Apple Network Server MacOS based ROMs found
- Thread starter eric
- Start date
No, we have this part. We need the full register dump. Thank you for trying it, though - proves it's not just my hardware, but what ROM did you do this with?
@ClassicHasClass you are booting into Mac OS using external SCSI?
Looking at ANS 1.1.20.1 Open Firmware in DingusPPC, there's some differences from 1.0.5 that might be strange. When it creates the
3:CMD_RESET_BUS
2:CMD_RESET_DEVICE
1:CMD_CLEAR_FIFO
DingusPPC expects a 0:CMD_NOP after 2:CMD_RESET_DEVICE so
Since you have a real ANS, can you tell me if this command works?
If it doesn't work, then try this:
I've attached a script to list partitions from a Apple Partition Map partitioned disks. Load and use it like this:
Looking at ANS 1.1.20.1 Open Firmware in DingusPPC, there's some differences from 1.0.5 that might be strange. When it creates the
53c94 device, it does a reset-scsi which 1.0.5 does not do.reset-scsi performs 3 commands:3:CMD_RESET_BUS
2:CMD_RESET_DEVICE
1:CMD_CLEAR_FIFO
DingusPPC expects a 0:CMD_NOP after 2:CMD_RESET_DEVICE so
53c94 doesn't work.Since you have a real ANS, can you tell me if this command works?
Code:
" scsi/sd@0" select-devIf it doesn't work, then try this:
Code:
dev scsi
: new-reset-scsi
." doing reset-scsi" cr
3 cmd! d# 500 ms
2 cmd!
0 cmd!
1 cmd!
;
' reset-scsi ' new-reset-scsi brpatch
reset-scsi
" scsi/sd@0" select-devI've attached a script to list partitions from a Apple Partition Map partitioned disks. Load and use it like this:
Code:
unselect-dev
dl \ press return to begin the download
\ then paste the contents of the text file (you won't see the contents because
\ echo is not used during download), then press Control-D to evaluate the contents.
list-partitions scsi/sd@0Yes, I'm booting it from external SCSI, but I'm out of town for a couple days - I will check for you when I get back.
Is this it ? This is the 2.26NT.
Also I was trying to find my serial to USB cable but can't seem to find it. Which one are you using?
r0 = DEADBEEF
r1 = 0004FF00
r2 = 000691B8
r3 = DEADBEEF
r4 = 00050000
r5 = FF809C50
r6 = DEADBEEF
r7 = DEADBEEF
r8 = 18A2D8C5
r9 = DEADBEEF
r10 = DEADBEEF
r11 = DEADBEEF
r12 = DEADBEEF
r13 = DEADBEEF
r14 = DEADBEEF
r15 = DEADBEEF
r16 = DEADBEEF
r17 = DEADBEEF
r18 = DEADBEEF
r19 = DEADBEEF
r20 = DEADBEEF
r21 = DEADBEEF
r22 = DEADBEEF
r23 = DEADBEEF
r24 = DEADBEEF
r25 = DEADBEEF
r26 = DEADBEEF
r27 = DEADBEEF
r28 = DEADBEEF
r29 = DEADBEEF
r30 = DEADBEEF
r31 = DEADBEEF
Special Regs:
IV = 00000300
SRR0 = 00050014
SRR1 = 00001071
CR = 00000000
LR = 00050004
CTR = DEADBEEF
XER = C000BE6F
DAR = 18A2D8C5
DSISR = 42000000
SDR1 = 03FE0000
That indeed looks like a full register dump, but what command did you use to generate it? Do you have the full transcript of what you did? I want to make sure we're capturing the right information.
I use a PL-2303 USB-to-serial dongle (an old iConcepts PDA adapter; it works with pretty much anything, even Mac OS 9).
I use a PL-2303 USB-to-serial dongle (an old iConcepts PDA adapter; it works with pretty much anything, even Mac OS 9).
This would imply .data section either wasn't loaded correctly or part of it got overwritten by something.
Any chance you could dump 0x13000 bytes at 0x50000? I can then compare against what *should* be there. Certainly I have no idea where 0x18A2D8C5 came from.
I'm working on getting both now. But I need to get a batter way to get commands to the ANS.
@ClassicHasClass thanks for sending the info about the iConcepts USB serial adapter. The only thing is, what 8 Pin Male to DB9 Female cable did you use? It would seem simple but it looks like different cables have different pinouts and I'm not sure which one we need for console access?
