, May 22, 2012 at 02:58:47PM -0700, Joe Perches wrote:
> > While there are several things I like about the
> > printk modifications, (binary header, delta time,
> > slightly better partial message deinterleaving,
> > global msg_id, kmsg is ok too), I am concerned
> > about the utility and expectations for the new
> > [v]printk_emit functions.
> > 
> > I think it is not really ready to be merged
> > at this time.
> > 
> > The commit sequencing was unclean.
> 
> Yes, but the build was never broken, the system always worked, and the
> end result was agreed apon by everyone as a nice addition.
> 
> > The original commit originally required KERN_CONT
> > and it was modified by another commit to return
> > to the current behavior.
> > 
> > What really are the expectations and true use-cases
> > for [v]printk_emit?
> 
> What's wrong with the existing use case?  Since when do we write
> use-cases for kernel functions?

When it's a generic function that appears to be used
to attach additional data to message logging,
and it takes up buffer space and currently reduces
the number of dev_<foo> messages in dmesg without
adding value.

What's it for and who's going to use it and more
importantly, why?

I do not want to link human readable messaging with
machine parsing of binary data.  I'm not sure it's
useful and possibly sets expectations for immutability.

> > How is it really better that what is available now?
> > 
> > Perhaps it would be better to respin all the
> > printk modifications without adding [v]printk_emit
> > and have the [v]printk_emit bits debated a bit more.
> 
> I'm always glad to review patches, but to just propose something that
> works to be reverted without a patch to replace the functionality, isn't
> ok.

It hasn't been committed yet in anything other than
-next.

Lots of stuff gets committed there and later removed
and hits Linus' mainline tree.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/