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.


Messages - Dave Yeo

Pages: [1] 2 3 ... 373
1
Events / Re: Warpstock 2025 USA - The end of my journey with ArcaOS
« on: November 12, 2025, 07:37:02 pm »
Security also becomes very important with a WiFi driver. Especially if being used in a public space.

2
Events / Re: Warpstock 2025 USA - The end of my journey with ArcaOS
« on: November 12, 2025, 12:41:37 am »
Seems to be an NDIS device but the only drivers I found were Windows 11, no Linux or BSD on a quick look on DuckDuckGo. You could try searching Google for vendor 0BB4 Device 0FFE but without a BSD driver I doubt that ArcaOS can build a driver for your device. You could also open a ticket and do a testlog network with the device plugged in and set for tethering, or I guess your device calls it modem mode.
Mine has WiFi hotspot, USB tethering, Bluetooth tethering and Ethernet tethering in the settings. The Ethernet tethering does look interesting, guess plug a USB ethernet dongle in and plug the other end into the computer.
Edit: Should be testlog generic if you open a ticket but with how few USB network devices that AOS supports I doubt that you will have success.

3
Events / Re: Warpstock 2025 USA - The end of my journey with ArcaOS
« on: November 09, 2025, 07:47:54 pm »
I looked into doing just that. At least with my phone, a 2022 Moto Edge, the share network presents itself as a wireless dongle. Forget the ID but it was barely supported in Linux and unsupported in the BSD's. I was hoping it would present as an Ethernet dongle. Perhaps other phone models do, with a supported ID.
The problem with WiFi adapters is the radio, namely turning it on and off. Also need a software stack to handle authorization etc. Unluckily there is no standard when it comes to the radio's so every WiFi adapter needs its own code for enabling the radio.
As for Arca Noae, I'd guess they just don't have the resources to do this stuff. I can't imagine that they're getting rich doing this. After overhead, paying IBM their share and such, there likely isn't that much profit in a license sale and it's hard to imagine that they're selling that many licenses.
Plus we're getting left behind in this 64bit universe. I personally think it is pretty amazing how far AOS has come.

4
General Discussion / Re: Spam Bots
« on: November 05, 2025, 11:46:54 pm »
Sounds like bots are downloading the themes, can't imagine enough users for that many downloads

5
General Discussion / Re: Spam Bots
« on: November 05, 2025, 04:45:51 pm »
The images don't display in FF/SM due to being webp which our browsers are too old to support.

6
General Discussion / Re: Spam Bots
« on: November 05, 2025, 08:59:26 am »
Well I went to their website, https://anubis.techaro.lol/ and it claimed to be testing the soul of my browser and I seemed to have passed as after a second or two the page loaded. This is with No-Script so no JavaScript was run.
Edit: This is SeaMonkey. Turning JS on (only one domain) did make the site work better and it works fine on our browsers, at least at a quick look.

7
Perhaps the best ftp daemon for OS/2 is available at ftp://ftp.pmoylan.org/FtpServer/index.html

8
Programming / Re: VisualAge COBOL 2.2 on ArcaOS or Warp 4.52
« on: October 24, 2025, 06:43:01 am »
Be careful with LIBPATH, likely it installed itself early in LIBPATH, move it to the end.
For VACPP, I use a script instead of letting it mess with config.sys, might be best to do the same.
I'll attach my script to give ideas. You should revert your config.sys to before you installed it and keep the new one under a different name so you know what you need.
Edit: Also note the use of the Toolkit and that some of it is installed by YUM or ANPM

9
Programming / Re: native mainframe mini-clone
« on: October 13, 2025, 07:06:38 pm »
Couple of things. There is a 64 bit LX format that NASM supports. Bird came up with the patch for VirtualBox.
Currently OS/2 can use PAE to a limited degree, mostly used as a ram disk. Things are setup by os2ldr before the kernel launches. Current os2ldr is based on QSINIT, from the latest blurb,
Code: [Select]
QSINIT - 2025-08-10

QSINIT had been updated (2025-08-10):

    Binary: ftp://212.12.30.18/public/QS/QS_LDR.ZIP
    Source Code and SDK:  ftp://212.12.30.18/public/QS/QS_SDK.ZIP

QSINIT is a small 32-bit something (you may call it an operating system, or a "small DOS"), that you can write applications for, and that may act as an OS/2 kernel bootloader.

The author of this software, Dmitry Zavalskov, has reserved the copyright to himself. The binary is released as freeware, the source code as freeware for non-commercial use.

Much of the kernel is actually 16 bit, with 16 bit code limited to 1 GB address space. OS/2 2.x before Warp V4 FP13 (Warp Server also supported high memory earlier) only supported 1 GB of address space with the kernel getting the upper half, so you'd have to work from the lower 512MBs (less as much is used by DLL's) and there are still API's that underneath are 16 bit and have to reside in low memory.

