Re: [PATCH] Fix broken mmiotrace due to dr6 by reference change

[K.Prasad]

On Sat, Jan 16, 2010 at 10:39:30AM +0100, Luca Barbieri wrote:
> Commit 62edab9056a6cf0c9207339c8892c923a5217e45 (from June 2009 but merged
> in 2.6.33) changes notify_die to pass dr6 by reference.
> 
> However, it forgets to fix the check for DR_STEP in kmmio.c, breaking
> mmiotrace. It also passes a wrong value to the post handler.
> 

Yes, looks like I missed it...must have hunted down all those registered
to the die notifier...to propagate the fix.

> This simple fix makes mmiotrace work again.
> 
> Please send to Linus for inclusion in 2.6.33
> 
> Signed-off-by: Luca Barbieri <luca@xxxxxxxxxxxxxxxxx>
Acked-by: K.Prasad <prasad@xxxxxxxxxxxxxxxxxx>

Thanks,
K.Prasad

> ---
>  arch/x86/mm/kmmio.c |    7 ++++---
>  1 files changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/mm/kmmio.c b/arch/x86/mm/kmmio.c
> index c0f6198..536fb68 100644
> --- a/arch/x86/mm/kmmio.c
> +++ b/arch/x86/mm/kmmio.c
> @@ -538,14 +538,15 @@ static int
>  kmmio_die_notifier(struct notifier_block *nb, unsigned long val, void *args)
>  {
>  	struct die_args *arg = args;
> +	unsigned long* dr6_p = (unsigned long *)ERR_PTR(arg->err);
> 
> -	if (val == DIE_DEBUG && (arg->err & DR_STEP))
> -		if (post_kmmio_handler(arg->err, arg->regs) == 1) {
> +	if (val == DIE_DEBUG && (*dr6_p & DR_STEP))
> +		if (post_kmmio_handler(*dr6_p, arg->regs) == 1) {
>  			/*
>  			 * Reset the BS bit in dr6 (pointed by args->err) to
>  			 * denote completion of processing
>  			 */
> -			(*(unsigned long *)ERR_PTR(arg->err)) &= ~DR_STEP;
> +			*dr6_p &= ~DR_STEP;
>  			return NOTIFY_STOP;
>  		}
> 
> -- 
> 1.6.3.3
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/