July 2026 - Oberon - mailman3.inf.ethz.ch

Re: [Oberon] Export of Oberon Mail.SendCmd().
by Joerg 30 Oct '26

30 Oct '26

Generally, re-using code instead of copying is a good approach. If you do so, then the whole code of Mail is imported in News. Possible but perhaps there is even a better approach to re-use code. What about this: try to separate all common code in an own module (e.g call it Communication.Mod or MsgBase.Mod or Utilities.Mod or so) and import this new module in Mail and in News. br Jörg > Am 14.06.2023 um 19:53 schrieb peter(a)easthope.ca: > > Hello again, > > Several procedures in Mail are used in Oberon News. Rg.: > Mail.ParseContent(), Mail.DecodeMessage() & Mail.GetSetting(). > > Is export of SendCmd() and use in News objectionable? > Ie. in Mail, change PROCEDURE SendCmd(...) to PROCEDURE SendCmd*(...). > > Thanks, ... P. > > - > mobile: +1 778 951 5147 > VoIP: +1 604 670 0140 > > -- > Oberon(a)lists.inf.ethz.ch mailing list for ETH Oberon and related systems > https://lists.inf.ethz.ch/mailman/listinfo/oberon

1 0

Re: [Oberon] clock_gettime
by peter＠easthope.ca 01 Aug '26

01 Aug '26

From: Felix Friedrich <felixf(a)ethz.ch> Date: Thu, 16 Jul 2026 08:15:45 +0200 > ... set AOSCONFIG="TraceModules 1" ... before starting A2? Yesterday without "TraceModules 1". https://easthope.ca/A2trap2026.07.15.txt Today with "TraceModules 1". https://easthope.ca/A2trap2026.07.16.txt This Linux with Busybox has limitations not present in most Linuxes. The failure occurs early in A2 startup and I should find a way to follow that? Thx, ... P. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioned Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

3 4

Re: [Oberon] A2 start. Was "clock_gettime."
by peter＠easthope.ca 31 Jul '26

31 Jul '26

From: Felix Friedrich <felixf(a)ethz.ch> Date: Wed, 22 Jul 2026 08:32:05 +0200 > As Michael wrote you probably need to > > export AOSCONFIG="TraceModules 1" Acknowledged and the 32 bit system produces output as you show although ending at "GC mode : metadata". For additional insight I thought of trying the Linux32 A2 on my 64 bit Linux EasyOS. Reviewed this Debian Multiarch page. https://wiki.debian.org/Multiarch/HOWTO%C2%A0%C2%A0 Then apt-get install "libc6:i386" ... The following additional packages will be installed: gcc-14-base:i386 libgcc-s1:i386 Suggested packages: glibc-doc:i386 locales:i386 libnss-nis:i386 libnss-nisplus:i386 Recommended packages: libidn2-0:i386 The following NEW packages will be installed: gcc-14-base:i386 libc6:i386 libgcc-s1:i386 0 upgraded, 3 newly installed, 0 to remove and 128 not upgraded. Need to get 2,862 kB of archives. After this operation, 13.1 MB of additional disk space will be used. ... Appeared to complete OK. Therefore. # cd A/A2/Linux32 # ./a2.sh ./a2.sh: line 2: 22231 Trace/breakpoint trap ./oberon run a2.txt Comments? Has anyone seen Git based Linux32 A2 working on a 32 bit system? On a 64 bit system with multiarch? Thanks, ... P.L. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioned Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

2 2

A new low-cost Oberon for RISC5 FPGA development board
by Chris Burrows 25 Jul '26

25 Jul '26

Applications have been written in Oberon to run on FPGA hardware using the new low-cost (EUR 50) Olimex GateMateA1-EVB open-source development board: https://www.olimex.com/Products/FPGA/GateMate/GateMateA1-EVB/open-source-ha… Not only is this board designed AND manufactured in Europe, the Cologne Chip GateMate FPGA device that it uses is also designed AND produced in Europe: https://www.colognechip.com/programmable-logic/gatemate/ Also remarkable is that the Verilog tools used to build the GateMate FPGA bitstreams are 1/10th the size of the previous Verilog tools we used (500 Mbytes vs. 10 Gbytes) and result in RISC5 Verilog build times that are approximately 10 time faster (1 minute vs. 10 minutes): https://github.com/YosysHQ/oss-cad-suite-build The Astrobe for RISC5 FPGA Oberon development system is an embedded software system comparable to Astrobe for Cortex-M / RP2xxx. i.e. Standalone RISC5 Oberon applications are compiled and statically linked on a PC. They are then uploaded via a serial link to the Olimex development board where they run independently of an operating system. For more information see: https://www.astrobe.com/RISC5 The free Astrobe for RISC5 software targeting the GateMate FPGA is curently undergoing final testing. A pre-release version is planned before the end of this month. Register your contact details if you want to get your hands on a copy as soon as it is available: https://www.astrobe.com/RISC5/register.htm -- Regards, Chris Burrows CFB Software https://www.astrobe.com/

