Dan Woodard <[EMAIL PROTECTED]> writes: > I installed the latest (unstable) version of crosshurd and was able to > boot and run native-install twice, but my ethernet adapter (RTL8139) was > not recognized. Do I have to recompile Mach from source with the driver > active?
This driver is enabled. The problem might be that this version your rtl8139 is not(yet) supported. There is a patch that might fix your problem. Is there some debian hacker willing to build a GNU Mach package with Alfred's patches included? I will have another look at my autoconf patch, to see if I can fix the current problems without changing that much. Marcus told me that adding extra drivers to gnumach-1-branch (upstream CVS) is not a problem. This is something that is not hard to do, so I will do that soon. > Sorry to be so uninformed, but do drivers have to be compiled into the > Mach microkernel binary file? If so, how is this different from the Linux > kernel? It works just like linux, although GNU Mach can't load modules. Use the right arguments for configure (--enable-rtl8139, for example) and run make. > i.e., from the gnu.org page on Hurd: > > "It is possible to develop and test new Hurd kernel components without > rebooting the machine (not even accidentally). Running your own kernel > components doesn't interfere with other users, and so no special system > privileges are required." This is partially true. With kernel components the person who wrote this meant: `components that normally run in kernelspace'. (Well, that is what I think ;)) An example is ext2fs, it runs in userspace. You can implement extra filesystems (translators) as user and run them as a user. Better examples are `ftpfs' and `pfinet' (the TCP/IP stack). In the far(?) future L4 will be used as microkernel. It does not have userspace drivers, etc. -- Marco
