Re: [PATCH] printk(): add KERN_CONT where needed

From: Andrew Morton <akpm@linux-foundation.org>
To: Kay Sievers <kay@vrfy.org>
Cc: Joe Perches <joe@perches.com>,
	linux-kernel@vger.kernel.org,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Len Brown <lenb@kernel.org>
Subject: Re: [PATCH] printk(): add KERN_CONT where needed
Date: Mon, 9 Apr 2012 16:08:03 -0700	[thread overview]
Message-ID: <20120409160803.27bc1376.akpm@linux-foundation.org> (raw)
In-Reply-To: <CAPXgP12jX37GLgMJQuDYEi6q5A2Ma3d+e+_sNkFVdYfwsdMrtQ@mail.gmail.com>

On Tue, 3 Apr 2012 05:00:10 +0200
Kay Sievers <kay@vrfy.org> wrote:

> > Maybe it'd be better to aggregate content rather like
> > printk does. __Aggregate until you get a newline or a
> > new KERN_<LEVEL>
> 
> The continuation printk() can can always go wrong when multiple
> threads do that in parallel. We can try to make it better with a
> per-cpu buffer, but I guess there will always be a situation where
> this can happen.

Maybe we can be a bit smarter.  For example, if `current' is unchanged
and __builtin_return_address(0) is unchanged, keep on buffering.

It's all a bit hacky, but weeding out all those thousands of printks
which never get printed anyway doesn't sound much fun either.

> -	printk(")");
> +	printk(KERN_CONT ")");

And I do think we should avoid doing it that way, if only because it
consumes 10 display columns and makes a mess.  Maybe use pr_cont()? 
But that implies that the affected code is using the pr_foo()
facilities, and a lot of it doesn't.  So maybe a new macro.

All a bit of a pain.