Re: [PATCH] CDC_WR_OVER

[Pat LaVarre <p.lavarre@ieee.org>]

> we can begin to move forward now by accepting or
> revising the patch of this e-mail?

As yet revising, Not accepting.

In fact somehow we have moved this profile/ features/ writability patch
of too many names from sr_mod.ko into cdrom.ko without actually
achieving writability for devices other than DVD-RAM connected via
ide-cd.ko.

> -	if (ide_cdrom_get_capabilities(drive, &cap))
> +	if (ide_cdrom_get_capabilities(drive, &cap)) {
> +		; /* FIXME: mask CDC_WR */
>  		return 0;
> +	}
>  
>  	if (cap.lock == 0)
>  		CDROM_CONFIG_FLAGS(drive)->no_doorlock = 1;
> @@ -3211,6 +3214,7 @@
>  
>  	nslots = ide_cdrom_probe_capabilities (drive);
>  
> +	; /* FIXME: also not ro if CDC_WR */
>  	if (CDROM_CONFIG_FLAGS(drive)->dvd_ram)
>  		set_disk_ro(drive->disk, 0);

Possibly some of those conjectured ide-cd FIXME are necessary.

Now substituting ATAPI for USB, even with the patch applied to
2.6.0-test9, still we see:

$ sudo sg_scan -i
/dev/sg0: scsi0 channel=0 id=0 lun=0 [em]  type=5
    Iomega    RRD               40.D [wide=0 sync=0 cmdq=0 sftre=0 pq=0x0]
$
$ sudo sg_dd of=/dev/sg0 if=/dev/zero bs=2k count=1
1+0 records in
1+0 records out
$
$ sudo sg_inq /dev/hdd | egrep 'id'
 Vendor identification: Iomega
 Product identification: RRD
$
$ sudo dd of=/dev/hdd if=/dev/zero bs=2k count=1
dd: writing `/dev/hdd': Operation not permitted
1+0 records in
0+0 records out
$
$ grep -i permitted /usr/include/asm/errno.h
#define EPERM            1      /* Operation not permitted */
$ grep EPERM drivers/ide/ide-cd.c
$ grep EPERM drivers/cdrom/cdrom.c
$

Pat LaVarre