On Aug 17, 2005, at 01:33:00, Matt Domsch wrote:
This is conceptually similar to how SCSI Generic (either /dev/sg or ioctl(SG_IO)) works (userspace passes in preformated SCSI CDBs and gets back the resultant CDBs and extended sense data). The sg driver doesn't look at the data being passed down to any great extent. It doesn't validate that the command will make sense to the end device.
This is not true anymore. Recently the SG driver obtained a basic form of SCSI command checking to prohibit vendor commands from those processes without CAP_RAW_IO, even if said process had full access to the device node itself. Cheers, Kyle Moffett -- Simple things should be simple and complex things should be possible -- Alan Kay - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
