U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) (08/18/89)
I bought a new Amiha 2000 this week, and looking through the schematics
in the back of the (German) user's manual, I noticed the familiar
(from my A1000) memory overlay line (OVL). It went from a CIA (as
is to be expected) to GARY.
This makes me wonder. The 2000 doesn't need a boot rom, does it? The
start of the kickstart ROM has the same structure as the boot rom
in the 1000. (I know; I disassembled it). Does the OVL line maybe
map a PART of the kickstart rom to location zero? Or is there a
boot rom anyway?
Freely_Distributable=Greetings(Not_For_Any_Commercial_Purpose)->
Olaf.Seibert;
Astounding News! By Express Via Norfolk! OS+2 is a trademark of
The Atlantic Crossed in Three Days! Olaf 'Rhialto' Seibert
[E.A. Poe, 1844] U211344@hnykun11.bitnetlaprade@ringworld.harris-atd.com (Ken Laprade) (08/19/89)
In article <8908181206.AA07159@jade.berkeley.edu> U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) writes: > ... >This makes me wonder. The 2000 doesn't need a boot rom, does it? The >start of the kickstart ROM has the same structure as the boot rom >in the 1000. (I know; I disassembled it). Does the OVL line maybe >map a PART of the kickstart rom to location zero? Or is there a >boot rom anyway? > ... The 2000 certainly needs a boot ROM (whether it is called 'boot' ROM or 'kickstart' ROM doesn't really matter). It also needs to give the 68000 the location of that ROM via the reset vector at location 0. Any 68000 based machine that wants to have low RAM (for the interrupt vectors) must somehow trick the 68000 into fetching the vectors from ROM until the RAM ones are ready. The Amiga does this with the OVL line controlled by the CIA. A related question: Is the kickstart disk (for 1000's) an exact image of the ROM version? Does it have the exact same startup code (including reset vector) as the 1000 boot PROMs? I.e., can I just blow PROMs from the kickstart disk and install them in the 1000 (with appropriate changes to disable the kickstart RAM)? -- Ken Laprade ARPA: laprade@trantor.harris-atd.com Harris Corporation - GISD UUCP: ...uunet!x102a!trantor!laprade Palm Bay, Florida voice: (407)727-4433
daveh@cbmvax.UUCP (Dave Haynie) (08/19/89)
in article <8908181206.AA07159@jade.berkeley.edu>, U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) says: > This makes me wonder. The 2000 doesn't need a boot rom, does it? The > start of the kickstart ROM has the same structure as the boot rom > in the 1000. (I know; I disassembled it). Does the OVL line maybe > map a PART of the kickstart rom to location zero? In fact. The base of the ROM ($00FC0000) is overlaid at location 0 when the OVL line is asserted. The same thing happens in A1000s; this overlay tells the system where the start of code is, in standard 68000 fashion. One interesting side effect of this ROM overlay, plus the way autoconfig works, is a noticable lack of RAM after RESET. Most of the time that's not a problem, but if you ever need to write a program that resets the machine and then does something other than jump to the start of ROM, it's a problem. I'd personally be interested in seeing a routine that would reliably do that on a 68000 machine; I don't think it's possible.... > Astounding News! By Express Via Norfolk! OS+2 is a trademark of > The Atlantic Crossed in Three Days! Olaf 'Rhialto' Seibert > [E.A. Poe, 1844] U211344@hnykun11.bitnet -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy We have no choice. We are, after all, professionals.
daveh@cbmvax.UUCP (Dave Haynie) (08/19/89)
in article <2559@trantor.harris-atd.com>, laprade@ringworld.harris-atd.com (Ken Laprade) says: > I.e., can I just blow PROMs from the kickstart disk and install them in the > 1000 (with appropriate changes to disable the kickstart RAM)? That's basically the way we made the ROMs. No code changes. In fact, if you have an MMU based machine and use SetCPU to boot a KickStart, it counts on the reset vector just like if it were a ROM. > Ken Laprade ARPA: laprade@trantor.harris-atd.com -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy We have no choice. We are, after all, professionals.
