September 2014 - Oberon - mailman3.inf.ethz.ch

Spartan 6 Kickstarter project
by Aubrey.McIntosh＠Alumni.UTexas.Net 12 Oct '14

12 Oct '14

There is a Kickstarter project to produce a Spartan 6 board. This reached its funding goal in 4 days and will run for about 3 more weeks. https://www.kickstarter.com/projects/1812459948/minispartan6-a-powerful-fpg… -- Aubrey McIntosh, Ph.D. 1502 Devon Circle Austin TX 78723-1814 (512)-348-7401

6 5

RISC5 clock speed and interrupt context switch
by skulski＠pas.rochester.edu 06 Oct '14

06 Oct '14

Hello: I am exploring the FPGA-based Project Oberon. Two questions popped up after reading the hardware portion of the new Project Oberon and the NW Technical Reports from the FPGA-related part of the personal website. 1. The RISC5 clock speed is 25 MHz. Can it be improved to the Microblaze ballpark that is over 100 MHz? 2. The interrupt HW system. A fast interrupt service can be achieved with a set of shadow registers for context switching, used on Analog Devices ADSP21xx series and SHARC. Has this been considered for the RISC5 interrupt system? I am asking because improving the performance in these two areas could make RISC5 (or its derivatives) an attractive option for deeply embedded signal processing. Regards, Wojtek

7 11

Nimrod is not Oberon
by Stefan Salewski 02 Oct '14

02 Oct '14

Yes, I am still subscribed to this list... And I noticed that "Oberon on FPGA" project, but have not read much about it. Recently, when I was looking about some news about Mozillas Rust language, I discovered Nimrod, a language which looks similar to Modula/Oberon, with some interesting new features. Nimrod is not very popular now, some people simple do not like the Wirthian style syntax. (And of course there are many other interesting new languages available now, Mozillas Rust with C syntax, Julia for numerical computing, Haskell -- not really new, but still modern...) But maybe someone of you may be interested in Nimrod. Of course it is no operating system, just a modern language. Keywords: Wirthian syntax, with Python indentation, system- and high-level programming, garbage-collector which still allows time critical applications, advanced generics, templates, high-level macros. Good support for parallel processing. Speed seems to be close to C. Currently Compiler generates C intermediate code. Currently there is no wikipedia entry (was rejected) so it is difficult to discover it. There was an article in DrDobbs Journal recently, and of course there is a homepage with a tutorial. http://nimrod-lang.org/ http://www.drdobbs.com/open-source/nimrod-a-new-systems-programming-languag… Not a high rank at http://langpop.corger.nl/ but Oberon is not listed at all ;-) Best regards Stefan Salewski

3 3

How is [non-gadget] MenuFrame darkened?
by eas lab 25 Sep '14

25 Sep '14

Since I could no longer manage the small fonts, last year I hit-out and got a BIG phillips 1920x1080. These days I use linux:Slackware:V13. This allows me to run multiple instances of LNO & LEO. Only after navigating through several programs, workspaces, terminal is it slowly evident that the display has become <unclear>. The problem is subtle with many programs, but obvious with our LNO & LEO: the MenuFrame is not <halfDark> as it was soon after boot. Without the MenuFrame marked, it's very difficult to work. Connecting the old CRT, shows it as it should be. So the problem is with the display. OTHO, running a different [not preferred] linux version, also is OK. The display is a monster, in that it has a menu-tree with dozens of leaves! Since the switch from initially-good to bad always happens [so far] while I'm navigating one of several apps. which don't show the symptom, it would need a research-project to debug, what state transition/s cause the problem. If someone could describe exactly *what* ETHO does to the MenuFrame, then I could perhaps have the correct wording for a google search, to solve this problem. == Chris Glur.

1 0

Version extentions.
by eas lab 20 Sep '14

20 Sep '14

As part of my attempt to recover/consolidated my decades of N-O files, I installed the latest [a few years old] LNO. Both the FrameBuffer & X version are improved over my previous version. Like an early version of ALO char("*") needs 2-fingers. Since "*" is the most used char, one wonders if/how the developer ever took the product for a serious drive. I.e. what tricks am I missing. --- Also reading J.Verhoeven's VruttenBoel ETHO webpages: shows "ETH Oberon (<ver 2.4.3>) for Linux x86", named as LNO. I should perhaps take blame for naming: "ETH Oberon (<ver 2.4.3>) for Linux x86" , LEO. Little wonder that Duke got confused and frustrated, and made more forcefull queries. He might have become a valuable user/contributer by now, if not expelled. The version extensions of ETHO: DOSbased, Native, LNO, LEO, ALO have been very successful IMO. What does this tell us about ETHO's essential design? IMO linux's version forking is a pity. == Chris Glur.

