March 2008 - Oberon - mailman3.inf.ethz.ch

startup time & Init strings.
by PETER EASTHOPE 19 Mar '08

19 Mar '08

Chris, cg> How long does your CF take to boot ETH-Oberon ? The Toshiba T2100, 80486, starts from the CF in 23 s. The Toshiba 205CDS, 100 MHz Pentium, starts from the hdd in 30 s. I've added this to http://carnot.yi.org/ExampleSystems.html . Thanks for the suggestion. I've never done any low level programming. Perhaps someone can help with decoding the Init string for VGA, "b81200cd10". Referring the to the One-Byte Opcode Map on page A-4 of the Intel 386 Programmer's Reference Manual, B8 is MOV eAX. What is eAX? Referring to MOV on page 17-114, "B8+ rw MOV reg16, imm16 2 [clocks], Move immediate word to register". So "b81200" means "put 00 into register 12"? Thanks, ... Peter E. http://carnot.yi.org/

1 0

Code Injection: The Movie
by Søren Renner 17 Mar '08

17 Mar '08

There were no movies on sites like youtube which show off Bluebottle or Oberon, so I made one. http://www.youtube.com/watch?v=Pe0ZdzO_urU

1 0

Re: Booters for ETH-Oberon ?
by Chris Glur 12 Mar '08

12 Mar '08

peasthope(a)shaw.ca wrote: > cg> ... OLPC ... > > Why not get an early model Pentium laptop or > recent model desktop? Consumers discard them. Yes, I've got several P6-type scrap boxes, but I want a minimum power portable for when I'm 'out of the office' and also I don't just restrict myself to what I personally need. > cg> How long does your CF take to boot ETH-Oberon ? > > Seconds. Will try to remember to time it this > weekend and answer on the Web page. That's what I'd expect, so I'm surprised that the OLPC which as you would know, as a MIT project hopes to sell millions to 3rd world governments for educating their kids, in places where there's no mains electricity supply; needs 2 minutes to boot it's minimum version of linux. > The computer engineering for CF is thanks to > J. Sedlacek, by the way. Good, that's how it's supposed to work: collaborative sharing of knowledge, by shared documentation. sven.stauber(a)inf.ethz.ch wrote: > The assembler source code of the Oberon boot loader, MBR > and Boot Manager are contained in the AosBuild.zip package > (http://www.bluebottle.ethz.ch/downloads/crazy/AosBuild.zip) > If you run AOS (or Oberon in AOS), you can use the stand-alone > assembler PCAAMD64.Mod to assemble the code. Thanks. Aubrey.McIntosh wrote: > I have an ongoing interest in boot sectors and booting Oberon (V4 > mostly). This interest spans all of the process beginning when > execution is transferred to 7C0:0000 until the user has control with a > mouse click with an application on the screen. It includes PXE. > I have done some reading related to this interest. > > I ran AOS from CF on the CV860A > <http://www.lex.com.tw:8080/product/CV860A.htm> for more > than a year, with reboots. This machine did not have any moving parts. > > I am currently engaged in a project to put an Oberon application on a > FLASH based SoC, the PIC32, although I don't know whether this will be > V4 based, or Component Pascal based. In support of this process, I am > studying the BootLoader and the power on starting sequence in this > processor. As an aside, I am really enthusiastic about the MIPS > processor since I learned about the PIC32. MIPS turns out to be in > the WiFi router next to my desk, and a host of other places. If that's from the PIC family of Microchip I've got bad experience. I hate it when they take a good wheelbarrow and extend it into a cadilac. They started out with minimum 14pin devices, when the PICforth compiler was appropriate for a device which I was considering. Then they kept on upgrading, so the whole initial idea became absurdly inappropriate. > What do you hope to do? I believe in part of the ideas of OLPC: that low power computing can revolutionise 3rd world education and hence development the same way as the cell phone *HAS* helped development. How do you generate the non-86x object code from oberon source ? > I do note, as an aside, that the state of an Oberon system seems to > be reflected in the linkages available off of Modules.module. If you > store this to disk on a native machine, and re-read it at boot time, > you should be back where you were. OK, does that 'state' include the tree of displayed viewers or just which modules were loaded at the time ? If I wanted to 'do' something that required a module that wasn't currently loaded, it would just auto-load. What matters to continue at my pre-power-down state is my 'data': equivalent to which-files-were-open in other OSs ? soren.renner wrote: > This discussion reminds me of how nice NO was. > I develop in Bluebottle now -- but I really miss the mouse chords. > Maybe they could be added back -- the modeless UI , might be > difficult but chording for cut/paste with a buffer ? > Also greatly missed is TextPopups. Thanks for mentioning that. I half suspected that BB/AOS had lost the main benefit of S3 [for me]. You've saved me the waisted effort of trying BB. == Chris Glur.

