Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - Andi B.

Pages: [1] 2 3 4
Hardware / Asrock B550M Pro4 - anyone tested this?
« on: July 09, 2024, 01:17:38 pm »
This board looks pretty good to me regarding all the interfaces I like. Anyone tested this in CSM/legacy mode with a Ryzen 7 with integrated graphic (5700G I wanna have)?
- Memory available to OS/2 > 3GB?
- Ryzen 7 graphic with Panorama and 3840x2160 working?
- M2 slots working for boot device?

I know there is a thread about the B450M Pro4 but B550M Pro4 I would like.

Off Topic discussions / Which Linux desktop for long time OS/2 users?
« on: July 07, 2024, 12:30:18 pm »
I've played a bit with Linux systems in the last months and I'm a bit familiar with Debian now. I've a NAS running Debian and OpenMediaFault which also serves my git repositories. I'm also running a Debian box with MATE which I connect with RDP from my main OS/2/eCS/ArcaOS system when I need a newer browser.

MATE is not a bad choice. But what I didn't find with any Linux I tried until now is, how can I convince the desktop to remember the window position of my programs. I understand there are people outside there who work on very small displays like with notebook and they 'full screen' every program the use. But I'm used to big displays. 2 times 24" at least or 42.5" are my standard displays. I need a window manager like the WPS who respect my needs. Not the needs of smartphone junkies or notebook users only. My mail window have to stay near top right. Chatzilla near bottom right (but not below or over some task bars). Browser windows near the center/bottom. Various other programs have to be placed on the different places I closed them last time.

Which Linux window manager or which setting can do that the way it was standard the last decades? Did I overlook some setting or helper program? On later Windozes I've to use a WindowManager utility for this. At least there is a solution. But on Linux?

Btw. RDP should work too. I remember that's the reason I didn't test KDE plasma and Gnome at first as I understand they use Wayland which TTBOMK does not work with RDP.

I'm trying to build a dll which should export some functions and global variables. I made this long time ago I think with our gcc3.xx but now want to do the same with our g++.exe (GCC) 9.2.0.

I think exporting symbols is disabled now by default so I added to the functions I want to export
Code: [Select]
__attribute__ ((dllexport)) which works fine. Now when I look at my dll with
Code: [Select]
lxlite /C:exp myown.dllI see the exported functions as I expected.

But my global variables still have the C naming convention leading _ which I don't like. F.i. I have -
Code: [Select]
00048 │ _NdpAttribute
00049 │ _NdpPropertiesInfo
00050 ┴ _NdpTypes
but I wanna have them without the _

The variables are defined that way -
Code: [Select]
__attribute__ ((dllexport)) extern const char *NdpTypes[];and I use g++ instead gcc. So I thought it should work the same as with the functions (they don't have the _).

Any thoughts?

Programming / PMPrintf package v2.80 and PMPrintf_abtrace code samples
« on: April 21, 2024, 07:43:39 pm »
I've packed together my latest code base of PMPrintf and packed it together as v2.80. See

PMPrintf sends standard printf formatted debug messages into a queue. If the PM viewer application (PMPrintf.exe) or CLI logger (PmPfLog.exe) is running it will display these queued data. Otherwise they go to NULL. Log files can be written to disk in parallel too.

If you never used PMPrintf I strongly suggest looking at the code and usage samples at These samples loads pmprintf.dll dynamically if it is available on the target system. If not your program will normally run and your debug messages goes to NULL. There is no need to make extra release and debug builds. The small overhead will be negligible most of the time. A programmers life can be so easy :-).

Have fun.

Programming / Parameter passing to dll - difference VAC / GCC - varargs
« on: January 20, 2024, 01:50:50 pm »
I've a dll which exports functions which I wanna call from VAC, GCC (and Watcom) compiled programs.

Problem is gcc seems to pass varargs different than VAC.
I've a function in the dll something like
Code: [Select]
ULONG         _System   xPmPrintf(char *Format, ...);which works when called by programs compiled with VAC3.65 but produces garbage when called by a gcc test program (except the first arg).

I found here that Dmitriy noted about a problem with gcc and varargs but can't find a solution.

Any hints?

Can someone please remind me where I find the ignore user setting? My KI / KNI / AI / NAI / useless non-related nonsense detector get triggered regularly.

