Re: [PATCH v2 RESEND] libata: if T_LENGTH is zero, dma direction should be DMA_NONE

[Damien Le Moal]

On 2021/12/14 23:45, George Kennedy wrote:
> Avoid data corruption by rejecting pass-through commands where
> T_LENGTH is zero (No data is transferred) and the dma direction
> is not DMA_NONE.
> 
> Cc: <stable@xxxxxxxxxxxxxxx>

George,

FYI, you should not send patches that need backporting to
stable@xxxxxxxxxxxxxxx. The above Cc: tag will do that automatically.

> Reported-by: syzkaller<syzkaller@xxxxxxxxxxxxxxxx>
> Signed-off-by: George Kennedy<george.kennedy@xxxxxxxxxx>
> ---
> Used the Maintainers suggested fix.

The usual way of writing this is something like:

Changes from v1:
* Blah

This way, if you need a v3, v4, etc, you can keep the previous changes
description and add new ones on top.

> 
>  drivers/ata/libata-scsi.c | 15 +++++++++++++--
>  1 file changed, 13 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
> index 1b84d55..313e947 100644
> --- a/drivers/ata/libata-scsi.c
> +++ b/drivers/ata/libata-scsi.c
> @@ -2859,8 +2859,19 @@ static unsigned int ata_scsi_pass_thru(struct ata_queued_cmd *qc)
>  		goto invalid_fld;
>  	}
>  
> -	if (ata_is_ncq(tf->protocol) && (cdb[2 + cdb_offset] & 0x3) == 0)
> -		tf->protocol = ATA_PROT_NCQ_NODATA;
> +	if ((cdb[2 + cdb_offset] & 0x3) == 0) {
> +		/*
> +		 * When T_LENGTH is zero (No data is transferred), dir should
> +		 * be DMA_NONE.
> +		 */
> +		if (scmd->sc_data_direction != DMA_NONE) {
> +			fp = 2 + cdb_offset;
> +			goto invalid_fld;
> +		}
> +
> +		if (ata_is_ncq(tf->protocol))
> +			tf->protocol = ATA_PROT_NCQ_NODATA;
> +	}
>  
>  	/* enable LBA */
>  	tf->flags |= ATA_TFLAG_LBA;

This look OK to me. Will apply.


-- 
Damien Le Moal
Western Digital Research