Re: [PATCH] for_each macros correctness

[Heiko Carstens <heiko.carstens@de.ibm.com>]

On Sun, Jan 26, 2014 at 08:54:18AM -0200, Jose Alonso wrote:
> 
> I observed that there are for_each macros that do an extra memory access
> beyond the defined area.
> Normally this does not cause problems.
> But, this can cause exceptions. For example: if the area is allocated at
> the end of a page and the next page is not accessible.
> 
> For correctness, I suggest changing the arguments of the 'for loop' like
> others 'for_each' do in the kernel.
> 
> files involved:
>    drivers/s390/cio/qdio.h
>    drivers/scsi/isci/host.h
>    drivers/sh/clk/core.c
>    include/linux/blk-mq.h
>    include/linux/shdma-base.h
>    sound/soc/sh/rcar/adg.c
> 
> 
> Signed-off-by: Jose Alonso <joalonsof@gmail.com>
> ---
> diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
> index 8acaae1..a563e4c 100644
> --- a/drivers/s390/cio/qdio.h
> +++ b/drivers/s390/cio/qdio.h
> @@ -359,14 +359,12 @@ static inline int multicast_outbound(struct qdio_q *q)
>  #define need_siga_sync_out_after_pci(q)	\
>  	(unlikely(q->irq_ptr->siga_flag.sync_out_after_pci))
> 
> -#define for_each_input_queue(irq_ptr, q, i)	\
> -	for (i = 0, q = irq_ptr->input_qs[0];	\
> -		i < irq_ptr->nr_input_qs;	\
> -		q = irq_ptr->input_qs[++i])
> -#define for_each_output_queue(irq_ptr, q, i)	\
> -	for (i = 0, q = irq_ptr->output_qs[0];	\
> -		i < irq_ptr->nr_output_qs;	\
> -		q = irq_ptr->output_qs[++i])
> +#define for_each_input_queue(irq_ptr, q, i)		\
> +	for (i = 0; i < irq_ptr->nr_input_qs &&		\
> +		({ q = irq_ptr->input_qs[i]; 1; }); i++)
> +#define for_each_output_queue(irq_ptr, q, i)		\
> +	for (i = 0; i < irq_ptr->nr_output_qs &&	\
> +		({ q = irq_ptr->output_qs[i]; 1; }); i++)

Hi Jose,

I generated a single commit of this part of your patch and applied
it to the s390 git tree.

Thanks!