* Nick Piggin wrote: > Move some fork related scheduler policy from fork.c to sched.c where > it belongs. this only makes sense if all the other fork-time initializations are done in sched.c too - these are scattered all around copy_process(). The attached patch (it compiles & boots) does this. All the lowlevel scheduler logic (that was done in fork.c) is now in sched.c - fork.c only sees the higher level primitives. I've also updated a couple of comments that relate to the affected code. Ingo