1 0

Re: [Oberon] Urgent help to recover ETHNO files.
by peter＠easthope.ca 19 Sep '14

19 Sep '14

> Vaguely, I remember tracing the location and structure of N-O's directory > within the partition/S. N-O's Partitions.ShowBlocks IDE0#00 0 2 ~ These might help, for a start at least. http://www.ethoberon.ethz.ch/bootmbr.html http://www.ethoberon.ethz.ch/betadocu.html Is the data human readable? If so you can connect the hdd to a working system. Then apply Partitions.PartitionToFile dev#part file [block blocks] ~. Then use a text editor on file to sort your data. > ... we used partitions to separate topics. > This worked quiet well, except that if you had 10 topics, and 2 backups, > that needed 30 partitions, just for data. An alternative is to define a syntax, such as in SGML or XML, and contain all the data in one or a few files. Regards, ... Peter E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 12 Tel +1 360 639 0202 http://carnot.yi.org/ Bcc: peter at easthope. ca

2 1

Re: [Oberon] Pitch from AST: Minix on Beagleboard
by Frans-Pieter Vonck 19 Sep '14

19 Sep '14

> What's the power consumption on these boards? http://en.wikipedia.org/wiki/BeagleBoard#Specifications_2 > Also it's been pointed out that full-linux on rPi is no good I agree that the Pi is to slow to use it as a full (flash) blown desktop. On the other hand, I made a videowall out of 15 raspberries connected to old crt tv's and it worked great.* I programmed the project in shell scripts over ssh networks. I was thinking of using oberon as the language for the main program. Especially of using the messageing architecture of the Oberon System for communicating with the video clients. In my quick fix however, in linux, I had to ensure a working state of the clients and the server I programmed reboots for every occuring anomality.** I love the idea of the single user, single task system on top of the linux threading system. On the other hand, Tanenbaum is also using a messaging system for Minix to ensure the state of the operating system. And the system can be run in single user mode. It would be great if Oberon07 was part of the Minix distribution. There is already a modula-2 compiler in the ACK (Amsterdam Compiler Kit).*** Greets, Frans-Pieter Vonck *I've looked into using native AOS for this video project, but the ready to use software and documentation for the Pi, made me decide not to. In honour of Jurg Gutknecht - I will keep on striving to do an artproject with AOS one day. ** Sounds familiar? https://www.hq.nasa.gov/alsj/a11/a11.1201-pa.html "Those guys were lucky" - Edsger Dijkstra. *** Check out Ceriel Jacobs work at the Free University http://www.cs.vu.nl/~ceriel/ > I've been well impressed by the rPi, as a potential replacement PC; but my everyday use is as a sound-player for TextToSpeech [I don't read big text any more; I lie down & have them read to me like a baby]. > > But rPi seems to use 100-times the power of the <Chinese button sized audio player>, which doesn't allow me to 'mark' the files as <read, to be repeated ...etc.> > > I'm getting interested in investigating some potential Oberon7 based projects/products [even if I'm forced to use M$loth tools]. > But I didn't get good info from Astobe's URL. > > Also it's been pointed out that full-linux on rPi is no good for even-nearly real time. Often rPi just <takes a coffee-break for > a few seconds> perhaps it's doing a garbage collection. > > OTOH probably MIMIX is much less of a hog? > But then the <simple single-loop concept> OS of ETHO is even better. But, I'm wondering how/why LinuxBasedETHO hangs the pid if I send a NNTP-post and the communication cuts before it returns. > Obviously, the <send procedure> of ETHO 'hold' the CPU, and the > published <round-robin task polling> is prevented. > LEO has a "threaded" version/capabilty, which I've never tested, because I prefer KeepItSimple. > > ==Chris Glur. > > > On 9/16/14, Frans-Pieter Vonck <fp(a)vonck.nl> wrote: >> Hello Oberoneurs, >> >> >> Here you can see Andy's Tanenbaum' s pitch at EW Neurenberg of the minix 3 >> system running on a >> beagleboard. >> http://youtu.be/vlOsy0PZZyc?t=57s >> >> It could be an interesting example for porting oberon sytem to the arm. Especially if you look at how they solved driver issues. >> >> By the way, everybody is invited for Tanenbaum's retirement farewell lecture at >> 23 October 2014 in the Aula of the Vrije Universiteit, >> http://www.cs.vu.nl/~ast/afscheid/index.php >> >> Guess he has now he has time enough for porting minix to the FPGA, like Niklaus! >> >> Paul, do you already know Kees Jongenburger? >> https://archive.fosdem.org/2014/schedule/event/minix_3_on_arm/ >> >> >> Greets, >> Frans-Pieter >> >> >> >> >> >> -- >> Oberon(a)lists.inf.ethz.ch mailing list for ETH Oberon and related systems https://lists.inf.ethz.ch/mailman/listinfo/oberon >> >

