Ben Hutchings wrote: > On Sun, 2012-12-16 at 11:14 +1100, [email protected] wrote:
>> - Why not use float instead of infinite-precision integer arithmetic? > > We cannot assume the presence of an FPU. (The kernel handles FPU > emulation for userland, but not for itself.) Also, the scheduler doesn't reset floating-point state when switching between kernel threads. Dealing with floating-point exceptions can be another source of pain. Jonathan -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/[email protected]
