Re: sched: restore sanity

From: Arjan van de Ven
Date: Sun Dec 20 2009 - 19:47:05 EST

On Sun, 20 Dec 2009 10:21:50 -0800
Joe Perches <joe@xxxxxxxxxxx> wrote:

> On Sun, 2009-12-20 at 19:17 +0100, Arjan van de Ven wrote:
> > On Sun, 20 Dec 2009 09:22:23 -0800 Joe Perches <joe@xxxxxxxxxxx>
> > wrote:
> > > What are the negatives of using pr_<level>?
> > pr_ is really just for "I am a driver and want a single line message
> > out in a standardized format".
> > Nothing wrong with that.
> >
> > But here you are changing fundamental kernel outputs in the style
> > of an oops message. Multiline, complex and machine parsed things...
> The prefixing was trivial to change.
> No other output was modified.

but you still have a "multiline message" (the atomic schedule oops like
dump) that now is half printk half pr_*... where pr_ is just a silly
wrapper around printk now that the prefixing etc is gone.

> I believe kernel log output is specifically _not_ guaranteed
> and should not be so guaranteed to remain stable across
> versions.

changing it gratuitously is bad though. There *are* scripts out there
parsing oopses. Yes they change as kernels change, but changing
the output of oopses for silly reasons is just hostile


Arjan van de Ven Intel Open Source Technology Centre
For development, discussion and tips for power savings,
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at