On Fri, Apr 8, 2011 at 18:14, Greg KH <greg@xxxxxxxxx> wrote: > On Fri, Apr 08, 2011 at 08:14:12AM -0700, James Bottomley wrote: >> So, while I agree totally with the above: udev and userspace is the way >> to go, I'm not totally opposed to having a non-invasive mechanism for >> indicating a user's preferred name for a device. ÂI think there are a >> couple of ways to do this: >> >> Â Â Â1. Entirely in userspace: just have udev consult a preferred name >> Â Â Â Â file and create say /dev/disk/by-preferred. ÂThen have all the >> Â Â Â Â tools that normally output device information do the same (i.e. >> Â Â Â Â since real name to preferred name is 1:1, they could all do a >> Â Â Â Â reverse lookup). >> Â Â Â2. have a writeable sysfs preferred_name field, either in the >> Â Â Â Â generic device or just in SCSI. ÂThe preferred name would be >> Â Â Â Â used by outbound only (i.e. kernel dev_printk messages and >> Â Â Â Â possibly /proc/partitions). ÂAll inbound uses of the device >> Â Â Â Â would come via the standard udev mechanisms >> Â Â Â Â (i.e. /dev/disk/by-preferred would be the usual symlink). ÂThis >> Â Â Â Â means from the kernel point of view, no renaming has happened. >> Â Â Â Â We'd just try to print out the preferred name in certain >> Â Â Â Â circumstances, which should solve most of the described problem. > > Either, or both, of those options are fine with me as well. I guess a generic sysfs file, that can carry a user-given name for a device, and which is looked-up with dev_printk() macro, and maybe /proc/partitions is all what we would need if we want to involve the kernel here. Kay -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
