Jump to content
223 posts in this topic

Recommended Posts

Well...

 

Here are the sources to start developing a driver for Marvell-Yukon, Syskonnect and other NICs

 

The other thread was going nowhere

 

Please post on advances, and not requierements on how much you need this driver!

 

Good Luck.

 

EDIT: added sources from NetBSD4

FreeBSD.zip

linux.zip

sk_netbsd4.zip

Link to comment
https://www.insanelymac.com/forum/topic/4851-marvell-support-effort/
Share on other sites

  • 2 weeks later...

I'm a few hundred lines into writing a driver for the yukon (I started it tonight). This is my first device driver, and the apple frameworks require sifting through dev docs, so it may take awhile. Also, I happen to be a student who prioritizes his studies over personal projects, so be patient. I'll post updates here periodically.

 

Will

It depends entirely on the device. There's a USB example with the ADC Reference Library, as well as a PCMCIA one. There are several audio examples, also. Since plain old PCI doesn't have any examples, I read the Writing PCI Drivers doc, among others I searched for in the ADC.

 

Basically, one has to write it from scratch, since the code structure is quite different from Apple's frameworks. Reference the other ports, certainly, for things like registers and such. There are sometimes notes about hardware bugs and workarounds, too. So, I guess what I'm saying is that when you sit down to port a driver, you should prepare yourself to write the thing from the bottom up. There is no simple method, of which I am aware.

 

When I am sure I have the foundation of the driver correct, and I have the licensing {censored} figured out, I'll post what I've done, so others may fill in the gaps if they so choose.

 

Will

When I am sure I have the foundation of the driver correct, and I have the licensing {censored} figured out, I'll post what I've done, so others may fill in the gaps if they so choose.

 

Cool, thanks for your work. I unfortunately can't code very well but I'm praying someone will figure it out soon. This is the one thing left before I can move my primary OS to OS X.

Well, I'm going to try to get basic features working, first, which, when I figure out how the hardware queues work, should be trivial. I'm still working on the registers. It's pretty convoluted. It'll take a while. The more code I read, the more complex and daunting this task is. One step at a time, though.

 

I know that didn't really answer your question, but I really don't know how long it'll take. It might take a week, or a month or more. Likely more than a week, since I have end-of-semester exams, projects, and crits.

Edited by altaic

since I can't code I won't flame anyone, and I'd rather thank the few people who have taken the taks of porting drivers.

 

But it is most dissapointing that after some 4 months of OsX this community has achieved very little in terms of porting to everyone.

 

We have been more succesful in cracking and hacking installs of OsX (a somewhat illegal activity) than acctually expanding the hardware base (a totally legal activity through opensource!)

 

Thus, OsX will still be a limited Os for few people

 

Please people, keep on trying!

Ok, I got reply from Syskonnect.

 

The good news is they will definitely support x86 platform. The bad news is that porting resources are not sheduled yet, so it's not going to be inmediate. Maybe the release of x86 machines on January will put them under some pressure.

 

Stay tuned, it ain't over yet...

Mortis you are right, but it is also true that the people who could really help us porting x86 application/drivers etc. and have experience (like Maxxuss or cremes for example) don't help us. I'm not saying that they are expected to, since is not their job, but they could do in 1 day much more than what we could do in 2 weeks. Is there a way to contact people them and ask them to give us a hand?

Mortis, I agree, I believe that everyone here is starting to agree with you. I'm not trying to {censored} about it, but I believe that it is sad that we've had projects devoted to Marvell and nVidia and everything and the fruits of the labor (in the nVidia case) are barely worth it. We all know there will eventually be nVidia drivers, it'll just be a matter of time, I understand that, but I still believe it would be great if we had it working. The most likely situation for it to be successful is if we were to get a lot more people working on it. Of course I, like yourself, don't know how to code drivers, so I can't exactly help.

Guest
This topic is now closed to further replies.
×
×
  • Create New...