On Wed, 19 May 2010 18:10:34 -0400, Jorden M <[email protected]> wrote: > On Wed, May 19, 2010 at 6:00 PM, Jens Staal <[email protected]> wrote: >> That the kernel API breaks that fast is only important if you want to >> keep >> up with Linus' Linux. Let's say that you find all the drivers you ever >> would >> like to use in version 2.6.35 - then you could basically just use modules >> from this release indefinitely. A major update to support newer drivers >> would only be done when it was considered needed. I have no idea how the >> other OSes who tries to exploit linux drivers are doing it, though. >> As for amount of work - probably lots and way over my competence. It was >> basically some random musings and I just liked to know how others thought >> about them. > > That helps narrow things but there's still a huge amount of code that > any driver can call out to. That would be a lot to port, and could be > impracticable as I'd bet that drivers call out to some pretty > Linux-kernel specific subsystems (sysfs, for one, which would not need > to be ported to Plan 9.) Yes it would be possible to `stub' all that > out, but you would have to make a potentially large framework just for > stubs, since preconditions and postconditions of your stub API calls > may have to be met to ensure that the drivers continue to function.
In order to pull this off you would wither have to work closely with the Plan 9 folks (9fans more specifically), or fork the code and go your own way. Problem is that if you propose this over on 9fans you will be hammered with reticule and insults. Just read through the 9fans archive for suggestions of doing anything like this and you will get a very clear picture of what I mean. I see two ways forward. 1) fork and move on, or 2) embed youself in the pla9 community and start writing drivers (saying nothing of your motivations). I made similar mistakes and paid for it. EBo -- -- To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/glendix?hl=en
