From owner-freebsd-threads@FreeBSD.ORG Sun Aug 24 14:13:38 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D58FB16A4C2 for ; Sun, 24 Aug 2003 14:13:38 -0700 (PDT) Received: from relay1.cris.net (relay1.cris.net [212.110.128.64]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6B1C743F93 for ; Sun, 24 Aug 2003 14:13:34 -0700 (PDT) (envelope-from phantom@phantom.cris.net) Received: from phantom.cris.net (root@phantom.cris.net [212.110.130.74]) by relay1.cris.net (8.12.6/8.12.6) with ESMTP id h7P0LNsN092982 for ; Mon, 25 Aug 2003 00:21:24 GMT Received: (from phantom@localhost) by phantom.cris.net (8.12.6/8.12.2) id h7OLKdoQ025606; Mon, 25 Aug 2003 00:20:39 +0300 (EEST) (envelope-from phantom) Date: Mon, 25 Aug 2003 00:20:39 +0300 From: Alexey Zelkin To: threads@freebsd.org Message-ID: <20030825002039.A25487@phantom.cris.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i X-Operating-System: FreeBSD 4.7-STABLE i386 Subject: jdk14 & libc_r / fork() X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Aug 2003 21:13:39 -0000 hi, I have hit into a problem recently while trying to reproduce some deadlocks in jdk14 port. Problem details are simple -- JDK calls fork() and got frozen because of storm of unhandled (i.e. ignored SIGBUSes). I have checked that SIGBUS ignorance is not a problem, but reason of SIGBUS is. Some investigations shown that while doing of libc_r's wrapper for fork() we try to cleanup pthread's stacks. Since we are using custom stacks in JVM (malloc()'ed ones) libc_r attempts to free() them. Most times everything goes right, but sometimes (very rarely) SIGBUS is being generated while free()'ing of stacks. more details: it's RELENG_4 from trace (sorry, i accidiently deleted complete trace), but from memory can restore following points: libc_r/uthread/uthread_fork.c:154 -- [..] libc/stdlib/malloc.c:606 -- Any ideas on reasons ? PS: BTW, it's looks very similar to "red zone" violation in protection, but since problem appears after __sys_fork() call - it should not be a problem. Am I right ? From owner-freebsd-threads@FreeBSD.ORG Sun Aug 24 23:54:59 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 37C2816A4BF; Sun, 24 Aug 2003 23:54:59 -0700 (PDT) Received: from axl.seasidesoftware.co.za (axl.seasidesoftware.co.za [196.31.7.201]) by mx1.FreeBSD.org (Postfix) with ESMTP id 87AB243F85; Sun, 24 Aug 2003 23:54:56 -0700 (PDT) (envelope-from sheldonh@starjuice.net) Received: from sheldonh by axl.seasidesoftware.co.za with local (Exim 4.22) id 19rBFb-0000FG-Dk; Mon, 25 Aug 2003 08:54:51 +0200 Date: Mon, 25 Aug 2003 08:54:51 +0200 From: Sheldon Hearn To: David Xu Message-ID: <20030825065451.GA852@starjuice.net> Mail-Followup-To: David Xu , "deischen@freebsd.org" , "freebsd-java@freebsd.org" , "threads@freebsd.org" References: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> User-Agent: Mutt/1.5.4i Sender: Sheldon Hearn cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 06:54:59 -0000 On (2003/08/21 17:05), David Xu wrote: > ULE does not work with KSE, I have uploaded a patch, please try it. > http://people.freebsd.org/~davidxu/kse/kse_ule.diff > You can not get cpu affinity for KSE process with this scheduler, > The patch only let KSE process work under ULE scheduler. Your patch fixes my jboss panic with libkse. > I have not fully tested it, for some reason, I won't commit the patch. I heartily endorse your patch. :-) Thanks, Sheldon. From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 00:44:19 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7731116A4BF; Mon, 25 Aug 2003 00:44:19 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5DBBD43FDD; Mon, 25 Aug 2003 00:43:59 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCRNJV; Mon, 25 Aug 2003 15:26:12 +0800 Message-ID: <3F49BF15.3090904@viatech.com.cn> Date: Mon, 25 Aug 2003 15:47:33 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sheldon Hearn References: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> <20030825065451.GA852@starjuice.net> In-Reply-To: <20030825065451.GA852@starjuice.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 07:44:19 -0000 Sheldon Hearn wrote: >On (2003/08/21 17:05), David Xu wrote: > > > >>ULE does not work with KSE, I have uploaded a patch, please try it. >>http://people.freebsd.org/~davidxu/kse/kse_ule.diff >>You can not get cpu affinity for KSE process with this scheduler, >>The patch only let KSE process work under ULE scheduler. >> >> > >Your patch fixes my jboss panic with libkse. > > > >>I have not fully tested it, for some reason, I won't commit the patch. >> >> > >I heartily endorse your patch. :-) > > Thanks, I am pleased to see libkse works. :-) However don't expect I will commit it, kernel scheduler interface is a bit weird, so I won't try to add another weird code. >Thanks, >Sheldon. > > > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 02:02:51 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D041E16A4C0; Mon, 25 Aug 2003 02:02:51 -0700 (PDT) Received: from axl.seasidesoftware.co.za (axl.seasidesoftware.co.za [196.31.7.201]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0324743F75; Mon, 25 Aug 2003 02:02:50 -0700 (PDT) (envelope-from sheldonh@starjuice.net) Received: from sheldonh by axl.seasidesoftware.co.za with local (Exim 4.22) id 19rDFP-0000ZK-IE; Mon, 25 Aug 2003 11:02:47 +0200 Date: Mon, 25 Aug 2003 11:02:47 +0200 From: Sheldon Hearn To: David Xu Message-ID: <20030825090247.GF852@starjuice.net> Mail-Followup-To: David Xu , "deischen@freebsd.org" , "freebsd-java@freebsd.org" , "threads@freebsd.org" References: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> <20030825065451.GA852@starjuice.net> <3F49BF15.3090904@viatech.com.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3F49BF15.3090904@viatech.com.cn> User-Agent: Mutt/1.5.4i Sender: Sheldon Hearn cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 09:02:52 -0000 On (2003/08/25 15:47), David Xu wrote: > >I heartily endorse your patch. :-) > > > Thanks, I am pleased to see libkse works. :-) > However don't expect I will commit it, kernel scheduler interface is a > bit weird, so > I won't try to add another weird code. Does Jeff know about your patch? Perhaps he'd be willing to review and commit, since he's the original author of SCHED_ULE. Ciao, Sheldon. From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 03:14:36 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 360B716A4BF; Mon, 25 Aug 2003 03:14:36 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id C400243FBF; Mon, 25 Aug 2003 03:14:29 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCRNZ8; Mon, 25 Aug 2003 17:56:49 +0800 Message-ID: <3F49E267.3070302@viatech.com.cn> Date: Mon, 25 Aug 2003 18:18:15 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sheldon Hearn References: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> <20030825065451.GA852@starjuice.net> <3F49BF15.3090904@viatech.com.cn> <20030825090247.GF852@starjuice.net> In-Reply-To: <20030825090247.GF852@starjuice.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 10:14:36 -0000 Sheldon Hearn wrote: >On (2003/08/25 15:47), David Xu wrote: > > > >>>I heartily endorse your patch. :-) >>> >>> >>> >>Thanks, I am pleased to see libkse works. :-) >>However don't expect I will commit it, kernel scheduler interface is a >>bit weird, so >>I won't try to add another weird code. >> >> > >Does Jeff know about your patch? Perhaps he'd be willing to review and >commit, since he's the original author of SCHED_ULE. > > No, I think he does not know the patch. Please don't commit it, there is a performance problem, after each time a thread is exited, I must exit a kse and then let scheduler reinitialized it. In general, I needn't manage a struct kse, it should be a scheduler internal data. >Ciao, >Sheldon. >_______________________________________________ >freebsd-threads@freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-threads >To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" > > > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 03:19:51 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 281DD16A4BF; Mon, 25 Aug 2003 03:19:51 -0700 (PDT) Received: from axl.seasidesoftware.co.za (axl.seasidesoftware.co.za [196.31.7.201]) by mx1.FreeBSD.org (Postfix) with ESMTP id 96A2444003; Mon, 25 Aug 2003 03:19:49 -0700 (PDT) (envelope-from sheldonh@starjuice.net) Received: from sheldonh by axl.seasidesoftware.co.za with local (Exim 4.22) id 19rERv-0000vy-LD; Mon, 25 Aug 2003 12:19:47 +0200 Date: Mon, 25 Aug 2003 12:19:47 +0200 From: Sheldon Hearn To: David Xu Message-ID: <20030825101947.GL852@starjuice.net> Mail-Followup-To: David Xu , "deischen@freebsd.org" , "threads@freebsd.org" , "freebsd-java@freebsd.org" References: <20030821165727.C1C0.DAVIDXU@viatech.com.cn> <20030825065451.GA852@starjuice.net> <3F49BF15.3090904@viatech.com.cn> <20030825090247.GF852@starjuice.net> <3F49E267.3070302@viatech.com.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3F49E267.3070302@viatech.com.cn> User-Agent: Mutt/1.5.4i Sender: Sheldon Hearn cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 10:19:51 -0000 On (2003/08/25 18:18), David Xu wrote: > >Does Jeff know about your patch? Perhaps he'd be willing to review and > >commit, since he's the original author of SCHED_ULE. > > > No, I think he does not know the patch. > Please don't commit it, there is a performance problem, after each time > a thread is exited, > I must exit a kse and then let scheduler reinitialized it. In general, I > needn't manage a struct kse, > it should be a scheduler internal data. Okay, so better to just use SCHED_4BSD for now, then. :-) Ciao, Sheldon. From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 11:04:13 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7B6D116A4BF for ; Mon, 25 Aug 2003 11:04:13 -0700 (PDT) Received: from freefall.freebsd.org (freefall.freebsd.org [216.136.204.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id DDE4A43FEC for ; Mon, 25 Aug 2003 11:04:04 -0700 (PDT) (envelope-from owner-bugmaster@freebsd.org) Received: from freefall.freebsd.org (peter@localhost [127.0.0.1]) by freefall.freebsd.org (8.12.9/8.12.9) with ESMTP id h7PI44Up030941 for ; Mon, 25 Aug 2003 11:04:04 -0700 (PDT) (envelope-from owner-bugmaster@freebsd.org) Received: (from peter@localhost) by freefall.freebsd.org (8.12.9/8.12.9/Submit) id h7PI43pC030935 for freebsd-threads@freebsd.org; Mon, 25 Aug 2003 11:04:03 -0700 (PDT) Date: Mon, 25 Aug 2003 11:04:03 -0700 (PDT) Message-Id: <200308251804.h7PI43pC030935@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: peter set sender to owner-bugmaster@freebsd.org using -f From: FreeBSD bugmaster To: freebsd-threads@FreeBSD.org Subject: Current problem reports assigned to you X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Aug 2003 18:04:13 -0000 Current FreeBSD problem reports Critical problems S Submitted Tracker Resp. Description ------------------------------------------------------------------------------- o [2000/06/13] kern/19247 threads uthread_sigaction.c does not do anything o [2002/01/16] kern/33951 threads pthread_cancel is ignored 2 problems total. Serious problems S Submitted Tracker Resp. Description ------------------------------------------------------------------------------- o [2001/01/19] bin/24472 threads libc_r does not honor SO_SNDTIMEO/SO_RCVT o [2001/01/25] bin/24632 threads libc_r delicate deviation from libc in ha o [2001/04/02] bin/26307 threads libc_r aborts when using the KDE media pl o [2001/10/31] bin/31661 threads pthread_kill signal handler doesn't get s o [2001/11/26] bin/32295 threads pthread dont dequeue signals o [2002/03/07] bin/35622 threads sigaltstack is missing in libc_r o [2002/06/27] bin/39922 threads [PATCH?] Threaded applications executed w o [2003/03/02] bin/48856 threads Setting SIGCHLD to SIG_IGN still leaves z o [2003/03/10] bin/49087 threads Signals lost in programs linked with libc a [2003/04/08] bin/50733 threads buildworld won't build, because of linkin o [2003/05/07] bin/51949 threads thread in accept cannot be cancelled o [2002/02/01] i386/34536 threads accept() blocks other threads o [2000/07/18] kern/20016 threads pthreads: Cannot set scheduling timer/Can o [2002/05/25] kern/38549 threads the procces compiled whith pthread stoppe o [2002/10/10] kern/43887 threads abnormal CPU useage when use pthread_mute o [2003/05/30] kern/52817 threads top(1) shows garbage for threaded process o [2000/08/26] misc/20861 threads libc_r does not honor socket timeouts o [2001/01/25] misc/24641 threads pthread_rwlock_rdlock can deadlock o [2002/08/04] misc/41331 threads Pthread library open sets O_NONBLOCK flag 19 problems total. Non-critical problems S Submitted Tracker Resp. Description ------------------------------------------------------------------------------- o [2001/09/09] bin/30464 threads pthread mutex attributes -- pshared o [2002/05/02] bin/37676 threads libc_r: msgsnd(), msgrcv(), pread(), pwri o [2000/05/25] misc/18824 threads gethostbyname is not thread safe o [2000/10/21] misc/22190 threads A threaded read(2) from a socketpair(2) f o [2002/07/16] misc/40671 threads pthread_cancel doesn't remove thread from 5 problems total. From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 20:42:26 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 87B8516A4BF; Mon, 25 Aug 2003 20:42:26 -0700 (PDT) Received: from mail.chesapeake.net (chesapeake.net [208.142.252.6]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5AEC043FBF; Mon, 25 Aug 2003 20:42:25 -0700 (PDT) (envelope-from jroberson@chesapeake.net) Received: from localhost (jroberson@localhost) by mail.chesapeake.net (8.11.6/8.11.6) with ESMTP id h7Q3fmv67730; Mon, 25 Aug 2003 23:41:49 -0400 (EDT) (envelope-from jroberson@chesapeake.net) Date: Mon, 25 Aug 2003 23:41:48 -0400 (EDT) From: Jeff Roberson To: David Xu In-Reply-To: <3F49E267.3070302@viatech.com.cn> Message-ID: <20030825234011.F12093-100000@mail.chesapeake.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 03:42:26 -0000 On Mon, 25 Aug 2003, David Xu wrote: > Sheldon Hearn wrote: > > >On (2003/08/25 15:47), David Xu wrote: > > > > > > > >>>I heartily endorse your patch. :-) > >>> > >>> > >>> > >>Thanks, I am pleased to see libkse works. :-) > >>However don't expect I will commit it, kernel scheduler interface is a > >>bit weird, so > >>I won't try to add another weird code. > >> > >> > > > >Does Jeff know about your patch? Perhaps he'd be willing to review and > >commit, since he's the original author of SCHED_ULE. > > > > > No, I think he does not know the patch. > Please don't commit it, there is a performance problem, after each time > a thread is exited, > I must exit a kse and then let scheduler reinitialized it. In general, I > needn't manage a struct kse, > it should be a scheduler internal data. > Why do you need to do adjustrunqueue() in sched_prio? I also don't understand the case in sched_switchout(). Can you please explain that? Other than that this patch seems fine. KSE is technically missing the proper sched api calls which is why ULE doesn't work with it. Cheers, Jeff From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 21:00:33 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 90AE916A4BF; Mon, 25 Aug 2003 21:00:33 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3F6C843FE0; Mon, 25 Aug 2003 21:00:15 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCRQ83; Tue, 26 Aug 2003 11:41:45 +0800 Message-ID: <3F4ADC3D.1080908@viatech.com.cn> Date: Tue, 26 Aug 2003 12:04:13 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Jeff Roberson References: <20030825234011.F12093-100000@mail.chesapeake.net> In-Reply-To: <20030825234011.F12093-100000@mail.chesapeake.