Opened 14 years ago
Closed 14 years ago
#2 closed task (fixed)
Decide on Odin SDK
| Reported by: | dmik | Owned by: | |
|---|---|---|---|
| Priority: | blocker | Milestone: | Crossroad |
| Component: | general | Version: | |
| Severity: | Keywords: | ||
| Cc: |
Description
We should decide if we want to use the Odin SDK for porting or create a native port of Java from scratch.
Change History (2)
comment:1 Changed 14 years ago by
comment:2 Changed 14 years ago by
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
After more thinking on the issue, it looks like the best way is to start off using the Odin SDK (e.g. mostly reusing the Windows code) and optimize the narrow places using the native OS/2 API as needed. This should shorten the road to the first working result of the Java 6 port which is important.
I have access to the Odin SVN now so I will start with that approach (first, #3 is to be completed).
Note: See
TracTickets for help on using
tickets.
Unless it appears to be very simple to do it without the Odin SDk, my recommendation is to use Odin even if it requires updating Odin. Innotek and Goldencode initially were going to partner for building Java 1.4. Goldencode decided it would take as much work to update Odin as to port Java natively. I remember listening to their presentation at Warpstock 2002. My thoughts then are as now. They may spend as much effort on Odin but then you have a base for other things, with native Java you get just Java. They claimed to be able to keep the port up as well natively as they could with Odin but it turned out that they couldn't even get 1.4.2 out and only had 1.4.1 available. Innotek on the other hand put out several updates from 1.4.1 through 1.4.2 and I can use their exe to run Java 1.5_07 (including the full SDK, not just the runtime).