Rayson Ho wrote: > On 7/27/07, Brian Gupta <[EMAIL PROTECTED]> wrote: >> , and for that >> matter why the design of MxN was being used. (It seemed a bit black >> box to me) > > I remember the "Dinosaurs OS book" talks about why M:N scales better > than 1:1. However, as hardware becomes more powerful, the difference > is not as big, yet 1:1 is a lot cleaner to implement, and has become > the default thread model in Solaris 9 and Linux 2.6.
Yep, Silberschatz's book (http://codex.cs.yale.edu/avi/os-book/os7/). Whose cover is only as mysterious as the Dragon book's :) It does sound a little black box, but if I'm not mistaken the idea was no to limit the number of user threads, despite the limited number of kernel threads. libthread decided which user thread would run and to which kernel thread they'd be assigned to. Allowing the existence of more user threads than kernel ones. So it makes sense to say that it scales well. Rafael _______________________________________________ perf-discuss mailing list [email protected]