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 04:00:33 -0000 Jeff Roberson wrote: >On Mon, 25 Aug 2003, David Xu wrote: > > > >>Sheldon Hearn wrote: >> >> >> >>>On (2003/08/25 15:47), David Xu wrote: >>> >>> >>> >>> >>> >>>>>I heartily endorse your patch. :-) >>>>> >>>>> >>>>> >>>>> >>>>> >>>>Thanks, I am pleased to see libkse works. :-) >>>>However don't expect I will commit it, kernel scheduler interface is a >>>>bit weird, so >>>>I won't try to add another weird code. >>>> >>>> >>>> >>>> >>>Does Jeff know about your patch? Perhaps he'd be willing to review and >>>commit, since he's the original author of SCHED_ULE. >>> >>> >>> >>> >>No, I think he does not know the patch. >>Please don't commit it, there is a performance problem, after each time >>a thread is exited, >>I must exit a kse and then let scheduler reinitialized it. In general, I >>needn't manage a struct kse, >>it should be a scheduler internal data. >> >> >> > >Why do you need to do adjustrunqueue() in sched_prio? I also don't >understand the case in sched_switchout(). Can you please explain that? > > adjustrunqueue maintains kg_last_assigned and related things, when a thread's priority is changed, the thread might no longer can be in scheduler's run queue, instead it will be in ksegrp's runqueue, because there is higher priority thread, and a KSE it attached should be detached now, and the KSE will attach to another higher priority thread, ULE ignores this requirement, as I can understand, ULE is only aware of 1:1 between KSE and thread. It would be nice if scheduler interface is thread aware but not kse aware. >Other than that this patch seems fine. KSE is technically missing the >proper sched api calls which is why ULE doesn't work with it. > >Cheers, >Jeff > >_______________________________________________ >freebsd-threads@freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-threads >To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" > > > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 21:08:38 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4772416A4BF; Mon, 25 Aug 2003 21:08:38 -0700 (PDT) Received: from mail.chesapeake.net (chesapeake.net [208.142.252.6]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3485543FBF; Mon, 25 Aug 2003 21:08:37 -0700 (PDT) (envelope-from jroberson@chesapeake.net) Received: from localhost (jroberson@localhost) by mail.chesapeake.net (8.11.6/8.11.6) with ESMTP id h7Q48VR79478; Tue, 26 Aug 2003 00:08:31 -0400 (EDT) (envelope-from jroberson@chesapeake.net) Date: Tue, 26 Aug 2003 00:08:31 -0400 (EDT) From: Jeff Roberson To: David Xu In-Reply-To: <3F4ADC3D.1080908@viatech.com.cn> Message-ID: <20030826000750.N12093-100000@mail.chesapeake.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 04:08:38 -0000 On Tue, 26 Aug 2003, David Xu wrote: > Jeff Roberson wrote: > > >On Mon, 25 Aug 2003, David Xu wrote: > > > > > > > >>Sheldon Hearn wrote: > >> > >> > >> > >>>On (2003/08/25 15:47), David Xu wrote: > >>> > >>> > >>> > >>> > >>> > >>>>>I heartily endorse your patch. :-) > >>>>> > >>>>> > >>>>> > >>>>> > >>>>> > >>>>Thanks, I am pleased to see libkse works. :-) > >>>>However don't expect I will commit it, kernel scheduler interface is a > >>>>bit weird, so > >>>>I won't try to add another weird code. > >>>> > >>>> > >>>> > >>>> > >>>Does Jeff know about your patch? Perhaps he'd be willing to review and > >>>commit, since he's the original author of SCHED_ULE. > >>> > >>> > >>> > >>> > >>No, I think he does not know the patch. > >>Please don't commit it, there is a performance problem, after each time > >>a thread is exited, > >>I must exit a kse and then let scheduler reinitialized it. In general, I > >>needn't manage a struct kse, > >>it should be a scheduler internal data. > >> > >> > >> > > > >Why do you need to do adjustrunqueue() in sched_prio? I also don't > >understand the case in sched_switchout(). Can you please explain that? > > > > > adjustrunqueue maintains kg_last_assigned and related things, when a > thread's priority is changed, > the thread might no longer can be in scheduler's run queue, instead it > will be in ksegrp's runqueue, > because there is higher priority thread, and a KSE it attached should be > detached now, and the KSE > will attach to another higher priority thread, ULE ignores this > requirement, as I can understand, > ULE is only aware of 1:1 between KSE and thread. > It would be nice if scheduler interface is thread aware but not kse aware. Yes, wouldn't it be nice.. I don't think it should be ksegrp aware either. oh well, it wasn't my design. Will you commit this patch? > > >Other than that this patch seems fine. KSE is technically missing the > >proper sched api calls which is why ULE doesn't work with it. > > > >Cheers, > >Jeff > > > >_______________________________________________ > >freebsd-threads@freebsd.org mailing list > >http://lists.freebsd.org/mailman/listinfo/freebsd-threads > >To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" > > > > > > > > > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 21:35:05 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5098016A4C0; Mon, 25 Aug 2003 21:35:05 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id 636A643F75; Mon, 25 Aug 2003 21:34:58 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCRRBP; Tue, 26 Aug 2003 12:16:31 +0800 Message-ID: <3F4AE462.3040808@viatech.com.cn> Date: Tue, 26 Aug 2003 12:38:58 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Jeff Roberson References: <20030826000750.N12093-100000@mail.chesapeake.net> In-Reply-To: <20030826000750.N12093-100000@mail.chesapeake.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 04:35:05 -0000 Jeff Roberson wrote: >>>> >>>> >>>Why do you need to do adjustrunqueue() in sched_prio? I also don't >>>understand the case in sched_switchout(). Can you please explain that? >>> >>> >>> >>> >>adjustrunqueue maintains kg_last_assigned and related things, when a >>thread's priority is changed, >>the thread might no longer can be in scheduler's run queue, instead it >>will be in ksegrp's runqueue, >>because there is higher priority thread, and a KSE it attached should be >>detached now, and the KSE >>will attach to another higher priority thread, ULE ignores this >>requirement, as I can understand, >>ULE is only aware of 1:1 between KSE and thread. >>It would be nice if scheduler interface is thread aware but not kse aware. >> >> > >Yes, wouldn't it be nice.. I don't think it should be ksegrp aware >either. oh well, it wasn't my design. > > SA process doesn't rely on kse and ksegrp because I introduced a kse_upcall structure, so I don't care someone drops kse or ksegrp and makes them as scheduler specific data structure. >Will you commit this patch? > > Will do. From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 21:58:12 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 246C516A4BF; Mon, 25 Aug 2003 21:58:12 -0700 (PDT) Received: from rwcrmhc12.comcast.net (rwcrmhc12.comcast.net [216.148.227.85]) by mx1.FreeBSD.org (Postfix) with ESMTP id 046FD43FFD; Mon, 25 Aug 2003 21:58:11 -0700 (PDT) (envelope-from julian@elischer.org) Received: from interjet.elischer.org ([12.233.125.100]) by attbi.com (rwcrmhc12) with ESMTP id <2003082604580301400i1cbue>; Tue, 26 Aug 2003 04:58:03 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id VAA27872; Mon, 25 Aug 2003 21:58:02 -0700 (PDT) Date: Mon, 25 Aug 2003 21:58:01 -0700 (PDT) From: Julian Elischer To: David Xu In-Reply-To: <3F4AE462.3040808@viatech.com.cn> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "deischen@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" cc: "threads@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 04:58:12 -0000 On Tue, 26 Aug 2003, David Xu wrote: > Jeff Roberson wrote: > > >>>> > >>>> > >>>Why do you need to do adjustrunqueue() in sched_prio? I also don't > >>>understand the case in sched_switchout(). Can you please explain that? > >>> > >>> > >>> > >>> > >>adjustrunqueue maintains kg_last_assigned and related things, when a > >>thread's priority is changed, > >>the thread might no longer can be in scheduler's run queue, instead it > >>will be in ksegrp's runqueue, > >>because there is higher priority thread, and a KSE it attached should be > >>detached now, and the KSE > >>will attach to another higher priority thread, ULE ignores this > >>requirement, as I can understand, > >>ULE is only aware of 1:1 between KSE and thread. > >>It would be nice if scheduler interface is thread aware but not kse aware. > >> > >> > > > >Yes, wouldn't it be nice.. I don't think it should be ksegrp aware > >either. oh well, it wasn't my design. > > > > > SA process doesn't rely on kse and ksegrp because I introduced a > kse_upcall structure, > so I don't care someone drops kse or ksegrp and makes them as scheduler > specific data structure. Well, this is not quite true. without KSEGRPS there is no possibility to make both process scope and system scope threads. process scope threads require a rendevous structure of some sort and it can not be the process. The fact that the 1:1 threads don't do this is why they can not do process-scope threads and system scope threads but are system scope only. MACH didn't have the additional concept of the KSEGRP and the contortions they had to go to to try do process scope threads (they eventually gave up) (I was a MACH user at that time) was incredible. > > >Will you commit this patch? > > > > > Will do. > > > > _______________________________________________ > freebsd-threads@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-threads > To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 22:18:45 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 26D4F16A4BF; Mon, 25 Aug 2003 22:18:45 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id EAF2D43FD7; Mon, 25 Aug 2003 22:18:35 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCRRGG; Tue, 26 Aug 2003 13:00:05 +0800 Message-ID: <3F4AEE98.4080607@viatech.com.cn> Date: Tue, 26 Aug 2003 13:22:32 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Julian Elischer References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 05:18:45 -0000 Julian Elischer wrote: >On Tue, 26 Aug 2003, David Xu wrote: > > > >>Jeff Roberson wrote: >> >> >> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>Why do you need to do adjustrunqueue() in sched_prio? I also don't >>>>>understand the case in sched_switchout(). Can you please explain that? >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>adjustrunqueue maintains kg_last_assigned and related things, when a >>>>thread's priority is changed, >>>>the thread might no longer can be in scheduler's run queue, instead it >>>>will be in ksegrp's runqueue, >>>>because there is higher priority thread, and a KSE it attached should be >>>>detached now, and the KSE >>>>will attach to another higher priority thread, ULE ignores this >>>>requirement, as I can understand, >>>>ULE is only aware of 1:1 between KSE and thread. >>>>It would be nice if scheduler interface is thread aware but not kse aware. >>>> >>>> >>>> >>>> >>>Yes, wouldn't it be nice.. I don't think it should be ksegrp aware >>>either. oh well, it wasn't my design. >>> >>> >>> >>> >>SA process doesn't rely on kse and ksegrp because I introduced a >>kse_upcall structure, >>so I don't care someone drops kse or ksegrp and makes them as scheduler >>specific data structure. >> >> > >Well, this is not quite true. >without KSEGRPS there is no possibility to make both >process scope and system scope threads. > >process scope threads require a rendevous structure of some sort >and it can not be the process. > >The fact that the 1:1 threads don't do this is why they can not >do process-scope threads and system scope threads but are system scope >only. > > OK, I just don't want to manage kse, what I want is to set a concurrent level for a ksegrp, for example, a process-scope thread's group has a concurrent level equals number of cpu, so please don't force me to create and destroy kse, make them transparent. >MACH didn't have the additional concept of the KSEGRP and the >contortions they had to go to to try do process scope threads (they >eventually gave up) (I was a MACH user at that time) was incredible. > > > >>>Will you commit this patch? >>> >>> >>> >>> >>Will do. >> >> >> >>_______________________________________________ >>freebsd-threads@freebsd.org mailing list >>http://lists.freebsd.org/mailman/listinfo/freebsd-threads >>To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" >> >> >> > >_______________________________________________ >freebsd-threads@freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-threads >To unsubscribe, send any mail to "freebsd-threads-unsubscribe@freebsd.org" > > > From owner-freebsd-threads@FreeBSD.ORG Mon Aug 25 22:32:25 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 667A116A4BF; Mon, 25 Aug 2003 22:32:25 -0700 (PDT) Received: from rwcrmhc12.comcast.net (rwcrmhc12.comcast.net [216.148.227.85]) by mx1.FreeBSD.org (Postfix) with ESMTP id A558543FBD; Mon, 25 Aug 2003 22:32:24 -0700 (PDT) (envelope-from julian@elischer.org) Received: from interjet.elischer.org ([12.233.125.100]) by attbi.com (rwcrmhc12) with ESMTP id <2003082605322301400f48dre>; Tue, 26 Aug 2003 05:32:24 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id WAA28178; Mon, 25 Aug 2003 22:32:23 -0700 (PDT) Date: Mon, 25 Aug 2003 22:32:22 -0700 (PDT) From: Julian Elischer To: David Xu In-Reply-To: <3F4AEE98.4080607@viatech.com.cn> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "deischen@freebsd.org" cc: "threads@freebsd.org" cc: Sheldon Hearn cc: "freebsd-java@freebsd.org" Subject: Re: vmark hangs with libthr and libkse X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Aug 2003 05:32:25 -0000 On Tue, 26 Aug 2003, David Xu wrote: > Julian Elischer wrote: > > >On Tue, 26 Aug 2003, David Xu wrote: > > > > > > > >>Jeff Roberson wrote: > >> > >> > OK, I just don't want to manage kse, what I want is to set a > concurrent level for a ksegrp, > for example, a process-scope thread's group has a concurrent level > equals number of cpu, > so please don't force me to create and destroy kse, make them transparent. > I gave patches to peter to do that a couple of months ago but I can't touch the "sched" interface without starting a commit war. after the patches only 3 files know about the kse structure. From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 09:35:30 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F0AC616A4BF for ; Wed, 27 Aug 2003 09:35:29 -0700 (PDT) Received: from rwcrmhc13.comcast.net (rwcrmhc13.comcast.net [204.127.198.39]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7FA4B43FCB for ; Wed, 27 Aug 2003 09:35:28 -0700 (PDT) (envelope-from julian@elischer.org) Received: from interjet.elischer.org ([12.233.125.100]) by attbi.com (rwcrmhc13) with ESMTP id <2003082716222801500d1jhne>; Wed, 27 Aug 2003 16:22:28 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id JAA45444 for ; Wed, 27 Aug 2003 09:22:27 -0700 (PDT) X-Received: from mx2.freebsd.org (mx2.freebsd.org [216.136.204.119]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id JAA45389 for ; Wed, 27 Aug 2003 09:17:11 -0700 (PDT) X-Received: from hub.freebsd.org (hub.freebsd.org [216.136.204.18]) by mx2.freebsd.org (Postfix) with ESMTP id 3216C55AEA for ; Wed, 27 Aug 2003 09:17:08 -0700 (PDT) (envelope-from owner-freebsd-hackers@freebsd.org) X-Received: by hub.freebsd.org (Postfix) id DD30116A4D7; Wed, 27 Aug 2003 09:17:07 -0700 (PDT) Delivered-To: julian@freebsd.org X-Received: from hub.freebsd.org (localhost [127.0.0.1]) by hub.freebsd.org (Postfix) with ESMTP id C555F16A4C1; Wed, 27 Aug 2003 09:17:07 -0700 (PDT) Delivered-To: freebsd-hackers@freebsd.org X-Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C956D16A4BF; Wed, 27 Aug 2003 09:16:35 -0700 (PDT) X-Received: from vexpert.dbai.tuwien.ac.at (vexpert.dbai.tuwien.ac.at [128.131.111.2]) by mx1.FreeBSD.org (Postfix) with ESMTP id 16FF843FDD; Wed, 27 Aug 2003 09:16:35 -0700 (PDT) (envelope-from gerald@pfeifer.com) X-Received: from [128.131.111.52] (naos [128.131.111.52]) by vexpert.dbai.tuwien.ac.at (Postfix) with ESMTP id 86B3D13793; Wed, 27 Aug 2003 18:16:33 +0200 (CEST) Date: Wed, 27 Aug 2003 18:16:30 +0200 (CEST) From: Gerald Pfeifer To: Scott Long In-Reply-To: <3F4CD6C8.7010103@freebsd.org> Message-ID: References: <3F4CD6C8.7010103@freebsd.