The Current State & Plans For Porting Linux/BSD Software To Redox OS

[Volta]

The discussion is centered around context switching latencies in relation to microkernel designs. seL4 needs to context switch, Qnx too. The fact they aren't general purpose ( whatever means, considering that today every operating system serves many tasks ) is irrelevant in this context. Just to note that Qnx Limited made a demo of its desktop-oriented operating system many years ago http://toastytech.com/guis/qnxdemo.html ....yes, it was usable a general purpose operating system. Obviously market dynamics made it impossible to dethrone Windows. But this is a common story, considering that even Linux hasn't been able to do it.

Context switching is the point. That's why seL4, QNX will always be toys outside their narrow usage. Linux achieved much more in this market.

No. It has been shown that L4Linux has smaller latencies than monolithic Linux. This fairy tail about slow microkernels was debunked when Liedtke developed L3.

Nothing was debunked. In what case? Was RT Linux used? The fairy tell is microkernels can be as fast as monolithic ones. Show me general purpose OS (not some simple layer where nearly nothing happens, so there are less context switches) using microkernel which has comparable performance to Linux.

Mach is slow, but because its architecture is subpar. This doesn't imply microkernels MUST BE slow!! Otherwise why someone would use OkL4 as a hypervisor on top of which to run Android? This happens a lot in infotainment systems.

As a simple layer? Why would anyone run Android on it if it's not slow?

[ppcorp]

as far as I know, it's not, while I don't know the point of it, apparently they get some pretty decent funding every now and then to push it along

Linus said this about Linux when he announced the project on usenet.