Programming / Where to find Toolkit2.0 - SOM compiler .sc files....
« on: December 23, 2023, 04:58:35 pm »
I've toolkit 2.1 and probably all newer ones. Even Watcom10.6 includes some .sc files. But the project I'm working on seems to need Toolkit v2.0.

I've no clue about SOM and classes and.... but there is one .csc file in my project which constantly bothers me. There should be a way to convert this .sc file to the newer .idl form. ctoi.cmd which is included in os2tk45. But it complains about a file in toolkit21
Code: [Select]
{0}[e:\_work\pmprintf\trunk] ctoi WPSTEST.CSC
"", line 440: error: Override method "somInitClass" not in base class.
"", line 444: error: Override method "somNew" not in base class.
"", line 448: error: Override method "somRenew" not in base class.
Number of errors: 3
error in converting WPSTEST.CSC

So I think I need toolkit20. Where can I find it? Any other ideas? Maybe I try to disable all these SOM things in pmprintf cause I guess nobody will ever use pmprintf with som classes anymore anyhow.

I connect to some Linux box via SSH. After figuring out all the keys stuff it works. But now I've updated to 'OpenSSH_8.4p1, OpenSSL 1.1.1s  1 Nov 2022' and see some problems/differences. As I'm not familiar with all these SSH and Linux stuff I want to ask here before opening a ticket.

Old version 'OpenSSH_5.3p1, OpenSSL 0.9.8k 25 Mar 2009' worked fine. I've colored dir (dir, ls -la, ls -ls, ...) but with new version 'OpenSSH_8.4p1, OpenSSL 1.1.1s  1 Nov 2022' the colors are gone. I know I've setup these colors in some conf file on the Linux box (I think) but don't remember where anymore. But there must be a difference between our two ssh versions. Cause I can connect with the old version at the same time as with the new ssh.exe version. Old has colored dir, new not. I'm using the same user of course.

I also noticed that with the old version 'nano' works while with the new version nano gives an 'Error opening terminal: os2.'.

Anyone knows what this 'terminal: os2. means? Or how to change? Or how to find out why the old version works better than the new one?

Additional info - the old ssh version was installed here in /usr/local/bin. No clue from where I got it but most probably not from the netlabs repository as it would have been in /usr/bin then.

I've tried FreeRDP to connect to different Linux boxes. It works but using FreeRDP the Linux host uses EN keyboard layout. All my Linux boxes does have German keyboards attached and work as expected when I'm sitting in front of them. My OS/2 system has a German keyboard too. But when I connect from my OS/2 system a Linux host via FreeRDP they get the English keys instead the German ones.

I don't see any setting in our FreeRDP port to translate the keys. Is there anything I'v to adjust on Linux (host) side?

Applications / LarsenCommander - new test version
« on: May 19, 2023, 12:07:08 pm »
Since a while I'm working on some problems in LarsenCommander again. There is still one problem I want to resolve before I make an official release. But as I'm preparing for holiday vacation I think it may be good to spread the current version to the public.

Main difference to older releases is - I reenabled big buffers for file copy operations. The advantage is much faster copy speed especially with big files. Copy speed is still not as fast as the copy command from a 4os2 command prompt but the difference is not as dramatic than before. I had to rework and extend the original dynamic buffer adaption code for this. It seems current copy speed is the maximum you can get when copying with an PM application.

The max. copy buffer size in old releases was determined by a simple algorithm and based on the setting in Options - Various - File Copy/move tab. I had to limit the buffer in some 1.7. release to 60k cause there was a subtle bug in ndfs DAV plugin which leads to some zero bytes in large files copied with the help of big buffers. The side effect was a limited copy speed at about 30MBytes/s I just discovered a few months ago. There's an old ticket about the ndpdav bug which is resolved in newer releases I think. So I think it's save to reenable big buffers again. For convenience I include my fixed ndpdav.dll in this lcmd*zip. But you probably don't need it. Current ndfs 3.1.6 also has changed in a way that such subtle plugin bugs shouldn't show up anymore. But be warned if you use old ndfs/ndpdav versions.