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Sender: owner-freebsd-hackers@freebsd.org Errors-To: owner-freebsd-hackers@freebsd.org ReSent-Date: Wed, 27 Aug 2003 09:22:18 -0700 (PDT) Resent-From: Julian Elischer Resent-To: threads@freebsd.org ReSent-Subject: Re: Call for thread testers ReSent-Message-ID: cc: hackers@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 16:35:30 -0000 On Wed, 27 Aug 2003, Scott Long wrote: > Packages that I would most like to see are: > [...] > Others? Wine? (ports/emulators/wine) Though I'm currently fighting to get the CVS version cooperate on 4-STABLE, given recent changes to the threading support in Wine. > The questions that I would like to have answered are: > > Does it take advantage of pthreads > Does it work with libkse and libthr > Does it operate correctly > Does it perform better/worse than with libc_r > Does it perform better/worse than under FreeBSD 4.x What would be extremely helpful, IMHO, would be guidelines how to use the various kinds of threading support (especially in the ports collection). Especially, how should we update ports so that they work both on 4.x and 5.x and, optionally, take advantage of the other threading packages? Is there some nice overview somewhere? Gerald PS: For example, our system compiler on -STABLE fails to build Wine if sources are compiled with -pthread, but linked without this option, while this works fine with stock GCC 3.2.3 (which is most likely due to problems in Wine, not FreeBSD or GCC, but still a problem). -- Gerald Pfeifer (Jerry) gerald@pfeifer.com http://www.pfeifer.com/gerald/ _______________________________________________ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org" From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 09:45:57 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7D2DD16A4BF; Wed, 27 Aug 2003 09:45:57 -0700 (PDT) Received: from perrin.nxad.com (internal.ext.nxad.com [69.1.70.251]) by mx1.FreeBSD.org (Postfix) with ESMTP id 89F9443FF7; Wed, 27 Aug 2003 09:45:54 -0700 (PDT) (envelope-from sean@nxad.com) Received: by perrin.nxad.com (Postfix, from userid 1001) id 003CF21058; Wed, 27 Aug 2003 09:45:53 -0700 (PDT) Date: Wed, 27 Aug 2003 09:45:53 -0700 From: Sean Chittenden To: Scott Long Message-ID: <20030827164553.GC51167@perrin.nxad.com> References: <3F4CD6C8.7010103@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3F4CD6C8.7010103@freebsd.org> X-PGP-Key: finger seanc@FreeBSD.org X-PGP-Fingerprint: 3849 3760 1AFE 7B17 11A0 83A6 DD99 E31F BC84 B341 X-Web-Homepage: http://sean.chittenden.org/ User-Agent: Mutt/1.5.4i cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 16:45:57 -0000 [moving to -threads] > This is kind of an unconventional call for help. As we approach the > release of 5.2, we'd really like to show off the performance and > stability of our new threading packages. So, I'm looking for people > to volunteer to go out and put some of the thread-capable enterprise > and desktop packages to the test. Packages that I would most like > to see are: > > Apache 2 > BIND 9 > OpenOffice > Mozilla > KDE desktop + Konq > Java 1.3/1.4 + Tomcat/etc > MySQL > Others? > > The questions that I would like to have answered are: > > Does it take advantage of pthreads > Does it work with libkse and libthr > Does it operate correctly > Does it perform better/worse than with libc_r > Does it perform better/worse than under FreeBSD 4.x What's the current status of KSE? Is the project page (http://www.FreeBSD.org/kse/) considered up to date by the KSE developers? Are the remaining bits for KSE either bug related or dealing with ARCH != x86? Are there any remaining/outstanding "gotchas"/bugs on the radar/problems that the KSE developers are aware of? I've heard that the konsole bug has been fixed and there's a patch floating around to fix the SIGABRTs for that program (was my biggest gripe last time I tested libkse). Just trying to figure out what my expectations should be with regards to testing at this point. Is KSE to the point that it's ready for performance testing or is stability still a concern? tia, -sc -- Sean Chittenden From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 10:07:03 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E824E16A4BF for ; Wed, 27 Aug 2003 10:07:03 -0700 (PDT) Received: from smtp.mho.com (smtp.mho.net [64.58.4.6]) by mx1.FreeBSD.org (Postfix) with SMTP id 0004643FA3 for ; Wed, 27 Aug 2003 10:07:02 -0700 (PDT) (envelope-from scottl@freebsd.org) Received: (qmail 34996 invoked by uid 1002); 27 Aug 2003 17:07:02 -0000 Received: from unknown (HELO freebsd.org) (64.58.1.252) by smtp.mho.net with SMTP; 27 Aug 2003 17:07:02 -0000 Message-ID: <3F4CE52A.3050409@freebsd.org> Date: Wed, 27 Aug 2003 11:06:50 -0600 From: Scott Long User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.3) Gecko/20030425 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sean Chittenden References: <3F4CD6C8.7010103@freebsd.org> <20030827164553.GC51167@perrin.nxad.com> In-Reply-To: <20030827164553.GC51167@perrin.nxad.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 17:07:04 -0000 Sean Chittenden wrote: > [moving to -threads] > > >>This is kind of an unconventional call for help. As we approach the >>release of 5.2, we'd really like to show off the performance and >>stability of our new threading packages. So, I'm looking for people >>to volunteer to go out and put some of the thread-capable enterprise >>and desktop packages to the test. Packages that I would most like >>to see are: >> >>Apache 2 >>BIND 9 >>OpenOffice >>Mozilla >>KDE desktop + Konq >>Java 1.3/1.4 + Tomcat/etc >>MySQL >>Others? >> >>The questions that I would like to have answered are: >> >>Does it take advantage of pthreads >>Does it work with libkse and libthr >>Does it operate correctly >>Does it perform better/worse than with libc_r >>Does it perform better/worse than under FreeBSD 4.x > > > What's the current status of KSE? Is the project page > (http://www.FreeBSD.org/kse/) considered up to date by the KSE > developers? Are the remaining bits for KSE either bug related or > dealing with ARCH != x86? Are there any remaining/outstanding > "gotchas"/bugs on the radar/problems that the KSE developers are aware > of? I'll review the page today and see about getting it updated if needed. > > I've heard that the konsole bug has been fixed and there's a patch > floating around to fix the SIGABRTs for that program (was my biggest > gripe last time I tested libkse). That patch was added to the kde port, correct? > > Just trying to figure out what my expectations should be with regards > to testing at this point. Is KSE to the point that it's ready for > performance testing or is stability still a concern? > David Xu has been doing a lot of bugfixing work over the summer, so stability should be good enough to start tests. What I'm really looking at is plain data collection. Not only are performance numbers nice, but success/failure/disaster reports are incredibly valuable in making the thread libraries sellable and credible. A simple, "Apache2 won't start when linked with libkse" will get people motivated to investigate and fix it. Scott From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 10:18:56 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id ADBD916A4BF; Wed, 27 Aug 2003 10:18:56 -0700 (PDT) Received: from perrin.nxad.com (internal.ext.nxad.com [69.1.70.251]) by mx1.FreeBSD.org (Postfix) with ESMTP id DD41543FAF; Wed, 27 Aug 2003 10:18:55 -0700 (PDT) (envelope-from sean@nxad.com) Received: by perrin.nxad.com (Postfix, from userid 1001) id 46CCE20F01; Wed, 27 Aug 2003 10:18:55 -0700 (PDT) Date: Wed, 27 Aug 2003 10:18:55 -0700 From: Sean Chittenden To: Scott Long Message-ID: <20030827171855.GA72355@perrin.nxad.com> References: <3F4CD6C8.7010103@freebsd.org> <20030827164553.GC51167@perrin.nxad.com> <3F4CE52A.3050409@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3F4CE52A.3050409@freebsd.org> X-PGP-Key: finger seanc@FreeBSD.org X-PGP-Fingerprint: 3849 3760 1AFE 7B17 11A0 83A6 DD99 E31F BC84 B341 X-Web-Homepage: http://sean.chittenden.org/ User-Agent: Mutt/1.5.4i cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 17:18:56 -0000 > >I've heard that the konsole bug has been fixed and there's a patch > >floating around to fix the SIGABRTs for that program (was my biggest > >gripe last time I tested libkse). > > That patch was added to the kde port, correct? Correct. > >Just trying to figure out what my expectations should be with > >regards to testing at this point. Is KSE to the point that it's > >ready for performance testing or is stability still a concern? > > David Xu has been doing a lot of bugfixing work over the summer, so > stability should be good enough to start tests. I've been watching him and the rest of the kse folks work with great interest. I was hoping for an all clear of sorts from them or possibly a list of what's remaining before they turn things over to the public for bug gnashing. -sc -- Sean Chittenden From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 10:46:48 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B000916A4C0; Wed, 27 Aug 2003 10:46:48 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id AE5DB43FF5; Wed, 27 Aug 2003 10:46:47 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7RHkjtp006999; Wed, 27 Aug 2003 13:46:45 -0400 (EDT) Date: Wed, 27 Aug 2003 13:46:45 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Sean Chittenden In-Reply-To: <20030827164553.GC51167@perrin.nxad.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 17:46:48 -0000 On Wed, 27 Aug 2003, Sean Chittenden wrote: > [moving to -threads] > > What's the current status of KSE? Is the project page > (http://www.FreeBSD.org/kse/) considered up to date by the KSE The only thing that really isn't quite up-to-date is "Efficient scope system threads" which is done and the MD functions for non-x86 platforms, which should add Jake for sparc64. AMD64 has been done by David and myself. Alpha userland is done (needs debugging) but kernel needs a couple of functions (Marcel) implemented. > developers? Are the remaining bits for KSE either bug related or > dealing with ARCH != x86? Are there any remaining/outstanding Mostly archs == ia64, alpha, and sparc. > "gotchas"/bugs on the radar/problems that the KSE developers are aware > of? No. I think libkse works better than libc_r at this point, and has more implemented (wrt scope process/system threads and concurrency level). > I've heard that the konsole bug has been fixed and there's a patch > floating around to fix the SIGABRTs for that program (was my biggest > gripe last time I tested libkse). > > Just trying to figure out what my expectations should be with regards > to testing at this point. Is KSE to the point that it's ready for > performance testing or is stability still a concern? As far as I can tell, it is stable enough for everything I've thrown at it. But we need people to use it with Apache, MySQL, etc, in more real-world environments and tell us what they find. It would be nice if we had our own test team that could do this for us. As developers, it's very hard to find the time to work on the implementation and to set up environments in which to test Apache, MySQL, etc. It's not just installing those ports; it's also knowing how to tune them, feed them with real data, etc. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 10:55:23 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7D54916A4DB; Wed, 27 Aug 2003 10:55:23 -0700 (PDT) Received: from perrin.nxad.com (internal.ext.nxad.com [69.1.70.251]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9E77443FEC; Wed, 27 Aug 2003 10:55:22 -0700 (PDT) (envelope-from sean@nxad.com) Received: by perrin.nxad.com (Postfix, from userid 1001) id 20F922105C; Wed, 27 Aug 2003 10:55:22 -0700 (PDT) Date: Wed, 27 Aug 2003 10:55:22 -0700 From: Sean Chittenden To: Daniel Eischen Message-ID: <20030827175522.GB72355@perrin.nxad.com> References: <20030827164553.GC51167@perrin.nxad.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-PGP-Key: finger seanc@FreeBSD.org X-PGP-Fingerprint: 3849 3760 1AFE 7B17 11A0 83A6 DD99 E31F BC84 B341 X-Web-Homepage: http://sean.chittenden.org/ User-Agent: Mutt/1.5.4i cc: threads@freebsd.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 17:55:23 -0000 > > "gotchas"/bugs on the radar/problems that the KSE developers are > > aware of? > > No. I think libkse works better than libc_r at this point, and has > more implemented (wrt scope process/system threads and concurrency > level). Very cool, this is what I'd been waiting to hear. *updates /usr/src and begins build* Thanks for the info! -sc -- Sean Chittenden From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 13:20:18 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B26EF16A56D for ; Wed, 27 Aug 2003 13:20:18 -0700 (PDT) Received: from pf178.bochnia.sdi.tpnet.pl (pf178.bochnia.sdi.tpnet.pl [217.97.94.178]) by mx1.FreeBSD.org (Postfix) with ESMTP id CC82E43FBD for ; Wed, 27 Aug 2003 13:20:17 -0700 (PDT) (envelope-from fahren@student.uci.agh.edu.pl) Received: by pf178.bochnia.sdi.tpnet.pl (Postfix, from userid 666) id E7947969; Wed, 27 Aug 2003 22:20:09 +0200 (CEST) Date: Wed, 27 Aug 2003 22:20:09 +0200 From: Maciej Freudenheim To: threads@freebsd.org Message-ID: <20030827202009.GA44225@piggie> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="1yeeQ81UyVL57Vl7" Content-Disposition: inline X-PGP-Key-URL: http://student.uci.agh.edu.pl/~fahren/fahren.gpg User-Agent: Mutt/1.5.4i Subject: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 20:20:18 -0000 --1yeeQ81UyVL57Vl7 Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hello! I was tuning xmms some time ago. Now i'm running it linked with libthr for about 2 weeks and i haven't observed any problems. Here are few numbers: libc_r.so.5 =3D> /usr/lib/libc_r.so.5 (0x28336000) 45197 fahren 98 0 17984K 14708K select 0:46 3.08% 3.08% xmms libc_r.so.5 =3D> /usr/lib/libkse.so.1 (0x28336000) 45261 fahren 98 0 17236K 14384K select 0:11 2.80% 2.78% xmms libc_r.so.5 =3D> /usr/lib/libthr.so.1 (0x28336000) 45347 fahren 97 0 17236K 14384K select 0:10 2.41% 2.39% xmms The tests were all done of course on the same version of xmms (cvs from=20 Tue Aug 12 2003) with the same settings and playing the same mp3 file=20 ('dream of mirrors' :) OS version was FreeBSD-current from Aug 23. So, > Does it work with libkse and libthr=20 Yes. > Does it operate correctly Yes. > Does it perform better/worse than with libc_r Better. 1:1 libthr is the winner. > Does it perform better/worse than under FreeBSD 4.x=20 Much better. Xmms is eating about 10% cpu on the same machine with FreeBSD 4.x. Hope it is useful to somebody :) fahren. --1yeeQ81UyVL57Vl7 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (FreeBSD) iD8DBQE/TRJ5iZPiEPhhqGARAtMgAJ92QXoqavxQoKV+BOTEBfid6GJ9lACgst56 AbNZTovgCMaHKw+tGcA8QLg= =75bn -----END PGP SIGNATURE----- --1yeeQ81UyVL57Vl7-- From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 14:06:20 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A7FE516A4BF; Wed, 27 Aug 2003 14:06:20 -0700 (PDT) Received: from silver.he.iki.fi (silver.he.iki.fi [193.64.42.241]) by mx1.FreeBSD.org (Postfix) with ESMTP id ECCEB43FCB; Wed, 27 Aug 2003 14:06:18 -0700 (PDT) (envelope-from pete@he.iki.fi) Received: from he.iki.fi (h81.vuokselantie10.fi [193.64.42.129]) by silver.he.iki.fi (8.12.9/8.11.4) with ESMTP id h7RL6E5L009984; Thu, 28 Aug 2003 00:06:15 +0300 (EEST) (envelope-from pete@he.iki.fi) Message-ID: <3F4D1D43.7080509@he.iki.fi> Date: Thu, 28 Aug 2003 00:06:11 +0300 From: Petri Helenius User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sean Chittenden References: <3F4CD6C8.7010103@freebsd.org> <20030827164553.GC51167@perrin.nxad.com> In-Reply-To: <20030827164553.GC51167@perrin.nxad.