grr@cbmvax.UUCP (George Robbins) (08/19/89)
In article <8908181206.AA07159@jade.berkeley.edu> U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) writes: > I bought a new Amiha 2000 this week, and looking through the schematics > in the back of the (German) user's manual, I noticed the familiar > (from my A1000) memory overlay line (OVL). It went from a CIA (as > is to be expected) to GARY. > > This makes me wonder. The 2000 doesn't need a boot rom, does it? The > start of the kickstart ROM has the same structure as the boot rom > in the 1000. (I know; I disassembled it). Does the OVL line maybe > map a PART of the kickstart rom to location zero? Or is there a > boot rom anyway? OVL says "Map some ROM to location 00xxxx, so that the power-up/reset can find it's way to the normal ROM location. The A500/A2000 don't need "boot ROM's", but you do need to get started somehow... -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
grr@cbmvax.UUCP (George Robbins) (08/19/89)
In article <2559@trantor.harris-atd.com> laprade@trantor.harris-atd.com (Ken Laprade) writes: > In article <8908181206.AA07159@jade.berkeley.edu> U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) writes: > > ... > >This makes me wonder. The 2000 doesn't need a boot rom, does it? ... > > A related question: Is the kickstart disk (for 1000's) an exact image > of the ROM version? Does it have the exact same startup code (including > reset vector) as the 1000 boot PROMs? I.e., can I just blow PROMs from > the kickstart disk and install them in the 1000 (with appropriate changes > to disable the kickstart RAM)? The kickstart disk is the same as the ROM image for the system ROM's, though normally we make ROM's by loading the diskette the software types provide into an A1000 type machine, and then use an emulator to generate S-record "memory dump" files which are burned into (multiple) PROM's which are sent to the ROM vendor. Somewhat round-about, but it does (theoretically) yield ROMS which are the same as the kickstart image... The ROM's / Kickstart image contains it's own startup code that is supposed to be completely independent of the code in the A1000 "Boot ROMs". The boot roms, are responsible for getting the WCS loaded, then jump to the beginning the same as a reset would. If you happen to have both boot roms and system roms, the boot roms perceive that the ROM/WCS area is alreadly loaded, and skip the attempt to load it from disk... -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
michael@maui.cs.ucla.edu (michael gersten) (08/24/89)
In article <7700@cbmvax.UUCP> daveh@cbmvax.UUCP (Dave Haynie) writes: >have an MMU based machine and use SetCPU to boot a KickStart, it counts on I can use Kickstart again if I get an MMU? Hmm...whats the cheapest MMU board out there? Michael
douglee@becker.UUCP (Doug Lee) (08/24/89)
In article <8908181206.AA07159@jade.berkeley.edu> U211344@HNYKUN11.BITNET (Olaf 'Rhialto' Seibert) writes: >I bought a new Amiha 2000 this week, and looking through the schematics >in the back of the (German) user's manual, I noticed the familiar >(from my A1000) memory overlay line (OVL). It went from a CIA (as >is to be expected) to GARY. > >This makes me wonder. The 2000 doesn't need a boot rom, does it? The >start of the kickstart ROM has the same structure as the boot rom >in the 1000. (I know; I disassembled it). Does the OVL line maybe >map a PART of the kickstart rom to location zero? Or is there a >boot rom anyway? > The reason for this is that the boot rom code must start at location zero. When you reset the 68000, this causes an exception to be generated, which looks for a jump vector at location 0. The first two words point to the supervisor stack pointer and the last two words to the program counter. I realize this is ambiguous as the last two words are not necessarily the second set of two, but this is paraphrased from the Motorola manual itself. When the machine is turned on, the OVR line causes the boot rom to be mapped into location zero. as soon as things get going, this bit is toggled by a write to high ram and you once again have ram at location zero. This is all necessary because the exception vectors go frm location zero to $255. On the Amiga, you want to be able to program these, so ram should start at 0 once it's booted. All this good stuff about the 68000 is described in the MC68000 Programmer's reference manual Pg 35-40 5th edition. > Olaf.Seibert; -- Doug Lee douglee@becker 416-461-5357
daveh@cbmvax.UUCP (Dave Haynie) (08/24/89)
in article <26664@shemp.CS.UCLA.EDU>, michael@maui.cs.ucla.edu (michael gersten) says: > In article <7700@cbmvax.UUCP> daveh@cbmvax.UUCP (Dave Haynie) writes: >>have an MMU based machine and use SetCPU to boot a KickStart, it counts on > I can use Kickstart again if I get an MMU? Sure enough, for what it's worth. If your a developer testing 1.4, it's pretty handy. If you're just trying to run stuff that doesn't work on the current OS, good luck -- programs broken enough to fail under 1.3 have a reasonably good chance of failing on 68020/68030 machines. > Hmm...whats the cheapest MMU board out there? I don't know. Currently, any 68030 board or the A2620 will give you the MMU you require. -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy We have no choice. We are, after all, professionals.