From owner-freebsd-alpha Sun Sep 10 0:49:44 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 7DECF37B424 for ; Sun, 10 Sep 2000 00:49:40 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8A7n8O19425 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sun, 10 Sep 2000 09:49:20 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8A7ngp58504; Sun, 10 Sep 2000 09:49:43 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8A7neX18932; Sun, 10 Sep 2000 09:49:40 +0200 (CEST) (envelope-from ticso) Date: Sun, 10 Sep 2000 09:49:39 +0200 From: Bernd Walter To: Peter Wemm Cc: Matthew Jacob , Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Message-ID: <20000910094939.A18867@cicely5.cicely.de> References: <20000909141749.A13527@cicely5.cicely.de> <200009100616.e8A6GpG76494@netplex.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009100616.e8A6GpG76494@netplex.com.au>; from peter@netplex.com.au on Sat, Sep 09, 2000 at 11:16:51PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 09, 2000 at 11:16:51PM -0700, Peter Wemm wrote: > AHA! I suspected the bwx/swiz mixup was implicated in this, and this > suggests it does have something to do with it. > > Can you try something like this?: cc -c -O -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -fformat-extensions -ansi -g -nostdinc -I- -I. -I../.. -I../../../include -D_KERNEL -include opt_global.h -elf -mno-fp-regs -ffixed-8 -Wa,-mev56 ../../alpha/pci/cia.c ../../alpha/pci/cia.c: In function `cia_attach': ../../alpha/pci/cia.c:493: invalid operands to binary == *** Error code 1 > > Index: cia.c > =================================================================== > RCS file: /home/ncvs/src/sys/alpha/pci/cia.c,v > retrieving revision 1.27 > diff -u -r1.27 cia.c > --- cia.c 2000/09/02 01:05:37 1.27 > +++ cia.c 2000/09/10 06:13:49 > @@ -485,14 +485,17 @@ > if (!platform.iointr) /* XXX */ > set_iointr(alpha_dispatch_intr); > > - if (cia_ispyxis) { > + if (cia_ispyxis) > snprintf(chipset_type, sizeof(chipset_type), "pyxis"); > + else > + snprintf(chipset_type, sizeof(chipset_type), "cia"); > + That's this line: > + if (cia_ispyxis || chipset_bwx || chipset == cia_bwx_chipset) { > chipset_bwx = 1; > chipset_ports = CIA_EV56_BWIO; > chipset_memory = CIA_EV56_BWMEM; > chipset_dense = CIA_PCI_DENSE; > } else { > - snprintf(chipset_type, sizeof(chipset_type), "cia"); > chipset_bwx = 0; > chipset_ports = CIA_PCI_SIO1; > chipset_memory = CIA_PCI_SMEM1; > > I am not certain that this is correct, but we are definately screwing up > the non-pyxis BWX case. This is a sledgehammer approach to see if bwx was > set up earlier on and leave it on, rather than partly resetting it back to > swiz mode but leaving the bwx chipset function pointers. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 1: 8:27 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from netplex.com.au (adsl-63-207-30-186.dsl.snfc21.pacbell.net [63.207.30.186]) by hub.freebsd.org (Postfix) with ESMTP id D841537B422 for ; Sun, 10 Sep 2000 01:08:23 -0700 (PDT) Received: from netplex.com.au (peter@localhost [127.0.0.1]) by netplex.com.au (8.11.0/8.9.3) with ESMTP id e8A88FG76895; Sun, 10 Sep 2000 01:08:15 -0700 (PDT) (envelope-from peter@netplex.com.au) Message-Id: <200009100808.e8A88FG76895@netplex.com.au> X-Mailer: exmh version 2.1.1 10/15/1999 To: Bernd Walter Cc: Matthew Jacob , Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <20000910094939.A18867@cicely5.cicely.de> Date: Sun, 10 Sep 2000 01:08:15 -0700 From: Peter Wemm Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Sat, Sep 09, 2000 at 11:16:51PM -0700, Peter Wemm wrote: > > AHA! I suspected the bwx/swiz mixup was implicated in this, and this > > suggests it does have something to do with it. > > > > Can you try something like this?: > > cc -c -O -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmiss ing-prototypes -Wpointer-arith -Winline -Wcast-qual -fformat-extensions -a nsi -g -nostdinc -I- -I. -I../.. -I../../../include -D_KERNEL -include opt _global.h -elf -mno-fp-regs -ffixed-8 -Wa,-mev56 ../../alpha/pci/cia.c > ../../alpha/pci/cia.c: In function `cia_attach': > ../../alpha/pci/cia.c:493: invalid operands to binary == > *** Error code 1 oops. :-) Change it to: if (cia_ispyxis || chipset_bwx) { > That's this line: > > + if (cia_ispyxis || chipset_bwx || chipset == cia_bwx_chipset) { ie: if the init code already set chipset_bwx, don't turn it off. To be most correct, it should probably be: Index: cia.c =================================================================== RCS file: /home/ncvs/src/sys/alpha/pci/cia.c,v retrieving revision 1.27 diff -c -r1.27 cia.c *** cia.c 2000/09/02 01:05:37 1.27 --- cia.c 2000/09/10 08:07:02 *************** *** 485,499 **** if (!platform.iointr) /* XXX */ set_iointr(alpha_dispatch_intr); ! if (cia_ispyxis) { snprintf(chipset_type, sizeof(chipset_type), "pyxis"); ! chipset_bwx = 1; chipset_ports = CIA_EV56_BWIO; chipset_memory = CIA_EV56_BWMEM; chipset_dense = CIA_PCI_DENSE; } else { - snprintf(chipset_type, sizeof(chipset_type), "cia"); - chipset_bwx = 0; chipset_ports = CIA_PCI_SIO1; chipset_memory = CIA_PCI_SMEM1; chipset_dense = CIA_PCI_DENSE; --- 485,500 ---- if (!platform.iointr) /* XXX */ set_iointr(alpha_dispatch_intr); ! if (cia_ispyxis) snprintf(chipset_type, sizeof(chipset_type), "pyxis"); ! else ! snprintf(chipset_type, sizeof(chipset_type), "cia"); ! ! if (chipset_bwx) { chipset_ports = CIA_EV56_BWIO; chipset_memory = CIA_EV56_BWMEM; chipset_dense = CIA_PCI_DENSE; } else { chipset_ports = CIA_PCI_SIO1; chipset_memory = CIA_PCI_SMEM1; chipset_dense = CIA_PCI_DENSE; Matt Jacob didn't think this was the problem, but the discovery that forcing BWX mode suggests otherwise. Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 12:29:49 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from dragon.nuxi.com (trang.nuxi.com [209.152.133.57]) by hub.freebsd.org (Postfix) with ESMTP id 6F3AA37B423 for ; Sun, 10 Sep 2000 12:29:47 -0700 (PDT) Received: (from obrien@localhost) by dragon.nuxi.com (8.9.3/8.9.1) id MAA82973; Sun, 10 Sep 2000 12:29:42 -0700 (PDT) (envelope-from obrien) Date: Sun, 10 Sep 2000 12:29:42 -0700 From: "David O'Brien" To: Istvan Gyenes Cc: alpha@freebsd.org Subject: Re: Noname hangs... Message-ID: <20000910122942.A82955@dragon.nuxi.com> Reply-To: obrien@freebsd.org References: <002f01c01a6f$9df5fd50$1ee14298@nala> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <002f01c01a6f$9df5fd50$1ee14298@nala>; from frts@simba.sch.bme.hu on Sat, Sep 09, 2000 at 04:58:51PM +0200 X-Operating-System: FreeBSD 5.0-CURRENT Organization: The NUXI BSD group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 09, 2000 at 04:58:51PM +0200, Istvan Gyenes wrote: > Can somebody tell me what the problem is? The controller is the onboard > NCR810 and the disk is a RZ26L. I thought the NoName didn't have an on-board controler -- rather you had to use a PCI card SCSI controler. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 12:39:16 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from post.mail.nl.demon.net (post-10.mail.nl.demon.net [194.159.73.20]) by hub.freebsd.org (Postfix) with ESMTP id C1DEE37B422; Sun, 10 Sep 2000 12:39:11 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #2) id 13YCwX-00032W-00; Sun, 10 Sep 2000 19:39:10 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e8AJe5t63389; Sun, 10 Sep 2000 21:40:05 +0200 (CEST) (envelope-from wkb) Date: Sun, 10 Sep 2000 21:40:05 +0200 From: Wilko Bulte To: "David O'Brien" Cc: Istvan Gyenes , alpha@freebsd.org Subject: Re: Noname hangs... Message-ID: <20000910214005.A63353@freebie.demon.nl> References: <002f01c01a6f$9df5fd50$1ee14298@nala> <20000910122942.A82955@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20000910122942.A82955@dragon.nuxi.com>; from obrien@freebsd.org on Sun, Sep 10, 2000 at 12:29:42PM -0700 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Sep 10, 2000 at 12:29:42PM -0700, David O'Brien wrote: > On Sat, Sep 09, 2000 at 04:58:51PM +0200, Istvan Gyenes wrote: > > Can somebody tell me what the problem is? The controller is the onboard > > NCR810 and the disk is a RZ26L. > > I thought the NoName didn't have an on-board controler -- rather you had > to use a PCI card SCSI controler. Not correct. Per HARDWARE.TXT: ... - expansion: 3 32 bit PCI slots (1 shared with ISA) 5 ISA slots (1 shared with PCI) - embedded Fast SCSI using a NCR/Symbios 53C810 chip ... -- Wilko Bulte wilko@freebsd.org Arnhem, the Netherlands To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 14: 4:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 41B1337B423; Sun, 10 Sep 2000 14:04:15 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id OAA07802; Sun, 10 Sep 2000 14:04:06 -0700 Date: Sun, 10 Sep 2000 14:00:18 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Wilko Bulte Cc: "David O'Brien" , Istvan Gyenes , alpha@FreeBSD.ORG Subject: Re: Noname hangs... In-Reply-To: <20000910214005.A63353@freebie.demon.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org WRT Multia, the notes are wrong. The NCR810 is this dopey little PCI plugin that then does a right angle bend providing another PCI slot. > On Sun, Sep 10, 2000 at 12:29:42PM -0700, David O'Brien wrote: > > On Sat, Sep 09, 2000 at 04:58:51PM +0200, Istvan Gyenes wrote: > > > Can somebody tell me what the problem is? The controller is the onboard > > > NCR810 and the disk is a RZ26L. > > > > I thought the NoName didn't have an on-board controler -- rather you had > > to use a PCI card SCSI controler. > > Not correct. Per HARDWARE.TXT: > ... > - expansion: 3 32 bit PCI slots (1 shared with ISA) > 5 ISA slots (1 shared with PCI) > - embedded Fast SCSI using a NCR/Symbios 53C810 chip > ... > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 14:26:49 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 4181937B423 for ; Sun, 10 Sep 2000 14:26:46 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id OAA07851; Sun, 10 Sep 2000 14:26:28 -0700 Date: Sun, 10 Sep 2000 14:22:40 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Peter Wemm Cc: Bernd Walter , Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <200009100808.e8A88FG76895@netplex.