Hardware support in JNode



Manufacturer Device name Test
(Many) Intel X86 OK
(Many) X86_64 temporarly not working


Manufacturer Device name Test
(Standard) IDE LBA Disk OK
(Standard) PCI IDE Controller OK


Manufacturer Device name Test
ATI Radeon OK
ATI Mach64 In development Does not work
NVidia RIVA TNT Should work
NVidia RIVA TNT2/TNT2 Pro Should work
NVidia RIVA TNT2 Ultra Should work
NVidia Vanta/Vanta LT Ok
NVidia RIVA TNT2 Model 64/Model 64 Pro Ok
NVidia Aladdin TNT2 Should work
NVidia GeForce 256 Ok
NVidia GeForce DDR Should work
NVidia GeForce2 MX 200 DDR Ok


Manufacturer Device name Test
(Standard) PS/2 Ok
(Standard) USB keyboard Not yet stable Ok
Microsoft Natural Keyboard Pro PS/2+USB Ok


Manufacturer Device name Test
(Many) NE2000 Ok
(Standard) Loopback Should work
3Com 3C905C_TXM Ok
3Com 3C905B [Cyclone] Ok
Broadcom BCM570x Does not work
RealTek 8029 PCI Ok
RealTek 8139 PCI Ok

Pointing device

Manufacturer Device name Test
(Standard) USB mouse Ok
Logitech WheelMouse PS/2 & USB Ok
Logitech Optical WheelMouse PS/2 & USB Ok


Manufacturer Device name Test
AMD Athlon Ok
AMD Duron Ok
AMD Athlon 64 Switch of MP detection (using mp=no) Ok
Intel Pentium 4 Ok
Intel Pentium Should work
Intel Pentium Pro Should work
Intel Pentium II Ok
Intel Pentium 3 Ok


Manufacturer Device name Test
(Standard) USB mouse Ok
(Standard) USB keyboard Not yet stable Ok
(Standard) USB HUB Ok
(Standard) UHCI host controller Ok
Logitech WheelMouse PS/2 & USB Ok
Logitech Optical WheelMouse PS/2 & USB Ok
Microsoft Natural Keyboard Pro PS/2+USB Ok

more platforms to support

I always think Java is great especially due to its portablity.
So why don't make java OS to run on MORE platforms than intel, intel, intel and compatibles with intel?
What about PowerPC, SPARC, ARM, MIPS?

It is a matter of resourcing and priorities

As Levente points out, the process for porting JNode to a new hardware is "well understood". The problem (of course) is that it is a lot of work, and would require someone with considerable skills in writing compilers, assembly language cores and so on. And this is not just a one-time cost because the JNode port would then need to be maintained.

We have to balance the effort/resources needed to create and maintain a JNode port against the effort/resources needed to do other tasks aimed at making JNode more usable on the x86 platform. The current active developer group is concentrating on the latter. IMO, this is the right thing to do. It is better to have JNode run fast with lots of functionality on the x86 platform (that 95+% of people use) rather than have it run slowly with less functionality on a wide range of platforms.

But, hey, if someone (or some company) with the right skills were to commit to creating and maintaining a JNode port to (say) PowerPC, I'm sure we'd be very supportive.

I agree

The JNode architecture is suitable for supporting more platforms and we could provide technical support for developers interested in porting JNode to new platforms.

Here is an abstract of what porting involves: http://www.jnode.org/node/512