Date: Fri, 20 Oct 1995 13:28:10 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: leisner@sdsp.mc.xerox.com (Marty Leisner) Cc: cimaxp1!jb@werple.net.au, leisner@sdsp.mc.xerox.com, hackers@FreeBSD.ORG, jb@cimlogic.com.au Subject: Re: NetBSD/FreeBSD (pthreads) Message-ID: <199510202028.NAA05261@phaeton.artisoft.com> In-Reply-To: <9510201855.AA17636@gnu.mc.xerox.com> from "Marty Leisner" at Oct 20, 95 11:55:04 am
next in thread | previous in thread | raw e-mail | index | archive | help
> Why would I want kernel level threads? > > 1) does a debugger understand userlevel threads? > 2) if I run a threaded application under strace, with kernel > levels threads is may make some sense...with user level threads > there's all this junk in the way... > 3) Anyone have good hard numbers about the differences between user/kernel > level threads on performance? Yeah. On an 8 processor box, a kernel-threaded app is 8 times as concurrent as a user threaded app (which by definition can only utilize one processor at a time). Consider also that thengs like statfs/fstatfs are potentially long-blocking operations that aren't select()'able. Debugging tools are another (almost completely unrelated) problem. > Like I said, the mit package is awfully clever... So's the AT&T/Sun stuff. The problem is that no one has bothered with implementing both types of clever in the same package. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510202028.NAA05261>