1 0

Pitch from AST: Minix on Beagleboard
by Frans-Pieter Vonck 19 Sep '14

19 Sep '14

Hello Oberoneurs, Here you can see Andy's Tanenbaum' s pitch at EW Neurenberg of the minix 3 system running on a beagleboard. http://youtu.be/vlOsy0PZZyc?t=57s It could be an interesting example for porting oberon sytem to the arm. Especially if you look at how they solved driver issues. By the way, everybody is invited for Tanenbaum's retirement farewell lecture at 23 October 2014 in the Aula of the Vrije Universiteit, http://www.cs.vu.nl/~ast/afscheid/index.php Guess he has now he has time enough for porting minix to the FPGA, like Niklaus! Paul, do you already know Kees Jongenburger? https://archive.fosdem.org/2014/schedule/event/minix_3_on_arm/ Greets, Frans-Pieter

2 1

ALO usage Log/Query ?
by eas lab 07 Sep '14

07 Sep '14

This rambling journey exposes some exercising of the current ALO version. A reason for making/keeping detailed logs, is confirmed by my difficulty to 'pick up the threads' of ... Tue, Jan 29, 2013 at 2:39 PM Subject: [Oberon] HOW2 get Token at cursor? <paulreed(a)paddedcell.com> ] Have a look at TextFrames.Call(), which calls Oberon.Call() after setting ] up the parameters to the command. ... Yes, as advised I've consulted PO.System.pdf, and even [listen while sleeping, via wily, which is a killer-ap for linux-ETHO-users] to: 4.4.3. Cursor management. === A certain project needs the user to be able to roam-the-screen with the mouse-in-right-hand and left-hand-press-any-key when the cursor is on specific tokens/words. BTW, I'm writing this all in ALO [in color - of course]:--------------- Real task to exercise ALO140731Z Trace co-ordinates of cursor <- Find "ursor" in *.Mod, where *.Mod is <loaded>. ?! but ALO will use *nix-X cursor control !? Binary file ./ALO.DisplayX11.Mod matches Binary file ./ARM.Display.Mod matches Display loaded. Display3 loaded. Display3.Mod <--no mention of "ursor" GfxDisplay.Mod <--no mention of "ursor" =?=> why/what difference: ARM.Display.Mod : Display.Mod =nX ARM.Display.Mod == ... (** Move hardware cursor to (x,y) *) PROCEDURE MoveSprite*(x,y : LONGINT); (** non-portable *) ... SYSTEM.PUT(index,51X); SYSTEM.PUT(data, CHR(x DIV 100H)); SYSTEM.PUT(index,53X); SYSTEM.PUT(data, CHR(y MOD 100H)); spriteX:=x; spriteY:=y END MoveSprite; =?=> HOW2 trace spriteX:=x; = Check Hello.Mod = Out.String("spriteX:=x"); Out.Int(spriteX,8); Out.Ln; BACKUP: ARM.Display.Mod => ARM.Display.ModORGNL =?=> Can't compile a running/used module ? -> ARMCompiler.Compile * == compiling Display pos 1055 err 152 symbol file of imported module not found -> System.Directory *Display* == Display.Sym -> System.Free Display == Display reference count not zero <- being USED ?!! !!!!!!!!!! Since the present system's compiler can't run except when the Display is active, how can this problem be solved ? Last year, to finally get output from <cusorPikToken>, I made a module, similar to an existing one that gave the X,Y values -- IIRC? == Chris Glur.

2 7

Urgent help to recover ETHNO files.
by eas lab 07 Sep '14

07 Sep '14

I've lost/broken the extended-partitions-chain of my valuable old NativeOberon IDE-disk. Which also contains the logs and instructions of how I previously fixed such! Because of NO's flat-file-sytem, we used partitions to separate topics. This worked quiet well, except that if you had 10 topics, and 2 backups, that needed 30 partitions, just for data. Vaguely, I remember tracing the location and structure of N-O's directory within the partition/S. N-O's Partitions.ShowBlocks IDE0#00 0 2 ~ which gives eg:-- IDE0#00 0 00000000 EB 3E C0 8E D0 BC 00 7C 8B F4 50 07 50 1F FB FC .>.....|..P.P... 00000010 BF 00 06 B9 00 01 F2 A5 EA 1D 06 00 00 BE BE 07 ................ 00000020 B3 04 80 3C 80 74 0E 80 3C 00 75 1C 83 C6 10 FE ...<.t..<.u..... ... can be conveniently duplicated by LEO's:- System.Execute dd if=/dev/hdb count=2 | hexdump -C It sound easy: * from saved partition data: you know the start & end sector of each partition. * using `dd` as above, either directly from *nix or more conveniently [to capture the logs] from LEO, you can examine any sector; or even do: <skip to SectorN, count Msectors | output in HexDumpFormat | but show only lines containing my searchKey>. So far the dataFlood has overwhelmed me. Q1. Is NO's file-directory started at a fixed-sector-count from the sector start? Where is it? Q2. Where's the info about NO's file dir? Many apparent NO:file-directory sections have been seen, which have convenient, multiple of 16-byte size; but their start locations differ between different partitions. As if the File-Dir is floating. For some years now, I've used only LEO, which has/needs no knowledge of my above questions. Under *nix old N-O files are most conveniently accessed by LNO. == Chris Glur.

1 0