> I'm somewhat reluctant to force userland to make function calls to do > things like sqrt if the compiler is now smart enough to inline them > -- it is bad to slow down 99% of people for the benefit of 1%. Sounds to me like an argument in favour for that fpuless-i386 port someone suggested. (Ideally, I think it would be best to make it possible to put that code in the kernel instead of userland and then build userland with FPU instructions, for the benefit of cases like installation disks where code size matters a lot more than performance of what little floating point is done. But of course whoever actually does it is going to get to make that call.) /~\ The ASCII der Mouse \ / Ribbon Campaign X Against HTML mouse%rodents.montreal.qc.ca@localhost / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B