lkml.org 
[lkml]   [2015]   [Aug]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 2/7] tty: core: Add tty_debug() for printk(KERN_DEBUG) messages
On Wed, Aug 05, 2015 at 02:33:37PM -0400, Peter Hurley wrote:
> On 07/23/2015 09:35 PM, Greg Kroah-Hartman wrote:
> > On Sun, Jul 12, 2015 at 10:49:08PM -0400, Peter Hurley wrote:
> >> Introduce tty_debug() macro to output uniform debug information for
> >> tty core debug messages (function name and tty name).
> >>
> >> Note: printk(KERN_DEBUG) is retained here over pr_debug() since
> >> messages can be enabled in non-DEBUG builds.
> >
> > But pr_debug() is the "standard" way to enable/disable debugging
> > messages, so I'd really like to see that be used here.
>
> Ok, I can do that; I'll roll Joe's suggestions in at that time.
>
> > Even better, this is a tty device, so it should be using dev_dbg(),
> > which gives us tons of good information built-in for the tty and can
> > properly be parsed by userspace tools to know exactly what device caused
> > what message at what point in time.
> >
> > So I'll take this for now, but moving it to use dev_dbg() would be best
> > eventually.
>
> The issue with using dev_dbg is that (SysV) ptys are not devices.

True :(

> However,
> if you'd prefer, I could rework this macro to format output like dev_dbg;
> ie., <driver> <tty name>: <fmt>

That would be good, as that's what userspace is expecting the messages
to look like.

thanks,

greg k-h


\
 
 \ /
  Last update: 2015-08-05 21:41    [W:0.157 / U:0.600 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site