On Thursday 24 May 2007 07:57, Mike Frysinger wrote: > i recall the i386/x86_64 port being updated with some code that > modifies itself at runtime ... basically you configure the kernel for > SMP support but at boot if the kernel detects just one processor, the > spinlock code and friends will modify themselves to be nops so there > is no runtime overhead compared with building the kernel for a > uniprocessor ... > > anyone know where these implementation details are ? http://lwn.net/Articles/164121/ tavi -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