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit cc: threads@freebsd.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 21:06:20 -0000 Sean Chittenden wrote: >>MySQL >> I´ve been running MySQL 4.0 on our test-boxes with libc_r mapped to libkse for a few months now, not too heavy a load but after the signals got fixed a few months ago I haven´t detected issues. >>Others? >> >> Is there a reason why libmap.conf is not enabled by default? >What's the current status of KSE? Is the project page >(http://www.FreeBSD.org/kse/) considered up to date by the KSE >developers? Are the remaining bits for KSE either bug related or >dealing with ARCH != x86? Are there any remaining/outstanding >"gotchas"/bugs on the radar/problems that the KSE developers are aware >of? > > I´ve been happy with KSE since mutexes got their pre-emption points, now SMP and SMPwHTT seem to really gain more speed above and beyond c_r and uniprocessor environment. Can´t wait to get the 8-way Opteron boxes to give them a spin. Pete From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 14:11:51 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3A1E516A4BF; Wed, 27 Aug 2003 14:11:51 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7554A43F75; Wed, 27 Aug 2003 14:11:50 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7RLBmtp013911; Wed, 27 Aug 2003 17:11:48 -0400 (EDT) Date: Wed, 27 Aug 2003 17:11:48 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Petri Helenius In-Reply-To: <3F4D1D43.7080509@he.iki.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=X-UNKNOWN Content-Transfer-Encoding: QUOTED-PRINTABLE cc: threads@freebsd.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 21:11:51 -0000 On Thu, 28 Aug 2003, Petri Helenius wrote: >=20 > I=B4ve been happy with KSE since mutexes got their pre-emption points, no= w=20 > SMP and SMPwHTT > seem to really gain more speed above and beyond c_r and uniprocessor=20 > environment. And it was very helpful to get Petri's feedback; we can always use more like that :) --=20 Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 15:07:07 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 80CC416A4BF; Wed, 27 Aug 2003 15:07:07 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id C512A43F75; Wed, 27 Aug 2003 15:07:06 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7RM76tp023470; Wed, 27 Aug 2003 18:07:06 -0400 (EDT) Date: Wed, 27 Aug 2003 18:07:06 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: threads@FreeBSD.org In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 22:07:07 -0000 [ redirected to @threads ] On Wed, 27 Aug 2003, Daniel Eischen wrote: > On Wed, 27 Aug 2003, Scott Long wrote: > > > All, > > > > This is kind of an unconventional call for help. As we approach the > > release of 5.2, we'd really like to show off the performance and > > stability of our new threading packages. So, I'm looking for people > > to volunteer to go out and put some of the thread-capable enterprise > > and desktop packages to the test. Packages that I would most like > > to see are: > > > > Apache 2 > > BIND 9 > > OpenOffice > > Mozilla > > KDE desktop + Konq > > Java 1.3/1.4 + Tomcat/etc > > MySQL > > Others? > > As to OpenOffice, Mozilla, and KDE/Konqueror, David and I have > been using them all with libkse with no noticable difference. > I don't know how you'd judge performance with these, since > it is more "how it feels". > > You shouldn't use jdk13 with libkse or libthr since it relies > on the internals of the thread library (libc_r). It is only useful > to benchmark jdk14 which, in theory, should work with all our > native thread libraries. > > > The questions that I would like to have answered are: > > > > Does it take advantage of pthreads > > Does it work with libkse and libthr > > Does it operate correctly > > Does it perform better/worse than with libc_r > > Does it perform better/worse than under FreeBSD 4.x > > One thing I have noticed about Mozilla, is that it does > create both scope process and scope system threads, which > was kind of nice to see. > > The ACE library is threaded and includes a set of tests; > some of which show performance: > > http://www.cs.wustl.edu/~schmidt/ACE.html > > David and I have a couple of other tests which we use to > measure performance of our threads libraries. One of these tests is crew, which is part of Dave Butenhof's threads tests (http://people.freebsd.org/~davidxu/crew/programs.tar.z) Usage: crew crew searches recursively for the given string in the directory. David and I use it as follows: $ crew node /usr/src/sys By default, it creates 4 worker threads to seach for the string. Here are the results for all thread libraries; they are somewhat surprising. The test was run a couple of times to fill up cache and those timing results were omitted. Run 1 Run 2 Run 3 ------------------------------------------------------ libc_r real 0m48.451s 0m48.835s 0m50.905s user 0m2.694s 0m2.745s 0m2.818s sys 0m8.996s 0m8.931s 0m8.863s ------------------------------------------------------ libthr real 1m2.553s 1m52.579s 1m50.306s user 0m3.423s 0m4.561s 0m4.397s sys 0m11.842s 0m27.099s 0m27.300s ------------------------------------------------------ libkse(M:N) real 0m49.175s 0m49.910s 0m51.033s user 0m2.840s 0m2.836s 0m2.880s sys 0m8.857s 0m8.857s 0m8.792s ------------------------------------------------------ libkse(1:1) real 0m20.240s 0m19.969s 0m19.981s user 0m2.741s 0m2.783s 0m2.858s sys 0m8.877s 0m8.832s 0m8.768s ------------------------------------------------------ All the above tests were run on a single CPU 850MHz PIII with 256MB RAM (Dell Lattitude C400). There seems to be something wrong with libthr; the times vary too much (by almost 100%). libkse in M:N mode seems to be on par with libc_r. But the real surprise is libkse built in 1:1 mode (add -DSYSTEM_SCOPE_ONLY to CFLAGS when building libpthread). -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 16:20:37 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3AC9E16A4BF; Wed, 27 Aug 2003 16:20:37 -0700 (PDT) Received: from freefall.freebsd.org (freefall.freebsd.org [216.136.204.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id 98CFE43FDF; Wed, 27 Aug 2003 16:20:36 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) Received: from localhost (davidxu@localhost [127.0.0.1]) by freefall.freebsd.org (8.12.9/8.12.9) with ESMTP id h7RNKXUp034104; Wed, 27 Aug 2003 16:20:34 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) From: David Xu To: deischen@FreeBSD.org, Daniel Eischen , threads@FreeBSD.org Date: Thu, 28 Aug 2003 07:23:35 +0800 User-Agent: KMail/1.5.2 References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308280723.35697.davidxu@FreeBSD.org> cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: davidxu@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 23:20:37 -0000 On Thursday 28 August 2003 06:07, Daniel Eischen wrote: > [ redirected to @threads ] > > David and I have a couple of other tests which we use to > > measure performance of our threads libraries. > > One of these tests is crew, which is part of Dave Butenhof's > threads tests (http://people.freebsd.org/~davidxu/crew/programs.tar.z) > > Usage: crew > > crew searches recursively for the given string in the directory. > David and I use it as follows: > > $ crew node /usr/src/sys > > By default, it creates 4 worker threads to seach for the string. > Here are the results for all thread libraries; they are somewhat > surprising. The test was run a couple of times to fill up cache > and those timing results were omitted. > > Run 1 Run 2 Run 3 > ------------------------------------------------------ > libc_r real 0m48.451s 0m48.835s 0m50.905s > user 0m2.694s 0m2.745s 0m2.818s > sys 0m8.996s 0m8.931s 0m8.863s > ------------------------------------------------------ > libthr real 1m2.553s 1m52.579s 1m50.306s > user 0m3.423s 0m4.561s 0m4.397s > sys 0m11.842s 0m27.099s 0m27.300s > ------------------------------------------------------ > libkse(M:N) real 0m49.175s 0m49.910s 0m51.033s > user 0m2.840s 0m2.836s 0m2.880s > sys 0m8.857s 0m8.857s 0m8.792s > ------------------------------------------------------ > libkse(1:1) real 0m20.240s 0m19.969s 0m19.981s > user 0m2.741s 0m2.783s 0m2.858s > sys 0m8.877s 0m8.832s 0m8.768s > ------------------------------------------------------ > > All the above tests were run on a single CPU 850MHz PIII > with 256MB RAM (Dell Lattitude C400). > > There seems to be something wrong with libthr; the times > vary too much (by almost 100%). libkse in M:N mode seems > to be on par with libc_r. But the real surprise is libkse > built in 1:1 mode (add -DSYSTEM_SCOPE_ONLY to CFLAGS when > building libpthread). Er, strange result, libkse M:N and libkse 1:1 should be identical in this test, I had always run this test, and their performance are always close enough, but 1:1 was always a bit slow than M:N, all tests are done under 4BSD scheduler, under ULE scheduler, SA process has performance problem, M:N is slower under ULE than 1:1. David Xu From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 16:57:20 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 106B816A4BF; Wed, 27 Aug 2003 16:57:20 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 48F8443FDD; Wed, 27 Aug 2003 16:57:19 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7RNvItp011938; Wed, 27 Aug 2003 19:57:18 -0400 (EDT) Date: Wed, 27 Aug 2003 19:57:18 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: David Xu In-Reply-To: <200308280723.35697.davidxu@FreeBSD.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@FreeBSD.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 23:57:20 -0000 On Thu, 28 Aug 2003, David Xu wrote: > On Thursday 28 August 2003 06:07, Daniel Eischen wrote: > > [ redirected to @threads ] > > > David and I have a couple of other tests which we use to > > > measure performance of our threads libraries. > > > > One of these tests is crew, which is part of Dave Butenhof's > > threads tests (http://people.freebsd.org/~davidxu/crew/programs.tar.z) > > > > Usage: crew > > > > crew searches recursively for the given string in the directory. > > David and I use it as follows: > > > > $ crew node /usr/src/sys > > > > By default, it creates 4 worker threads to seach for the string. > > Here are the results for all thread libraries; they are somewhat > > surprising. The test was run a couple of times to fill up cache > > and those timing results were omitted. > > > > Run 1 Run 2 Run 3 > > ------------------------------------------------------ > > libc_r real 0m48.451s 0m48.835s 0m50.905s > > user 0m2.694s 0m2.745s 0m2.818s > > sys 0m8.996s 0m8.931s 0m8.863s > > ------------------------------------------------------ > > libthr real 1m2.553s 1m52.579s 1m50.306s > > user 0m3.423s 0m4.561s 0m4.397s > > sys 0m11.842s 0m27.099s 0m27.300s > > ------------------------------------------------------ > > libkse(M:N) real 0m49.175s 0m49.910s 0m51.033s > > user 0m2.840s 0m2.836s 0m2.880s > > sys 0m8.857s 0m8.857s 0m8.792s > > ------------------------------------------------------ > > libkse(1:1) real 0m20.240s 0m19.969s 0m19.981s > > user 0m2.741s 0m2.783s 0m2.858s > > sys 0m8.877s 0m8.832s 0m8.768s > > ------------------------------------------------------ > > > > All the above tests were run on a single CPU 850MHz PIII > > with 256MB RAM (Dell Lattitude C400). > > > > There seems to be something wrong with libthr; the times > > vary too much (by almost 100%). libkse in M:N mode seems > > to be on par with libc_r. But the real surprise is libkse > > built in 1:1 mode (add -DSYSTEM_SCOPE_ONLY to CFLAGS when > > building libpthread). > > Er, strange result, libkse M:N and libkse 1:1 should be identical > in this test, I had always run this test, and their performance > are always close enough, but 1:1 was always a bit slow than M:N, > all tests are done under 4BSD scheduler, under ULE scheduler, > SA process has performance problem, M:N is slower under ULE than 1:1. Yes, perhaps my kernel was a bit out of date. I also had forgotten I had libc_r mapped to libkse with libmap.conf, so the libc_r tests were actually using libkse! I re-ran the tests on a different box, single PIII 800MHz, 512MB RAM. They look better, although libthr still doesn't give consistent results. Run 1 Run 2 Run 3 ----------------------------------------------------------- libc_r real 0m13.739s 0m13.739s 0m13.882s user 0m3.330s 0m3.302s 0m3.394s sys 0m9.858s 0m9.893s 0m9.820s ----------------------------------------------------------- libkse(M:N) real 0m11.977s 0m12.199s 0m12.097s user 0m3.248s 0m3.081s 0m2.857s sys 0m8.190s 0m8.517s 0m8.575s ----------------------------------------------------------- libkse(1:1) real 0m11.972s 0m12.044s 0m12.035s user 0m3.198s 0m2.980s 0m3.183s sys 0m8.244s 0m8.480s 0m8.282s ----------------------------------------------------------- libthr real 0m34.180s 0m16.193s 0m34.119s user 0m5.075s 0m3.874s 0m5.255s sys 0m28.286s 0m11.626s 0m28.038s ----------------------------------------------------------- libkse(1:1) and libkse(M:N) are about equal, and slightly better than libc_r. I can't explain libthr results. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 16:58:56 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0005916A4BF for ; Wed, 27 Aug 2003 16:58:55 -0700 (PDT) Received: from ns2.gnf.org (ns2.gnf.org [63.196.132.68]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8258443FD7 for ; Wed, 27 Aug 2003 16:58:54 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from EXCHCLUSTER01.lj.gnf.org (exch01.lj.gnf.org [172.25.10.19]) by ns2.gnf.org (8.12.8p1/8.12.8) with ESMTP id h7RNwooq056463 for ; Wed, 27 Aug 2003 16:58:50 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from roark.gnf.org ([172.25.24.15]) by EXCHCLUSTER01.lj.gnf.org with Microsoft SMTPSVC(5.0.2195.5329); Wed, 27 Aug 2003 16:58:53 -0700 Received: from roark.gnf.org (localhost [127.0.0.1]) by roark.gnf.org (8.12.9/8.12.9) with ESMTP id h7RNwraH004728 for ; Wed, 27 Aug 2003 16:58:53 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: (from gtetlow@localhost) by roark.gnf.org (8.12.9/8.12.9/Submit) id h7RNwr37004727 for threads@FreeBSD.org; Wed, 27 Aug 2003 16:58:53 -0700 (PDT) (envelope-from gtetlow) Date: Wed, 27 Aug 2003 16:58:53 -0700 From: Gordon Tetlow To: threads@FreeBSD.org Message-ID: <20030827235853.GH695@roark.gnf.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="rCwQ2Y43eQY6RBgR" Content-Disposition: inline User-Agent: Mutt/1.4.1i X-Habeas-SWE-1: winter into spring X-Habeas-SWE-2: brightly anticipated X-Habeas-SWE-3: like Habeas SWE (tm) X-Habeas-SWE-4: Copyright 2002 Habeas (tm) X-Habeas-SWE-5: Sender Warranted Email (SWE) (tm). The sender of this X-Habeas-SWE-6: email in exchange for a license for this Habeas X-Habeas-SWE-7: warrant mark warrants that this is a Habeas Compliant X-Habeas-SWE-8: Message (HCM) and not spam. Please report use of this X-Habeas-SWE-9: mark in spam to . X-OriginalArrivalTime: 27 Aug 2003 23:58:53.0872 (UTC) FILETIME=[2B7B2B00:01C36CF7] Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Aug 2003 23:58:56 -0000 --rCwQ2Y43eQY6RBgR Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Aug 27, 2003 at 10:05:28AM -0600, Scott Long wrote: > All, >=20 > This is kind of an unconventional call for help. As we approach the > release of 5.2, we'd really like to show off the performance and > stability of our new threading packages. So, I'm looking for people > to volunteer to go out and put some of the thread-capable enterprise > and desktop packages to the test. Packages that I would most like > to see are: >=20 > Apache 2 > BIND 9 > OpenOffice > Mozilla > KDE desktop + Konq > Java 1.3/1.4 + Tomcat/etc > MySQL > Others? I've found that privoxy is a good program to test threads. It does alot of network io too. ports/www/privoxy. Just point your web browser to use it. > The questions that I would like to have answered are: >=20 > Does it take advantage of pthreads Yes. > Does it work with libkse and libthr Sort of. > Does it operate correctly libthr: tends to segfault (sig11). Occasional DNS resolution issues. libkse: tends to segfault (sig11), I've gotten my box to crash once too (BAD!). ^C still doesn't kill the process.