Apache 1.3.4 on Windows NT 3.1

Yes, I know it’s crazy old, totally useless, but it did mostly compile.

Apache 1.3.4 running on Windows NT 3.1 Advanced Server

Apache 1.3.4 running on Windows NT 3.1 Advanced Server

Assuming it’s hasn’t been crashed or hacked it should be online here:


Unlike Serweb 0.3, Apache is HTTP 1.1 compliant, which means that I can put it behind haproxy, and enjoy the fact it doesn’t need a dedicated IP address.

Although I can’t imagine anyone wanting it, here is the binary/source and my htdoc dump.  Download it here: apache-NT-31.zip & unzip.exe

Apache console mode

Apache console mode

I had to pull out some stuff, like some of the service features, so it really only runs as a console app.

I’ve compiled it with /Zi meaning full debug and no optimization.  If you want to re-compile you’ll probably want either the Win32 SDK, or Visual C++ 1.0 32bit, and replace the headers and libraries from the Windows NT 3.5 SDK.  Much like trying to build GCC 2.6.3 on Windows NT.

Also in a silly way, thanks to Qemu, I’m now running both OS/2 & Windows NT on the same server, running Linux.

Just got a Roland Soundcanvas SC-88!

Roland Soundcanvas SC-88

Roland Soundcanvas SC-88

As annoying as it is that Windows 10 won’t drive my el-cheapo USB-MIDI interface adapter, I have to say that even loading Windows XP into VMWare Player, and then loading DOSBox is a pleasant enough experience when connecting to the Roland.

Pluses include the built in power supply, RCA jacks, and of course it just sounds snappier compared to the Yamaha MU-80.

I’ll have to capture samples from both, to compare and contrast.

I have to say that for a mere 2,400 Yen I’m really enjoying this fine piece of hardware!

I’m unsure of the difference between the SC-88, the SC-88 Pro, the SC-88VL and the SC-88ST, and of course the SC-88ST Pro.  I just know I’m lucky that mine arrived OK and in working order!  Thanks ps38152000, whomever you are.

Starting on an ELF cross compiler for Windows

I’m using Slackware 4.0 as a starting point, so it’s Binutils 2.9.1 and GCC .. I verified that I can build a static hello world executable, and it runs! …

However Linux 2.0.40 has the same issue, it starts to decompress, and triggers a reboot in both Qemu and PCem.  Going in circles I guess.  I suppose the next step is to use the exact version they have in Slackware to see if Qemu can actually run that pre-built kernel, and if I can create one via cross compiling.

I should add that on Debian 7.1 I got GCC running, and it too produces the exact same thing.

Not that I think anyone cares, but here is my pre-built toolchain with some source (The binutils was built under Linux, using a MinGW cross compilerelfgcc_2.0.40.7z

Windows NT 3.1 & KVM

No go.

No go.

I don’t know what I was expecting, but I thought I’d try to install Windows NT 3.1 Advanced Server in a KVM virtual machine.  No doubt the processor is just too new.  The -cpu 486 / -cpu pentium flags didn’t help things out at all.  However using Qemu has it running just fine.

I also had this crazy idea that haproxy could front HTTP 1.1 requests into serweb so I could go back to having a Windows NT 3.1 web server.  Naturally that didn’t work.

502 Bad Gateway

The server returned an invalid or incomplete response.

Oh well.

The useless update, is that I managed to get Apache 1.3.4 to compile and run on Windows NT 3.1!

Apache 1.3.4 on Windows NT 3.1

Cross compiled Linux 1.0.9!

Linux 1.0.9 running!

Linux 1.0.9 running!

After getting Linux 0.98 to compile, I thought I’d take a stab at Linux 1.0.  I vaugely recall when it was released, and I just remember a much larger push to 1.1.  So I guess it really comes as no surprise that in the Linux kernel archives, there is simply the 1.0 tar, and 9 patch files.

I went ahead, and patched up the release, and then tried to build with GCC 2.3.3.  This however proved not to be up to the task, as 2.3.3 has issues with some of the assembly macros, so delving into the readme shows that you need to use GCC 2.4.5 or higher.  Since I wanted to keep at least the tools on par, I went ahead and build 2.4.5, and once more again used the gcc driver from 2.6.3.  I further ended up relying on headers, and checking tool versions from Debian 0.91, which also revealed that they were still using GAS 1.38 back then.

One interesting note while building piggback, which takes the compressed system object, and wraps it in an object file, is that it directly uses the magic “0x00640107”, which is for a later “Linux/i386 impure executable (OMAGIC)” filetype.  But because my binutils is so ancient, I needed to change it to “0x00000107” so that the linker would recognize it as a “386 executable not stripped” file.  As always when having no idea what I was doing, it was easier to have it make an empty object file, set the type for 12345678 and look for where it occurs in the data stream, and just match it with a known object file.  As you can see, it worked.

I don’t know if it is of any interest, but the kernel source, along with a binary is available to download linux-, and the same goes for GCC gcc-

And of course, you’ll want the latest download, which includes the pre-built tools, qemu, and build environment to get you started.

User Mode Linux revisited (UML) aka SLiRP networking

So my uh ‘friend’ that got into trouble when he found out that his ‘dedicated’ machine turned out to be a VM which he couldn’t launch nested KVM VM’s, and instead found that User Mode Linux (UML), would allow them to run their touchy ancient Linux application in a psudo VM/Container.  Well they finally bit the bullet and decided to move to something better.

And by better, it was cheaper.  And why was it cheaper?  Because it is even a more restricted VM.


So naturally the panic call was made, because TUN/TAP networking was not permitted in this new VM.  So what to do.

Well, keeping in mind how Qemu gets around this problem, it binds in a copy of SLiRP.  And it turns out that UML can actually call SLiRP directly!  So cool we have an ‘out’.  First things first, we need SLiRP on the host machine.  I’m old, so that means I build it from source.That means I’m downloading slirp-1.0.16.tar.gz, along with the 1.0.17 patch.  I’m not sure if I need to go into how to extract source, patch, running configure and compiling.

One thing of note is that you really really really want to set the “FULL_BOLT” option either in the Makefile, or in config.h

With SLiRP built, I just copy it into /usr/local/bin .. I’m sure there is packages and stuff out there, but heh I’m old.

OK next up I make a small script to call SLiRP, in this case, I’m going to redirect port 80 directly into the VM.  And for a test port 2323 which then goes into port 23 (why not ssh? .. sigh don’t go there).

So my script looks like this:

/usr/local/bin/slirp “redir 80 80” “redir 23 2323”

Pretty simple right?  I’m using a script as there will be more than one VM, so relying on .slirprc isn’t a solution for me.

./linux-2.6.24-rc7 ubd0=junk.ubda eth0=slirp,,/virtual/sl.sh

And away we go!

Inside the VM we can configure it with the usual SLiRP config:

ifconfig eth0
route add default gw

And now we can access the internal http server!

Add in some magic to /etc/resolv.conf such as:


and it’ll automatically use whatever the host is configured to do.

Torbjörn Granlund’s Excellent resource on running free OS’s on Qemu

Ever get tired of x86 on x86?  yeah me too.

How to solve that problem?

Simple, grab QEMU, and jump off into all those cool RISC processors of the 1990’s that were going to save us all from the WINTEL hegemony!

Lots of instructions, samples, images, and hints here:


It’s really more comprehensive than I’ve sat down to do, so yeah it’s awesome!

Supported platforms include: