Date: Tue, 26 Feb 2002 20:43:57 -0500 (EST) From: Robert Watson <rwatson@FreeBSD.org> To: David Malone <dwmalone@maths.tcd.ie> Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/i386/i386 exception.s genassym.c machdep.c mp_machdep.c mpapic.c swtch.s vm_machdep.c src/sys/i386/include cpufunc.h pcb.h src/sys/i386/isa apic_vector.s clock.c icu_vector.s intr_machdep.c intr_machdep.h npx.c src/sys/kern ... Message-ID: <Pine.NEB.3.96L.1020226203319.38595G-100000@fledge.watson.org> In-Reply-To: <20020226232607.GA33515@walton.maths.tcd.ie>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 26 Feb 2002, David Malone wrote: > We'd all also like to help with getting SMP-NG working for 5.0, but we > can't do that while the ideas are all sitting in John's head. Unless > other people start working on sorting out -current then we'll just have > to push back 5.0 again. At least we made some progress with lock > pushdown and Julian's work in the last 6 months. Sigh. I originally sent this e-mail privately, but under the circumstances it seems appropriate to respond publically. There is a persisting myth that no work happens on SMPng outside of John's Perforce branches. This appears not actually to be the case. In the past month, two important components of FreeBSD have moved to fine-grained locking in the main tree: struct fd/file, and process groups. In addition, over the last few months, several other important pieces of infrastructure have gained support for fine grained locks, including taskqueues. In addition, there's an in-progress implementation of TTY locking that is being contributed by a third-party developer, and hopefully will be done soon. Even I managed to lock something -- the jail sub-system. Most of the basic synchronization primitives are in pretty final form, and are carefully documented in section 9 manpages. Nothing is preventing anyone from picking up any of the additional important tasks on the SMPng work-queue, including fine-grained locking of pipes, components of VFS, NFS, sections of the device driver system, the socket code, several network protocols, and more. In fact, there are plenty more items that can be added to the work-queue. It is true that some things people may work on will be closely related to John's work targets, but it is not correct to assert that no progress (or even that the majority of work items) can be approached without John doing the work. Robert N M Watson FreeBSD Core Team, TrustedBSD Project robert@fledge.watson.org NAI Labs, Safeport Network Services To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96L.1020226203319.38595G-100000>