Re: 2.6.20 BUG: soft lockup detected on CPU#0!

[Andrew Morton <akpm@linux-foundation.org>]

On Thu, 8 Feb 2007 10:56:12 +0100 Ingo Molnar <mingo@elte.hu> wrote:

> 
> * Eric Dumazet <dada1@cosmosbay.com> wrote:
> 
> > This reminds me the current problem in close_files()
> >  code, where we trigger soft lockup quite regularly.
> > 
> > Is there any chance/interest we can solve the issue Andrew had with 
> > this patch ?
> > 
> > http://lkml.org/lkml/2006/5/2/273
> 
> yes - the -rt patch included the patch below for more than 2 years. 
> (note that this one is even more finegrained)
> 
> 	Ingo
> 
> Index: linux/kernel/exit.c
> ===================================================================
> --- linux.orig/kernel/exit.c
> +++ linux/kernel/exit.c
> @@ -431,8 +433,10 @@ static void close_files(struct files_str
>  		while (set) {
>  			if (set & 1) {
>  				struct file * file = xchg(&fdt->fd[i], NULL);
> -				if (file)
> +				if (file) {
>  					filp_close(file, files);
> +					cond_resched();
> +				}
>  			}
>  			i++;
>  			set >>= 1;

That doesn't hang like the other patch did on 2.6.17-rc3.

Very mysterious.