1 1

Re: [Oberon] A2 start. Was "clock_gettime"
by peter＠easthope.ca 19 Jul '26

19 Jul '26

From: Michael Schierl <schierlm(a)gmx.de> Date: Sat, 18 Jul 2026 14:09:29 +0200 > Depending on shell, ... # echo $0 /bin/bash > ... you may have to > "export AOSCONFIG" so that subprocesses can see it. Three console transcripts now. https://easthope.ca/A2trap2026.07.15.txt https://easthope.ca/A2trap2026.07.16.txt https://easthope.ca/A2trap2026.07.18.txt A2trap2026.07.18.txt has "export AOSCONFIG". The console for A2 Linux64 has these two lines, GC mode : metadata WORK: mounted whereas Linux32 shows "GC mode : metadata" but no "WORK: mounted". Will look there for a snag there. Thx, ... P. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioned Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

1 0

Re (2): clock_gettime
by peter＠easthope.ca 15 Jul '26

15 Jul '26

Hello Felix & all, From: Felix Friedrich <felixf(a)ethz.ch> Date: Wed, 15 Jul 2026 07:51:57 +0200 > The procedure variables offered by Linux.Unix.Mod should (but > possibly don't) form the minimal set of procedures required to run > the system. Not certain I understand. You suggest Linux.Unix.Mod may be invoking a Linux procedure not available in the extant system; correct? > Do you have more output after that Trap 11 line? Ref. https://easthope.ca/A2trap2026.07.15.txt Four traps. A trap for each core in the machine here? > Maybe in this case an easy fix is to offer a different Timer module? Unfamiliar territory for me. Appears I should study the evolution of Linux clocks from the 32 bit machines to 64 bit. Then try to modify Linux.Unix.Mod to invoke procedures accordingly. Significant project. https://linux.die.net/man/2/clock_getres has "Note for SMP systems". Thanks, ... P.L. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioning Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

1 0

clock_gettime
by peter＠easthope.ca 15 Jul '26

15 Jul '26

Hi, ================================= Starting A2 system. Unix.Dlsym: entry 'clock_gettime' not found Unix.Dlsym: entry 'clock_getres' not found GC mode : metadata LinuxA2 Gen. 32-bit, Apr 19 2023 2026/07/05 06:04 Trap 11 (Segmentation violation) ================================= Clock access in an older Linux differs from contemporary Linux. I wonder whether Linux.Unix.Mod can determine what is available before invoking something certain to fail. Then invoke something which works. Linux.Unix.Mod has this. Dlsym( libc, "clock_gettime", ADDRESSOF(clock_gettime) ); ... clock_gettime- : PROCEDURE {C} (clk_id: INTEGER; tp: ADDRESS): INTEGER; Ideas? Thanks, ... P.L. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioning Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

3 2

Re: [Oberon] clock_gettime
by peter＠easthope.ca 12 Jul '26

12 Jul '26

>From A2/source/Linux.Unix.Mod, CLOCK_MONOTONIC* = 1; CLOCK_MONOTONIC_RAW* = 4; CLOCK_PROCESS_CPUTIME_ID* = 2; CLOCK_REALTIME* = 0; CLOCK_THREAD_CPUTIME_ID* = 3; ... clock_gettime- : PROCEDURE {C} (clk_id: INTEGER; tp: ADDRESS): INTEGER; clock_getres- : PROCEDURE {C} (clk_id: INTEGER; res: ADDRESS): INTEGER; Web searches reported that the Linux kernel introduced CLOCK_MONOTONIC in 2.5.63, CLOCK_MONOTONIC_RAW in 2.6.28, CLOCK_PROCESS_CPUTIME_ID in 2.6.12, CLOCK_REALTIME in 2.6.0, CLOCK_THREAD_CPUTIME_ID in 2.6.12, clock_gettime in 2.6 and clock_getres in 2.6. Whereas in a 32 bit machine here, uname -a reports, Linux puppypc 2.6.35.13_xo1.5-20110713.2219.olpc.da7074b_Puppy . So the clock calls should work. Ideas? Thanks, ... P. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioning Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

1 0

Re (2): How is RTOS involved in running Linux on RISC-V?
by peter＠easthope.ca 08 Jul '26

08 Jul '26

From: Liam Proven <lproven(a)gmail.com> Date: Sat, 4 Jul 2026 16:20:57 +0100 > If this is Google's automatic-idiocy bot, DO NOT TRUST IT. Sometimes I encounter false information on the path to truthful information. Critical thinking is essential. Thx, ... P. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 <= decommissioning Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

2 1

How is RTOS involved in running Linux on RISC-V?
by peter＠easthope.ca 04 Jul '26

04 Jul '26

Hi, Google explained that the two can coexist on a multi-core RISC-V with RTOS handling some hardware level tasks. For more verbiage, ask Google. TTFN, ... P.L. -- mobile: +1 778 951 5147 VoIP: +1 778 508 0020 Bcc: peter at e a s t h o p e dot c a projects: en.wikibooks.org/wiki/User:PeterEasthope

3 2