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Well, no, I forced non-bwx mode.... just got back from being away...I'll resume where I was with this info... > Bernd Walter wrote: > > On Sat, Sep 09, 2000 at 11:16:51PM -0700, Peter Wemm wrote: > > > AHA! I suspected the bwx/swiz mixup was implicated in this, and this > > > suggests it does have something to do with it. > > > > > > Can you try something like this?: > > > > cc -c -O -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmiss > ing-prototypes -Wpointer-arith -Winline -Wcast-qual -fformat-extensions -a > nsi -g -nostdinc -I- -I. -I../.. -I../../../include -D_KERNEL -include opt > _global.h -elf -mno-fp-regs -ffixed-8 -Wa,-mev56 ../../alpha/pci/cia.c > > ../../alpha/pci/cia.c: In function `cia_attach': > > ../../alpha/pci/cia.c:493: invalid operands to binary == > > *** Error code 1 > > oops. :-) > > Change it to: if (cia_ispyxis || chipset_bwx) { > > > That's this line: > > > + if (cia_ispyxis || chipset_bwx || chipset == cia_bwx_chipset) { > > ie: if the init code already set chipset_bwx, don't turn it off. > To be most correct, it should probably be: > > Index: cia.c > =================================================================== > RCS file: /home/ncvs/src/sys/alpha/pci/cia.c,v > retrieving revision 1.27 > diff -c -r1.27 cia.c > *** cia.c 2000/09/02 01:05:37 1.27 > --- cia.c 2000/09/10 08:07:02 > *************** > *** 485,499 **** > if (!platform.iointr) /* XXX */ > set_iointr(alpha_dispatch_intr); > > ! if (cia_ispyxis) { > snprintf(chipset_type, sizeof(chipset_type), "pyxis"); > ! chipset_bwx = 1; > chipset_ports = CIA_EV56_BWIO; > chipset_memory = CIA_EV56_BWMEM; > chipset_dense = CIA_PCI_DENSE; > } else { > - snprintf(chipset_type, sizeof(chipset_type), "cia"); > - chipset_bwx = 0; > chipset_ports = CIA_PCI_SIO1; > chipset_memory = CIA_PCI_SMEM1; > chipset_dense = CIA_PCI_DENSE; > --- 485,500 ---- > if (!platform.iointr) /* XXX */ > set_iointr(alpha_dispatch_intr); > > ! if (cia_ispyxis) > snprintf(chipset_type, sizeof(chipset_type), "pyxis"); > ! else > ! snprintf(chipset_type, sizeof(chipset_type), "cia"); > ! > ! if (chipset_bwx) { > chipset_ports = CIA_EV56_BWIO; > chipset_memory = CIA_EV56_BWMEM; > chipset_dense = CIA_PCI_DENSE; > } else { > chipset_ports = CIA_PCI_SIO1; > chipset_memory = CIA_PCI_SMEM1; > chipset_dense = CIA_PCI_DENSE; > > Matt Jacob didn't think this was the problem, but the discovery that > forcing BWX mode suggests otherwise. > > Cheers, > -Peter > -- > Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au > "All of this is for nothing if we don't go to the stars" - JMS/B5 > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 14:38:34 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from post.mail.nl.demon.net (post-11.mail.nl.demon.net [194.159.73.21]) by hub.freebsd.org (Postfix) with ESMTP id 3004A37B422; Sun, 10 Sep 2000 14:38:32 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #4) id 13YEo2-000F95-00; Sun, 10 Sep 2000 21:38:30 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e8ALdQd63870; Sun, 10 Sep 2000 23:39:26 +0200 (CEST) (envelope-from wkb) Date: Sun, 10 Sep 2000 23:39:26 +0200 From: Wilko Bulte To: Matthew Jacob Cc: "David O'Brien" , Istvan Gyenes , alpha@FreeBSD.ORG Subject: Re: Noname hangs... Message-ID: <20000910233926.B63825@freebie.demon.nl> References: <20000910214005.A63353@freebie.demon.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: ; from mjacob@feral.com on Sun, Sep 10, 2000 at 02:00:18PM -0700 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Sep 10, 2000 at 02:00:18PM -0700, Matthew Jacob wrote: > > WRT Multia, the notes are wrong. The NCR810 is this dopey little PCI plugin This dopy little thing: http://www.tcja.nl/~wilko/ncr_hack.html? I fail to see what is wrong about the description though?? NoName != Multia. Please elaborate > that then does a right angle bend providing another PCI slot. > > > > On Sun, Sep 10, 2000 at 12:29:42PM -0700, David O'Brien wrote: > > > On Sat, Sep 09, 2000 at 04:58:51PM +0200, Istvan Gyenes wrote: > > > > Can somebody tell me what the problem is? The controller is the onboard > > > > NCR810 and the disk is a RZ26L. > > > > > > I thought the NoName didn't have an on-board controler -- rather you had > > > to use a PCI card SCSI controler. > > > > Not correct. Per HARDWARE.TXT: > > ... > > - expansion: 3 32 bit PCI slots (1 shared with ISA) > > 5 ISA slots (1 shared with PCI) > > - embedded Fast SCSI using a NCR/Symbios 53C810 chip > > ... > > > > ---end quoted text--- -- Wilko Bulte wilko@freebsd.org Arnhem, the Netherlands To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 14:42:56 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id ED7EA37B506; Sun, 10 Sep 2000 14:42:52 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id OAA07909; Sun, 10 Sep 2000 14:42:51 -0700 Date: Sun, 10 Sep 2000 14:39:03 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Wilko Bulte Cc: "David O'Brien" , Istvan Gyenes , alpha@FreeBSD.ORG Subject: Re: Noname hangs... In-Reply-To: <20000910233926.B63825@freebie.demon.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org O> On Sun, Sep 10, 2000 at 02:00:18PM -0700, Matthew Jacob wrote: > > > > WRT Multia, the notes are wrong. The NCR810 is this dopey little PCI plugin > > This dopy little thing: http://www.tcja.nl/~wilko/ncr_hack.html? > > I fail to see what is wrong about the description though?? NoName != Multia. > > Please elaborate Matt == Dopey. I keep on forgetting AxpPCI33 is different from Multia. Sorry. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 15:38: 4 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 10B8537B423 for ; Sun, 10 Sep 2000 15:38:00 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id PAA08060; Sun, 10 Sep 2000 15:37:31 -0700 Date: Sun, 10 Sep 2000 15:33:43 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Peter Wemm , Bernd Walter , Christian Weisgerber Cc: freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <200009100808.e8A88FG76895@netplex.com.au> Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="1136674726-598072241-968625223=:23495" Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org 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. --1136674726-598072241-968625223=:23495 Content-Type: TEXT/PLAIN; charset=US-ASCII The following diffs worked for me. I'm in a hurry, otherwise I'd analyze this more closely. The main ideas here are 1) determination of type of chipset should be done cia_probe 2) Pyxis *should* imply BWX. But they're not the same thing, really. 3) (for now) chipset_bwx is the gating item, and it's a true/false value, while the ivar (which really should just be removed) is a uintptr_t. 4) While I was at it, the correct test for implver is < ALPHA_IMPLVER_EV5 (I assume that ALPHA_IMPLVER_EV6 has BWX). The cia.c file didn't change in the smp merge, so this should be orthogonal to the change (I haven't yet run with top of tree). Can somebody test the attached diff against an SX or an LX to make sure it's right? If so, I'll commit it, or Peter can commit it if he's around, etc... Much as I despise BWX, this encourages me to add this into rawhide. -matt --1136674726-598072241-968625223=:23495 Content-Type: TEXT/PLAIN; charset=US-ASCII; name="cia.c.diff" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename="cia.c.diff" SW5kZXg6IGFscGhhL3BjaS9jaWEuYw0KPT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PQ0KUkNTIGZpbGU6IC9ob21lL25jdnMvc3JjL3N5cy9hbHBoYS9wY2kvY2lh LmMsdg0KcmV0cmlldmluZyByZXZpc2lvbiAxLjI3DQpkaWZmIC11IC1yMS4y NyBjaWEuYw0KLS0tIGFscGhhL3BjaS9jaWEuYwkyMDAwLzA5LzAyIDAxOjA1 OjM3CTEuMjcNCisrKyBhbHBoYS9wY2kvY2lhLmMJMjAwMC8wOS8xMCAyMjoy NTo0Mw0KQEAgLTM1NSwzMSArMzU1LDcgQEANCiAJaWYgKGluaXR0ZWQpIHJl dHVybjsNCiAJaW5pdHRlZCA9IDE7DQogDQotCWNpYV9yZXYgPSBSRUdWQUwo Q0lBX0NTUl9SRVYpICYgUkVWX01BU0s7DQotDQotCS8qDQotCSAqIERldGVy bWluZSBpZiB3ZSBoYXZlIGEgUHl4aXMuICBPbmx5IHR3byBzeXN0eXBlcyBj YW4NCi0JICogaGF2ZSB0aGlzOiB0aGUgRUIxNjQgc3lzdHlwZSAoQWxwaGFQ QzE2NExYIGFuZCBBbHBoYVBDMTY0U1gpDQotCSAqIGFuZCB0aGUgREVDX1NU NTUwIHN5c3R5cGUgKE1pYXRhKS4NCi0JICovDQotCWlmICgoaHdycGItPnJw Yl90eXBlID09IFNUX0VCMTY0ICYmDQotCSAgICAgKGh3cnBiLT5ycGJfdmFy aWF0aW9uICYgU1ZfU1RfTUFTSykgPj0gU1ZfU1RfQUxQSEFQQzE2NExYXzQw MCkgfHwNCi0JICAgIGh3cnBiLT5ycGJfdHlwZSA9PSBTVF9ERUNfNTUwKQ0K LQkJY2lhX2lzcHl4aXMgPSBUUlVFOw0KLQllbHNlDQotCQljaWFfaXNweXhp cyA9IEZBTFNFOw0KLQkNCi0JLyoNCi0JICogQUxDT1IvQUxDT1IyIFJldmlz aW9ucyA+PSAyIGFuZCBQeXhpcyBoYXZlIHRoZSBDTkZHIHJlZ2lzdGVyLg0K LQkgKi8NCi0JaWYgKGNpYV9yZXYgPj0gMiB8fCBjaWFfaXNweXhpcykNCi0J CWNpYV9jb25maWcgPSBSRUdWQUwoQ0lBX0NTUl9DTkZHKTsNCi0JZWxzZQ0K LQkJY2lhX2NvbmZpZyA9IDA7DQotDQotCWlmIChhbHBoYV9pbXBsdmVyKCkg IT0gQUxQSEFfSU1QTFZFUl9FVjUNCi0JICAgIHx8IGFscGhhX2FtYXNrKEFM UEhBX0FNQVNLX0JXWCkNCi0JICAgIHx8ICEoY2lhX2NvbmZpZyAmIENORkdf QldFTikpIHsNCisJaWYgKGNoaXBzZXRfYnd4ID09IDApIHsNCiAJCXN3aXpf aW5pdF9zcGFjZSgmaW9fc3BhY2Uuc3dpeiwgS1YoQ0lBX1BDSV9TSU8xKSk7 DQogCQlzd2l6X2luaXRfc3BhY2VfaGFlKCZtZW1fc3BhY2Uuc3dpeiwgS1Yo Q0lBX1BDSV9TTUVNMSksDQogCQkJCSAgICBjaWFfc3dpel9zZXRfaGFlX21l bSwgMCk7DQpAQCAtNDEzLDE3ICszODksNDggQEANCiAJcGNpX2luaXRfcmVz b3VyY2VzKCk7DQogCWlzYV9pbml0X2ludHIoKTsNCiAJY2lhX2luaXRfc2dt YXAoKTsNCisNCisJY2lhX3JldiA9IFJFR1ZBTChDSUFfQ1NSX1JFVikgJiBS RVZfTUFTSzsNCisNCisJLyoNCisJICogRGV0ZXJtaW5lIGlmIHdlIGhhdmUg YSBQeXhpcy4gIE9ubHkgdHdvIHN5c3R5cGVzIGNhbg0KKwkgKiBoYXZlIHRo aXM6IHRoZSBFQjE2NCBzeXN0eXBlIChBbHBoYVBDMTY0TFggYW5kIEFscGhh UEMxNjRTWCkNCisJICogYW5kIHRoZSBERUNfU1Q1NTAgc3lzdHlwZSAoTWlh dGEpLg0KKwkgKi8NCisJaWYgKChod3JwYi0+cnBiX3R5cGUgPT0gU1RfRUIx NjQgJiYNCisJICAgICAoaHdycGItPnJwYl92YXJpYXRpb24gJiBTVl9TVF9N QVNLKSA+PSBTVl9TVF9BTFBIQVBDMTY0TFhfNDAwKSB8fA0KKwkgICAgaHdy cGItPnJwYl90eXBlID09IFNUX0RFQ181NTApDQorCQljaWFfaXNweXhpcyA9 IFRSVUU7DQorCWVsc2UNCisJCWNpYV9pc3B5eGlzID0gRkFMU0U7DQorCQ0K Kw0KIA0KLQlpZiAoYWxwaGFfaW1wbHZlcigpICE9IEFMUEhBX0lNUExWRVJf RVY1DQotCSAgICB8fCBhbHBoYV9hbWFzayhBTFBIQV9BTUFTS19CV1gpDQot CSAgICB8fCAhKGNpYV9jb25maWcgJiBDTkZHX0JXRU4pKQ0KKwkvKg0KKwkg KiBBTENPUi9BTENPUjIgUmV2aXNpb25zID49IDIgYW5kIFB5eGlzIGhhdmUg dGhlIENORkcgcmVnaXN0ZXIuDQorCSAqLw0KKwlpZiAoY2lhX3JldiA+PSAy IHx8IGNpYV9pc3B5eGlzKQ0KKwkJY2lhX2NvbmZpZyA9IFJFR1ZBTChDSUFf Q1NSX0NORkcpOw0KKwllbHNlDQorCQljaWFfY29uZmlnID0gMDsNCisNCisJ aWYgKChhbHBoYV9pbXBsdmVyKCkgPCBBTFBIQV9JTVBMVkVSX0VWNSkgfHwN CisJICAgIChhbHBoYV9hbWFzayhBTFBIQV9BTUFTS19CV1gpICE9IDApIHx8 DQorCSAgICAoY2lhX2NvbmZpZyAmIENORkdfQldFTikgPT0gMCkgew0KIAkJ dXNlX2J3eCA9IDA7DQorCX0gZWxzZSB7DQorCQl1c2VfYnd4ID0gMTsNCisJ fQ0KIA0KKwlpZiAoY2lhX2lzcHl4aXMpIHsNCisJCWlmICh1c2VfYnd4ID09 IDApIHsNCisJCQlwcmludGYoIlBZWElTIGJ1dCBub3QgQldYP1xuIik7DQor CQl9DQorCX0NCisNCiAJZGV2aWNlX2FkZF9jaGlsZChkZXYsICJwY2liIiwg MCk7DQogCWRldmljZV9zZXRfaXZhcnMoZGV2LCAodm9pZCAqKXVzZV9id3gp Ow0KLQkNCi0JY2hpcHNldF9id3ggPSB1c2VfYnd4Ow0KLQ0KKwljaGlwc2V0 X2J3eCA9IHVzZV9id3ggPSAodXNlX2J3eCA9PSAodWludHB0cl90KSAxKTsN CiAJcmV0dXJuIDA7DQogfQ0KIA0KQEAgLTQ4NSwxNCArNDkyLDE0IEBADQog CWlmICghcGxhdGZvcm0uaW9pbnRyKQkvKiBYWFggKi8NCiAJCXNldF9pb2lu dHIoYWxwaGFfZGlzcGF0Y2hfaW50cik7DQogDQotCWlmIChjaWFfaXNweXhp cykgew0KLQkJc25wcmludGYoY2hpcHNldF90eXBlLCBzaXplb2YoY2hpcHNl dF90eXBlKSwgInB5eGlzIik7DQorCWlmIChjaGlwc2V0X2J3eCkgew0KKwkJ c25wcmludGYoY2hpcHNldF90eXBlLCBzaXplb2YoY2hpcHNldF90eXBlKSwg ImNpYS9id3giKTsNCiAJCWNoaXBzZXRfYnd4ID0gMTsNCiAJCWNoaXBzZXRf cG9ydHMgPSBDSUFfRVY1Nl9CV0lPOw0KIAkJY2hpcHNldF9tZW1vcnkgPSBD SUFfRVY1Nl9CV01FTTsNCiAJCWNoaXBzZXRfZGVuc2UgPSBDSUFfUENJX0RF TlNFOw0KIAl9IGVsc2Ugew0KLQkJc25wcmludGYoY2hpcHNldF90eXBlLCBz aXplb2YoY2hpcHNldF90eXBlKSwgImNpYSIpOw0KKwkJc25wcmludGYoY2hp cHNldF90eXBlLCBzaXplb2YoY2hpcHNldF90eXBlKSwgImNpYS9zd2l6Iik7 DQogCQljaGlwc2V0X2J3eCA9IDA7DQogCQljaGlwc2V0X3BvcnRzID0gQ0lB X1BDSV9TSU8xOw0KIAkJY2hpcHNldF9tZW1vcnkgPSBDSUFfUENJX1NNRU0x Ow0KSW5kZXg6IGNhbS9zY3NpL3Njc2lfc2EuYw0K --1136674726-598072241-968625223=:23495-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 17: 8:58 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 4AB0837B423 for ; Sun, 10 Sep 2000 17:08:56 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id RAA08350 for ; Sun, 10 Sep 2000 17:08:56 -0700 Date: Sun, 10 Sep 2000 17:05:08 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: alpha@FreeBSD.ORG Subject: so much for *that* idea..... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I did a build from more or less top of tree- I had had a clean buildworld, so I built a kernel and did the installworld and did mergemaster and did disklabel -B, and got: >>>b (boot dqa0.0.0.11.0 -flags a) block 0 of dqa0.0.0.11.0 is a valid boot block reading 15 blocks from dqa0.0.0.11.0 bootstrap code read in base = 180000, image_start = 0, image_bytes = 1e00 initializing HWRPB at 2000 initializing page table at 172000 initializing machine state setting affinity to the primary CPU jumping to bootstrap code Loading /boot/loader Console: SRM firmware console VMS PAL rev: 0x1000900010115 OSF PAL rev: 0x1000800020117 Switch to OSF PAL code succeeded. FreeBSD/alpha SRM disk boot, Revision 1.0 (mjacob@farrago.feral.com, Fri Sep 8 07:16:18 PDT 2000) Memory: 262144 k - halted CPU 0 halt code = 2 kernel stack not valid halt PC = 200000000 for my pains. Tsk. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 17:29:56 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 71A9637B423; Sun, 10 Sep 2000 17:29:54 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id RAA28533; Sun, 10 Sep 2000 17:29:38 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009110029.RAA28533@pike.osd.bsdi.com> Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <20000909172204.A17373@cicely5.cicely.de> from Bernd Walter at "Sep 9, 2000 05:22:04 pm" To: Bernd Walter Date: Sun, 10 Sep 2000 17:29:38 -0700 (PDT) Cc: Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Sat, Sep 09, 2000 at 02:17:49PM +0200, Bernd Walter wrote: > > On Fri, Sep 08, 2000 at 02:55:56PM -0700, Matthew Jacob wrote: > > > > > > This is what is broken for me too. I have not been able to really trace it > > > down yet. I don't have the halt switch set up. > > > > If I hardcode cia_pcib_read_config() and cia_pcib_write_config() to > > use bwx the machine boots at least into singleuser - not tested more yet. > > What is the special deal with the swiz stuff? > > As far As I understood it hangs on the first width 2 read access which is the > > second acces to the first available device. While the width 4 acces to probe > > for existence succeded. > > > > > > > pci0: physical bus=0 > > > > > cia_pcib_read_config: b=0, s=0, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=1, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=2, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=3, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=4, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=5, f=0, reg=0, width=4 > > > > > cia_pcib_read_config: b=0, s=5, f=0, reg=0, width=2 > > I was able to use it but got a panic after starting cvsup: > fatal kernel trap: > > trap entry = 0x2 (memory management fault) > a0 = 0x0 > a1 = 0x1 > a2 = 0x0 > pc = 0xfffffc00005168c8 > ra = 0xfffffc0000516f70 > curproc = 0xfffffe000a39c440 > pid = 2476, comm = cvsup > > Stopped at mtx_exit_hard+0x108: ldq t1,0(s1) <0x0> > db> trace > mtx_exit_hard() at mtx_exit_hard+0x108 > userret() at userret+0x1d0 > trap() at trap+0x828 > XentMM() at XentMM+0x20 > (null)() at 0x120270038 > > I don't know if it is because of my changes, the SMPng code or > something else. > I would asume it's because of SMPng Yes, I'm trying to work on this one, but I'm still trying to get kgdb working. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 17:53:29 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.inka.de (quechua.inka.de [212.227.14.2]) by hub.freebsd.org (Postfix) with ESMTP id 89AB137B424 for ; Sun, 10 Sep 2000 17:53:27 -0700 (PDT) Received: from ganerc.mips.inka.de (uucp@) by mail.inka.de with local-bsmtp id 13YHqg-0003qr-03; Mon, 11 Sep 2000 02:53:26 +0200 Received: (from daemon@localhost) by ganerc.mips.inka.de (8.11.0/8.11.0) id e8B0Ujj98934 for freebsd-alpha@freebsd.org; Mon, 11 Sep 2000 02:30:45 +0200 (CEST) (envelope-from daemon) From: naddy@mips.inka.de (Christian Weisgerber) Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Date: 11 Sep 2000 02:30:44 +0200 Message-ID: <8ph93k$30j8$1@ganerc.mips.inka.de> References: <200009100808.e8A88FG76895@netplex.com.au> To: freebsd-alpha@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > The following diffs worked for me. Works for my PC164, too, against PRE_SMPNG. -- Christian "naddy" Weisgerber naddy@mips.inka.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 23: 9:34 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 2BBA037B422; Sun, 10 Sep 2000 23:09:31 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8B69LF15516 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Mon, 11 Sep 2000 08:09:23 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8B69Kp64059; Mon, 11 Sep 2000 08:09:20 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8B69Gh53877; Mon, 11 Sep 2000 08:09:16 +0200 (CEST) (envelope-from ticso) Date: Mon, 11 Sep 2000 08:09:16 +0200 From: Bernd Walter To: John Baldwin Cc: Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Message-ID: <20000911080915.A53861@cicely5.cicely.de> References: <20000909172204.A17373@cicely5.cicely.de> <200009110029.RAA28533@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009110029.RAA28533@pike.osd.bsdi.com>; from jhb@pike.osd.bsdi.com on Sun, Sep 10, 2000 at 05:29:38PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Sep 10, 2000 at 05:29:38PM -0700, John Baldwin wrote: > Bernd Walter wrote: > > On Sat, Sep 09, 2000 at 02:17:49PM +0200, Bernd Walter wrote: > > I was able to use it but got a panic after starting cvsup: > > fatal kernel trap: > > > > trap entry = 0x2 (memory management fault) > > a0 = 0x0 > > a1 = 0x1 > > a2 = 0x0 > > pc = 0xfffffc00005168c8 > > ra = 0xfffffc0000516f70 > > curproc = 0xfffffe000a39c440 > > pid = 2476, comm = cvsup > > > > Stopped at mtx_exit_hard+0x108: ldq t1,0(s1) <0x0> > > db> trace > > mtx_exit_hard() at mtx_exit_hard+0x108 > > userret() at userret+0x1d0 > > trap() at trap+0x828 > > XentMM() at XentMM+0x20 > > (null)() at 0x120270038 > > > > I don't know if it is because of my changes, the SMPng code or > > something else. > > I would asume it's because of SMPng > > Yes, I'm trying to work on this one, but I'm still trying to get kgdb > working. I can't reproduce it anymore with a recent kernel and Peters latest PC164 patch. There were some promissing commits from Doug Rabson during these days. It happened with source from 8th sept and now I'm using source from 10th. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sun Sep 10 23:38:23 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 1BDF137B422; Sun, 10 Sep 2000 23:38:21 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8B6cCF18018 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Mon, 11 Sep 2000 08:38:15 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8B6c8p64198; Mon, 11 Sep 2000 08:38:12 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8B6c8v53959; Mon, 11 Sep 2000 08:38:08 +0200 (CEST) (envelope-from ticso) Date: Mon, 11 Sep 2000 08:38:07 +0200 From: Bernd Walter To: John Baldwin Cc: Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Message-ID: <20000911083807.A53937@cicely5.cicely.de> References: <20000909172204.A17373@cicely5.cicely.de> <200009110029.RAA28533@pike.osd.bsdi.com> <20000911080915.A53861@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000911080915.A53861@cicely5.cicely.de>; from ticso on Mon, Sep 11, 2000 at 08:09:15AM +0200 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Sep 11, 2000 at 08:09:15AM +0200, Bernd Walter wrote: > I can't reproduce it anymore with a recent kernel and Peters latest PC164 > patch. > There were some promissing commits from Doug Rabson during these days. > It happened with source from 8th sept and now I'm using source from 10th. Forget this statement - it's still there I got a panic right now. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 1:51:59 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from finch-post-11.mail.demon.net (finch-post-11.mail.demon.net [194.217.242.39]) by hub.freebsd.org (Postfix) with ESMTP id D961937B423; Mon, 11 Sep 2000 01:51:56 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-11.mail.demon.net with esmtp (Exim 2.12 #1) id 13YPJj-000NIA-0B; Mon, 11 Sep 2000 08:51:55 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id JAA02988; Mon, 11 Sep 2000 09:53:08 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Mon, 11 Sep 2000 09:52:42 +0100 (BST) From: Doug Rabson To: Bernd Walter Cc: John Baldwin , Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <20000911080915.A53861@cicely5.cicely.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 11 Sep 2000, Bernd Walter wrote: > On Sun, Sep 10, 2000 at 05:29:38PM -0700, John Baldwin wrote: > > Bernd Walter wrote: > > > On Sat, Sep 09, 2000 at 02:17:49PM +0200, Bernd Walter wrote: > > > I was able to use it but got a panic after starting cvsup: > > > fatal kernel trap: > > > > > > trap entry = 0x2 (memory management fault) > > > a0 = 0x0 > > > a1 = 0x1 > > > a2 = 0x0 > > > pc = 0xfffffc00005168c8 > > > ra = 0xfffffc0000516f70 > > > curproc = 0xfffffe000a39c440 > > > pid = 2476, comm = cvsup > > > > > > Stopped at mtx_exit_hard+0x108: ldq t1,0(s1) <0x0> > > > db> trace > > > mtx_exit_hard() at mtx_exit_hard+0x108 > > > userret() at userret+0x1d0 > > > trap() at trap+0x828 > > > XentMM() at XentMM+0x20 > > > (null)() at 0x120270038 > > > > > > I don't know if it is because of my changes, the SMPng code or > > > something else. > > > I would asume it's because of SMPng > > > > Yes, I'm trying to work on this one, but I'm still trying to get kgdb > > working. > > I can't reproduce it anymore with a recent kernel and Peters latest PC164 > patch. > There were some promissing commits from Doug Rabson during these days. > It happened with source from 8th sept and now I'm using source from 10th. That particular panic is the one I saw before removing the calls to mtx_enter/mtx_exit from interrupt(). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 8: 8:18 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id 6DB0237B443 for ; Mon, 11 Sep 2000 08:08:15 -0700 (PDT) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id LAA00012; Mon, 11 Sep 2000 11:08:13 -0400 (EDT) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.0/8.9.1) id e8BF8Dh44575; Mon, 11 Sep 2000 11:08:13 -0400 (EDT) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Mon, 11 Sep 2000 11:08:13 -0400 (EDT) To: mjacob@feral.com Cc: alpha@FreeBSD.ORG Subject: Re: so much for *that* idea..... In-Reply-To: References: X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs Lucid Message-ID: <14780.62432.185970.793919@grasshopper.cs.duke.edu> Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob writes: > > I did a build from more or less top of tree- I had had a clean buildworld, so > I built a kernel and did the installworld and did mergemaster and did > disklabel -B, and got: <...> > halt code = 2 > kernel stack not valid halt > PC = 200000000 > > for my pains. Tsk. This typically means you're faulting very early in the boot process. Thus is typically caused by trap'ing & then calling vm_map_lookup() out of vm_fault() prior to setting up the vm maps. The vm_map_lookup() faults & you get caught in an infinate recursion ending in a ksp not valid. First thing I'd try is temporarily making the ALPHA_MMCSR_ACCESS case in trap() always fatal for a kernel-mode trap. This should catch the initial trap & give you a clue as to what's going wrong. Sorry for my delayed reply. My wife & I just moved into our new place. No phone service yet, much less DSL. Blah. Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 8:40:15 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id ADDE637B423 for ; Mon, 11 Sep 2000 08:40:12 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id IAA10592; Mon, 11 Sep 2000 08:40:10 -0700 Date: Mon, 11 Sep 2000 08:36:20 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Andrew Gallatin Cc: alpha@FreeBSD.ORG Subject: Re: so much for *that* idea..... In-Reply-To: <14780.62432.185970.793919@grasshopper.cs.duke.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 11 Sep 2000, Andrew Gallatin wrote: > > Matthew Jacob writes: > > > > I did a build from more or less top of tree- I had had a clean buildworld, so > > I built a kernel and did the installworld and did mergemaster and did > > disklabel -B, and got: > > <...> > > > halt code = 2 > > kernel stack not valid halt > > PC = 200000000 > > > > for my pains. Tsk. > > This typically means you're faulting very early in the boot process. > Thus is typically caused by trap'ing & then calling vm_map_lookup() > out of vm_fault() prior to setting up the vm maps. The > vm_map_lookup() faults & you get caught in an infinate recursion > ending in a ksp not valid. > > First thing I'd try is temporarily making the ALPHA_MMCSR_ACCESS case > in trap() always fatal for a kernel-mode trap. This should catch the > initial trap & give you a clue as to what's going wrong. Yup- that's a good trick to try. > Sorry for my delayed reply. My wife & I just moved into our new > place. No phone service yet, much less DSL. Blah. It wasn't delayed- not at all. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 9:31:52 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 69B9D37B423; Mon, 11 Sep 2000 09:31:49 -0700 (PDT) Received: from cicely5.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8BGVls03689; Mon, 11 Sep 2000 18:31:47 +0200 (MET DST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8BGVkK54640; Mon, 11 Sep 2000 18:31:46 +0200 (CEST) (envelope-from ticso) Date: Mon, 11 Sep 2000 18:30:20 +0200 From: Bernd Walter To: Doug Rabson Cc: John Baldwin , Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Message-ID: <20000911183020.B53937@cicely5.cicely.de> References: <20000911080915.A53861@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from dfr@nlsystems.com on Mon, Sep 11, 2000 at 09:52:42AM +0100 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Sep 11, 2000 at 09:52:42AM +0100, Doug Rabson wrote: > That particular panic is the one I saw before removing the calls to > mtx_enter/mtx_exit from interrupt(). Yes - I also saw it before. But at the time of my writing never after. At least now I'm shure it's still there. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 10:29: 1 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from simba.sch.bme.hu (simba.sch.bme.hu [152.66.227.11]) by hub.freebsd.org (Postfix) with ESMTP id E90F237B422 for ; Mon, 11 Sep 2000 10:28:58 -0700 (PDT) Received: by simba.sch.bme.hu id TAA00839; Mon, 11 Sep 2000 19:28:53 +0200 (MET DST) Message-ID: <002801c01c15$69a277c0$1ee14298@nala> From: "Istvan Gyenes" To: References: <002f01c01a6f$9df5fd50$1ee14298@nala> <20000910122942.A82955@dragon.nuxi.com> <20000910214005.A63353@freebie.demon.nl> Subject: Re: Noname hangs... Date: Mon, 11 Sep 2000 19:26:24 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Thanks all who responded. The SCSI cable was bad .. :-( -- frts To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 10:56:23 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 8E15337B424 for ; Mon, 11 Sep 2000 10:56:20 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id KAA11109 for ; Mon, 11 Sep 2000 10:56:20 -0700 Date: Mon, 11 Sep 2000 10:52:30 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: alpha@FreeBSD.ORG Subject: Re: so much for *that* idea..... In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This isn't the kernel. This is the newer boot loader. Whatever was checked in after 9/7 busted me. Replacing /boot/loader with /boot/loader.old and symlinking /kernel -> /boot/kernel/kernel.ko and /modules -> /boot/kernel did the right thing for me so I could boot. There are other breakages too (mfs is broken- EFAULT, ignored my loader.conf to load ispfw), but at least this got me to multiuser. -matt > > I did a build from more or less top of tree- I had had a clean buildworld, so > I built a kernel and did the installworld and did mergemaster and did > disklabel -B, and got: > > >>>b > (boot dqa0.0.0.11.0 -flags a) > block 0 of dqa0.0.0.11.0 is a valid boot block > reading 15 blocks from dqa0.0.0.11.0 > bootstrap code read in > base = 180000, image_start = 0, image_bytes = 1e00 > initializing HWRPB at 2000 > initializing page table at 172000 > initializing machine state > setting affinity to the primary CPU > jumping to bootstrap code > Loading /boot/loader > Console: SRM firmware console > VMS PAL rev: 0x1000900010115 > OSF PAL rev: 0x1000800020117 > Switch to OSF PAL code succeeded. > > FreeBSD/alpha SRM disk boot, Revision 1.0 > (mjacob@farrago.feral.com, Fri Sep 8 07:16:18 PDT 2000) > Memory: 262144 k > - > halted CPU 0 > > halt code = 2 > kernel stack not valid halt > PC = 200000000 > > for my pains. Tsk. > > -matt > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 11: 0:22 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id 29D8E37B422 for ; Mon, 11 Sep 2000 11:00:19 -0700 (PDT) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id OAA03984; Mon, 11 Sep 2000 14:00:16 -0400 (EDT) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.0/8.9.1) id e8BI0GI44833; Mon, 11 Sep 2000 14:00:16 -0400 (EDT) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Mon, 11 Sep 2000 14:00:15 -0400 (EDT) To: mjacob@feral.com Cc: freebsd-alpha@freebsd.org Subject: Re: so much for *that* idea..... In-Reply-To: References: X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs Lucid Message-ID: <14781.7447.587262.648587@grasshopper.cs.duke.edu> Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob writes: > > This isn't the kernel. This is the newer boot loader. Whatever was checked in > after 9/7 busted me. Replacing /boot/loader with /boot/loader.old > and symlinking /kernel -> /boot/kernel/kernel.ko and /modules -> /boot/kernel > did the right thing for me so I could boot. > Speaking of which.. I, for one, wish the whole /boot/kernel/kernel.ko mess was backed out. If I wanted to play "kernel, kernel, where's the damned kernel" I'd run linux. The kernel belongs in the root filesystem, damn it. Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 11:10:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id D79B437B422 for ; Mon, 11 Sep 2000 11:10:13 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id LAA11190; Mon, 11 Sep 2000 11:10:12 -0700 Date: Mon, 11 Sep 2000 11:06:21 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Andrew Gallatin Cc: freebsd-alpha@freebsd.org Subject: Re: so much for *that* idea..... In-Reply-To: <14781.7447.587262.648587@grasshopper.cs.duke.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Matthew Jacob writes: > > > > This isn't the kernel. This is the newer boot loader. Whatever was checked in > > after 9/7 busted me. Replacing /boot/loader with /boot/loader.old > > and symlinking /kernel -> /boot/kernel/kernel.ko and /modules -> /boot/kernel > > did the right thing for me so I could boot. > > > > Speaking of which.. > > I, for one, wish the whole /boot/kernel/kernel.ko mess was backed out. > If I wanted to play "kernel, kernel, where's the damned kernel" I'd > run linux. > > The kernel belongs in the root filesystem, damn it. It's *in* the root filesystem. But hidden. I chided David about this already- mostly from the point of view of- "Gee- I don't recall discussion about this". He claimed "we all discussed this at last year's FreeBSDcon". -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 13: 7:55 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 75AA237B423 for ; Mon, 11 Sep 2000 13:07:53 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id NAA60493 for freebsd-alpha@FreeBSD.org; Mon, 11 Sep 2000 13:07:52 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009112007.NAA60493@pike.osd.bsdi.com> Subject: Re: cvs commit: src/sys/pci pcisupport.c To: freebsd-alpha@FreeBSD.org Date: Mon, 11 Sep 2000 13:07:52 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > That particular panic is the one I saw before removing the calls to > mtx_enter/mtx_exit from interrupt(). I've debugged it some more and it looks like a bug somewhere in the mutex code. We are releasing the Giant lock when it is uncontested, but mtx_exit() is still calling mtx_exit_hard(), which it should only be calling for contested locks. As a result, when we try to pull a process off the blocked list so we can give it the mutex and schedule it to run, we dereference a NULL pointer. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 13:18:45 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from netplex.com.au (adsl-63-207-30-186.dsl.snfc21.pacbell.net [63.207.30.186]) by hub.freebsd.org (Postfix) with ESMTP id 1AF9A37B423; Mon, 11 Sep 2000 13:18:42 -0700 (PDT) Received: from netplex.com.au (peter@localhost [127.0.0.1]) by netplex.com.au (8.11.0/8.9.3) with ESMTP id e8BKIIG84366; Mon, 11 Sep 2000 13:18:18 -0700 (PDT) (envelope-from peter@netplex.com.au) Message-Id: <200009112018.e8BKIIG84366@netplex.com.au> X-Mailer: exmh version 2.1.1 10/15/1999 To: Bernd Walter Cc: John Baldwin , Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) In-Reply-To: <20000911080915.A53861@cicely5.cicely.de> Date: Mon, 11 Sep 2000 13:18:18 -0700 From: Peter Wemm Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Sun, Sep 10, 2000 at 05:29:38PM -0700, John Baldwin wrote: > > Bernd Walter wrote: > > > On Sat, Sep 09, 2000 at 02:17:49PM +0200, Bernd Walter wrote: > > > I was able to use it but got a panic after starting cvsup: > > > fatal kernel trap: > > > > > > trap entry = 0x2 (memory management fault) > > > a0 = 0x0 > > > a1 = 0x1 > > > a2 = 0x0 > > > pc = 0xfffffc00005168c8 > > > ra = 0xfffffc0000516f70 > > > curproc = 0xfffffe000a39c440 > > > pid = 2476, comm = cvsup > > > > > > Stopped at mtx_exit_hard+0x108: ldq t1,0(s1) <0x0> > > > db> trace > > > mtx_exit_hard() at mtx_exit_hard+0x108 > > > userret() at userret+0x1d0 > > > trap() at trap+0x828 > > > XentMM() at XentMM+0x20 > > > (null)() at 0x120270038 > > > > > > I don't know if it is because of my changes, the SMPng code or > > > something else. > > > I would asume it's because of SMPng > > > > Yes, I'm trying to work on this one, but I'm still trying to get kgdb > > working. > > I can't reproduce it anymore with a recent kernel and Peters latest PC164 > patch. Err, what about my patch? Cannot reproduce what? Did you mean to say that you tried my patch and the PC164 booted up and didn't lockup or panic? Please be specific can you boot past the previous hang with my patch or not? (Yes, I know you posted later on with a panic, but that bug is different). Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 13:29:59 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.swissonline.ch (mail.swissonline.ch [62.2.32.83]) by hub.freebsd.org (Postfix) with ESMTP id 9DEE437B422 for ; Mon, 11 Sep 2000 13:29:46 -0700 (PDT) Received: from ofehr.com ([62.2.106.137]) by mail.swissonline.ch (8.9.3/8.9.3) with ESMTP id WAA29415 for ; Mon, 11 Sep 2000 22:29:38 +0200 (MET DST) Message-ID: <39BD4077.6823F3B1@ofehr.com> Date: Mon, 11 Sep 2000 20:28:39 +0000 From: Oliver Fehr X-Mailer: Mozilla 4.75 [en] (X11; U; FreeBSD 4.1-STABLE alpha) X-Accept-Language: en, de-CH MIME-Version: 1.0 To: freebsd-alpha@FreeBSD.org Subject: dev_net.c broken? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Could it be that dev_net.c is broken? Doing make gives me (dev_net.c dates 09/10/00) /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:200: conflicting types fo r `bootp' /usr/src/sys/boot/alpha/netboot/../../../../lib/libstand/net.h:117: previous dec laration of `bootp' /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c: In function `net_getpara ms': /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:222: too many arguments t o function `bootp' *** Error code 1 Stop in /usr/src/sys/boot/alpha/netboot. *** Error code 1 Stop in /usr/src/sys/boot/alpha. *** Error code 1 Stop in /usr/src/sys/boot. *** Error code 1 Stop in /usr/src/sys. *** Error code 1 Stop in /usr/src. *** Error code 1 Stop in /usr/src. *** Error code 1 Stop in /usr/src. -- Cheers Oliver Fehr E-mail:Oliver.Fehr@ofehr.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 13:37:36 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 72DF537B422; Mon, 11 Sep 2000 13:37:30 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id NAA61735; Mon, 11 Sep 2000 13:37:29 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009112037.NAA61735@pike.osd.bsdi.com> Subject: Bug in spinlocks? To: dfr@FreeBSD.org Date: Mon, 11 Sep 2000 13:37:29 -0700 (PDT) Cc: alpha@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hmm, is there any reason why we aren't disabling interrupts on the alpha when we obtain spinlocks? In the i386 code we use the following macro: /* Get a spin lock, handle recursion inline (as the less common case) */ #define _getlock_spin_block(mp, tid, type) do { \ u_int _mtx_fl = read_eflags(); \ disable_intr(); \ if (atomic_cmpset_int(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _mtx_fl); \ else \ (mp)->mtx_savefl = _mtx_fl; \ } while (0) on the alpha it is almost the same: /* * Get a spin lock, handle recusion inline (as the less common case) */ #define _getlock_spin_block(mp, tid, type) do { \ u_int _ipl = alpha_pal_rdps() & ALPHA_PSL_IPL_MASK; \ if (atomic_cmpset_64(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _ipl); \ else { \ alpha_mb(); \ (mp)->mtx_saveipl = _ipl; \ } \ } while (0) Note that in the alpha we are just saving and restoring the ipl, but we aren't actually changing it to disable interrupts. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 13:46:16 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 2DDC937B423; Mon, 11 Sep 2000 13:46:13 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8BKjwQ24032 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Mon, 11 Sep 2000 22:46:00 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8BKjxp67311; Mon, 11 Sep 2000 22:45:59 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8BKjs754979; Mon, 11 Sep 2000 22:45:54 +0200 (CEST) (envelope-from ticso) Date: Mon, 11 Sep 2000 22:45:54 +0200 From: Bernd Walter To: Peter Wemm Cc: John Baldwin , Matthew Jacob , dfr@FreeBSD.ORG, Christian Weisgerber , freebsd-alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/pci pcisupport.c (fwd) Message-ID: <20000911224554.A54830@cicely5.cicely.de> References: <20000911080915.A53861@cicely5.cicely.de> <200009112018.e8BKIIG84366@netplex.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009112018.e8BKIIG84366@netplex.com.au>; from peter@netplex.com.au on Mon, Sep 11, 2000 at 01:18:18PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Sep 11, 2000 at 01:18:18PM -0700, Peter Wemm wrote: > Err, what about my patch? Cannot reproduce what? Did you mean to say that > you tried my patch and the PC164 booted up and didn't lockup or panic? > Please be specific can you boot past the previous hang with my patch or > not? Your patch is working fine for what it was mentioned for. The system paniced later but this seems to be a completely unrelated problem. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 16:15:40 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from cgi.sstar.com (cgi.sstar.com [209.205.176.12]) by hub.freebsd.org (Postfix) with ESMTP id 67DE537B423 for ; Mon, 11 Sep 2000 16:15:34 -0700 (PDT) Received: from bluto.jimking.net (bluto.jimking.net [216.54.255.8]) by cgi.sstar.com (8.9.3/8.9.3) with ESMTP id SAA58043; Mon, 11 Sep 2000 18:15:26 -0500 (CDT) (envelope-from jim@jimking.net) Received: from marble (marble.lgc.com [134.132.228.4]) by bluto.jimking.net (8.9.3/8.9.3) with SMTP id SAA03560; Mon, 11 Sep 2000 18:15:24 -0500 (CDT) (envelope-from jim@jimking.net) Message-ID: <006301c01c46$2a9ca010$04e48486@marble> From: "Jim King" To: "Oliver Fehr" , References: <39BD4077.6823F3B1@ofehr.com> Subject: Re: dev_net.c broken? Date: Mon, 11 Sep 2000 18:15:23 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Oliver Fehr wrote: > Could it be that dev_net.c is broken? Doing make gives me (dev_net.c > dates 09/10/00) > > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:200: conflicting > types fo > r `bootp' > /usr/src/sys/boot/alpha/netboot/../../../../lib/libstand/net.h:117: > previous dec > laration of `bootp' > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c: In function > `net_getpara > ms': > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:222: too many > arguments t > o function `bootp' > *** Error code 1 > > Stop in /usr/src/sys/boot/alpha/netboot. > *** Error code 1 > > Stop in /usr/src/sys/boot/alpha. > *** Error code 1 > > Stop in /usr/src/sys/boot. > *** Error code 1 > > Stop in /usr/src/sys. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > Stop in /usr/src. I got the same thing today. :-( Jim To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 17:28:52 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 1058237B422 for ; Mon, 11 Sep 2000 17:28:51 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id RAA71076 for alpha@FreeBSD.org; Mon, 11 Sep 2000 17:28:49 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009120028.RAA71076@pike.osd.bsdi.com> Subject: Mutex's aren't recursing To: alpha@FreeBSD.org Date: Mon, 11 Sep 2000 17:28:49 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hmm, well, after debugging some, it seems that the mutexes on the alpha are broken wrt recursion. To see the code I'm using, grab the patchset http://www.FreeBSD.org/~jhb/patches/alpha.ithreads.patch. The code in question are the mtx_enter and mtx_enter_hard functions in sys/alpha/include/mutex.h and sys/alpha/alpha/syncy_machdep.c, respectively. The problem seems to be that when we recurse on a lock, we aren't marking the lock as recursed, and we aren't incrementing the recursion count. However, the code we are using to do this is almost exactly the same between i386 and alpha (and it works on i386). As a result, when the "inner" function releases Giant, it doesn't see that Giant is recursed, so it releases it directly. When the "outer" function then releases Giant, it is already free, resulting in an assertion failure if INVARIANTS is on, or in a panic later on as reported by other people on here. The following snippet of KTR output show's Giant being acquired twice but the recursion field (`r') staying at zero: 590 0:019738556 cpu0 machine/mutex.h.511 REL Giant [0xfffffc0000666170] at ../../alpha/alpha/interrupt.c:123 r=0 589 0:019685941 cpu0 machine/mutex.h.472 GOT Giant [0xfffffc0000666170] at ../../alpha/alpha/interrupt.c:121 r=0 588 0:019681066 cpu0 ../../alpha/alpha/interrupt.c.115 clock interrupt 587 0:019665286 cpu0 machine/mutex.h.472 GOT Giant [0xfffffc0000666170] at ../../alpha/alpha/ipl_funcs.c:138 r=0 586 0:019611469 cpu0 machine/mutex.h.511 REL malloc [0xfffffc000065f650] at ../../kern/kern_malloc.c:283 r=0 585 0:019563381 cpu0 machine/mutex.h.472 GOT malloc [0xfffffc000065f650] at ../../kern/kern_malloc.c:157 r=0 I've stared at the mutex code in question but don't see where it is breaking. Anyone else see anything that might be wrong? -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Mon Sep 11 23:46:36 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id B006337B422 for ; Mon, 11 Sep 2000 23:46:34 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id XAA13864 for ; Mon, 11 Sep 2000 23:46:34 -0700 Date: Mon, 11 Sep 2000 23:46:15 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: alpha@FreeBSD.ORG Subject: FWIW....top o' tree worked for me on a buildworld Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I made it through buildworld using a top 'o tree kernel- all w/o crashing. Now- if we could only get the loader fixed..... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 2:13:37 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-32.mail.demon.net (anchor-post-32.mail.demon.net [194.217.242.90]) by hub.freebsd.org (Postfix) with ESMTP id 8244A37B422; Tue, 12 Sep 2000 02:13:34 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-32.mail.demon.net with esmtp (Exim 2.12 #1) id 13Ym88-00006Y-0W; Tue, 12 Sep 2000 10:13:29 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id KAA07706; Tue, 12 Sep 2000 10:14:29 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Tue, 12 Sep 2000 10:13:48 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: dfr@FreeBSD.org, alpha@FreeBSD.org Subject: Re: Bug in spinlocks? In-Reply-To: <200009112037.NAA61735@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 11 Sep 2000, John Baldwin wrote: > Hmm, is there any reason why we aren't disabling interrupts on the alpha > when we obtain spinlocks? In the i386 code we use the following macro: > > /* Get a spin lock, handle recursion inline (as the less common case) */ > #define _getlock_spin_block(mp, tid, type) do { \ > u_int _mtx_fl = read_eflags(); \ > disable_intr(); \ > if (atomic_cmpset_int(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ > mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _mtx_fl); \ > else \ > (mp)->mtx_savefl = _mtx_fl; \ > } while (0) > > on the alpha it is almost the same: > > /* > * Get a spin lock, handle recusion inline (as the less common case) > */ > > #define _getlock_spin_block(mp, tid, type) do { \ > u_int _ipl = alpha_pal_rdps() & ALPHA_PSL_IPL_MASK; \ > if (atomic_cmpset_64(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ > mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _ipl); \ > else { \ > alpha_mb(); \ > (mp)->mtx_saveipl = _ipl; \ > } \ > } while (0) > > Note that in the alpha we are just saving and restoring the ipl, but we > aren't actually changing it to disable interrupts. Looks like a major brain failure on my part. Try these (untested) changes: Index: mutex.h =================================================================== RCS file: /home/ncvs/src/sys/alpha/include/mutex.h,v retrieving revision 1.5 diff -u -r1.5 mutex.h --- mutex.h 2000/09/09 23:18:47 1.5 +++ mutex.h 2000/09/12 09:13:25 @@ -223,9 +223,9 @@ extern char STR_IEN[]; extern char STR_IDIS[]; #endif /* MTX_STRS */ -#define ASS_IEN MPASS2((alpha_pal_rdps & ALPHA_PSL_IPL_MASK) \ +#define ASS_IEN MPASS2((alpha_pal_rdps() & ALPHA_PSL_IPL_MASK) \ == ALPHA_PSL_IPL_HIGH, STR_IEN) -#define ASS_IDIS MPASS2((alpha_pal_rdps & ALPHA_PSL_IPL_MASK) \ +#define ASS_IDIS MPASS2((alpha_pal_rdps() & ALPHA_PSL_IPL_MASK) \ != ALPHA_PSL_IPL_HIGH, STR_IDIS) #endif /* INVARIANTS */ @@ -326,7 +326,7 @@ */ #define _getlock_spin_block(mp, tid, type) do { \ - u_int _ipl = alpha_pal_rdps() & ALPHA_PSL_IPL_MASK; \ + u_int _ipl = alpha_pal_swpipl(ALPHA_PSL_IPL_HIGH); \ if (atomic_cmpset_64(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _ipl); \ else { \ @@ -544,8 +544,8 @@ * Simple assembly macros to get and release non-recursive spin locks */ #define MTX_ENTER(lck) \ - call_pal PAL_OSF1_rdps; \ - and v0, ALPHA_PSL_IPL_MASK, v0; \ + ldiq a0, ALPHA_PSL_IPL_HIGH; \ + call_pal PAL_OSF1_swpipl; \ 1: ldq_l a0, lck+MTX_LOCK; \ cmpeq a0, MTX_UNOWNED, a1; \ beq a1, 1b; \ @@ -553,9 +553,7 @@ stq_c a0, lck+MTX_LOCK; \ beq a0, 1b; \ mb; \ - stl v0, lck+MTX_SAVEIPL; \ - ldq a0, ALPHA_PSL_IPL_HIGH; \ - call_pal PSL_OSF1_swpipl + stl v0, lck+MTX_SAVEIPL #define MTX_EXIT(lck) \ mb; \ -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 10:31:50 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from post.mail.nl.demon.net (post-10.mail.nl.demon.net [194.159.73.20]) by hub.freebsd.org (Postfix) with ESMTP id 27E0737B423 for ; Tue, 12 Sep 2000 10:31:48 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #2) id 13YtuM-0006Tz-00; Tue, 12 Sep 2000 17:31:46 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e8CHWm801739; Tue, 12 Sep 2000 19:32:48 +0200 (CEST) (envelope-from wkb) Date: Tue, 12 Sep 2000 19:32:47 +0200 From: Wilko Bulte To: Matthew Jacob Cc: alpha@freebsd.org Subject: Re: FWIW....top o' tree worked for me on a buildworld Message-ID: <20000912193247.F1010@freebie.demon.nl> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: ; from mjacob@feral.com on Mon, Sep 11, 2000 at 11:46:15PM -0700 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Sep 11, 2000 at 11:46:15PM -0700, Matthew Jacob wrote: > > I made it through buildworld using a top 'o tree kernel- all w/o crashing. > > Now- if we could only get the loader fixed..... Hmm. I have an AS2100 at work building world for me. The last couple of days the dreaded Alcor lockup kept me from booting a -current kernel on AS1000A (which has a lot quicker CPU). It is now using a 4.1-R kernel (don't tell me this isn't a brilliant idea. I know. But I lost all my working -current kernels. AS2100 is panicing on a -current kernel (pagefault during T2 probe IIRC). Alpha is not a lot of fun these days :-( -- Wilko Bulte wilko@freebsd.org Arnhem, the Netherlands