lkml.org 
[lkml]   [2008]   [Jun]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 0/8] dynamic debug
On Mon, Jun 16, 2008 at 11:26:12AM +0900, Takashi Nishiie wrote:
> Jason Baron wrote:
> >Each kernel sub-system seems to have its own way of dealing with
> >debugging statements. Some of these methods include 'dprintk',
> >'pr_debug', 'dev_debug', 'DEBUGP'. There are also a myriad of
> >ways of enabling these statements.
>
> I propose to replace 'Pr_debug', 'Dev_debug', and 'DEBUGP' with
> kernel markers. SystemTap is used for the output of the log.
>
> I propose to make it to the function to output only specified
> kernel markers as a log and the function in a word like LTTng of
> a simple version by using the framework and kernel markers of
> ftrace if the log is output by using debugfs.
>
> Thank you,
>

perhaps markers could be used to replace 'pr_debug', 'dev_debug', and 'DEBUGP'
but i have yet to see patches for that...

In a number of ways, these dynamic debug patches differ from markers:

-Markers have a pre-defined format string and arguments list, whereas debug
statements have a 'printk' format
-these patches are built around per-module debugging, which is largely implicit
whereas markers explicitly define sets of related markers.
-these patches allow 'flags' and levels to be set per-module, markers do not
have this concept.
-these patches are tied into a procfs control file, whereas markers are
controlled by kernel modules which register handlers.

These two patchsets are really addressing different problems afaict.

thanks,

-Jason





\
 
 \ /
  Last update: 2008-06-16 20:23    [W:0.108 / U:2.672 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site