3 2

Re: [Oberon] Booters for ETH-Oberon ?
by PETER EASTHOPE 09 Mar '08

09 Mar '08

Chris, cg> ... the amount of mechanical-engineering that you had to do to fit it ... Not really that bad, now that the details are documented. Think of the hours most of us have spent changing drives and moving data. Support for a PCMCIA-CF adapter is an alternative; certain to consume more time than installation of an ATA-CF adapter. cg> * there are possiblilities of inconsistencies at several stages: the IDE's C,H,S, the BIOS, lilo/Grub..etc. M$loth wants to kill any co-resident foreigners ...etc. A few simple policies prevent most of the trouble. * If a MS system is needed, put it in the first partition. * If the machine has no systems other than Oberon and Aos, use BootMan. If Linux or DOS or Windows is present use Grub. * If you want to boot Oberon or Aos from a drive other than the first, open http://www.oberon.ethz.ch/betadocu.html and apply the instructions under "Partitions.ShowBlocks ... Troubleshooting a boot problem ..." cg> I've only ever had one ETH-Oberon that didn't need a fd0 to boot. Diskettes are mostly for installation and maintenance now. Booting from a diskette shouldn't be a routine. cg> ... OLPC ... Why not get an early model Pentium laptop or recent model desktop? Consumers discard them. cg> How long does your CF take to boot ETH-Oberon ? Seconds. Will try to remember to time it this weekend and answer on the Web page. The computer engineering for CF is thanks to J. Sedlacek, by the way. Regards, ... Peter E. http://carnot.yi.org/

1 0

Re: Oberon Digest, Vol 54, Issue 1
by Søren Renner 09 Mar '08

09 Mar '08

This discussion reminds me of how nice NO was. I develop in Bluebottle now -- but I really miss the mouse chords. Maybe they could be added back -- the modeless UI , might be difficult but chording for cut/paste with a buffer ? Also greatly missed is TextPopups.

1 0

Booters for ETH-Oberon ?
by Chris Glur 08 Mar '08

08 Mar '08

Sven Stauber wrote: > The boot loader is located in the partition where the file > system is located (not in the MBR!). The assembler source > code of the Oberon Bootloader is contained in the release > (BBL.Asm). Also, the sources of MBR (OBEMBR4.ASM) and the > boot manager (BootManager.Asm) are available. Have you seen these ? Where can I get them ? ------------- Peter E. wrote:- > The Toshiba T2100 laptop boots from the CF card. > The modifications are described in detail. > Open my shop page. http://carnot.yi.org/ > At the bottom of the page open Example computer > systems. Scroll down to T2100. Yes I had read your CF installation, but I was so overwhelmed by the amount of mechanical-engineering that you had to do to fit it that I didn't realise that you actually booted from the CF. Bootloaders are complex/problematic: * you often need to work at asm-level. N-O's Decoder.Mod is not perfect & was removed from the packages ? * there are possiblilities of inconsistencies at several stages: the IDE's C,H,S, the BIOS, lilo/Grub..etc. M$loth wants to kill any co-resident foreigners ...etc. I've only ever had one ETH-Oberon that didn't need a fd0 to boot. I'm also interested in soid-state-disks in connection with the OLPC [one laptop per child] project, which I've been following. My initial criticism seems flawed since even if the x86 CPU uses much more power than an ARM, the 0.8Watts used is a small part of the power needed by *any* significant LCD-display. And the OLPC has got a specially power-economic display. The OLPC takes 2 minutes to boot linux. How long does your CF take to boot ETH-Oberon ? I guess ETH-Oberon would not shine on the OLPC because OLPC is not conveniently mouse-chordable -- AFIAK ? I'm including the extract below from OLPC, because it's the first confirmation of a principle which is central to me, that I've seen: identifying entities by their 'recentcy' is more usefull than eg. alphabetically by name [and is handled nicely by mc, which I mentioned]. That's why stacks are usefull. Perhaps different users have different thinking methods ? Evidently many people prefer to search alphabetically for ThatFunnyNamed873File than 'the one I used yesterday' ? http://www.olpctalks.com/walter_bender/ walter_bender_olpc_meeting.html > then another thing that we're doing, is we have this concept of a > journal. Now when you had asked about GTK and the file menu, > hierarchical file systems are great-- nothing wrong with it-- but it's > not the only way to do things, and we think that for young children, > there's a lot of cognitive studies about this [...] used to be a work > with Seymour Papert and Piaget. They studied this in great detail. > > We decided that the primary mechanism by which the children will > interact with their work, will interact with their computer, is > through a temporal interface rather than a hierarchical interface. > > So what we do is we maintain a journal or a diary of all the > interactions with a laptop. And so the laptop automatically > accumulates a portfolio of their work. And they use that diary to go > back to return to something they were doing, so in this case this is > an entry. These are the people they were drawing with. Here's some > comments in the diary. You can go to the previous entries. You can use > this diary entry to re-engage, to re-launch the activity. It sort of > adds a sense of persistence from day to day. If you've got to recall a > project, you don't have to reassemble everything. You click it on the > diary and all the pieces come back together. == Chris Glur.

4 3