Re: [PATCH] sd: Try READ CAPACITY 16 first for all SCSI-3 devices

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Tue, 2008-12-23 at 09:47 -0700, Matthew Wilcox wrote:
> New features are being added to the READ CAPACITY 16 results, so we want
> to try to issue it in preference to READ CAPACITY 10.  Unfortunately,
> some USB devices hang when they see a READ CAPACITY 16, so we limit
> our chances of causing a hang by restricting this command to devices
> which claim conformance to SCSI-3.  USB devices are currently limited
> to claiming at most SCSI-2 conformance.
> 
> Of course, it's entirely legitimate for devices to not implement READ
> CAPACITY 16, so this patch also includes a fallback to READ CAPACITY 10
> for SCSI-3 devices.
> 
> Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
> 
> ---
>  drivers/scsi/sd.c |    7 +++++--
>  1 files changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
> index f244349..25a923b 100644
> --- a/drivers/scsi/sd.c
> +++ b/drivers/scsi/sd.c
> @@ -1424,11 +1424,14 @@ sd_read_capacity(struct scsi_disk *sdkp, unsigned char *buffer)
>  	int sector_size;
>  	struct scsi_device *sdp = sdkp->device;
>  
> -	/* Force READ CAPACITY(16) when PROTECT=1 */
> -	if (scsi_device_protection(sdp)) {
> +	if (sdp->scsi_level > SCSI_2) {

Were you going to update this?  Martin Petersen already found an early
SCSI-3 device that crashes upon receiving Read Capacity(16).

James