Seems at a base we need
DE-9 pin 2 (RxD) <-> Mini-DIN-8 pin 3 (TxD−)
DE-9 pin 3 (TxD) <-> Mini-DIN-8 pin 5 (RxD−)
DE-9 pin 5 (GND) <-> Mini-DIN-8 pin 4 (GND)
?
I'm not sure what the wiring is (I can buzz it out if you really want), but it's just an off-the-shelf Mac serial MiniDIN-8 to DE-9 null modem. There are tons of sellers on eBay.
Thanks for that info, no problem I figured it would be fine but when I looked on amazon, a bunch of the cables had bad reviews saying they were wired wrong. I'll just search for null modem.
Some USB serial port adapters serial port wiring examples at https://forums.macrumors.com/threads/serial-app-to-old-world-of.2429025/post-33239745
Just a quick update on trying out Windows NT with some AI tool assisted help. I was able to patch enough memory to get the VENEER.EXE to boot to this
Then that errored out with
Did some more work with patching memory and got to
At this point, too much typing was required to keep going so again, I'll just wait until I can connect to the serial console.
Then that errored out with
Did some more work with patching memory and got to
At this point, too much typing was required to keep going so again, I'll just wait until I can connect to the serial console.
That is very promising, but we really need a full transcript of what you're doing, or it will be tricky to replicate.
@joevt
Code:
0 > unselect-dev ok
0 > dl ok
0 > list-partitions scsi/sd@0
Device isn't there! unable to open the list-partitions device
ok
0 > list-partitions scsi/sd@0,0
Device isn't there! unable to open the list-partitions device
ok
0 > see list-partitions
: list-partitions
bl parse open-dev ?dup if
to ih (list-partitions)
else
" unable to open the list-partitions device" type cr
then
; ok
0 > " scsi/sd@0" select-dev
Device isn't there!can't OPEN device
ok
0 > dev scsi ok
0 > dl ok
0 > ' reset-scsi ' new-reset-scsi brpatch ok
0 > reset-scsi doing reset-scsi
ok
0 > " scsi/sd@0" select-dev
Device isn't there!can't OPEN device
ok
0 > dev scsi ok
0 > ls
FF8461F0: /sd@0,0
FF846FD8: /st@0,0
ok
0 > dev sd@0,0 ok
0 > words
load write read seek close open
write-blocks read-blocks max-transfer block-size dma-sync dma-map-out
dma-map-in dma-free dma-alloc sense-buffer timed-spin set-timeout
error? retry-cmd set-tid-lun db-rem write-10 write-6
read-10 read-6 read-capacity max-blks offset-hi offset-lo
disk-label deblocker my-block-size
ok
0 > " scsi/sd@0,0" select-dev
Device isn't there!can't OPEN device
ok
Last edited:
It's been a long day, I should have thought of that.
Code:
0 > list-partitions scsi/sd@2,0
# Start Size Name Type CPU Sigs 0 1 2 Volume
01 00000001 0000003F Apple Apple_partition_map PM PM PM
02 00000040 00000036 Macintosh Apple_Driver43 68000 ` . N
03 00000076 0000004A Macintosh Apple_Driver43 68000 ` A
04 000000C0 00000200 Macintosh Apple_Driver_IOKit
05 000002C0 00000200 Patch Partition Apple_Patches
06 000004C0 003FFB2F untitled Apple_HFS LK BD Shiner9
07 003FFFEF 00000007 Extra Apple_Free
okOh, and
" scsi/sd@2,0" select-dev does work.
Last edited:
Thanks for this. Looks like DingusPPC needs some fixes for 53c94.
I think
,0 is for the LUN number? Not needed on most devices since they have a single LUN.Open Firmware has a
show-children for each built-in SCSI controllers.Use it like this:
Code:
\ 53c94 (TNT or ANS)
" scsi" select-dev show-children
\ mesh (TNT) or 53c825@11 (ANS)
" scsi-int" select-dev show-children
53c825@12 (ANS)
" scsi-int2" select-dev show-childrenOpen Firmware 1.1.20.1 adds simpler commands to do the above:
Code:
probe-scsi0
probe-scsi1
probe-scsi2The command should list multiple LUNs if they exist.
A device that might include multiple LUNs is a CD changer?
I have seen optical disk juke boxes that used multiple LUNs, but I think it depends on how the manufacturer wanted to configure it. That is, not every disk changer may use multiple LUNs. Some might occupy multiple SCSI IDs.