Re: [RFC][PATCH] tracing: Print nasty banner when trace_printk() is in use

[Steven Rostedt <rostedt@goodmis.org>]

On Wed, 28 May 2014 19:22:39 +0200
Peter Zijlstra <peterz@infradead.org> wrote:

> On Wed, May 28, 2014 at 01:14:40PM -0400, Steven Rostedt wrote:
> > 
> > trace_printk() is used to debug fast paths within the kernel. Places
> > that gets called in any context (interrupt or NMI) or thousands of
> > times a second. Something you do not want to do with a printk().
> > 
> > In order to make it completely lockless as it needs a temporary buffer
> > to handle some of the string formatting, a page is created per cpu for
> > every context (four per cpu; normal, softirq, irq, NMI).
> > 
> > Since trace_printk() should only be used for debugging purposes,
> > there's no reason to waste memory on these buffers on a production
> > system. That means, trace_printk() should never be used unless a
> > developer is debugging their kernel. There's macro magic to allocate
> > the buffers if trace_printk() is used anywhere in the kernel.
> > 
> > To help enforce that trace_printk() isn't used outside of development,
> > when it is used, a nasty banner is displayed on bootup (or when a module
> > is loaded that uses trace_printk() and the kernel core does not).
> > 
> > Here's the banner:
> > 
> >  ****************************************
> >  ** NOTICE NOTICE NOTICE NOTICE NOTICE **
> >  ** trace_printk() being used.         **
> >  ** Allocating extra memory for it     **
> >  ****************************************
> > 
> > Hmm, maybe I should add "Not for production use" to scare people even
> > more?
> 
> Does that really stop people from doing stupid? Wouldn't it be better to

Scary banners usually do. Perhaps this isn't scary enough.

> make sure nobody merges a trace_printk() user in mainline? You can set
> up a commit hook and check for +.*trace_printk or so.

Of course, but that requires me to monitory it. It may be too late when
I notice it.

Nothing prevents me from doing both :-)

-- Steve