Peter Cordes wrote: > Ok, since you asked for it, you're old fashioned. Happy now? :)
Yeah, I feel a lot happier now!<g> > Both chips run the same user-space binaries, so running gcc on the same > source code will give identical binaries. I take your word for it, I haven't tried and/or compared the actual binary that gets produced. I have looked closely and checked the cross-compiler and do know that it does not produce the same code, even if the target is the same (x86 vs ppc hosts used to produce an 823 ppc target). That confused me some as in theory it should produce the same target, at least I thought. Anyway, I haven't used ppc enough to even know some of the minor similarities or differences, some which you have pointed out. > However, unless you have a good reason to compile the kernel on another > machine (e.g. the machine you want the kernel for won't boot until you get > the kernel compiled), it's certainly easier to just build the kernel on the > machine that will run it. This is just for convenience in managing > kernel config files and modules. Yes, I think we are in complete agreement on this. Yesterday I was in a meeting and our division manager walked in with a new G4 Titanium, which looked nice sitting there, but he didn't have an airport in it yet, so he couldn't use the wireless...I would have given him my airport, but heck, I just got a home wireless gateway and will be using the Pismo with it...good time to look at why that airport module won't compile though...;-) -- Alan DuBoff Software Orchestration, Inc.