Lcmd also uses high memory for the copy buffer now. So if you watch your memory consumption (f.i. with above512.cmd or theseus) you may find a bit less memory usage from the scarce lower shared mem. Unfortunately this does not fix another long standing problem - copying big trees of directories sometimes leads to eat up a lot (sometimes all) lower shared memory. This may eventually bring your system down until you kill lcmd. Finally I can reproduce this problem but unfortunately I don't have a fix for this right now. If you ever run into this bug in the past, sorry there's no fix in this release for that.

In this archive I've also included the current available translations. When you want to use lcmd with another language or make your own translation look into the locale directory.

I've also checked in all current sources at the official home of LarsenCommander -

Have fun :-)

Anyone running Injoys Firewall VPN feature?

Or how to you connect securely into your home LAN from outside world?

It seems to be possible to configure an AVM Fritzbox for this task. Is there a simple how to to get this working? I mean connection from the outside with a notebook running OS/2 (or Linux or Win) to my internal OS/2 server and ...

I can start the daemon 'git daemon --reuseaddr --base-path=/git_repositories/ /git_rerepositories/', I can sort of connect from the other machine with 'git clone git@server:/git_repositories/test' but get
Code: [Select]
Cloning into 'test'...
fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.

My goal is to have git repositories on my server running OS/2. At first I started with ssh which seems to be the usual way to connect to a git server. But can't get this running. See the other thread -,3175.0.html

Now I tried with git daemon which is in netlabs.exp. But can't get this to work either. Seems I've to setup a linux server instead...

I'm thinking about buying a HP OfficeJet Pro 7740 printer but it only supports 'HP PCL 3 GUI, HP PCL 3 Enhanced'. Usually I have printers with postscript and/or PCL5(+) support. No I wonder if I can use the features of this A3 printing engine. I guess scanner will work the same way as it does with my HP OfficeJet Pro 475dw (smb, ftp?, email). Working fax isn't that important. But I need high resolution print outs on A3 page sizes (11 x 17"). Color, duplex and tray selection.

I'm pretty sure ported programs will work good enough with the help of cups. But how about the standard OS/2 drivers? PCL 3 seems pretty old but back at the days when OS/2 drivers where written I fear it was limited to 300dpi and maybe even color didn't work work PCL3. Any knows?

Applications / OpenOffice default Font Arial -> Calibri?
« on: January 29, 2023, 10:41:37 am »
How can I set the default font to Calibri? Usually f.i. in Calc font is set to Arial. I think this is the same for all platforms. Is there a way to set the default to Arial? Do I have to change the default templates for that?

I just had a curious problem with my T540p. I've different eCS / OS/2 / ArcaOS installation next to Win10/7 on this Thinkpad. I now decided to make an additional NTFS partition available for Win and never thought this will lead to such catastrophic problem. Steps I did -
- booted OS/2 and with DFSee I created a new NTFS partition in den middle of the disk where there was some free space
- booted W10 and formated the newly partition with NTFS

A few days afterwards I wanted to boot one of the OS/2 partitions. But booting hangs after checking drive E:. Drive D: was checked before but the other JFS partitions afterwards were not checked (M: V: L: T: O: R: P: ...). The system simply hangs.

As all my installations do have JFS /autocheck:* I can't start any of the installed systems. Moreover this system can't be booted by the ArcaOS stick (known issue for this T540p). I can't boot from an old eCS CD too. I also didn't find any ALT-F1 cmd line working. What works was booting DOS from DFSee stick and changes partition type of the partition AFTER the new NTFS partition from 06 (FAT) to 07 and fix/align the one partition AFTER the NTFS one.

I think when I formated the NTFS partition from W10 this does also overwrite some stuff of the next partition behind. Or at least crippled some of the partition info of all these logical partitions. And afterwards OS/2 can't handle that and simply stopped booting. Although I do all my partitioning stuff with DFSee, Win always finds some way to destroy something. Or say it in other words, does some alignment on the disk which OS/2 can't deal with.

Now I've my systems booting again as usual. But I can't access the partition after the NTFS one anymore. Probably I've to delete it. Never thought that a simple format from W10 of an already aligned partition (with DFSee running from OS/2) can make such damage. Maybe I should let some gap around OS/2 partitions in future whenever I prepare a partition for Win.

Pages: [1] 2 3 4