10
General Discussion / LEO mirror
« on: October 10, 2025, 05:02:11 am »
Old timers might remember that we had a couple of file sites, Hobbes and LEO (link everything online). At one point LEO had a hard drive crash and vanished along with all its content, which was mostly OS/2 related. Still find links to it here and there.
Anyways, found a mirror of it, https://ftpmirror.infania.net/sites/ftp.leo.org/ perhaps others should mirror it.

11
Applications / Re: General Software Testing
« on: October 10, 2025, 01:25:49 am »
Mandel.c needs #define INCL_DOSSEMAPHORES before os2.h. Makefile needs some minor mods, -Zmts removed and -Zomf added in its place on the Mandel.exe recipe.
I will attach my work.
Note that it runs fine run from the source directory besides a warning about the lack of a map file. With the map file(s) gives a blank screen. I think as it expects only 256 colours
Edit:, yes the map files seem to be different palettes. I ran out of memory while playing with it so it may be worth adding a -Zhigh-mem where you have the -Zomf. Guess the check for map file could be removed or changed to a parameter. If I knew how, it could query the number of colours and only use the map file with 256 colours.
Edit2: With the map files, if you save the image, the gif looks correct and varies based on the map file.
Edit3: Updated source again, add stdlib.h a few places, fixed make clean, bigger stack for _beginthread.
Now get warnings like,
Code: [Select]
gcc    -c -o Mandel.o Mandel.c
Mandel.c: In function 'MandelWndProc':
Mandel.c:170:20: warning: passing argument 1 of '_beginthread' from incompatible pointer type [-Wincompatible-pointer-types]
  170 |      _beginthread (SaveThread, 0, 4096, &pfdFileDialog.szFullFile);
      |                    ^~~~~~~~~~
      |                    |
      |                    void (__attribute__((system)) *)(char *) {aka void (*)(char *)}

Mandel.h has,
Code: [Select]
void APIENTRY SaveThread (char *path);Not sure of the fix.

12
Article Discussions / Re: Games with older SDL and FSLIB
« on: October 04, 2025, 06:26:35 am »
Can you please now check Pingus:
- https://github.com/Pingus/pingus

Hi Martin, I've given up on Pingus. The latest has a hard requirement of OpenGL, went back a couple of versions to before they used OpenGL. It has a dependency on Boost, https://www.boost.org/ which does claim to support OS/2 and likely did at one point. I can build it but it crashes with a sigabort. Using the latest GCC and a debug build the problem is clear and I can't see how to fix.
Code: [Select]
src/engine/jamgram.cpp: In function 'int yyparse()':
src/engine/jamgram.cpp:2282:18: warning: 'void free(void*)' called on unallocated object 'yyssa' [-Wfree-nonheap-object]
src/engine/jamgram.cpp:1152:16: note: declared here

Building is done. To install, run:

    ./b2 install --prefix=<DIR>


[H:\tmp\boost_1_89_0\tools\build]b2
terminate called after throwing an instance of 'std::logic_error'
  what():  basic_string: construction from null is not valid

Killed by SIGABRT
pid=0x4442 ppid=0x3226 tid=0x0001 slot=0x0011 pri=0x0200 mc=0x0001 ps=0x0010
H:\TMP\BOOST_1_89_0\TOOLS\BUILD\B2.EXE
Creating 4442_01.TRP
Moved 4442_01.TRP to W:\var\log\app\68e0392b-4442_01-B2-exceptq.txt

The trp does point to jamgram.cpp being the issue. Even tried an earlier version, same problem.

13
Applications / Re: Environment for building ffmpeg
« on: September 30, 2025, 06:18:56 pm »
Good to hear. The iconv thing really should be a cast. Unluckily GCC has changed how casts are done and I forget the specifics.
Depending on what else you might do, you might want to put poll.h back. I had a compile fail yesterday due to a missing poll.h
Still strange you had the math problem.

14
Applications / Re: Environment for building ffmpeg
« on: September 29, 2025, 01:20:10 am »
Ok, did a git reset --hard to make the source the same as upstream and using your configure arguments and GCC 15.2.0 with poll.h moved out of the way to use FFmpeg's poll implementation, I had one build break, iconv wanting a static char* and being fed a char* (or the other way around). Since you didn't mention this, I figure you don't have iconv-devel installed. I added --disable-iconv to the configure line and the build finished fine. No patching mathops or such.

15
Applications / Re: Environment for building ffmpeg
« on: September 28, 2025, 10:38:14 pm »
Strange. Search for CLOSESOCKET in config.h, should be
Code: [Select]
#define HAVE_CLOSESOCKET 0

If it is #define HAVE_CLOSESOCKET 1, change it for now.
Now that my question about which GCC you are using has accidentally been answered, I'll try with GCC 15.2.0 and get a better idea.
How are you setting the environment for15.2.0?

Pages: [1] 2 3 ... 373