OS/2 1.x Improvement Ideas: Difference between revisions
Jump to navigation
Jump to search
Dr. Shuppet (talk | contribs) |
Dr. Shuppet (talk | contribs) |
||
| Line 5: | Line 5: | ||
* 64-bit RAMdisk. This is more of a challenge, it was done on ArcaOS, which is 32-bit OS/2, and unlike the former is mostly useless. | * 64-bit RAMdisk. This is more of a challenge, it was done on ArcaOS, which is 32-bit OS/2, and unlike the former is mostly useless. | ||
* A web server and/or client would be interesting. | * A web server and/or client would be interesting. | ||
* Porting more tools to analyze binaries. | |||
Latest revision as of 08:14, 15 January 2023
This is a list of ideas that could be implemented to improve the OS/2 1.x experience, of various kinds of difficulty and required knowledge.
- INT 21h support, for WLO to be able to run most Windows 3.0 applications. Many of them use INT 21h, it improved with 3.1, but that also added interfaces not implemented by WLO. Not sure how possible that is, since it would require ring 0 modifications, which are limited in OS/2 (to research: how limited? Do portions of drivers run in ring 0?).
- 32-bit RAMdisk. Shouldn't be too hard if the driver is able to set code segments to 386 mode, it should mostly just work. Given how much memory IBM software requires, this might be quite useful (though you might argue you can just plug in a 32-bit kernel...).
- 64-bit RAMdisk. This is more of a challenge, it was done on ArcaOS, which is 32-bit OS/2, and unlike the former is mostly useless.
- A web server and/or client would be interesting.
- Porting more tools to analyze binaries.