7 messages in org.freebsd.freebsd-platformsRe: To share or not share ? (was: Som...| From | Sent On | Attachments |
|---|---|---|
| John S. Dyson | Mar 17, 1997 10:00 pm | |
| John Birrell | Mar 17, 1997 10:38 pm | |
| Pedro Giffuni | Mar 18, 1997 12:09 am | |
| Terry Lambert | Mar 18, 1997 9:46 am | |
| Terry Lambert | Mar 18, 1997 9:51 am | |
| Terry Lambert | Mar 18, 1997 4:41 pm | |
| Pedro Giffuni | Mar 18, 1997 6:25 pm |
| Subject: | Re: To share or not share ? (was: Someone working on a SPARC version?) | |
|---|---|---|
| From: | John Birrell (jb...@cimlogic.com.au) | |
| Date: | Mar 17, 1997 10:38:32 pm | |
| List: | org.freebsd.freebsd-platforms | |
Pedro Giffuni wrote:
Of course, it would be very stupid from myself not to admit that we need to modify our tree following the NetBSD example (BTW, IMO, the best time to do it is ASAP, can someone illustrate me on what are the clear objectives behing 3.0-current ?). But we should protect our evolved code (LKMs and devices) from being swapped because another OS has a prettier structure than ours.
I think it is unrealistic to hold out hope that the kernel designs will ever be compatible enough to share code "easily" --- without hacking. With the exception of device drivers, I don't care about that 8-). The practical thing to do is to use the system that is already available for the architecture(s) you want to use. I use FreeBSD for i386 and NetBSD for Alpha.
For me, though, the big win would be sharing libraries and their header files because this reduces the amount of code my company has to support. From libraries and header files, the ports tree naturally follows.
When (if) I get time I'll try building NetBSD's libc on FreeBSD and try building ports against that. That should give me some idea of how incompatible the interfaces between userlands and the kernels are.
Regards,