Contributed by jolan on from the nearing-release-time dept.
(Comments are closed)
OpenBSD Journal
Contributed by jolan on from the nearing-release-time dept.
(Comments are closed)
Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]
By kokamomi (83.227.181.37) on
Comments
By phessler (64.173.147.27) on
Comments
By Jasper (80.60.145.215) on
I wonder what the "theme" of the 3.8 release will be, _will_ it really become hummpa??
By Roo (84.9.60.116) on
By Anonymous Coward (212.104.129.221) on
By yurii (192.115.26.60) yo@jct.ac.il on
By Anonymous Coward (70.109.176.42) on
Searched, but did not find.
Comments
By Brad (204.101.180.70) brad at comstyle dot com on
By Anonymous Coward (64.207.229.210) on
Comments
By djm@ (203.58.120.11) on
Comments
By Anonymous Coward (67.67.137.82) on
Comments
By Anonymous Coward (195.224.109.30) on
By Anonymous Coward (142.109.90.79) on
By Janne Johansson (130.237.91.134) on
place?
int
func(humppa_t finnish, char *songname) {
...
}
By Miod Vallat (82.255.98.183) miod@ on
By Anonymous Coward (195.224.109.30) on
Bugger !
Comments
By Anonymous Coward (130.189.246.177) on
Comments
By phessler (208.201.244.164) on
Comments
By Anonymous Coward (84.134.33.236) on
After some recent discussion on misc@ (IIRC), IMO either goals.html should be adjusted ("Provide a good cross compile/development platform") or someone[1] should try to set up such an environment.
It's not about building *packages* on a cross environment, but about developing on it, and then let the final build run on the native platform.
[1] I'd *love* to work on this.
Comments
By Anonymous Coward (66.44.1.203) on
With my limited experience with cross-GCC on other platforms, basically you just install binutils, headers, compilers and libraries to another PREFIX. So then from there is of course the idea of somehow working that new toolset into the BSD make system.
I wouldn't imagine it too hard, even on OpenBSD, if you just play with some variables.
Comments
By Nick Holland (68.43.117.34) nick@holland-consulting.net on http://www.openbsd.org/faq/
We are quite aware of the results.
There are lots of reasons OpenBSD is a native-build system. It may take a little longer, but we feel it is worth it. Search the mail list archives for this (often repeated) topic and why it is a non-starter with the developers on the OpenBSD project.
And personally, as someone who oversees mac68k builds, you aren't going to get one tiny bit of sympathy from me on your 200MHz machines. :)
Nick.
Comments
By Anonymous Coward (66.44.1.143) on
I wasn't suggesting that OpenBSD should do what NetBSD does. I understand the argument that compiling the system is a good stress test, etc.
However, someone asked the question, so I replied to it. All I'm saying is that even though the OBSD people are against it, it would probably be trivial to cross-compile to begin with.
I wasn't complaining about slow builds, I was replying to someone else's complaints. I happen to have an M68K system, by the way, so I know what you mean.
Comments
By henning (80.86.183.129) henning@ on
the funny thing is, that those who make claims like this are the ones that never tried it. maybe because those that tried it know that statements like these are wrong.
By Matthias Kilian (84.134.24.244) on
I didn't really complain, I even don't want to cross-build and then run that frankenbinaries on the target. I just would like a cross-environment for the inital work on new ports (i.e. new stuff in /usr/ports).
For example, first port *big-fat-piece-of-software-written-in-c++* on and for some fast platform, then cross-build it for a slower platform, and if it looks fine, do the remaining work natively right on the target platform.
And of course I know that such an environment is *not* trivial, as Henning already said. Building a cross-chain is easy, but *using* it the right way, for example not confusing host and target, is difficult. If in doubt, ask some of the folks working with embedded systems (regardless wether on *BSD, or Linux, or whatever).
Anyways, I'll try to hack on some cross-environment whenever there's nothing better to do. I may (and most probably will) give up some day, but who cares?
By Anonymous Coward (195.224.109.30) on
Considering that some of the solwest platforms are also some of the cheapest ( not always though, ) it might be more productive for development, and remain within the "build for it's own arch" guidelines by having multiple machines building the source.
Hmmm..... I might take a crack at this ....
Comments
By Andrew Dalgleish (203.100.230.217) on
You can build the kernel and most ports, but building the system bombs out.
Some of the Makefiles need patching, some of the problems come from within make(1) itself.
As far as using distcc to cross-compile, it kinda works.
By Anonymous Coward (195.224.109.30) on
By jasper (80.60.145.215) on
By Nick Holland (68.43.117.34) nick@holland-consulting.net on http://www.openbsd.org/faq/
Comments
By Johan M:son (213.114.133.92) on
packages for the arm platform, so chances are the stuff you need/want is
not pre-built (hence need to recompile).
By Anonymous Coward (195.224.109.30) on
By Anonymous Coward (24.107.59.254) on
Comments
By Anonymous Coward (64.81.36.140) on
Comments
By Brad (204.101.180.70) brad at comstyle dot com on
By Shane (203.20.79.132) on
CD-RW DVD-+RW floppy USB stick/drive HDD Tape net boot Choose your poison.
By chrissnell (66.25.86.139) on http://chrissnell.com
By Anonymous Coward (195.224.109.30) on
from misc@ 2005-08-22
By Anonymous Coward (213.9.211.12) on