5 Feb
2013
5 Feb
'13
6:28 p.m.
A version of the handler sources can be found here:
http://www.ibiblio.org/pub/academic/computer-science/history/pdp-8/os78/v4/
-chuck
On Mon, Feb 4, 2013 at 10:10 PM, Rick Murphy <rick at rickmurphy.net> wrote:
At 02:21 PM 2/4/2013, Charles wrote:
Excerpted: (there are also the expected SYS, DSK, BAT, TTY and this
build has RXA0 and RXA1 and DTA0 installed
although unfortunately I
don't have the hardware).
# NAME TYPE MODE SIZ BLK KIND U V ENT USER
10 (6442) 37 RWF 22+ A 120
11 (6542) 37 RWF 22+ A 114
12 (6642) 37 RWF 22+ A 107
13 (6742) 37 RWF 22+ A 100
14 (6443) 37 RWF 23+ A 120
15 (6543) 37 RWF 23+ A 114
16 (6643) 37 RWF 23+ A 107
17 (6743) 37 RWF 23+ A 100
So, apparently eight distinct User types for the A-D drives on disk 0
and disk 1.
That's the first thing I wanted to verify - interesting that your copy of
RESORC hasn't been patched to know the RL02 device names, but that's
cosmetic. Since each drive has a different "NAME" value they're not
colliding.
The number in the "name" column is the sum of the two sixbit words of the
device name, with the high bit forced on if the second word is nonzero.
R2A0 is 6222 4120 sixbit, sum 2342, or 4000 = 6342. R2B0 is 6442. R2A1 is
6443.
The "BLK" column - "22+" for drive zero, and "23+" for
drive 1, means that
the driver for disk zero is on block 22 of the system disk (and is a
two-page driver, indicated by the "+"), and the driver for disk 1 is on
block 23. So there's two distinct drivers for the two drives.
Not sure what the "ENT" column is, but they seem to have
the same pattern repeated twice? Is this normal?
Yes, that's normal. Each "drive" (segment of the disk) has a different
entry point into the common driver. That's what the "ENT" column tells you.
Apparently the RL02 driver can only handle one physical drive unit.
Here's the interpretation:
Entry 10 (6442) is for R2B0, entry at driver start + 120
Entry 11 (6542) is for R2C0, 114
Entry 12 (6642) is for R2D0, 107
Entry 13 (6742) is for R2E0, 100
Entry 14 (6443) is for R2B1, 120
Entry 15 (6543) is for R2C1, 114
Entry 16 (6643) is for R2D1, 107
Entry 17 (6743) is for R2E1, 100
Contrast this with my SIMH instance with four RK05s:
# NAME TYPE MODE SIZ BLK KIND U V ENT USER
05 RKA0 RK8E RWF 3248 20 RK05 0 A 20
06 RKB0 RK8E RWF 3248 20 RK05 0 A 21
07 RKA1 RK8E RWF 3248 20 RK05 1 A 22
10 RKB1 RK8E RWF 3248 20 RK05 1 A 23
11 RKA2 RK8E RWF 3248 20 RK05 2 A 24
12 RKB2 RK8E RWF 3248 20 RK05 2 A 25
13 RKA3 RK8E RWF 3248 20 RK05 3 A 26
14 RKB3 RK8E RWF 3248 20 RK05 3 A 27
17 DTA0 TD8E RWF 737 22+ TD8 0 D 10
There's one driver instance on block 20, single page driver.
The "U" column has the unit number, and the entry offset is different.
Here it's one driver for four physical and 8 logical disks. I don't have
the source for the RL02 driver, but apparently they've hardcoded the unit
number into the driver source. Weird.
-Rick