=20 Occasional DNS resolution issues. Here's the output from the console: Aug 27 10:22:41 Privoxy(134991872) Request: images.slashdot.org/greendot.gif Aug 27 10:22:41 Privoxy(134979584) Request: images.slashdot.org/pix.gif Aug 27 10:22:46 Privoxy(135041024) Error: could not resolve hostname images= .slashdot.org ^^^^--- This is bad, I have no problems resolving this normally. ^CAug 27 10:22:56 Privoxy(135163904) Info: exiting by signal 2 .. bye ^CAug 27 10:23:35 Privoxy(135151616) Info: exiting by signal 2 .. bye ^CAug 27 10:23:35 Privoxy(135110656) Info: exiting by signal 2 .. bye ^CAug 27 10:23:36 Privoxy(135065600) Info: exiting by signal 2 .. bye ^CAug 27 10:23:36 Privoxy(135053312) Info: exiting by signal 2 .. bye ^CAug 27 10:23:36 Privoxy(134742016) Info: exiting by signal 2 .. bye ^^^--- I guess I killed the individual threads. ^C^C^C ^C^C^C ^C^C^C ^^^--- Nothing left to kill? load: 0.35 cmd: privoxy 1257 [KSE] 0.06u 0.09s 0% 3424k ^^^--- I hit ^T here. > Does it perform better/worse than with libc_r When it worked, it worked better under thr and kse. > Does it perform better/worse than under FreeBSD 4.x ENOTTESTED In short, we aren't there yet. -gordon --rCwQ2Y43eQY6RBgR Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (FreeBSD) iD8DBQE/TUW9Ru2t9DV9ZfsRAiFWAJ9lsYJIgw6/+42RG0+3rqXWusLX3gCdH91h ROSeJ04lEwuXCJO/PhGu4bw= =N+kU -----END PGP SIGNATURE----- --rCwQ2Y43eQY6RBgR-- From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 17:05:25 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E1DC916A4BF for ; Wed, 27 Aug 2003 17:05:25 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2B63E43FE9 for ; Wed, 27 Aug 2003 17:05:25 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7S05Otp013227; Wed, 27 Aug 2003 20:05:24 -0400 (EDT) Date: Wed, 27 Aug 2003 20:05:24 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Gordon Tetlow In-Reply-To: <20030827235853.GH695@roark.gnf.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 00:05:26 -0000 On Wed, 27 Aug 2003, Gordon Tetlow wrote: > On Wed, Aug 27, 2003 at 10:05:28AM -0600, Scott Long wrote: > > All, > > > > This is kind of an unconventional call for help. As we approach the > > release of 5.2, we'd really like to show off the performance and > > stability of our new threading packages. So, I'm looking for people > > to volunteer to go out and put some of the thread-capable enterprise > > and desktop packages to the test. Packages that I would most like > > to see are: > > > > Apache 2 > > BIND 9 > > OpenOffice > > Mozilla > > KDE desktop + Konq > > Java 1.3/1.4 + Tomcat/etc > > MySQL > > Others? > > I've found that privoxy is a good program to test threads. It does > alot of network io too. ports/www/privoxy. Just point your web browser > to use it. > > > The questions that I would like to have answered are: > > > > Does it take advantage of pthreads > Yes. > > Does it work with libkse and libthr > Sort of. > > Does it operate correctly > libthr: tends to segfault (sig11). Occasional DNS resolution issues. > libkse: tends to segfault (sig11), I've gotten my box to crash > once too (BAD!). ^C still doesn't kill the process. > Occasional DNS resolution issues. Here's the output from the > console: > > Aug 27 10:22:41 Privoxy(134991872) Request: images.slashdot.org/greendot.gif > Aug 27 10:22:41 Privoxy(134979584) Request: images.slashdot.org/pix.gif > Aug 27 10:22:46 Privoxy(135041024) Error: could not resolve hostname images.slashdot.org > ^^^^--- This is bad, I have no problems resolving this normally. > ^CAug 27 10:22:56 Privoxy(135163904) Info: exiting by signal 2 .. bye How old is your kernel? SCHED_ULE or SCHED_4BSD? It could be that name resolution is giving it problems because it's not thread-safe (in libc)... -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 17:11:25 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0BFB616A4BF; Wed, 27 Aug 2003 17:11:25 -0700 (PDT) Received: from ns2.gnf.org (ns2.gnf.org [63.196.132.68]) by mx1.FreeBSD.org (Postfix) with ESMTP id C399643F93; Wed, 27 Aug 2003 17:11:21 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from EXCHCLUSTER01.lj.gnf.org (exch01.lj.gnf.org [172.25.10.19]) by ns2.gnf.org (8.12.8p1/8.12.8) with ESMTP id h7S0BHoq056717; Wed, 27 Aug 2003 17:11:17 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from roark.gnf.org ([172.25.24.15]) by EXCHCLUSTER01.lj.gnf.org with Microsoft SMTPSVC(5.0.2195.5329); Wed, 27 Aug 2003 17:11:21 -0700 Received: from roark.gnf.org (localhost [127.0.0.1]) by roark.gnf.org (8.12.9/8.12.9) with ESMTP id h7S0BKaH004970; Wed, 27 Aug 2003 17:11:20 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: (from gtetlow@localhost) by roark.gnf.org (8.12.9/8.12.9/Submit) id h7S0BK2C004969; Wed, 27 Aug 2003 17:11:20 -0700 (PDT) (envelope-from gtetlow) Date: Wed, 27 Aug 2003 17:11:20 -0700 From: Gordon Tetlow To: deischen@freebsd.org Message-ID: <20030828001120.GI695@roark.gnf.org> References: <20030827235853.GH695@roark.gnf.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="16qp2B0xu0fRvRD7" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.1i X-Habeas-SWE-1: winter into spring X-Habeas-SWE-2: brightly anticipated X-Habeas-SWE-3: like Habeas SWE (tm) X-Habeas-SWE-4: Copyright 2002 Habeas (tm) X-Habeas-SWE-5: Sender Warranted Email (SWE) (tm). The sender of this X-Habeas-SWE-6: email in exchange for a license for this Habeas X-Habeas-SWE-7: warrant mark warrants that this is a Habeas Compliant X-Habeas-SWE-8: Message (HCM) and not spam. Please report use of this X-Habeas-SWE-9: mark in spam to . X-OriginalArrivalTime: 28 Aug 2003 00:11:21.0146 (UTC) FILETIME=[E8E421A0:01C36CF8] cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 00:11:25 -0000 --16qp2B0xu0fRvRD7 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Aug 27, 2003 at 08:05:24PM -0400, Daniel Eischen wrote: > On Wed, 27 Aug 2003, Gordon Tetlow wrote: >=20 > > On Wed, Aug 27, 2003 at 10:05:28AM -0600, Scott Long wrote: > >=20 > > > The questions that I would like to have answered are: > > >=20 > > > Does it take advantage of pthreads > > Yes. > > > Does it work with libkse and libthr > > Sort of. > > > Does it operate correctly > > libthr: tends to segfault (sig11). Occasional DNS resolution issues. > > libkse: tends to segfault (sig11), I've gotten my box to crash > > once too (BAD!). ^C still doesn't kill the process.=20 > > Occasional DNS resolution issues. Here's the output from the > > console: > >=20 > > Aug 27 10:22:41 Privoxy(134991872) Request: images.slashdot.org/greendo= t.gif > > Aug 27 10:22:41 Privoxy(134979584) Request: images.slashdot.org/pix.gif > > Aug 27 10:22:46 Privoxy(135041024) Error: could not resolve hostname im= ages.slashdot.org > > ^^^^--- This is bad, I have no problems resolving this normally. > > ^CAug 27 10:22:56 Privoxy(135163904) Info: exiting by signal 2 .. bye >=20 > How old is your kernel? SCHED_ULE or SCHED_4BSD? FreeBSD roark.gnf.org 5.1-CURRENT FreeBSD 5.1-CURRENT #0: Thu Aug 21 09:39:= 38 PDT 2003 gtetlow@roark.gnf.org:/local/usr.obj/local/usr.src/sys/ROAR= K i386 SCHED_4BSD > It could be that name resolution is giving it problems because > it's not thread-safe (in libc)... Probably, but it is still a problem that needs to be addressed. Is it possible to wrap the resolver in a umtx or some such other structure? -gordon --16qp2B0xu0fRvRD7 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (FreeBSD) iD8DBQE/TUioRu2t9DV9ZfsRAg0oAJ9aeldtNy+ItHPAR4xdkBywKJ6GNgCgpTXD aiJx60jVOJFeXENDAzdU5ME= =HtZQ -----END PGP SIGNATURE----- --16qp2B0xu0fRvRD7-- From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 17:26:35 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 00FC416A4C1; Wed, 27 Aug 2003 17:26:35 -0700 (PDT) Received: from freefall.freebsd.org (freefall.freebsd.org [216.136.204.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id C8FCA43FB1; Wed, 27 Aug 2003 17:26:33 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) Received: from localhost (davidxu@localhost [127.0.0.1]) by freefall.freebsd.org (8.12.9/8.12.9) with ESMTP id h7S0QVUp039470; Wed, 27 Aug 2003 17:26:32 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) From: David Xu To: Daniel Eischen Date: Thu, 28 Aug 2003 08:29:32 +0800 User-Agent: KMail/1.5.2 References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308280829.32757.davidxu@FreeBSD.org> cc: threads@FreeBSD.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: davidxu@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 00:26:35 -0000 On Thursday 28 August 2003 07:57, Daniel Eischen wrote: > Yes, perhaps my kernel was a bit out of date. I also had > forgotten I had libc_r mapped to libkse with libmap.conf, > so the libc_r tests were actually using libkse! I re-ran > the tests on a different box, single PIII 800MHz, 512MB RAM. > They look better, although libthr still doesn't give consistent > results. > > Run 1 Run 2 Run 3 > ----------------------------------------------------------- > libc_r real 0m13.739s 0m13.739s 0m13.882s > user 0m3.330s 0m3.302s 0m3.394s > sys 0m9.858s 0m9.893s 0m9.820s > ----------------------------------------------------------- > libkse(M:N) real 0m11.977s 0m12.199s 0m12.097s > user 0m3.248s 0m3.081s 0m2.857s > sys 0m8.190s 0m8.517s 0m8.575s > ----------------------------------------------------------- > libkse(1:1) real 0m11.972s 0m12.044s 0m12.035s > user 0m3.198s 0m2.980s 0m3.183s > sys 0m8.244s 0m8.480s 0m8.282s > ----------------------------------------------------------- > libthr real 0m34.180s 0m16.193s 0m34.119s > user 0m5.075s 0m3.874s 0m5.255s > sys 0m28.286s 0m11.626s 0m28.038s > ----------------------------------------------------------- > > libkse(1:1) and libkse(M:N) are about equal, and slightly > better than libc_r. I can't explain libthr results. Use a kernel without witness compiled in, libthr should be faster than this result. But I always can not finish this test for libthr on my SMP machine, in most time, it will deadlock, so I can not give you a reliable result. David Xu From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 18:10:42 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 47DDA16A4BF; Wed, 27 Aug 2003 18:10:42 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 71E8243FBF; Wed, 27 Aug 2003 18:10:39 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7S1Actp024147; Wed, 27 Aug 2003 21:10:38 -0400 (EDT) Date: Wed, 27 Aug 2003 21:10:38 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: David Xu In-Reply-To: <200308280829.32757.davidxu@FreeBSD.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@FreeBSD.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 01:10:42 -0000 On Thu, 28 Aug 2003, David Xu wrote: > On Thursday 28 August 2003 07:57, Daniel Eischen wrote: > > Yes, perhaps my kernel was a bit out of date. I also had > > forgotten I had libc_r mapped to libkse with libmap.conf, > > so the libc_r tests were actually using libkse! I re-ran > > the tests on a different box, single PIII 800MHz, 512MB RAM. > > They look better, although libthr still doesn't give consistent > > results. > > > > Run 1 Run 2 Run 3 > > ----------------------------------------------------------- > > libc_r real 0m13.739s 0m13.739s 0m13.882s > > user 0m3.330s 0m3.302s 0m3.394s > > sys 0m9.858s 0m9.893s 0m9.820s > > ----------------------------------------------------------- > > libkse(M:N) real 0m11.977s 0m12.199s 0m12.097s > > user 0m3.248s 0m3.081s 0m2.857s > > sys 0m8.190s 0m8.517s 0m8.575s > > ----------------------------------------------------------- > > libkse(1:1) real 0m11.972s 0m12.044s 0m12.035s > > user 0m3.198s 0m2.980s 0m3.183s > > sys 0m8.244s 0m8.480s 0m8.282s > > ----------------------------------------------------------- > > libthr real 0m34.180s 0m16.193s 0m34.119s > > user 0m5.075s 0m3.874s 0m5.255s > > sys 0m28.286s 0m11.626s 0m28.038s > > ----------------------------------------------------------- > > > > libkse(1:1) and libkse(M:N) are about equal, and slightly > > better than libc_r. I can't explain libthr results. > > Use a kernel without witness compiled in, libthr should be faster > than this result. Well, what got me was the variance in time. Most of the time it takes twice as long. And it doesn't seem to hang for a few seconds at any one point; it plods along consistently, just twice as slow overall. It could be mutex contention and false wakeups or something... > But I always can not finish this test for libthr on my SMP machine, > in most time, it will deadlock, so I can not give you a reliable result. It hangs in my thread yield test also; don't know why. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Wed Aug 27 18:14:30 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 98E4616A4BF for ; Wed, 27 Aug 2003 18:14:30 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id C0AD343F75 for ; Wed, 27 Aug 2003 18:14:29 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7S1EStp024804; Wed, 27 Aug 2003 21:14:29 -0400 (EDT) Date: Wed, 27 Aug 2003 21:14:28 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Gordon Tetlow In-Reply-To: <20030828001120.GI695@roark.gnf.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 01:14:30 -0000 On Wed, 27 Aug 2003, Gordon Tetlow wrote: > On Wed, Aug 27, 2003 at 08:05:24PM -0400, Daniel Eischen wrote: > > On Wed, 27 Aug 2003, Gordon Tetlow wrote: > > > > > On Wed, Aug 27, 2003 at 10:05:28AM -0600, Scott Long wrote: > > > > > > > The questions that I would like to have answered are: > > > > > > > > Does it take advantage of pthreads > > > Yes. > > > > Does it work with libkse and libthr > > > Sort of. > > > > Does it operate correctly > > > libthr: tends to segfault (sig11). Occasional DNS resolution issues. > > > libkse: tends to segfault (sig11), I've gotten my box to crash > > > once too (BAD!). ^C still doesn't kill the process. > > > Occasional DNS resolution issues. Here's the output from the > > > console: > > > > > > Aug 27 10:22:41 Privoxy(134991872) Request: images.slashdot.org/greendot.gif > > > Aug 27 10:22:41 Privoxy(134979584) Request: images.slashdot.org/pix.gif > > > Aug 27 10:22:46 Privoxy(135041024) Error: could not resolve hostname images.slashdot.org > > > ^^^^--- This is bad, I have no problems resolving this normally. > > > ^CAug 27 10:22:56 Privoxy(135163904) Info: exiting by signal 2 .. bye > > > > How old is your kernel? SCHED_ULE or SCHED_4BSD? > > FreeBSD roark.gnf.org 5.1-CURRENT FreeBSD 5.1-CURRENT #0: Thu Aug 21 09:39:38 PDT 2003 gtetlow@roark.gnf.org:/local/usr.obj/local/usr.src/sys/ROARK i386 > > SCHED_4BSD OK, I've installed www/privoxy and have libmap.conf pointed at libkse. It's working as far as I can tell; using the default config file. I've pointed a Solaris Netscape7 at it, opened a few windows and browsed a bit. How do I get it to fail? It hasn't so far. I know I've got it working correctly, because I can see the log file it is generating. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 02:21:49 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 092E516A4BF; Thu, 28 Aug 2003 02:21:49 -0700 (PDT) Received: from heron.mail.pas.earthlink.net (heron.mail.pas.earthlink.net [207.217.120.189]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3880043FFD; Thu, 28 Aug 2003 02:21:48 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-2ivfk56.dialup.mindspring.com ([165.247.208.166] helo=mindspring.com) by heron.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19sIyN-0003KL-00; Thu, 28 Aug 2003 02:21:44 -0700 Message-ID: <3F4DC971.498DD8F0@mindspring.com> Date: Thu, 28 Aug 2003 02:20:49 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Gordon Tetlow References: <20030827235853.GH695@roark.gnf.org> <20030828001120.GI695@roark.gnf.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a43846a7df0d5c44a7cf89723c0a97a53f2601a10902912494350badd9bab72f9c350badd9bab72f9c cc: deischen@freebsd.org cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 09:21:49 -0000 Gordon Tetlow wrote: > > It could be that name resolution is giving it problems because > > it's not thread-safe (in libc)... > > Probably, but it is still a problem that needs to be addressed. Is it > possible to wrap the resolver in a umtx or some such other structure? The proper way to fix this, and the Netscape LDAP client libraries, until they are made reentrant, is to implement stub functions which lock a request queue, queue requests with monotonically increasing ID's, and then use a single library spawned worker thread in order to serially satisfy the queued requests (also queueing responses at the same time, using the ID corresponding to the request, so that it can be demux'ed back to the caller. Each caller would block on a per caller data structure condition variable, using an expandable pool of these structures in order to support an arbitrary number of calling threads. Microsoft calls this "apartment model threading". It's pretty trivial to do this; for an employer a while back, I did this for the Netscape LDAP client libraries to make them thread-safe. BTW: in case it wasn't obvious from the above, a mutex is exactly the wrong way to do this. Also: you may want to look into porting the Darwin lookupd; there was also a university which offered up their own version of a lookupd, the last time this question came up, which included caching to speed duplicate requests (an obvious win, for an http proxy server!), but no one took them up on their offer and committed the code. The main problem with a "thread safe resolver with the ability to handle an arbitrary number of client threads" is that you end up needing "an arbitrary number of UDP (and TCP, if the responses get large) sockets", which sucks. Better to serialize through one thread in an apartment model, and then expand the number of threads in the apartment to some administrative maximum concurrency, when the resolver code is rewritten. -- Terry From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 02:25:39 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A6BFE16A4BF; Thu, 28 Aug 2003 02:25:39 -0700 (PDT) Received: from heron.mail.pas.earthlink.net (heron.mail.pas.earthlink.net [207.217.120.189]) by mx1.FreeBSD.org (Postfix) with ESMTP id DB25943FD7; Thu, 28 Aug 2003 02:25:38 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-2ivfk56.dialup.mindspring.com ([165.247.208.166] helo=mindspring.com) by heron.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19sJ2A-0003oj-00; Thu, 28 Aug 2003 02:25:38 -0700 Message-ID: <3F4DCA5A.753F2E8C@mindspring.com> Date: Thu, 28 Aug 2003 02:24:42 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: deischen@FreeBSD.org References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a43846a7df0d5c44a74fbe1bfde1bb3165667c3043c0873f7e350badd9bab72f9c350badd9bab72f9c cc: threads@FreeBSD.org cc: David Xu Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 09:25:39 -0000 Daniel Eischen wrote: > > Use a kernel without witness compiled in, libthr should be faster > > than this result. > > Well, what got me was the variance in time. Most of the time > it takes twice as long. And it doesn't seem to hang for a few > seconds at any one point; it plods along consistently, just > twice as slow overall. It could be mutex contention and > false wakeups or something... > > > But I always can not finish this test for libthr on my SMP machine, > > in most time, it will deadlock, so I can not give you a reliable result. > > It hangs in my thread yield test also; don't know why. I never saw if Julian's IPI patch was committed or not. I would expect a lot of CPU wakeup latency without it, if you HLT in the idle loop... -- Terry From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 02:27:08 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4FA4F16A4BF for ; Thu, 28 Aug 2003 02:27:08 -0700 (PDT) Received: from royk.itea.ntnu.no (royk.itea.ntnu.no [129.241.190.230]) by mx1.FreeBSD.org (Postfix) with ESMTP id F2FCC43FDD for ; Thu, 28 Aug 2003 02:27:06 -0700 (PDT) (envelope-from morten@rodal.no) Received: from localhost (localhost [127.0.0.1]) by royk.itea.ntnu.no (Postfix) with ESMTP id 7150F7777F for ; Thu, 28 Aug 2003 11:27:05 +0200 (CEST) Received: from slurp.rodal.no (m200h.studby.ntnu.no [129.241.135.200]) by royk.itea.ntnu.no (Postfix) with ESMTP id 3B71477755 for ; Thu, 28 Aug 2003 11:27:05 +0200 (CEST) Received: (from morten@localhost) by slurp.rodal.no (8.12.9/8.12.9/Submit) id h7S9R4MR001738 for threads@freebsd.org; Thu, 28 Aug 2003 11:27:04 +0200 (CEST) (envelope-from morten) Date: Thu, 28 Aug 2003 11:27:04 +0200 From: Morten Rodal To: threads@freebsd.org Message-ID: <20030828092703.GB1222@slurp.rodal.no> References: <20030827202009.GA44225@piggie> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030827202009.GA44225@piggie> X-Content-Scanned: with sophos and spamassassin at mailgw.ntnu.no. Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 09:27:08 -0000 On Wed, Aug 27, 2003 at 10:20:09PM +0200, Maciej Freudenheim wrote: > Hello! > > I was tuning xmms some time ago. Now i'm running it linked with libthr > for about 2 weeks and i haven't observed any problems. > > > Does it operate correctly > Yes. > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) and on my laptop (P4 1.7GHz). Unfortunately I must report that the results are better with libc_r. When xmms starts to play any type of sound I can hear spikes and other distortions in the sound. This happens regardless of the load of the machine, and it does not help to increase the buffer size. However, after a while the spikes get less frequent. I will try libthr when I get home later today and see if it works better than libkse. -- Morten Rodal From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 06:17:48 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 870E116A4BF for ; Thu, 28 Aug 2003 06:17:48 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9E67143FE1 for ; Thu, 28 Aug 2003 06:17:47 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7SDHktp025785; Thu, 28 Aug 2003 09:17:46 -0400 (EDT) Date: Thu, 28 Aug 2003 09:17:46 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Morten Rodal In-Reply-To: <20030828092703.GB1222@slurp.rodal.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 13:17:48 -0000 On Thu, 28 Aug 2003, Morten Rodal wrote: > On Wed, Aug 27, 2003 at 10:20:09PM +0200, Maciej Freudenheim wrote: > > Hello! > > > > I was tuning xmms some time ago. Now i'm running it linked with libthr > > for about 2 weeks and i haven't observed any problems. > > > > > Does it operate correctly > > Yes. > > > > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > results are better with libc_r. I use ogle-gui to play DVDs and xmcd (with playback to sound device) to play CDs and they both work just fine with libkse. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 07:58:05 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E866116A4BF; Thu, 28 Aug 2003 07:58:05 -0700 (PDT) Received: from silver.he.iki.fi (silver.he.iki.fi [193.64.42.241]) by mx1.FreeBSD.org (Postfix) with ESMTP id 75FAB43FE1; Thu, 28 Aug 2003 07:58:04 -0700 (PDT) (envelope-from pete@he.iki.fi) Received: from he.iki.fi (h81.vuokselantie10.fi [193.64.42.129]) by silver.he.iki.fi (8.12.9/8.11.4) with ESMTP id h7SEvr5L017539; Thu, 28 Aug 2003 17:57:54 +0300 (EEST) (envelope-from pete@he.iki.fi) Message-ID: <3F4E186E.2010509@he.iki.fi> Date: Thu, 28 Aug 2003 17:57:50 +0300 From: Petri Helenius User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Daniel Eischen References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit cc: threads@freebsd.org cc: Scott Long Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 14:58:06 -0000 Daniel Eischen wrote: >And it was very helpful to get Petri's feedback; we can always >use more like that :) > > > Thanks, although I only write perfect code :-), what´s the status of gdb or other debugging tools for kse and/or thr ? Sometimes it seems that the core generated makes sense but at times it seems to be more or less useless. Pete From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 09:25:23 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 20F8A16A4BF for ; Thu, 28 Aug 2003 09:25:23 -0700 (PDT) Received: from perrin.nxad.com (internal.ext.nxad.com [69.1.70.251]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9464843FE1 for ; Thu, 28 Aug 2003 09:25:22 -0700 (PDT) (envelope-from sean@nxad.com) Received: by perrin.nxad.com (Postfix, from userid 1001) id EC7A321058; Thu, 28 Aug 2003 09:25:21 -0700 (PDT) Date: Thu, 28 Aug 2003 09:25:21 -0700 From: Sean Chittenden To: Morten Rodal Message-ID: <20030828162521.GC83759@perrin.nxad.com> References: <20030827202009.GA44225@piggie> <20030828092703.GB1222@slurp.rodal.no> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030828092703.GB1222@slurp.rodal.no> X-PGP-Key: finger seanc@FreeBSD.org X-PGP-Fingerprint: 3849 3760 1AFE 7B17 11A0 83A6 DD99 E31F BC84 B341 X-Web-Homepage: http://sean.chittenden.org/ User-Agent: Mutt/1.5.4i cc: threads@freebsd.org Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 16:25:23 -0000 > > Hello! > > > > I was tuning xmms some time ago. Now i'm running it linked with libthr > > for about 2 weeks and i haven't observed any problems. > > > > > Does it operate correctly > > Yes. > > > > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > results are better with libc_r. What scheduler are you using? Try 4BSD instead of ULE. -sc -- Sean Chittenden From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 16:49:44 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 46BCB16A4BF for ; Thu, 28 Aug 2003 16:49:44 -0700 (PDT) Received: from swan.mail.pas.earthlink.net (swan.mail.pas.earthlink.net [207.217.120.123]) by mx1.FreeBSD.org (Postfix) with ESMTP id C6D3043FDF for ; Thu, 28 Aug 2003 16:49:43 -0700 (PDT) (envelope-from kwsn@earthlink.net) Received: from sdn-ap-004caburbp0118.dialsprint.net ([63.184.32.118] helo=[172.16.0.1]) by swan.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 19sWWJ-0002BV-00; Thu, 28 Aug 2003 16:49:39 -0700 From: Jon Kuster To: Morten Rodal In-Reply-To: <20030828092703.GB1222@slurp.rodal.no> References: <20030827202009.GA44225@piggie> <20030828092703.GB1222@slurp.rodal.no> Content-Type: text/plain Message-Id: <1062114576.97641.5.camel@jonnyv.kwsn.lan> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.4 Date: Thu, 28 Aug 2003 16:49:36 -0700 Content-Transfer-Encoding: 7bit cc: threads@freebsd.org Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Aug 2003 23:49:44 -0000 On Thu, 2003-08-28 at 02:27, Morten Rodal wrote: > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > results are better with libc_r. > > When xmms starts to play any type of sound I can hear spikes and other > distortions in the sound. This happens regardless of the load of the > machine, and it does not help to increase the buffer size. However, > after a while the spikes get less frequent. > > I will try libthr when I get home later today and see if it works > better than libkse. I have the same experience. When I use libkse, I get distortions. They're hard to describe, but sort of like the digital equivalent of dragging a needle across a record. I can provide a short clip if anyone's interested. libthr fares a little better, but I get a bunch of clicks instead, which is at least less jarring. libc_r still works the best for xmms. This is on my XP2100+, SCHED_4BSD, and using libmap to change threading libs. From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 17:12:59 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6A7BB16A4BF; Thu, 28 Aug 2003 17:12:59 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4F60343FCB; Thu, 28 Aug 2003 17:12:56 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7T0Cttp023320; Thu, 28 Aug 2003 20:12:55 -0400 (EDT) Date: Thu, 28 Aug 2003 20:12:55 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: kan@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-559023410-851401618-1062115975=:20795" cc: threads@freebsd.org cc: obrien@freebsd.org Subject: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 00:12:59 -0000 This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. ---559023410-851401618-1062115975=:20795 Content-Type: TEXT/PLAIN; charset=US-ASCII Attached is a patch to remove -pthread from gcc. It's been deprecated for 2.5 years and it should finally be put to rest. Also, it will cause us problems now with multiple thread libraries from which to choose. The only problem is that the patch would take config.gcc off the vendor branch. I'm not sure how else to fix it. I've built world with these patches. I haven't built any ports, though I expect many to break. But it's got to happen sometime, and we've waited too long already. Here is a link to some email over 2 years ago: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=641401+0+archive/2001/freebsd-current/20010218.freebsd-current -- Dan Eischen ---559023410-851401618-1062115975=:20795 Content-Type: TEXT/PLAIN; charset=US-ASCII; name="gcc.diffs" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename="gcc.diffs" SW5kZXg6IGNvbmZpZy5nY2MNCj09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0NClJD UyBmaWxlOiAvb3B0L0ZyZWVCU0QvY3ZzL3NyYy9jb250cmliL2djYy9jb25m aWcuZ2NjLHYNCnJldHJpZXZpbmcgcmV2aXNpb24gMS4xLjEuNw0KZGlmZiAt dSAtcjEuMS4xLjcgY29uZmlnLmdjYw0KLS0tIGNvbmZpZy5nY2MJMTEgSnVs IDIwMDMgMDM6NDA6NDggLTAwMDAJMS4xLjEuNw0KKysrIGNvbmZpZy5nY2MJ MjggQXVnIDIwMDMgMTY6NTA6MjkgLTAwMDANCkBAIC00ODQsMTEgKzQ4NCwx MSBAQA0KIAl4bm8pIGZic2RfdG1fZmlsZT0iJHtmYnNkX3RtX2ZpbGV9IGZy ZWVic2QtbnRoci5oIjs7DQogCXggfCB4eWVzIHwgeHB0aHJlYWRzIHwgeHBv c2l4KQ0KIAkJdGhyZWFkX2ZpbGU9J3Bvc2l4Jw0KLQkJdG1ha2VfZmlsZT0i JHt0bWFrZV9maWxlfSB0LWZyZWVic2QtdGhyZWFkIg0KIAkJIyBCZWZvcmUg NS4wLCBGcmVlQlNEIGNhbid0IGJpbmQgc2hhcmVkIGxpYnJhcmllcyB0byAt bGMNCiAJCSMgd2hlbiAib3B0aW9uYWxseSIgdGhyZWFkZWQgdmlhIHdlYWsg cHRocmVhZF8qIGNoZWNrcy4NCiAJCWNhc2UgJG1hY2hpbmUgaW4NCiAJCSot Ki1mcmVlYnNkWzM0XSB8ICotKi1mcmVlYnNkWzM0XS4qKQ0KKwkJCXRtYWtl X2ZpbGU9IiR7dG1ha2VfZmlsZX0gdC1mcmVlYnNkLXRocmVhZCINCiAJCQl0 bWFrZV9maWxlPSIke3RtYWtlX2ZpbGV9IHQtc2xpYmdjYy1ub2xjLW92ZXJy aWRlIjs7DQogCQllc2FjDQogCQk7Ow0KSW5kZXg6IGNvbmZpZy9mcmVlYnNk LXNwZWMuaA0KPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQ0KUkNTIGZpbGU6IC9v cHQvRnJlZUJTRC9jdnMvc3JjL2NvbnRyaWIvZ2NjL2NvbmZpZy9mcmVlYnNk LXNwZWMuaCx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuOQ0KZGlmZiAtdSAt cjEuOSBmcmVlYnNkLXNwZWMuaA0KLS0tIGNvbmZpZy9mcmVlYnNkLXNwZWMu aAkyNCBBdWcgMjAwMyAxNjo1NjozMSAtMDAwMAkxLjkNCisrKyBjb25maWcv ZnJlZWJzZC1zcGVjLmgJMjggQXVnIDIwMDMgMTc6MTQ6MDggLTAwMDANCkBA IC0xMjcsNDkgKzEyNywxMyBAQA0KICAgJXtzaGFyZWQ6Y3J0ZW5kUy5vJXN9 IFwNCiAgIGNydG4ubyVzICINCiANCi0vKiBQcm92aWRlIGEgTElCX1NQRUMg YXBwcm9wcmlhdGUgZm9yIEZyZWVCU0QgYXMgY29uZmlndXJlZCBhbmQgYXMN Ci0gICByZXF1aXJlZCBieSB0aGUgdXNlci1sYW5kIHRocmVhZCBtb2RlbC4g IEJlZm9yZSBfX0ZyZWVCU0RfdmVyc2lvbg0KLSAgIDUwMDAxNiwgc2VsZWN0 IHRoZSBhcHByb3ByaWF0ZSBsaWJjLCBkZXBlbmRpbmcgb24gd2hldGhlciB3 ZSdyZQ0KLSAgIGRvaW5nIHByb2ZpbGluZyBvciBuZWVkIHRocmVhZHMgc3Vw cG9ydC4gIEF0IF9fRnJlZUJTRF92ZXJzaW9uDQotICAgNTAwMDE2IGFuZCBs YXRlciwgd2hlbiB0aHJlYWRzIHN1cHBvcnQgaXMgcmVxdWVzdGVkIGluY2x1 ZGUgYm90aA0KLSAgIC1sYyBhbmQgLWxjX3IgaW5zdGVhZCBvZiBvbmx5IC1s Y19yLiAgVG8gbWFrZSBtYXR0ZXJzIGludGVyZXN0aW5nLA0KLSAgIHdlIGNh bid0IGFjdHVhbGx5IHVzZSBfX0ZyZWVCU0RfdmVyc2lvbiBwcm92aWRlZCBi eSA8b3NyZWxkYXRlLmg+DQotICAgZGlyZWN0bHkgc2luY2UgaXQgYnJlYWtz IGNyb3NzLWNvbXBpbGluZy4gIEFzIGEgZmluYWwgdHdpc3QsIG1ha2UNCi0g ICBpdCBhIGhhcmQgZXJyb3IgaWYgLXB0aHJlYWQgaXMgcHJvdmlkZWQgb24g dGhlIGNvbW1hbmQgbGluZSBhbmQgZ2NjDQotICAgd2FzIGNvbmZpZ3VyZWQg d2l0aCAtLWRpc2FibGUtdGhyZWFkcyAodGhpcyB3aWxsIGhlbHAgYXZvaWQg YnVnDQotICAgcmVwb3J0cyBmcm9tIHVzZXJzIGNvbXBsYWluaW5nIGFib3V0 IHRocmVhZGluZyB3aGVuIHRoZXkNCi0gICBtaXNjb25maWd1cmVkIHRoZSBn Y2MgYm9vdHN0cmFwIGJ1dCBhcmUgbGF0ZXIgY29uc3VsdGluZyBGcmVlQlNE DQotICAgbWFudWFsIHBhZ2VzIHRoYXQgcmVmZXIgdG8gdGhlIG15dGhpY2Fs IC1wdGhyZWFkIG9wdGlvbikuICAqLw0KLQ0KIC8qIFByb3ZpZGUgYSBMSUJf U1BFQyBhcHByb3ByaWF0ZSBmb3IgRnJlZUJTRC4gIEp1c3Qgc2VsZWN0IHRo ZSBhcHByb3ByaWF0ZQ0KLSAgIGxpYmMsIGRlcGVuZGluZyBvbiB3aGV0aGVy IHdlJ3JlIGRvaW5nIHByb2ZpbGluZyBvciBuZWVkIHRocmVhZHMgc3VwcG9y dC4NCisgICBsaWJjLCBkZXBlbmRpbmcgb24gd2hldGhlciB3ZSdyZSBkb2lu ZyBwcm9maWxpbmcgb3Igbm90Lg0KICAgIChzaW11bGFyIHRvIHRoZSBkZWZh dWx0LCBleGNlcHQgbm8gLWxnLCBhbmQgbm8gLXApLiAgKi8NCiANCi0jaWZk ZWYgRkJTRF9OT19USFJFQURTDQogI2RlZmluZSBGQlNEX0xJQl9TUEVDICIJ CQkJCQkJXA0KLSAgJXtwdGhyZWFkOiAlZVRoZSAtcHRocmVhZCBvcHRpb24g aXMgb25seSBzdXBwb3J0ZWQgb24gRnJlZUJTRCB3aGVuIGdjYyBcDQotaXMg YnVpbHQgd2l0aCB0aGUgLS1lbmFibGUtdGhyZWFkcyBjb25maWd1cmUtdGlt ZSBvcHRpb24ufQkJXA0KKyAgJXtwdGhyZWFkOiAlZVRoZSAtcHRocmVhZCBv cHRpb24gaXMgZGVwcmVjYXRlZC59CQkJXA0KICAgJXshc2hhcmVkOgkJCQkJ CQkJXA0KICAgICAleyFwZzogLWxjfQkJCQkJCQkJXA0KICAgICAle3BnOiAg LWxjX3B9CQkJCQkJCVwNCiAgIH0iDQotI2Vsc2UNCi0jaW5jbHVkZSA8c3lz L3BhcmFtLmg+DQotI2lmIF9fRnJlZUJTRF92ZXJzaW9uID49IDUwMDAxNg0K LSNkZWZpbmUgRkJTRF9MSUJfU1BFQyAiCQkJCQkJCVwNCi0gICV7IXNoYXJl ZDoJCQkJCQkJCVwNCi0gICAgJXshcGc6ICV7cHRocmVhZDotbGNfcn0gLWxj fQkJCQkJXA0KLSAgICAle3BnOiAgJXtwdGhyZWFkOi1sY19yX3B9IC1sY19w fQkJCQkJXA0KLSAgfSINCi0jZWxzZQ0KLSNkZWZpbmUgRkJTRF9MSUJfU1BF QyAiCQkJCQkJCVwNCi0gICV7IXNoYXJlZDoJCQkJCQkJCVwNCi0gICAgJXsh cGc6CQkJCQkJCQlcDQotICAgICAgJXshcHRocmVhZDotbGN9CQkJCQkJCVwN Ci0gICAgICAle3B0aHJlYWQ6LWxjX3J9fQkJCQkJCQlcDQotICAgICV7cGc6 CQkJCQkJCQlcDQotICAgICAgJXshcHRocmVhZDotbGNfcH0JCQkJCQkJXA0K LSAgICAgICV7cHRocmVhZDotbGNfcl9wfX0JCQkJCQlcDQotICB9Ig0KLSNl bmRpZg0KLSNlbmRpZg0K ---559023410-851401618-1062115975=:20795-- From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 21:26:14 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 553C616A4BF for ; Thu, 28 Aug 2003 21:26:14 -0700 (PDT) Received: from web13406.mail.yahoo.com (web13406.mail.yahoo.com [216.136.175.64]) by mx1.FreeBSD.org (Postfix) with SMTP id CA43343F3F for ; Thu, 28 Aug 2003 21:26:13 -0700 (PDT) (envelope-from giffunip@yahoo.com) Message-ID: <20030829042613.80458.qmail@web13406.mail.yahoo.com> Received: from [200.91.194.13] by web13406.mail.yahoo.com via HTTP; Fri, 29 Aug 2003 05:26:13 BST Date: Fri, 29 Aug 2003 05:26:13 +0100 (BST) From: "=?iso-8859-1?q?Pedro=20F.=20Giffuni?=" To: Daniel Eischen MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 04:26:14 -0000 Hi guys; First of all congratulations to all the people involved in the new thread support! I'm not understanding what the new world order will be, but please try to agree something with the autoconf people: from my porting experience, autoconf is not interested on what PTHREAD_LIBS says; it just tries to link with pthread and if that fails it assumes the platform doesn't support threads at all. Hmm.. Also in case someone wants to test math stuff, the math/superlu_mt port has some tests that can be useful. (I don't have current and I can't really upgrade right now.) cheers, Pedro. ________________________________________________________________________ Want to chat instantly with your online friends? Get the FREE Yahoo! Messenger http://uk.messenger.yahoo.com/ From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 22:28:20 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9C2AA16A4BF for ; Thu, 28 Aug 2003 22:28:20 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id E319543FA3 for ; Thu, 28 Aug 2003 22:28:19 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7T5SJtp012884; Fri, 29 Aug 2003 01:28:19 -0400 (EDT) Date: Fri, 29 Aug 2003 01:28:19 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: "=?iso-8859-1?q?Pedro=20F.=20Giffuni?=" In-Reply-To: <20030829042613.80458.qmail@web13406.mail.yahoo.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 05:28:20 -0000 On Fri, 29 Aug 2003, [iso-8859-1] Pedro F. Giffuni wrote: > Hi guys; > > First of all congratulations to all the people involved in the new thread > support! > > I'm not understanding what the new world order will be, but please try to agree > something with the autoconf people: from my porting experience, autoconf is not > interested on what PTHREAD_LIBS says; it just tries to link with pthread and if > that fails it assumes the platform doesn't support threads at all. -pthread is currently a compiler switch, and we (FreeBSD) are one of the very few OSes that have such an option. It was only there to prevent linking to libc in FreeBSD-[34]; otherwise it would not exist. Do not confuse -pthread with -lpthread or -lthr/-lkse/-llinuxthreads/-lwhatever. If autoconf has been taught that FreeBSD has -pthread, then it will have to be untaught. Eliminating -pthread will expose those ports that use it and then they can be fixed. If left in and we try to link with libpthread,libkse,libthr, etc, we will end up with ports that link to one of those libraries _and_ libc_r. Just try copying libkse or libthr to libpthread _without_ changing PTHREAD_LIBS (so you still use libc_r) and rebuilding X or KDE. Those ports will end up detecting libpthread and linking to it, and they will still also link to libc_r. This will not work but you won't know it until you try running the port and wacky weirdness ensues. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Thu Aug 28 22:51:32 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C703C16A4BF for ; Thu, 28 Aug 2003 22:51:32 -0700 (PDT) Received: from web13403.mail.yahoo.com (web13403.mail.yahoo.com [216.136.175.61]) by mx1.FreeBSD.org (Postfix) with SMTP id A850443FE9 for ; Thu, 28 Aug 2003 22:51:31 -0700 (PDT) (envelope-from giffunip@yahoo.com) Message-ID: <20030829055131.73565.qmail@web13403.mail.yahoo.com> Received: from [200.91.194.13] by web13403.mail.yahoo.com via HTTP; Fri, 29 Aug 2003 06:51:31 BST Date: Fri, 29 Aug 2003 06:51:31 +0100 (BST) From: "=?iso-8859-1?q?Pedro=20F.=20Giffuni?=" To: deischen@FreeBSD.org In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 05:51:32 -0000 --- Daniel Eischen wrote: ... > > -pthread is currently a compiler switch, and we (FreeBSD) are > one of the very few OSes that have such an option. It was only > there to prevent linking to libc in FreeBSD-[34]; otherwise > it would not exist. Do not confuse -pthread with -lpthread > or -lthr/-lkse/-llinuxthreads/-lwhatever. If autoconf has > been taught that FreeBSD has -pthread, then it will have to > be untaught. > OK I got it now, everything should work fine then. Right now there is no policy if the ports tree should try to build everything threaded. Some ports (at least my ports, when possible) have a USE_THREADS option which I guess should be default at a future time. FWIW, the cad/tochnog port gave me less than 5% performance hit when build with libc_r wrt the non-threaded version (on an UP system - I was just curious). best regards, Pedro. ________________________________________________________________________ Want to chat instantly with your online friends? Get the FREE Yahoo! Messenger http://uk.messenger.yahoo.com/ From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 01:28:46 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8F26616A4BF; Fri, 29 Aug 2003 01:28:46 -0700 (PDT) Received: from exchhz01.viatech.com.cn (ip-167-164-97-218.anlai.com [218.97.164.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id 01EC843FDD; Fri, 29 Aug 2003 01:28:43 -0700 (PDT) (envelope-from davidxu@viatech.com.cn) Received: from viatech.com.cn (ip-240-1-168-192.rev.dyxnet.com [192.168.1.240]) by exchhz01.viatech.com.cn with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RKXCR9RQ; Fri, 29 Aug 2003 16:10:54 +0800 Message-ID: <3F4F0FAA.30403@viatech.com.cn> Date: Fri, 29 Aug 2003 16:32:42 +0800 From: David Xu User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5b) Gecko/20030723 Thunderbird/0.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: deischen@FreeBSD.org References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: "Pedro F. Giffuni" cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 08:28:46 -0000 Daniel Eischen wrote: >-pthread is currently a compiler switch, and we (FreeBSD) are >one of the very few OSes that have such an option. > > > When will you commit it ? David Xu From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 03:34:14 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D0A1416A4BF; Fri, 29 Aug 2003 03:34:14 -0700 (PDT) Received: from pool-151-200-10-97.res.east.verizon.net (pool-151-200-247-85.res.east.verizon.net [151.200.247.85]) by mx1.FreeBSD.org (Postfix) with ESMTP id 787D143FE1; Fri, 29 Aug 2003 03:34:13 -0700 (PDT) (envelope-from mtm@identd.net) Received: from kokeb.ambesa.net (f8ikd13ufhzrh0o2@localhost [127.0.0.1]) id h7TAYCl8018596; Fri, 29 Aug 2003 06:34:12 -0400 (EDT) (envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.9/8.12.9/Submit) id h7TAYBW3018595; Fri, 29 Aug 2003 06:34:11 -0400 (EDT) (envelope-from mtm) Date: Fri, 29 Aug 2003 06:34:11 -0400 From: Mike Makonnen To: Daniel Eischen Message-ID: <20030829103411.GA18513@kokeb.ambesa.net> References: <200308280723.35697.davidxu@FreeBSD.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.1i X-Operating-System: FreeBSD/5.1-CURRENT (i386) cc: threads@freebsd.org cc: Scott Long cc: David Xu Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 10:34:14 -0000 On Wed, Aug 27, 2003 at 07:57:18PM -0400, Daniel Eischen wrote: > > Yes, perhaps my kernel was a bit out of date. I also had > forgotten I had libc_r mapped to libkse with libmap.conf, > so the libc_r tests were actually using libkse! I re-ran > the tests on a different box, single PIII 800MHz, 512MB RAM. > They look better, although libthr still doesn't give consistent > results. > > Run 1 Run 2 Run 3 > ----------------------------------------------------------- > libc_r real 0m13.739s 0m13.739s 0m13.882s > user 0m3.330s 0m3.302s 0m3.394s > sys 0m9.858s 0m9.893s 0m9.820s > ----------------------------------------------------------- > libkse(M:N) real 0m11.977s 0m12.199s 0m12.097s > user 0m3.248s 0m3.081s 0m2.857s > sys 0m8.190s 0m8.517s 0m8.575s > ----------------------------------------------------------- > libkse(1:1) real 0m11.972s 0m12.044s 0m12.035s > user 0m3.198s 0m2.980s 0m3.183s > sys 0m8.244s 0m8.480s 0m8.282s > ----------------------------------------------------------- > libthr real 0m34.180s 0m16.193s 0m34.119s > user 0m5.075s 0m3.874s 0m5.255s > sys 0m28.286s 0m11.626s 0m28.038s > ----------------------------------------------------------- > > libkse(1:1) and libkse(M:N) are about equal, and slightly > better than libc_r. I can't explain libthr results. Libthr is built with debugging options by default. Still, though it shouldn't be taking such a big performance hit. I suspect the real cause has something to do with the general "unfinished" state of the library in the sense that I haven't bothered with optimizations so far. Also, the fact that syncronization on locks in the kernel use sched_lock may be pessimizing performance. Cheers. -- Mike Makonnen | GPG-KEY: http://www.identd.net/~mtm/mtm.asc mtm@identd.net | Fingerprint: 00E8 61BC 0D75 7FFB E4D3 6BF1 B239 D010 3215 D418 mtm@FreeBSD.Org| FreeBSD - Unleash the Daemon ! From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 06:26:06 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C7F9C16A4BF; Fri, 29 Aug 2003 06:26:06 -0700 (PDT) Received: from silver.he.iki.fi (silver.he.iki.fi [193.64.42.241]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7265343FE9; Fri, 29 Aug 2003 06:26:05 -0700 (PDT) (envelope-from pete@he.iki.fi) Received: from he.iki.fi (h81.vuokselantie10.fi [193.64.42.129]) by silver.he.iki.fi (8.12.9/8.11.4) with ESMTP id h7TDQ35L027338; Fri, 29 Aug 2003 16:26:04 +0300 (EEST) (envelope-from pete@he.iki.fi) Message-ID: <3F4F5467.1050404@he.iki.fi> Date: Fri, 29 Aug 2003 16:25:59 +0300 From: Petri Helenius User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 X-Accept-Language: en-us, en MIME-Version: 1.0 To: nectar@FreeBSD.org, freebsd-threads@FreeBSD.org Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Subject: threads and openssl X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 13:26:06 -0000 Is there a plan to resolve bin/51205: openssl in base system is not compiled thread safe or has it been resolved but the pr has been left unattended? It sounds like it would make sense to have the library thread safe, specially in 5-CURRENT where the threads have a performance gain. Pete From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 06:35:23 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4762A16A4BF for ; Fri, 29 Aug 2003 06:35:23 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 351C743F3F for ; Fri, 29 Aug 2003 06:35:22 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7TDZEtp025593; Fri, 29 Aug 2003 09:35:21 -0400 (EDT) Date: Fri, 29 Aug 2003 09:35:14 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: David Xu In-Reply-To: <3F4F0FAA.30403@viatech.com.cn> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "Pedro F. Giffuni" cc: freebsd-threads@freebsd.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 13:35:23 -0000 On Fri, 29 Aug 2003, David Xu wrote: > Daniel Eischen wrote: > > >-pthread is currently a compiler switch, and we (FreeBSD) are > >one of the very few OSes that have such an option. > > > > > > > When will you commit it ? I haven't heard from the gcc maintainers (kan, obrien) yet, so as soon as I get feedback from them (or timeout after a few days). The patch would take one of the files off the vendor branch, so I'd like kan/obrien to see if there is another way to do it. I think we can turn libkse back into libpthread after -pthread is removed. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 10:33:32 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6926C16A4BF for ; Fri, 29 Aug 2003 10:33:32 -0700 (PDT) Received: from puffin.mail.pas.earthlink.net (puffin.mail.pas.earthlink.net [207.217.120.139]) by mx1.FreeBSD.org (Postfix) with ESMTP id B4E1843FE9 for ; Fri, 29 Aug 2003 10:33:31 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-2ivfjas.dialup.mindspring.com ([165.247.205.92] helo=mindspring.com) by puffin.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19sn4C-0007j8-00; Fri, 29 Aug 2003 10:29:45 -0700 Message-ID: <3F4F8D58.5F4523AE@mindspring.com> Date: Fri, 29 Aug 2003 10:28:56 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: "Pedro F. Giffuni" References: <20030829042613.80458.qmail@web13406.mail.yahoo.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a4cd51410bfa1d93b0144b42c7ecc30728a7ce0e8f8d31aa3f350badd9bab72f9c350badd9bab72f9c cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 17:33:32 -0000 "Pedro F. Giffuni" wrote: > I'm not understanding what the new world order will be, but please try to agree > something with the autoconf people: from my porting experience, autoconf is not > interested on what PTHREAD_LIBS says; it just tries to link with pthread and if > that fails it assumes the platform doesn't support threads at all. This is why autoconf sucks: it tries to identify the platform and make the code conform to that, rather than identifying the platform capabilities (note: different from "does -pthread work?") and using that. This is why imake and xmkmf can compile every X application that uses them, while autoconf and automake can only do it for the platforms they have been explicitly told about. The autoconf program balkanizes things, while imake unites them. I wish autoconf would die. -- Terry From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 10:34:16 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7A60C16A4BF; Fri, 29 Aug 2003 10:34:16 -0700 (PDT) Received: from puffin.mail.pas.earthlink.net (puffin.mail.pas.earthlink.net [207.217.120.139]) by mx1.FreeBSD.org (Postfix) with ESMTP id DF10C43FA3; Fri, 29 Aug 2003 10:34:15 -0700 (PDT) (envelope-from tlambert2@mindspring.com) Received: from user-2ivfjas.dialup.mindspring.com ([165.247.205.92] helo=mindspring.com) by puffin.mail.pas.earthlink.net with asmtp (SSLv3:RC4-MD5:128) (Exim 3.33 #1) id 19sn8X-0000lD-00; Fri, 29 Aug 2003 10:34:13 -0700 Message-ID: <3F4F8E63.FCBFB517@mindspring.com> Date: Fri, 29 Aug 2003 10:33:23 -0700 From: Terry Lambert X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: "Pedro F. Giffuni" References: <20030829055131.73565.qmail@web13403.mail.yahoo.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-ELNK-Trace: b1a02af9316fbb217a47c185c03b154d40683398e744b8a4cd51410bfa1d93b06b05f084229f94ee350badd9bab72f9c350badd9bab72f9c350badd9bab72f9c cc: deischen@FreeBSD.org cc: freebsd-threads@FreeBSD.org Subject: Re: Removing -pthread from gcc X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 17:34:16 -0000 "Pedro F. Giffuni" wrote: > --- Daniel Eischen wrote: > > -pthread is currently a compiler switch, and we (FreeBSD) are > > one of the very few OSes that have such an option. It was only > > there to prevent linking to libc in FreeBSD-[34]; otherwise > > it would not exist. Do not confuse -pthread with -lpthread > > or -lthr/-lkse/-llinuxthreads/-lwhatever. If autoconf has > > been taught that FreeBSD has -pthread, then it will have to > > be untaught. > > OK I got it now, everything should work fine then. No, it's going to break every place that the conf file "knows" it's FreeBSD, and spews out -pthread into the generated Makefile's from the Makefile.in, because it "knows" that FreeBSD has -pthread. This is going to have to be changed on an application-by-application basis, since the damn thing is recognizing the platform, rather than the platform capabilities. Pick 10 autoconf-using, threads-using ports at random, and go into the platform support subdirectories, and note that the FreeBSD entry has explicit settings for using -pthread. -- Terry From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 10:38:18 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8575916A4BF for ; Fri, 29 Aug 2003 10:38:18 -0700 (PDT) Received: from signal.itea.ntnu.no (signal.itea.ntnu.no [129.241.190.231]) by mx1.FreeBSD.org (Postfix) with ESMTP id 49C4C43FE9 for ; Fri, 29 Aug 2003 10:38:17 -0700 (PDT) (envelope-from morten@rodal.no) Received: from localhost (localhost [127.0.0.1]) by signal.itea.ntnu.no (Postfix) with ESMTP id B1EDD6393E for ; Fri, 29 Aug 2003 19:38:15 +0200 (CEST) Received: from slurp.rodal.no (m200h.studby.ntnu.no [129.241.135.200]) by signal.itea.ntnu.no (Postfix) with ESMTP id 6F31E63934 for ; Fri, 29 Aug 2003 19:38:15 +0200 (CEST) Received: (from morten@localhost) by slurp.rodal.no (8.12.9/8.12.9/Submit) id h7THcEAJ024528 for threads@freebsd.org; Fri, 29 Aug 2003 19:38:14 +0200 (CEST) (envelope-from morten) Date: Fri, 29 Aug 2003 19:38:14 +0200 From: Morten Rodal To: threads@freebsd.org Message-ID: <20030829173813.GA23148@slurp.rodal.no> References: <20030827202009.GA44225@piggie> <20030828092703.GB1222@slurp.rodal.no> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="3V7upXqbjpZ4EhLz" Content-Disposition: inline In-Reply-To: <20030828092703.GB1222@slurp.rodal.no> X-Content-Scanned: with sophos and spamassassin at mailgw.ntnu.no. Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 17:38:18 -0000 --3V7upXqbjpZ4EhLz Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Aug 28, 2003 at 11:27:04AM +0200, Morten Rodal wrote: > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > results are better with libc_r. >=20 > When xmms starts to play any type of sound I can hear spikes and other > distortions in the sound. This happens regardless of the load of the > machine, and it does not help to increase the buffer size. However, > after a while the spikes get less frequent. >=20 > I will try libthr when I get home later today and see if it works > better than libkse. >=20 I am happy(?) to report that the clicks, spikes and other digital noises are still present using libc_r. After careful listening I can still occationaly hear them, but not nearly as frequent as with libkse or libthr. Someone on current@ have reported problems with sound, so it might not be threading related after all. Now all I have to do is get Mozilla Firebird working with libkse and I'm all set :) --=20 Morten Rodal --3V7upXqbjpZ4EhLz Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (FreeBSD) iD8DBQE/T4+FbWe1Cy11WVsRAjYWAJ0d/BpBBp6ukRrGqBH+Bw3IBNy7BACfemfa vv57WmZVFUGE1k6lktmFx2I= =f/A1 -----END PGP SIGNATURE----- --3V7upXqbjpZ4EhLz-- From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 15:02:46 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E013816A4C0; Fri, 29 Aug 2003 15:02:46 -0700 (PDT) Received: from freefall.freebsd.org (freefall.freebsd.org [216.136.204.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id 22C5843F85; Fri, 29 Aug 2003 15:02:46 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) Received: from localhost (davidxu@localhost [127.0.0.1]) by freefall.freebsd.org (8.12.9/8.12.9) with ESMTP id h7TM2hUp070806; Fri, 29 Aug 2003 15:02:45 -0700 (PDT) (envelope-from davidxu@FreeBSD.org) From: David Xu To: Morten Rodal , threads@FreeBSD.org Date: Sat, 30 Aug 2003 06:05:45 +0800 User-Agent: KMail/1.5.2 References: <20030829173813.GA23148@slurp.rodal.no> In-Reply-To: <20030829173813.GA23148@slurp.rodal.no> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308300605.45307.davidxu@FreeBSD.org> Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: davidxu@FreeBSD.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 22:02:47 -0000 On Saturday 30 August 2003 01:38, Morten Rodal wrote: > On Thu, Aug 28, 2003 at 11:27:04AM +0200, Morten Rodal wrote: > > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > > results are better with libc_r. > > > > When xmms starts to play any type of sound I can hear spikes and other > > distortions in the sound. This happens regardless of the load of the > > machine, and it does not help to increase the buffer size. However, > > after a while the spikes get less frequent. > > > > I will try libthr when I get home later today and see if it works > > better than libkse. > > I am happy(?) to report that the clicks, spikes and other digital > noises are still present using libc_r. After careful listening I can > still occationaly hear them, but not nearly as frequent as with libkse > or libthr. > > Someone on current@ have reported problems with sound, so it might not > be threading related after all. Now all I have to do is get Mozilla > Firebird working with libkse and I'm all set :) Note that I ever had the same problem as yours, but with lastest sound and pcm driver change in kernel, I rarely have the problem. also click and other problems in XMMS may be resulted by its timing code, I ever worked on a DVD player project and the project leader told me timing code in DVD Player is very diffcult to get it right on a time sharing system, something like nanosleep() won't works well in such system especially when you have several threads, this might be true, and can be applied to XMMS too, because libkse and libthr are true threading libraries, this may also explore some XMMS timing code problem. David Xu From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 15:45:24 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A0A6016A4C0; Fri, 29 Aug 2003 15:45:24 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 500E643F93; Fri, 29 Aug 2003 15:45:23 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7TMjMtp003429; Fri, 29 Aug 2003 18:45:22 -0400 (EDT) Date: Fri, 29 Aug 2003 18:45:22 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: David Xu In-Reply-To: <200308300605.45307.davidxu@FreeBSD.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org Subject: Re: Call for thread testers: XMMS X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 22:45:24 -0000 On Sat, 30 Aug 2003, David Xu wrote: > On Saturday 30 August 2003 01:38, Morten Rodal wrote: > > On Thu, Aug 28, 2003 at 11:27:04AM +0200, Morten Rodal wrote: > > > I have tried to run xmms with libkse on both a SMP (Dual P2 300MHz) > > > and on my laptop (P4 1.7GHz). Unfortunately I must report that the > > > results are better with libc_r. > > > > > > When xmms starts to play any type of sound I can hear spikes and other > > > distortions in the sound. This happens regardless of the load of the > > > machine, and it does not help to increase the buffer size. However, > > > after a while the spikes get less frequent. > > > > > > I will try libthr when I get home later today and see if it works > > > better than libkse. > > > > I am happy(?) to report that the clicks, spikes and other digital > > noises are still present using libc_r. After careful listening I can > > still occationaly hear them, but not nearly as frequent as with libkse > > or libthr. > > > > Someone on current@ have reported problems with sound, so it might not > > be threading related after all. Now all I have to do is get Mozilla > > Firebird working with libkse and I'm all set :) > > Note that I ever had the same problem as yours, but with lastest sound > and pcm driver change in kernel, I rarely have the problem. > also click and other problems in XMMS may be resulted by its timing code, > I ever worked on a DVD player project and the project leader told > me timing code in DVD Player is very diffcult to get it right on a time > sharing system, something like nanosleep() won't works well in such system > especially when you have several threads, this might be true, and can be > applied to XMMS too, because libkse and libthr are true threading > libraries, this may also explore some XMMS timing code problem. I have built xmms and have tried playing an mp3 and can hear the same "scratches". I think it is a problem in the way that xmms uses threading (feeding the audio device). I don't experience this problem with ogle playing DVDs, which is also multi-threaded. Using top, it looks like xmms has one thread that is always running. One of the other threads is always lower on the top totem pole, and this is the thread that is in pcmwr. Since I see these threads in top, they must be created as scope system threads, so any priority they have is ignored by the kernel. It might be interesting to see how xmms fares if its threads were created as scope process. -- Dan Eischen From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 15:49:15 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A92F216A4BF; Fri, 29 Aug 2003 15:49:15 -0700 (PDT) Received: from ns1.gnf.org (ns1.gnf.org [63.196.132.67]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4A7AA43FEC; Fri, 29 Aug 2003 15:49:10 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from EXCHCLUSTER01.lj.gnf.org (exch02.lj.gnf.org [172.25.10.20]) by ns1.gnf.org (8.12.8p1/8.12.8) with ESMTP id h7TMn8tF061991; Fri, 29 Aug 2003 15:49:08 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: from roark.gnf.org ([172.25.24.15]) by EXCHCLUSTER01.lj.gnf.org with Microsoft SMTPSVC(5.0.2195.5329); Fri, 29 Aug 2003 15:49:09 -0700 Received: from roark.gnf.org (localhost [127.0.0.1]) by roark.gnf.org (8.12.9/8.12.9) with ESMTP id h7TMn9aH030274; Fri, 29 Aug 2003 15:49:09 -0700 (PDT) (envelope-from gtetlow@gnf.org) Received: (from gtetlow@localhost) by roark.gnf.org (8.12.9/8.12.9/Submit) id h7TMn9h5030273; Fri, 29 Aug 2003 15:49:09 -0700 (PDT) (envelope-from gtetlow) Date: Fri, 29 Aug 2003 15:49:09 -0700 From: Gordon Tetlow To: deischen@freebsd.org Message-ID: <20030829224909.GW695@roark.gnf.org> References: <20030828001120.GI695@roark.gnf.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="vy3x0i7GCC7mAH4A" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.1i X-Habeas-SWE-1: winter into spring X-Habeas-SWE-2: brightly anticipated X-Habeas-SWE-3: like Habeas SWE (tm) X-Habeas-SWE-4: Copyright 2002 Habeas (tm) X-Habeas-SWE-5: Sender Warranted Email (SWE) (tm). The sender of this X-Habeas-SWE-6: email in exchange for a license for this Habeas X-Habeas-SWE-7: warrant mark warrants that this is a Habeas Compliant X-Habeas-SWE-8: Message (HCM) and not spam. Please report use of this X-Habeas-SWE-9: mark in spam to . X-OriginalArrivalTime: 29 Aug 2003 22:49:10.0069 (UTC) FILETIME=[C290FE50:01C36E7F] cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 22:49:15 -0000 --vy3x0i7GCC7mAH4A Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Aug 27, 2003 at 09:14:28PM -0400, Daniel Eischen wrote: >=20 > OK, I've installed www/privoxy and have libmap.conf pointed at libkse. > It's working as far as I can tell; using the default config file. > I've pointed a Solaris Netscape7 at it, opened a few windows and > browsed a bit. How do I get it to fail? It hasn't so far. >=20 > I know I've got it working correctly, because I can see the > log file it is generating. Hmm, for me it was easy, I just needed to go to www.ucomics.com/calvinandhobbes (or some other complicated (ie graphic intensive) page) and it did it on it's own. Not a whole lot to it. -gordon --vy3x0i7GCC7mAH4A Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (FreeBSD) iD8DBQE/T9hlRu2t9DV9ZfsRAvqvAJ9kBG34SGHKid5Hjn2YcWiFkbhDhwCeIORz uFaPaEM5ZcVNnRhIxhJD88E= =TYZf -----END PGP SIGNATURE----- --vy3x0i7GCC7mAH4A-- From owner-freebsd-threads@FreeBSD.ORG Fri Aug 29 16:09:31 2003 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1592B16A4BF for ; Fri, 29 Aug 2003 16:09:31 -0700 (PDT) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0485643FCB for ; Fri, 29 Aug 2003 16:09:30 -0700 (PDT) (envelope-from eischen@vigrid.com) Received: from mail.pcnet.com (mail.pcnet.com [204.213.232.4]) by mail.pcnet.com (8.12.8/8.12.1) with ESMTP id h7TN9Ttp007521; Fri, 29 Aug 2003 19:09:29 -0400 (EDT) Date: Fri, 29 Aug 2003 19:09:29 -0400 (EDT) From: Daniel Eischen X-Sender: eischen@pcnet5.pcnet.com To: Gordon Tetlow In-Reply-To: <20030829224909.GW695@roark.gnf.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: threads@freebsd.org Subject: Re: Call for thread testers X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: deischen@freebsd.org List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Aug 2003 23:09:31 -0000 On Fri, 29 Aug 2003, Gordon Tetlow wrote: > On Wed, Aug 27, 2003 at 09:14:28PM -0400, Daniel Eischen wrote: > > > > OK, I've installed www/privoxy and have libmap.conf pointed at libkse. > > It's working as far as I can tell; using the default config file. > > I've pointed a Solaris Netscape7 at it, opened a few windows and > > browsed a bit. How do I get it to fail? It hasn't so far. > > > > I know I've got it working correctly, because I can see the > > log file it is generating. > > Hmm, for me it was easy, I just needed to go to > www.ucomics.com/calvinandhobbes (or some other complicated (ie graphic > intensive) page) and it did it on it's own. Not a whole lot to it. Nope, I can't reproduce any error. I went to ucomics, and launched a couple of other konqueror windows at the same time going to other graphic intensive sites, and can't reproduce any problem. -- Dan Eischen