Stephan Assmus wrote:
I AM developing in Haiku, since months! I am writing this email in Haiku too. Why do you ignore the part where I said it runs just fine for me?! How often do I have to repeat this? Haiku may run better or worse depending on your hardware. Isn't it obvious enough? Haiku may use different drivers on your machine compared to mine. Therefore we are not running the exact same code!!
How often are you rebuilding? :)
Besides, running on Linux isn't viable; I'm developing kernel drivers.Again, you didn't listen to me. Go right back to what I wrote. You can develop kernel drivers from Linux, build an image and network boot a second machine. I don't care if you think it would be acceptable in 2003 only. First you tell me Haiku runs so bad for you that you cannot do work, I try to help and present an intermediate solution, now you say it's selfhosting. What now? I wouldn't call four crashes per hour "selfhosting". The network boot solution is actually a faster solution before your driver is stable, because if you crash the kernel you need to reboot. But if you have your IDE in Linux, you don't need to wait until that is loaded each time you need to reboot Haiku. So, even if Haiku ran perfectly, the network boot solution would still be better.
Yes, I did read that. Maybe I should make my position plainer. I've used over a dozen different incarnations of Linux, all the way from Slackware to Peanut to SMP. The only one I liked even a little was Lindows/Linspire because It Just Worked.
The solution isn't to develop from another OS but to fix this one. I should point out here that the report I made (and canceled when I detected a HDD problem) has just been reopened. Haiku cannot make a CD and install from it as of r30790. That's definite.
It isn't necessarily that the code used for the processor and drivers is different, but we are using the OS differently.
The crash I reported earlier isn't related to r30825 at all, I just went right back to r30782 ans tried a clean build (totally deleting generated/objects) and I'm still getting problems with `copyattr` throwing an exception and shutting down the build. This raises two issues:1. Why can no-one else verify this yet?Yes, very good question.
Never mind, I'll track it down and point someone at the solution later on today. Then I can get back to some real work.
2. How can it remain totally undiscovered until a clean build is done?It's ever so slightly possible that I have a corrupt source file, but the odds are against it. Just to check that I'm deleting the `copyattr` sources and doing another `svn up`.Apparently, there were multiple problems, some of which are already addressed. Ingo commited a fix to copyattr last night, that deals with permissions and attributes, I don't know if it's related to your problem. Then the p7zip archive apparently has wrong permissions for some files. That has not been fixed yet. Don't know if you have that even in your optional packages.
What I can verify is that, apart from the problems with Path.cpp from around r30790, Haiku does rebuld and reinstall itself from CD as at r30820. There were a heap of changes to libraries after that, and I expect the problem will be found there.
Rob
