zlacker

Floppinux – An Embedded Linux on a Single Floppy, 2025 Edition

submitted by Galaxy+(OP) on 2026-02-03 04:33:25 | 252 points 190 comments
[view article] [source] [go to bottom]

NOTE: showing posts with links only show all posts
◧◩
16. userbi+H8[view] [source] [discussion] 2026-02-03 05:59:14
>>hilti+T3
MenuetOS/KolibriOS:

>>38059961

>>27249075

That was 1999 and I never saw anything like that afterwards.

Now you have ;-)

◧◩◪◨
19. yjftsj+0a[view] [source] [discussion] 2026-02-03 06:10:30
>>zx8080+p8
https://kernel.org/ says 6.12 is still a supported LTS, so you could just run that.
◧◩
26. xkriva+Od[view] [source] [discussion] 2026-02-03 06:46:31
>>hilti+T3
xwoaf-rebuild matches that

https://web.archive.org/web/20240901115514/https://pupngo.dk...

◧◩
27. endgam+Zd[view] [source] [discussion] 2026-02-03 06:47:58
>>sockbo+o5
You might have some luck applying isohybrid(1) to the period-correct .iso image, making it bootable by other means: https://manpages.debian.org/stretch/syslinux-utils/isohybrid...
◧◩
43. vaylia+kl[view] [source] [discussion] 2026-02-03 07:49:42
>>hn_thr+I3
https://en.wikipedia.org/wiki/Floppy_disk

It's basically what people used before USB sticks. But it was also the storage medium that software was sold on, before CD-ROMs became widespread.

51. mobili+Nn[view] [source] 2026-02-03 08:08:26
>>Galaxy+(OP)
25 years ago i used floppyfw

https://www.zelow.no/floppyfw/

to setup small router on 486 with 12 MB ram and run flawless. Later i get Linksys WRT54GL and decommissioned that machine.

◧◩
59. jabl+jr[view] [source] [discussion] 2026-02-03 08:36:34
>>yjftsj+S7
Someone was still working on some minor cleanups in August 2025: https://lore.kernel.org/lkml/20250825163545.39303-1-andriy.s...

(That mail also mentions the floppy driver is "basically orphaned" though. But evidently it's still there and builds.)

Maybe you're thinking of the floppy tape (ftape) driver, which was removed back in the 2.6.20 kernel. Though there's a project keeping an out-of-tree version of it working with recent kernels at https://github.com/dbrant/ftape

◧◩◪◨
62. hilti+4s[view] [source] [discussion] 2026-02-03 08:42:13
>>pjmlp+Dp
Or controlling the heating and AC systems at 19 schools under its jurisdiction using a system that sends out commands over short-wave radio frequencies

https://www.popularmechanics.com/technology/infrastructure/a...

◧◩◪◨⬒
66. hilti+os[view] [source] [discussion] 2026-02-03 08:44:27
>>blackh+kk
I love this story where a C64 in Poland rans a Auto repair shop.

https://www.popularmechanics.com/technology/gadgets/a23139/c...

◧◩◪
85. UncleS+dz[view] [source] [discussion] 2026-02-03 09:34:27
>>gattil+7g
There was also "Tom's Root Boot" distro that fitted on a floppy:

https://en.wikipedia.org/wiki/Tomsrtbt

◧◩◪
97. M95D+jH[view] [source] [discussion] 2026-02-03 10:39:31
>>iberat+6m
loadlin ?

https://youpibouh.thefreecat.org/loadlin/

◧◩
108. mrighe+8X[view] [source] [discussion] 2026-02-03 12:31:16
>>sockbo+o5
It seems that both OpenBSD [1] and NetBSD [2] still support i386, for example here [3] you can find the image for a USB stick.

I expect at least the base system (including X) to work without big issues (if your hardware is supported), for extra packages you may need a bit of luck.

[1] https://www.openbsd.org/plat.html

[2] https://wiki.netbsd.org/ports/

[3] https://wiki.netbsd.org/ports/i386/

118. mlacks+3o1[view] [source] 2026-02-03 15:07:28
>>Galaxy+(OP)
Reminds me of my first linux distro called damnsmall linux. I think this was used as a first attempt to port linux to the gamecube, but the main team driving the effort ended up going with Gentoo instead.

From the main page:

As with most things in the GNU/Linux community, this project continues to stand on the shoulders of giants. I am just one guy without a CS degree, so for now, this project is based on antiX 23 i386. AntiX is a fantastic distribution that I think shares much of the same spirit as the original DSL project. AntiX shares pedigree with MEPIS and also leans heavily on the geniuses at Debian. So, this project stands on the shoulders of giants. In other words, DSL 2024 is a humble little project!

Though it may seem comparably ridiculous that 700MB is small in 2024 when DSL was 50MB in 2002, I’ve done a lot of hunting to find small footprint applications, and I had to do some tricks to get a workable desktop into the 700MB limit. To get the size down the ISO currently reduced full language support for German, English, French, Spanish, Portuguese and Brazilian Portuguese (de_DE, en_AU, en_GB, en_US, es_ES, fr_FR, es_ES, pt_PT, & pt_BR ). I had to strip the source codes, many man pages, and documentation out. I do provide a download script that will restore all the missing files, and so far, it seems to be working well.

https://www.damnsmalllinux.org/

◧◩
126. leiden+Pz1[view] [source] [discussion] 2026-02-03 15:55:51
>>sockbo+o5
Try Plop Boot Manager: https://www.plop.at/en/bootmanagers.html

It can boot from a floppy or from a CD drive, and it lets you chainload into a live usb even on old computers.

I used it to boot from CD from a floppy in an old Pentium MMX and it worked great (although slow, of course)

136. st_gol+gT1[view] [source] 2026-02-03 17:14:02
>>Galaxy+(OP)
Have you tested this on an actual 486?

Sadly, it does not seem to boot on my 486 DX2, I even stuffed 32M of RAM into the machine (8*4M, maximum the mainboard supports), more than the recommended 20M.

I have copied the floppy image from the site. It churns for about a minute and a half, loading kernel and initrd, then says "Booting kernel failed: Invalid Argument" and drops into SYSLINUX prompt.

EDIT: I tried a few more floppies to rule that out as the cause of the problem.

Here are some screenshots: https://imgur.com/a/floppinux-0-3-1-Mdh1c0w

EDIT 2: I cloned SYSLINUX, checked out the specific commit and did some prodding around.

The function `bios_boot_linux` in `com32/lib/syslinux/load_linux.c` initializes errno to EINVAL. Besides sanity checking the header of the kernel image, there are a few other error paths that also `goto bail;` without changing errno.

Those other error paths all seem to be related to handling the memory map. I know that the BIOS in my machine does not support the E820h routine. I have a hunch that this might be the reason why it fails.

The website has an image gallery where people ran it on actual hardware: https://krzysztofjankowski.com/floppinux/floppinux-in-the-wi...

Most of those machines seem to be newer systems which probably support E820h, except for another 486 DX2 with a similar vintage as mine, that also failed to boot.

◧◩◪◨⬒⬓⬔⧯
140. lprove+e22[view] [source] [discussion] 2026-02-03 17:49:47
>>anthk+1H1
> The Mac didn't exist in Europe

That is absolutely not a valid generalisation.

I worked on Macs from the start of my career in 1988. They were the standard computer for state schools in education here in the Isle of Man in the late 1980s and early 1990s.

The Isle of Man's national travel company ran on a Mac database, Omnis, and later moved to Windows to keep using Omnis.

It's still around:

https://www.omnis.net/

I supported dozens of Mac-using clients in London through the 1990s and they were the standard platform in some businesses. Windows NT Server had good MacOS support from the very first version, 3.1, and Macs could access Windows NT Server shares over the built-in Appleshare client, and store Mac files complete with their Resource Forks on NTFS volumes. From 1993 onwards this made mixed Mac/PC networks much easier.

I did subcontracted Mac support for a couple of friends of mine's consultancy businesses because they were Windows guys and didn't "speak Mac".

Yes, they were very strong in print, graphics, design, photography, etc. but not only in those markets. Richer types used them as home computers. I also worked on Macs in the music and dance businesses and other places.

Macs were always there.

Maybe you didn't notice but they always were. Knowing PC/Mac integration was a key career skill for me, and the rise of OS X made the classic MacOS knowledge segue into more general Unix/Windows integration work.

Some power users defected to Windows NT between 1993 and 2001 but then it reversed and grew much faster: from around 2001, PowerMacs started to become a credible desktop workstation for power users because of OS X. From 2006, Macintel boxes became more viable in general business use because the Intel chips meant you could run Windows in a VM at full speed for one or two essential Windows apps. They ran IE natively and WINE started to make OS X feasible for some apps with no need for a Windows licence.

In other words, the rise of OS X coincided with the rise of Linux as a viable server and GUI workstation.

◧◩◪◨⬒⬓
158. dmitry+NZ2[view] [source] [discussion] 2026-02-03 22:11:40
>>incanu+oZ2
https://web.archive.org/web/20151223083654/https://www.debia...

Sarge dropped i386, Squeeze i486

◧◩
160. anthk+b53[view] [source] [discussion] 2026-02-03 22:40:10
>>incanu+yJ1
http://delicate-linux.net/ This. add 8-16MB of RAM and you will happily run X.
◧◩
162. 1vuio0+rs3[view] [source] [discussion] 2026-02-04 00:53:40
>>sockbo+o5
"There is a herculean effort on the part of package maintainers to build software for distros, and no one has been building 32 bit version of software for years, even if it is possible to build from source."

This statement must be Linux-only

Pre-compiled packages for i386 are still available for all versions of NetBSD including the current one

I still compile software for i386 from pkgsrc

https://ftp.netbsd.org/pub/pkgsrc/current/

NB. I'm not interested in graphical software, I prefer VGA textmode

◧◩◪◨⬒⬓⬔
175. anthk+8N4[view] [source] [discussion] 2026-02-04 12:19:37
>>kelnos+DN3
Amigans are already using AmiSSL and AmiGemini (and some web browsers) perfectly fine in m68k CPU's recreated with FPGA's.

You can do modern TLS stuff with a machine from the 90's if you cut own the damn JavaScript and run services from https://farside.link or gemini://gemi.dev proxying the web to Gemini.

◧◩◪
177. st_gol+MR4[view] [source] [discussion] 2026-02-04 12:53:50
>>snac+sE4
It's a DataExpert OPTI-495SX: https://theretroweb.com/motherboards/s/dataexpert-opti-495sx

It is running some AMI BIOS variant with a copyright date of 1992, I currently don't have the exact version string around to compare with the ROM dumps on retroweb. vbindiff says the "F" and "M" images are identical and the "H" only has a few 1-byte differences, mostly typos in ASCII strings.

I've written a small boot sector program once that tries out memory and CPU information gathering techniques, so I know the INT 15h, E820h, E801h are not implemented but INT 12h and INT 15h AH=88h return something sane. When I have more than 16M installed, the later reports the full 31M of HIMEM, but I'm not sure how the ISA memory hole factors into this.

From what I saw glancing at the scanning code yesterday, syslinux 6.x should fall back onto AH=88h if AX=E820/E801 doesn't work. It's interesting to know that this worked in older SYSLINUX, I'm curious to check out what changed.

◧◩◪◨
179. szszrk+wX4[view] [source] [discussion] 2026-02-04 13:31:14
>>tensil+Mx
that is just one step from Floppotron...

https://silent.org.pl/home/2022/06/13/the-floppotron-3-0/

◧◩◪◨
181. snac+3g6[view] [source] [discussion] 2026-02-04 19:40:40
>>st_gol+MR4
Yeah, even if syslinux 6.x “falls back” when E820/E801 aren’t there, AH=88h is just a number, not a hole aware map. It can’t describe ISA/ROM/reserved regions, so the loader still might not be able to build something it trusts. If you want to prove it yourself, stick a log right around syslinux_memmap_find() and dump the memmap entries it thinks it has on that BIOS. Its interesting to notice that syslinux 6.03 had a fix for your same issue: "Fix hand over to kernel: "Booting kernel failed: Invalid argument" error." https://www.linuxquestions.org/questions/slackware-14/%5Bann...

Based on the POST strings of your motherboard versus mine (https://theretroweb.com/motherboards/s/fic-486-jal-rev-c), we both have an AMI BIOS so I might be able to run a similar test for my board. You're right that syslinux 6.x should fall back if E820 doesn't work but that hasn't been my experience on my motherboard hence the reversion to 4.x, I can reliably boot with 4-16MB of RAM.

◧◩◪
186. genewi+hwa[view] [source] [discussion] 2026-02-05 23:32:53
>>amne+IL
i ran my whole house network off a laptop with the specs of a raspberry pi 2 for a really long time. I finally broke and moved it to a VM because the laptop's built in port and USB were finally too slow to route traffic, 11mbit USB! It took a decade+[1] of "innovation" in the US before i could finally buy internet faster than 11mbit. IIRC i switched to VM based IPCop in ~2007.

[1] My first broadband connection was in 1998 at 768/768 kbit symmetrical. My first megabit speed connection was in 2006 or 2007. in 2010 or 2011 we got VDSL and it was 16 whole megabits. Now i have 300mbit on a good day, and 150mbit on a bad day.

I literally wrote the guide on how to use old hardware with VM tech to route your house, first with ipcop[2], then generically[3], and just this week i wrote a guide on how to get ipv6 working with starlink and dd-wrt[4].

i've been in this a long time.

[2]https://web.archive.org/web/20220323223325/https://www.dslre...

[3]https://web.archive.org/web/20131214075417/https://www.dslre...

and the dd-wrt starlink one from this week:

[4]https://nextcloud.projectftm.com/index.php/s/4iScqZbrfYiNcKy

ETA: it is hilarious how much pushback i got about doing all of this in a VM, just scant years before "you should just use a VM for that" became the default answer, and a decade before "just put it in a k8s cluster and pay someone a quarter million a year to babysit it" became a thing...

also ipcop booted and installed off a single floppy forever

[go to top]