On (03/02/16 17:04), Jan Kara wrote:
> On Wed 02-03-16 23:34:15, Sergey Senozhatsky wrote:
> > > I am looking forward to have the console printing offloaded
> > > into the workqueues. Then printk() will become consistently
> > > "fast" operation and will cause less surprises like this.
> >
> > I'm all for it. I need this rework badly. If Jan is too busy at
> > the moment, which I surely can understand, then I'll be happy to
> > help ("pick up the patches". please, don't get me wrong).
>
> So I'm rather busy with other stuff currently so if you can pick up my
> patches and finish them, it would be good. I think I have addressed all the
> comments you had to the previous version, except for handling the case
> where all the workers are too busy - maybe using a dedicated workqueue with
> a rescueue worker instead of system_wq would solve this issue.
>
> I've sent the current version of patches I have to you including the patch
> 3/3 which I use for debugging and testing whether the async printing really
> helps avoiding softlockups.
great, thank you! will take a look.
and yes, I was thinking about using printk's own workqueue with a
rescue thread bit set (Petr Mladek also proposed this).
-ss
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>