Messages in this thread | | | Subject | Re: [PATCH v2] init: Scream bloody murder if interrupts are enabled too early | From | Steven Rostedt <> | Date | Fri, 08 Mar 2013 17:35:00 -0500 |
| |
On Fri, 2013-03-08 at 13:55 -0800, Andrew Morton wrote: > On Fri, 08 Mar 2013 09:48:12 -0500 Steven Rostedt <rostedt@goodmis.org> wrote: > > > Also, while examining other warnings in init/main.c, I found two other > > locations that deserve a bloody murder scream if their conditions are > > hit, and updated them accordingly. > > So the whole effect of this patch is to spew a larger volume of > unuseful stuff (such as uninteresting stack traces) onto the console so > people are more likely to notice it?
Yep!
> Hmpf. I think I prefer > > akpm3:/usr/src/25> banner "OMG\!\!" > ####### # # ##### ### ### > # # ## ## # # ### ### > # # # # # # # ### ### > # # # # # # #### # # > # # # # # # > # # # # # # ### ### > ####### # # ##### ### ### >
I rather print out this: http://www.100mb.nl/
> > Maybe we should reduce the amount of goop we printk out at boot > instead ;)
But I seldom read the boot messages, that's what tools are for ;-)
ktest.pl keys off of a stack trace, and so does the tip automated tools, and I'm sure other tools do too. Not sure anything looks for a simple "This is a bug, don't do that".
-- Steve
| |