Date: Mon, 30 Oct 2006 11:23:24 -0800 From: Greg Lewis <glewis@eyesbeyond.com> To: Daniel Eischen <eischen@vigrid.com> Cc: Paul Allen <nospam@ugcs.caltech.edu>, Lev Serebryakov <lev@freebsd.org>, Robert Watson <rwatson@freebsd.org>, Julian Elischer <julian@elischer.org>, current@freebsd.org Subject: Re: KSE, libpthread & libthr: almost newbie question Message-ID: <20061030192324.GA24251@misty.eyesbeyond.com> In-Reply-To: <Pine.GSO.4.64.0610301141550.23983@sea.ntplx.net> References: <4542896D.1050001@elischer.org> <20061027231642.GJ30707@riyal.ugcs.caltech.edu> <45429703.8070305@elischer.org> <20061028104741.Q69980@fledge.watson.org> <45442A35.2030803@elischer.org> <20061029090309.T27107@fledge.watson.org> <20061029182108.GA46604@misty.eyesbeyond.com> <45458C7E.50104@elischer.org> <20061030163327.GA22419@misty.eyesbeyond.com> <Pine.GSO.4.64.0610301141550.23983@sea.ntplx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Oct 30, 2006 at 11:44:35AM -0500, Daniel Eischen wrote: > On Mon, 30 Oct 2006, Greg Lewis wrote: > >On Sun, Oct 29, 2006 at 09:24:14PM -0800, Julian Elischer wrote: > >>Greg Lewis wrote: > >>>If you really want to know, just send the running process a SIGQUIT and > >>>it will dump the currently running threads to stdout. But yes, 1.4 and > >>>1.5 > >>>both use "native" threads which correspond 1:1 with OS threads (plus > >>>there are threads the JVM creates itself, as you note). The JVM threads > >>>include garbage collection and AWT event handlers at least. > >>> > >> > >>I gather it doesn't use libpthread, but rather just the syscalls? > > > >No, it does use libpthread (or libthr, or libc_r if you so choose). What > >I'm saying is that the JVM maps a single Java thread to a single <pthread > >library of your choice> thread. How that maps to a kernel thread is > >then defined by the threading library. > > > >The point is that the JVM doesn't do any internal M:N business itself, > >which was the original point under discussion IIRC. > > Does the JVM specify system or process scope threads when it does > its mapping? Or does it not use pthread_attr_setscope() at all? > (I know this doesn't apply to libthr or libc_r, only libpthread.) It doesn't call pthread_attr_setscope() by default. The code suggests that it is called to set threads to system scope when profiling is turned on. -- Greg Lewis Email : glewis@eyesbeyond.com Eyes Beyond Web : http://www.eyesbeyond.com Information Technology FreeBSD : glewis@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20061030192324.GA24251>