<cut> > Strange indeed. And note that if the code were correct, your > surmise about the init_fpu() path being "logically the correct" > one would no longer be true - we'd be saving the FPU state of > the current process for no good reason. > > The more I look at the FPU management code, the more I marvel > that it even gives an appearance of working... Maybe no one has stress tested it enough up till now. As soon as we put he mips boards in QA we found a few fpu problems immediately. I hope those were it and we weren't just scratching the surface. Pete
