Re: [RFC] introduce pr_cont macro

[Ingo Molnar <mingo@elte.hu>]

* Harvey Harrison <harvey.harrison@gmail.com> wrote:

> On Thu, 2009-02-26 at 04:03 +0100, Ingo Molnar wrote:
> > * Cyrill Gorcunov <gorcunov@gmail.com> wrote:
> > 
> > > We cover all log-levels by pr_... macros except
> > > KERN_CONT one. Add it for convenience.
> > > 
> > > Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
> > > ---
> > > 
> > > I think start printing with some pr_... macro
> > > and continue with printk(KERN_CONT ...) look
> > > not that clear -- better to continue with
> > > same pr_... slogan. Thoughts? I hope I didn't
> > > miss anything.
> > > 
> > >  include/linux/kernel.h |    2 ++
> > >  1 file changed, 2 insertions(+)
> > > 
> > > Index: linux-2.6.git/include/linux/kernel.h
> > > ===================================================================
> > > --- linux-2.6.git.orig/include/linux/kernel.h
> > > +++ linux-2.6.git/include/linux/kernel.h
> > > @@ -370,6 +370,8 @@ static inline char *pack_hex_byte(char *
> > >          printk(KERN_NOTICE pr_fmt(fmt), ##__VA_ARGS__)
> > >  #define pr_info(fmt, ...) \
> > >          printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__)
> > > +#define pr_cont(fmt, ...) \
> > > +        printk(KERN_CONT pr_fmt(fmt), ##__VA_ARGS__)
> > 
> > Makes sense.
> > 
> > Acked-by: Ingo Molnar <mingo@elte.hu>
> > 
> 
> Don't you want to omit the pr_fmt() from the KERN_CONT case?

indeed, good catch. It should be:

#define pr_cont(fmt, ...) \
	printk(KERN_CONT ##__VA_ARGS__)

To not introduce stale subsystem tags in the printout. Perhaps 
it should even be:

#define __pr_cont(fmt, ...) \
	printk(KERN_CONT ##__VA_ARGS__)

To make it stand out more clearly that this is special.

	Ingo