Except for, apparently, the people who write kernel code:
(edited dmesg output)
.
hda: HITACHI CDR-8130, ATAPI CDROM drive
.
SCSI device sda: hdwr sector= 512 bytes. Sectors= 8925000 [4357 MB] [4.4 GB]
SCSI device sdb: hdwr sector= 512 bytes. Sectors= 8467200 [4134 MB] [4.1 GB]
.
Partition check:
sda: sda1 sda2
sdb: sdb1 sdb2
.
lp1 at 0x0378, (polling)
.
tty00 at 0x03f8 (irq = 4) is a 16550A
tty01 at 0x02f8 (irq = 3) is a 16550A
tty03 at 0x02e8 (irq = 3) is a 16550A
.
eth0: DEC 21040 Tulip at 0xdc00, 00 80 c6 03 6e 9b, IRQ 10
.
If my scsi disks are /dev/scsi/disk/0, /dev/scsi/disk/1, /dev/scsi/disk/2,
I'll still need to do mapping between this kernel-provided namespace
and my namespace to make sense of the kernel messages. If, as there is
when I'm running a 2.3.x kernel, a devfs present, the existance of this
namespace would does not invalidate my custom namespace, except that I
can then do similarly to what Solaris and Irix do and convert
/dev/scsi/disk/0 into a symlink to /hw/sda
To do any sort of device notifier, whether it be devfs or the
unworkable devd kludge, there needs to be a kernel namespace. We
already have two of them (the human-readable names listed above and
the major/minor black magic that's the traditional Unix approach),
so the badness of them is not a given.
____
david parsons \bi/ without a namespace, it's really hard to talk to
\/ anything.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/