[Devel] Re: [PATCH] Report that kernel is tainted if there were an OOPS before (v3)

[Randy Dunlap]

On Thu, 07 Jun 2007 16:51:50 +0400 Pavel Emelianov wrote:

> If the kernel OOPSed or BUGed then it probably should
> be considered as tainted. Thus, all subsequent OOPSes 
> and SysRq dumps will report the tainted kernel. This 
> saves a lot of time explaining oddities in the calltraces.
> 
> The previous version was buggy and reported the kernel 
> to be tainted at the very first oops as was noticed by 
> Dave Jones in the report from Antonino Daplas.
> 
> Compilation is checked for i386, x86_64 and ia64 since 
> I have no all the others at hands :)
> 
> Signed-off-by: Pavel Emelianov <xemul at openvz.org>
> Cc: Randy Dunlap <randy.dunlap at oracle.com>

Acked-by: Randy Dunlap <randy.dunlap at oracle.com>

I would change the doc. comment below:


> ---
> 
> diff --git a/Documentation/oops-tracing.txt b/Documentation/oops-tracing.txt
> index 7d5b60d..610e234 100644
> --- a/Documentation/oops-tracing.txt
> +++ b/Documentation/oops-tracing.txt
> @@ -237,6 +237,8 @@ characters, each representing a particul
>    7: 'U' if a user or user application specifically requested that the
>       Tainted flag be set, ' ' otherwise.
>  
> +  8: 'D' if the kernel has died recently, i.e. there was an OOPS or BUG.

something like:

+  8: 'D' if the kernel has already died, i.e. there was a prior OOPS or BUG.

> +
>  The primary reason for the 'Tainted: ' string is to tell kernel
>  debuggers if this is a clean kernel or if anything unusual has
>  occurred.  Tainting is permanent: even if an offending module is

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***