lkml.org 
[lkml]   [2014]   [Jun]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC][PATCH 0/3] x86/nmi: Print all cpu stacks from NMI safely
On Wed 2014-06-25 15:04:41, Konstantin Khlebnikov wrote:
> On Wed, Jun 25, 2014 at 2:01 PM, Jiri Kosina <jkosina@suse.cz> wrote:
> > On Tue, 24 Jun 2014, Konstantin Khlebnikov wrote:
>
> I like that context-depend printk routing from this patchset.
> I'm just not quite happy with this straight forward implementation.
> Rerouting flow into different buffers might serve different purposes,
> it must make code simpler instead of adding yet another knot.
> For example printk continuation might use it too.

Note that we could not share the buffer between normal and NMI
context easily. We would end with the same locking problems as we have now.

In fact, any sharing is complicated when NMI is involved, especially
when we need to access the data from both NMI and normal context.
And we need to copy them into the main ring buffer in a normal context.

This patchset works only because arch_trigger_all_cpu_backtrace() is
called from normal context. Also all other calls are ignored until the
previous call is finished, see "backtrace_flag".


Best Regards,
Petr


\
 
 \ /
  Last update: 2014-06-25 14:41    [W:0.059 / U:0.724 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site