To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 13:32:24 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-34.mail.demon.net (anchor-post-34.mail.demon.net [194.217.242.92]) by hub.freebsd.org (Postfix) with ESMTP id 87D6037B424 for ; Tue, 12 Sep 2000 13:32:17 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-34.mail.demon.net with esmtp (Exim 2.12 #1) id 13Ywiu-000N6E-0Y; Tue, 12 Sep 2000 21:32:09 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id VAA09773; Tue, 12 Sep 2000 21:33:12 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Tue, 12 Sep 2000 21:32:23 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@freebsd.org Subject: Re: Mutex's aren't recursing In-Reply-To: <200009120028.RAA71076@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 11 Sep 2000, John Baldwin wrote: > Hmm, well, after debugging some, it seems that the mutexes on the alpha are > broken wrt recursion. To see the code I'm using, grab the patchset > http://www.FreeBSD.org/~jhb/patches/alpha.ithreads.patch. The code in > question are the mtx_enter and mtx_enter_hard functions in > sys/alpha/include/mutex.h and sys/alpha/alpha/syncy_machdep.c, respectively. > The problem seems to be that when we recurse on a lock, we aren't marking the > lock as recursed, and we aren't incrementing the recursion count. However, > the code we are using to do this is almost exactly the same between i386 > and alpha (and it works on i386). As a result, when the "inner" function > releases Giant, it doesn't see that Giant is recursed, so it releases it > directly. When the "outer" function then releases Giant, it is already free, > resulting in an assertion failure if INVARIANTS is on, or in a panic later > on as reported by other people on here. > > The following snippet of KTR output show's Giant being acquired twice but > the recursion field (`r') staying at zero: > > 590 0:019738556 cpu0 machine/mutex.h.511 REL Giant [0xfffffc0000666170] at ../../alpha/alpha/interrupt.c:123 r=0 > 589 0:019685941 cpu0 machine/mutex.h.472 GOT Giant [0xfffffc0000666170] at ../../alpha/alpha/interrupt.c:121 r=0 > 588 0:019681066 cpu0 ../../alpha/alpha/interrupt.c.115 clock interrupt > 587 0:019665286 cpu0 machine/mutex.h.472 GOT Giant [0xfffffc0000666170] at ../../alpha/alpha/ipl_funcs.c:138 r=0 > 586 0:019611469 cpu0 machine/mutex.h.511 REL malloc [0xfffffc000065f650] at ../../kern/kern_malloc.c:283 r=0 > 585 0:019563381 cpu0 machine/mutex.h.472 GOT malloc [0xfffffc000065f650] at ../../kern/kern_malloc.c:157 r=0 > > I've stared at the mutex code in question but don't see where it is > breaking. Anyone else see anything that might be wrong? I sent you some mail yesterday about this. I got the constraints wrong for the inline assembler in atomic_cmpset. I disassembled some of the code in interrupt.c which is trying to enter the mutex. You can clearly see that it is misusing t1 as both an input and output to the inline. 0xfffffc00005542c4 : lda t1,8(zero) 0xfffffc00005542c8 : ldq t0,64(t7) 0xfffffc00005542cc : ldq_l t1,0(s1) 0xfffffc00005542d0 : cmpeq t1,t1,t2 0xfffffc00005542d4 : beq t2,0xfffffc00005542e4 0xfffffc00005542d8 : mov t0,t1 0xfffffc00005542dc : stq_c t1,0(s1) I'm just about to start testing this patch which should fix the problem and also provides efficent forms for atomic_{add,subtract,set,clear}_{32,64}. I also fixed the interrupt problems with the spin locks (I think). Index: atomic.h =================================================================== RCS file: /home/ncvs/src/sys/alpha/include/atomic.h,v retrieving revision 1.3 diff -u -r1.3 atomic.h --- atomic.h 2000/09/06 11:20:53 1.3 +++ atomic.h 2000/09/12 20:19:05 @@ -44,15 +44,149 @@ void atomic_add_16(volatile u_int16_t *, u_int16_t); void atomic_subtract_16(volatile u_int16_t *, u_int16_t); -void atomic_set_32(volatile u_int32_t *, u_int32_t); -void atomic_clear_32(volatile u_int32_t *, u_int32_t); -void atomic_add_32(volatile u_int32_t *, u_int32_t); -void atomic_subtract_32(volatile u_int32_t *, u_int32_t); - -void atomic_set_64(volatile u_int64_t *, u_int64_t); -void atomic_clear_64(volatile u_int64_t *, u_int64_t); -void atomic_add_64(volatile u_int64_t *, u_int64_t); -void atomic_subtract_64(volatile u_int64_t *, u_int64_t); +static __inline void atomic_set_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "bis %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_clear_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "bic %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_add_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "addl %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_subtract_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "subl %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_set_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "bis %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_clear_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "bic %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_add_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "addq %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_subtract_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "subq %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} #define atomic_set_char atomic_set_8 #define atomic_clear_char atomic_clear_8 @@ -95,7 +229,7 @@ ".section .text3,\"ax\"\n" /* improve branch prediction */ "3:\tbr 1b\n" /* try again */ ".previous\n" - : "=&r" (ret), "=r" (temp), "=m" (*p) + : "=&r" (ret), "=&r" (temp), "=m" (*p) : "r" (cmpval), "r" (newval), "m" (*p) : "memory"); @@ -123,7 +257,7 @@ ".section .text3,\"ax\"\n" /* improve branch prediction */ "3:\tbr 1b\n" /* try again */ ".previous\n" - : "=&r" (ret), "=r" (temp), "=m" (*p) + : "=&r" (ret), "=&r" (temp), "=m" (*p) : "r" (cmpval), "r" (newval), "m" (*p) : "memory"); Index: mutex.h =================================================================== RCS file: /home/ncvs/src/sys/alpha/include/mutex.h,v retrieving revision 1.5 diff -u -r1.5 mutex.h --- mutex.h 2000/09/09 23:18:47 1.5 +++ mutex.h 2000/09/12 09:13:25 @@ -223,9 +223,9 @@ extern char STR_IEN[]; extern char STR_IDIS[]; #endif /* MTX_STRS */ -#define ASS_IEN MPASS2((alpha_pal_rdps & ALPHA_PSL_IPL_MASK) \ +#define ASS_IEN MPASS2((alpha_pal_rdps() & ALPHA_PSL_IPL_MASK) \ == ALPHA_PSL_IPL_HIGH, STR_IEN) -#define ASS_IDIS MPASS2((alpha_pal_rdps & ALPHA_PSL_IPL_MASK) \ +#define ASS_IDIS MPASS2((alpha_pal_rdps() & ALPHA_PSL_IPL_MASK) \ != ALPHA_PSL_IPL_HIGH, STR_IDIS) #endif /* INVARIANTS */ @@ -326,7 +326,7 @@ */ #define _getlock_spin_block(mp, tid, type) do { \ - u_int _ipl = alpha_pal_rdps() & ALPHA_PSL_IPL_MASK; \ + u_int _ipl = alpha_pal_swpipl(ALPHA_PSL_IPL_HIGH); \ if (atomic_cmpset_64(&(mp)->mtx_lock, MTX_UNOWNED, (tid)) == 0) \ mtx_enter_hard(mp, (type) & MTX_HARDOPTS, _ipl); \ else { \ @@ -544,8 +544,8 @@ * Simple assembly macros to get and release non-recursive spin locks */ #define MTX_ENTER(lck) \ - call_pal PAL_OSF1_rdps; \ - and v0, ALPHA_PSL_IPL_MASK, v0; \ + ldiq a0, ALPHA_PSL_IPL_HIGH; \ + call_pal PAL_OSF1_swpipl; \ 1: ldq_l a0, lck+MTX_LOCK; \ cmpeq a0, MTX_UNOWNED, a1; \ beq a1, 1b; \ @@ -553,9 +553,7 @@ stq_c a0, lck+MTX_LOCK; \ beq a0, 1b; \ mb; \ - stl v0, lck+MTX_SAVEIPL; \ - ldq a0, ALPHA_PSL_IPL_HIGH; \ - call_pal PSL_OSF1_swpipl + stl v0, lck+MTX_SAVEIPL #define MTX_EXIT(lck) \ mb; \ -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 14:40: 5 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 789FF37B42C for ; Tue, 12 Sep 2000 14:40:00 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8CLdq102331 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Tue, 12 Sep 2000 23:39:54 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8CLdqp70945; Tue, 12 Sep 2000 23:39:53 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8CLdqs56576; Tue, 12 Sep 2000 23:39:52 +0200 (CEST) (envelope-from ticso) Date: Tue, 12 Sep 2000 23:39:51 +0200 From: Bernd Walter To: Doug Rabson Cc: John Baldwin , alpha@FreeBSD.ORG Subject: Re: Mutex's aren't recursing Message-ID: <20000912233951.A56554@cicely5.cicely.de> References: <200009120028.RAA71076@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from dfr@nlsystems.com on Tue, Sep 12, 2000 at 09:32:23PM +0100 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Sep 12, 2000 at 09:32:23PM +0100, Doug Rabson wrote: > I sent you some mail yesterday about this. I got the constraints wrong for > the inline assembler in atomic_cmpset. I disassembled some of the code in > interrupt.c which is trying to enter the mutex. You can clearly see that > it is misusing t1 as both an input and output to the inline. > > 0xfffffc00005542c4 : lda t1,8(zero) > 0xfffffc00005542c8 : ldq t0,64(t7) > 0xfffffc00005542cc : ldq_l t1,0(s1) > 0xfffffc00005542d0 : cmpeq t1,t1,t2 > 0xfffffc00005542d4 : beq t2,0xfffffc00005542e4 > 0xfffffc00005542d8 : mov t0,t1 > 0xfffffc00005542dc : stq_c t1,0(s1) > > I'm just about to start testing this patch which should fix the problem > and also provides efficent forms for > atomic_{add,subtract,set,clear}_{32,64}. I also fixed the interrupt > problems with the spin locks (I think). Your patch does the following on a PC164 to me: 1. The mutex change will hang my system every second boot during PCI device printings. 2. Both changes panic the system on boot: [...] da12: Fixed Direct Access SCSI-2 device da12: 5.000MB/s transfers (5.000MHz, offset 7) da12: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da14 at sym2 bus 0 target 3 lun 0 da14: Fixed Direct Access SCSI-CCS device da14: 3.300MB/s transfers da14: 170MB (349770 512 byte sectors: 64H 32S/T 170C) Mounting root from ufs:/dev/da0a fatal kernel trap: trap entry = 0x2 (memory management fault) a0 = 0x0 a1 = 0x1 a2 = 0x0 pc = 0xfffffc0000517458 ra = 0xfffffc0000518fbc curproc = 0xfffffe0008e02720 pid = 7, comm = init Stopped at mtx_exit_hard+0x118: ldq t1,0(s1) <0x0> db> trace mtx_exit_hard() at mtx_exit_hard+0x118 child_return() at child_return+0x9c exception_return() at exception_return -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 14:43:33 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 0889137B43C for ; Tue, 12 Sep 2000 14:43:31 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id OAA08606; Tue, 12 Sep 2000 14:43:24 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009122143.OAA08606@pike.osd.bsdi.com> Subject: Re: Mutex's aren't recursing In-Reply-To: from Doug Rabson at "Sep 12, 2000 09:32:23 pm" To: Doug Rabson Date: Tue, 12 Sep 2000 14:43:24 -0700 (PDT) Cc: alpha@freebsd.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > I sent you some mail yesterday about this. I got the constraints wrong for > the inline assembler in atomic_cmpset. I disassembled some of the code in > interrupt.c which is trying to enter the mutex. You can clearly see that > it is misusing t1 as both an input and output to the inline. Thanks! I'll try this out. Also, I just discovered that swtch.s didn't import sched_lock, so cpu_switch was modifying a private copy of sched_lock instead of the global variable (I think). I'll have some feedback on this in just a sec. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 15:12:13 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-32.mail.demon.net (anchor-post-32.mail.demon.net [194.217.242.90]) by hub.freebsd.org (Postfix) with ESMTP id 418E637B422 for ; Tue, 12 Sep 2000 15:12:11 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-32.mail.demon.net with esmtp (Exim 2.12 #1) id 13YyHh-000JVV-0W; Tue, 12 Sep 2000 23:12:09 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id XAA10106; Tue, 12 Sep 2000 23:13:16 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Tue, 12 Sep 2000 23:12:26 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@freebsd.org Subject: Re: Mutex's aren't recursing In-Reply-To: <200009122143.OAA08606@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 12 Sep 2000, John Baldwin wrote: > Doug Rabson wrote: > > I sent you some mail yesterday about this. I got the constraints wrong for > > the inline assembler in atomic_cmpset. I disassembled some of the code in > > interrupt.c which is trying to enter the mutex. You can clearly see that > > it is misusing t1 as both an input and output to the inline. > > Thanks! I'll try this out. Also, I just discovered that swtch.s didn't import > sched_lock, so cpu_switch was modifying a private copy of sched_lock instead > of the global variable (I think). I'll have some feedback on this in just a > sec. It doesn't work - I'm still working on it. I have a new version of cmpset which does appear to work for recursion but oddly makes the kernel panic almost instantly we try to get into single user mode. I think its probably right after a context switch since I can see mi_switch() releasing Giant a few times but not regaining it in the new process. I'm in the middle of building a new kernel with more KTRs enabled. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 15:12:51 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from finch-post-10.mail.demon.net (finch-post-10.mail.demon.net [194.217.242.38]) by hub.freebsd.org (Postfix) with ESMTP id 36C7137B424 for ; Tue, 12 Sep 2000 15:12:45 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-10.mail.demon.net with esmtp (Exim 2.12 #1) id 13YyIE-000Ac0-0A; Tue, 12 Sep 2000 22:12:42 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id XAA10125; Tue, 12 Sep 2000 23:14:15 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Tue, 12 Sep 2000 23:13:25 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@freebsd.org Subject: Re: Mutex's aren't recursing In-Reply-To: <200009122143.OAA08606@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 12 Sep 2000, John Baldwin wrote: > Doug Rabson wrote: > > I sent you some mail yesterday about this. I got the constraints wrong for > > the inline assembler in atomic_cmpset. I disassembled some of the code in > > interrupt.c which is trying to enter the mutex. You can clearly see that > > it is misusing t1 as both an input and output to the inline. > > Thanks! I'll try this out. Also, I just discovered that swtch.s didn't import > sched_lock, so cpu_switch was modifying a private copy of sched_lock instead > of the global variable (I think). I'll have some feedback on this in just a > sec. BTW, this is my current version of atomic.h: Index: atomic.h =================================================================== RCS file: /home/ncvs/src/sys/alpha/include/atomic.h,v retrieving revision 1.3 diff -u -r1.3 atomic.h --- atomic.h 2000/09/06 11:20:53 1.3 +++ atomic.h 2000/09/12 20:46:23 @@ -44,15 +44,149 @@ void atomic_add_16(volatile u_int16_t *, u_int16_t); void atomic_subtract_16(volatile u_int16_t *, u_int16_t); -void atomic_set_32(volatile u_int32_t *, u_int32_t); -void atomic_clear_32(volatile u_int32_t *, u_int32_t); -void atomic_add_32(volatile u_int32_t *, u_int32_t); -void atomic_subtract_32(volatile u_int32_t *, u_int32_t); - -void atomic_set_64(volatile u_int64_t *, u_int64_t); -void atomic_clear_64(volatile u_int64_t *, u_int64_t); -void atomic_add_64(volatile u_int64_t *, u_int64_t); -void atomic_subtract_64(volatile u_int64_t *, u_int64_t); +static __inline void atomic_set_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "bis %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_clear_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "bic %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_add_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "addl %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_subtract_32(volatile u_int32_t *p, u_int32_t v) +{ + u_int32_t temp; + + __asm __volatile ( + "1:\tldl_l %0, %2\n\t" /* load old value */ + "subl %0, %3, %0\n\t" /* calculate new value */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_set_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "bis %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_clear_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "bic %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_add_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "addq %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} + +static __inline void atomic_subtract_64(volatile u_int64_t *p, u_int64_t v) +{ + u_int64_t temp; + + __asm __volatile ( + "1:\tldq_l %0, %2\n\t" /* load old value */ + "subq %0, %3, %0\n\t" /* calculate new value */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 2f\n\t" /* spin if failed */ + "mb\n\t" /* drain to memory */ + ".section .text3,\"ax\"\n" /* improve branch prediction */ + "2:\tbr 1b\n" /* try again */ + ".previous\n" + : "=&r" (temp), "=m" (*p) + : "m" (*p), "r" (v) + : "memory"); +} #define atomic_set_char atomic_set_8 #define atomic_clear_char atomic_clear_8 @@ -82,20 +216,20 @@ static __inline u_int32_t atomic_cmpset_32(volatile u_int32_t* p, u_int32_t cmpval, u_int32_t newval) { - u_int32_t ret, temp; + u_int32_t ret; __asm __volatile ( - "1:\tldl_l %1, %5\n\t" /* load old value */ - "cmpeq %1, %3, %0\n\t" /* compare */ + "1:\tldl_l %0, %4\n\t" /* load old value */ + "cmpeq %0, %2, %0\n\t" /* compare */ "beq %0, 2f\n\t" /* exit if not equal */ - "mov %4, %1\n\t" /* value to store */ - "stl_c %1, %2\n\t" /* attempt to store */ - "beq %1, 3f\n\t" /* if it failed, spin */ + "mov %3, %0\n\t" /* value to store */ + "stl_c %0, %1\n\t" /* attempt to store */ + "beq %0, 3f\n\t" /* if it failed, spin */ "2:\n" /* done */ ".section .text3,\"ax\"\n" /* improve branch prediction */ "3:\tbr 1b\n" /* try again */ ".previous\n" - : "=&r" (ret), "=r" (temp), "=m" (*p) + : "=&r" (ret), "=m" (*p) : "r" (cmpval), "r" (newval), "m" (*p) : "memory"); @@ -110,20 +244,20 @@ static __inline u_int64_t atomic_cmpset_64(volatile u_int64_t* p, u_int64_t cmpval, u_int64_t newval) { - u_int64_t ret, temp; + u_int64_t ret; __asm __volatile ( - "1:\tldq_l %1, %5\n\t" /* load old value */ - "cmpeq %1, %3, %0\n\t" /* compare */ + "1:\tldq_l %0, %4\n\t" /* load old value */ + "cmpeq %0, %2, %0\n\t" /* compare */ "beq %0, 2f\n\t" /* exit if not equal */ - "mov %4, %1\n\t" /* value to store */ - "stq_c %1, %2\n\t" /* attempt to store */ - "beq %1, 3f\n\t" /* if it failed, spin */ + "mov %3, %0\n\t" /* value to store */ + "stq_c %0, %1\n\t" /* attempt to store */ + "beq %0, 3f\n\t" /* if it failed, spin */ "2:\n" /* done */ ".section .text3,\"ax\"\n" /* improve branch prediction */ "3:\tbr 1b\n" /* try again */ ".previous\n" - : "=&r" (ret), "=r" (temp), "=m" (*p) + : "=&r" (ret), "=m" (*p) : "r" (cmpval), "r" (newval), "m" (*p) : "memory"); -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 15:14: 0 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-32.mail.demon.net (anchor-post-32.mail.demon.net [194.217.242.90]) by hub.freebsd.org (Postfix) with ESMTP id 39A7037B42C for ; Tue, 12 Sep 2000 15:13:58 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-32.mail.demon.net with esmtp (Exim 2.12 #1) id 13YyJQ-000JY0-0W; Tue, 12 Sep 2000 23:13:57 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id XAA10132; Tue, 12 Sep 2000 23:15:30 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Tue, 12 Sep 2000 23:14:40 +0100 (BST) From: Doug Rabson To: Bernd Walter Cc: John Baldwin , alpha@freebsd.org Subject: Re: Mutex's aren't recursing In-Reply-To: <20000912233951.A56554@cicely5.cicely.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 12 Sep 2000, Bernd Walter wrote: > On Tue, Sep 12, 2000 at 09:32:23PM +0100, Doug Rabson wrote: > > I sent you some mail yesterday about this. I got the constraints wrong for > > the inline assembler in atomic_cmpset. I disassembled some of the code in > > interrupt.c which is trying to enter the mutex. You can clearly see that > > it is misusing t1 as both an input and output to the inline. > > > > 0xfffffc00005542c4 : lda t1,8(zero) > > 0xfffffc00005542c8 : ldq t0,64(t7) > > 0xfffffc00005542cc : ldq_l t1,0(s1) > > 0xfffffc00005542d0 : cmpeq t1,t1,t2 > > 0xfffffc00005542d4 : beq t2,0xfffffc00005542e4 > > 0xfffffc00005542d8 : mov t0,t1 > > 0xfffffc00005542dc : stq_c t1,0(s1) > > > > I'm just about to start testing this patch which should fix the problem > > and also provides efficent forms for > > atomic_{add,subtract,set,clear}_{32,64}. I also fixed the interrupt > > problems with the spin locks (I think). > > Your patch does the following on a PC164 to me: > 1. The mutex change will hang my system every second boot during PCI device > printings. I still don't understand this one fully. I think its orthogonal to the mutex problems. > 2. Both changes panic the system on boot: > [...] I see that too - I'm trying to debug it now. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Tue Sep 12 16:23:12 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id D0BF637B423 for ; Tue, 12 Sep 2000 16:23:10 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id QAA12225; Tue, 12 Sep 2000 16:23:03 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009122323.QAA12225@pike.osd.bsdi.com> Subject: Re: cvs commit: src/sys/alpha/alpha trap.c In-Reply-To: from Doug Rabson at "Sep 12, 2000 11:58:12 pm" To: Doug Rabson Date: Tue, 12 Sep 2000 16:23:03 -0700 (PDT) Cc: alpha@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > On Tue, 12 Sep 2000, Doug Rabson wrote: > > > dfr 2000/09/12 15:47:10 PDT > > > > Modified files: > > sys/alpha/alpha trap.c > > Log: > > Merge changes from the i386 port to allow userret() to be called both > > with and without holding the Giant mutex. > > With this change, my UP test box boots and appears happy. We will see if > it manages to complete its 'make world'. Can you turn on the Giant enter and exit's in interrupt() now? In my test code I can now see recursion happening ok, but the sched_lock recursion is doing some funny things. After my fix to swtch.s, recursion for sched_lock is being saved across cpu_switch properly, but our mutex code seems to be goofing up on mtx_enter still: 138 0:034559493 cpu0 machine/mutex.h.510 REL sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:813 r=0 137 0:034508805 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:785 r=0 136 0:032610555 cpu0 machine/mutex.h.471 GOT Giant [0xfffffc00006664a0] at ../../kern/kern_synch.c:958 r=0 135 0:032560177 cpu0 machine/mutex.h.510 REL Giant [0xfffffc00006664a0] at ../../alpha/alpha/interrupt.c:123 r=0 134 0:032509499 cpu0 machine/mutex.h.471 GOT Giant [0xfffffc00006664a0] at ../../alpha/alpha/interrupt.c:121 r=0 133 0:032504810 cpu0 ../../alpha/alpha/interrupt.c.115 clock interrupt 132 0:032450423 cpu0 machine/mutex.h.510 REL sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:956 r=1 At event 132, we release a lock with a recursion count of 1, so we set the recursion depth to 0, and clear the MTX_RECURSED flag, but we still own the lock. However, when we knab the sched_lock again at 137, we don't recurse properly, and it isn't marked as recursed. Thus, when we try to do the last mtx_exit() (in await() in this case) we don't own the mutex. :( However, in other places we are recursing on it ok: 118 0:031513861 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:898 r=1 117 0:031431686 cpu0 ../../kern/kern_synch.c.897 mi_switch: old proc 0xfffffe0009530720 (pid 16, intr6: isa), schedlock 0xfffffe 116 0:031423046 cpu0 ../../alpha/alpha/interrupt.c.581 ithd_loop pid 16: done 115 0:031374550 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../alpha/alpha/interrupt.c:577 r=0 > -- > Doug Rabson Mail: dfr@nlsystems.com > Nonlinear Systems Ltd. Phone: +44 20 8348 3944 -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 1:42:51 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from finch-post-10.mail.demon.net (finch-post-10.mail.demon.net [194.217.242.38]) by hub.freebsd.org (Postfix) with ESMTP id 944CE37B423 for ; Wed, 13 Sep 2000 01:42:48 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-10.mail.demon.net with esmtp (Exim 2.12 #1) id 13Z87x-000KQZ-0A; Wed, 13 Sep 2000 08:42:46 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id JAA12032; Wed, 13 Sep 2000 09:43:54 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Wed, 13 Sep 2000 09:42:57 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@FreeBSD.org Subject: Re: cvs commit: src/sys/alpha/alpha trap.c In-Reply-To: <200009122323.QAA12225@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 12 Sep 2000, John Baldwin wrote: > Doug Rabson wrote: > > On Tue, 12 Sep 2000, Doug Rabson wrote: > > > > > dfr 2000/09/12 15:47:10 PDT > > > > > > Modified files: > > > sys/alpha/alpha trap.c > > > Log: > > > Merge changes from the i386 port to allow userret() to be called both > > > with and without holding the Giant mutex. > > > > With this change, my UP test box boots and appears happy. We will see if > > it manages to complete its 'make world'. > > Can you turn on the Giant enter and exit's in interrupt() now? I have done this already, mainly to increase the pressure on Giant. I still think that its not the right thing to do, even for SMP machines. Also, there is clearly a race for recursing a mutex if it can be recursed from an irq context since the "m->mtx_recurse++" statement could be interrupted. > In my > test code I can now see recursion happening ok, but the sched_lock recursion > is doing some funny things. After my fix to swtch.s, recursion for sched_lock > is being saved across cpu_switch properly, but our mutex code seems to be > goofing up on mtx_enter still: > > 138 0:034559493 cpu0 machine/mutex.h.510 REL sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:813 r=0 > 137 0:034508805 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:785 r=0 > 136 0:032610555 cpu0 machine/mutex.h.471 GOT Giant [0xfffffc00006664a0] at ../../kern/kern_synch.c:958 r=0 > 135 0:032560177 cpu0 machine/mutex.h.510 REL Giant [0xfffffc00006664a0] at ../../alpha/alpha/interrupt.c:123 r=0 > 134 0:032509499 cpu0 machine/mutex.h.471 GOT Giant [0xfffffc00006664a0] at ../../alpha/alpha/interrupt.c:121 r=0 > 133 0:032504810 cpu0 ../../alpha/alpha/interrupt.c.115 clock interrupt > 132 0:032450423 cpu0 machine/mutex.h.510 REL sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:956 r=1 > > At event 132, we release a lock with a recursion count of 1, so we set the > recursion depth to 0, and clear the MTX_RECURSED flag, but we still own the > lock. However, when we knab the sched_lock again at 137, we don't recurse > properly, and it isn't marked as recursed. Thus, when we try to do the > last mtx_exit() (in await() in this case) we don't own the mutex. :( Hmm. At #132, it appears to be enabling interrupts too which is bad. > > However, in other places we are recursing on it ok: > > 118 0:031513861 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../kern/kern_synch.c:898 r=1 > 117 0:031431686 cpu0 ../../kern/kern_synch.c.897 mi_switch: old proc 0xfffffe0009530720 (pid 16, intr6: isa), schedlock 0xfffffe > 116 0:031423046 cpu0 ../../alpha/alpha/interrupt.c.581 ithd_loop pid 16: done > 115 0:031374550 cpu0 machine/mutex.h.471 GOT sched lock [0xfffffc00006662d0] at ../../alpha/alpha/interrupt.c:577 r=0 I'll start staring at the spin lock code and see if I can spot something. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 3:50: 6 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from freefall.freebsd.org (freefall.FreeBSD.org [216.136.204.21]) by hub.freebsd.org (Postfix) with ESMTP id DC68C37B43E for ; Wed, 13 Sep 2000 03:50:00 -0700 (PDT) Received: (from gnats@localhost) by freefall.freebsd.org (8.9.3/8.9.2) id DAA32194; Wed, 13 Sep 2000 03:50:00 -0700 (PDT) (envelope-from gnats@FreeBSD.org) Received: from ldah.math.uni.lodz.pl (ldah.math.uni.lodz.pl [212.191.65.29]) by hub.freebsd.org (Postfix) with ESMTP id 72A0437B422 for ; Wed, 13 Sep 2000 03:44:08 -0700 (PDT) Received: (from dagoon@localhost) by ldah.math.uni.lodz.pl (8.11.0/8.9.3) id e8DAh1p00587; Wed, 13 Sep 2000 12:43:01 +0200 (CEST) (envelope-from dagoon) Message-Id: <200009131043.e8DAh1p00587@ldah.math.uni.lodz.pl> Date: Wed, 13 Sep 2000 12:43:01 +0200 (CEST) From: Marcin Gryszkalis Reply-To: dagoon@ldah.math.uni.lodz.pl To: FreeBSD-gnats-submit@freebsd.org X-Send-Pr-Version: 3.2 Subject: alpha/21247: edquota fails with alignment error Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Number: 21247 >Category: alpha >Synopsis: edquota fails with alignment error >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-alpha >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Wed Sep 13 03:50:00 PDT 2000 >Closed-Date: >Last-Modified: >Originator: Marcin Gryszkalis >Release: FreeBSD 4.1-STABLE alpha >Organization: Math Dept. of University of Lodz >Environment: AlphaStation 500, 2 scsi disks, >Description: Whole second disk is /home and it's the only partition with quota. After edquota user it goes like this: "/tmp/EdP.aKn468" 3L, 132C written pid 468 (edquota): unaligned access: va=0x11ffb2bc pc=0x1600b288c ra=0x1600b2810 op=stq pid 468 (edquota): unaligned access: va=0x11ffb2cc pc=0x1600b288c ra=0x1600b2810 op=stq pid 468 (edquota): unaligned access: va=0x11ffb2c4 pc=0x1600b288c ra=0x1600b2810 op=stq edquota: /home: cannot change current allocation I wrote a short program to test quotactl() and it works ok, so the porblem is in edquota. >How-To-Repeat: >Fix: >Release-Note: >Audit-Trail: >Unformatted: To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 9:10:10 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id EA09C37B43E for ; Wed, 13 Sep 2000 09:10:06 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id JAA20308 for ; Wed, 13 Sep 2000 09:10:05 -0700 Date: Wed, 13 Sep 2000 09:06:07 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: alpha@FreeBSD.ORG Subject: interesting new one Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org checking for core dump...savecore: no core dump Doing additional network setup: ntpdate ntpd portmap ypbind. Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map too small: 6266880 total allocated syncing disks... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 9:56:50 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 6AECE37B423 for ; Wed, 13 Sep 2000 09:56:49 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id JAA37112; Wed, 13 Sep 2000 09:56:41 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009131656.JAA37112@pike.osd.bsdi.com> Subject: Re: interesting new one In-Reply-To: from Matthew Jacob at "Sep 13, 2000 09:06:07 am" To: mjacob@feral.com Date: Wed, 13 Sep 2000 09:56:41 -0700 (PDT) Cc: alpha@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > checking for core dump...savecore: no core dump > Doing additional network setup: ntpdate ntpd portmap ypbind. > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map > too small: 6266880 total allocated Kernel and world out of sync. Specifically all the NFS stuff (or possibly portmap). Disable NFS and you will boot fine. Then you need to build a new world of course. > syncing disks... -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 10:25: 5 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 887B537B43C for ; Wed, 13 Sep 2000 10:25:00 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8DHOsj28531 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Wed, 13 Sep 2000 19:24:56 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8DHOqp73603; Wed, 13 Sep 2000 19:24:56 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8DHOq557641; Wed, 13 Sep 2000 19:24:52 +0200 (CEST) (envelope-from ticso) Date: Wed, 13 Sep 2000 19:24:51 +0200 From: Bernd Walter To: Matthew Jacob Cc: alpha@FreeBSD.ORG Subject: Re: interesting new one Message-ID: <20000913192451.A57625@cicely5.cicely.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from mjacob@feral.com on Wed, Sep 13, 2000 at 09:06:07AM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, Sep 13, 2000 at 09:06:07AM -0700, Matthew Jacob wrote: > > checking for core dump...savecore: no core dump > Doing additional network setup: ntpdate ntpd portmap ypbind. > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map > too small: 6266880 total allocated > > syncing disks... I had this when I boot a kernel from 28th Aug with a recent world. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 10:30:49 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 823EE37B422 for ; Wed, 13 Sep 2000 10:30:43 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id KAA20585; Wed, 13 Sep 2000 10:30:41 -0700 Date: Wed, 13 Sep 2000 10:30:20 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: John Baldwin Cc: alpha@FreeBSD.ORG Subject: Re: interesting new one In-Reply-To: <200009131656.JAA37112@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 13 Sep 2000, John Baldwin wrote: > Matthew Jacob wrote: > > > > checking for core dump...savecore: no core dump > > Doing additional network setup: ntpdate ntpd portmap ypbind. > > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map > > too small: 6266880 total allocated > > Kernel and world out of sync. Specifically all the NFS stuff (or possibly > portmap). Disable NFS and you will boot fine. Then you need to build a new > world of course. Ow. Okay. Hmm. Interesting bootstrapping problem since all of my source is NFS mounted... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 10:37:36 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from netplex.com.au (adsl-63-207-30-186.dsl.snfc21.pacbell.net [63.207.30.186]) by hub.freebsd.org (Postfix) with ESMTP id 7AE7E37B43C for ; Wed, 13 Sep 2000 10:37:34 -0700 (PDT) Received: from netplex.com.au (peter@localhost [127.0.0.1]) by netplex.com.au (8.11.0/8.9.3) with ESMTP id e8DHbUG94853; Wed, 13 Sep 2000 10:37:31 -0700 (PDT) (envelope-from peter@netplex.com.au) Message-Id: <200009131737.e8DHbUG94853@netplex.com.au> X-Mailer: exmh version 2.1.1 10/15/1999 To: mjacob@feral.com Cc: John Baldwin , alpha@FreeBSD.ORG Subject: Re: interesting new one In-Reply-To: Date: Wed, 13 Sep 2000 10:37:30 -0700 From: Peter Wemm Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > On Wed, 13 Sep 2000, John Baldwin wrote: > > > Matthew Jacob wrote: > > > > > > checking for core dump...savecore: no core dump > > > Doing additional network setup: ntpdate ntpd portmap ypbind. > > > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem _map > > > too small: 6266880 total allocated > > > > Kernel and world out of sync. Specifically all the NFS stuff (or possibly > > portmap). Disable NFS and you will boot fine. Then you need to build a ne w > > world of course. > > Ow. Okay. Hmm. Interesting bootstrapping problem since all of my source is NF S > mounted... Heh. That's why "conventional wisdom" is to NOT use modules while doing kernel development except for specific drivers that you are working on and loading/unloading manually. echo "NO_MODULES=too_dangerous" >> /etc/make.conf Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 10:39:20 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 9740E37B42C for ; Wed, 13 Sep 2000 10:39:17 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id KAA20670; Wed, 13 Sep 2000 10:39:10 -0700 Date: Wed, 13 Sep 2000 10:38:49 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Peter Wemm Cc: John Baldwin , alpha@FreeBSD.ORG Subject: Re: interesting new one In-Reply-To: <200009131737.e8DHbUG94853@netplex.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Matthew Jacob wrote: > > > > On Wed, 13 Sep 2000, John Baldwin wrote: > > > > > Matthew Jacob wrote: > > > > > > > > checking for core dump...savecore: no core dump > > > > Doing additional network setup: ntpdate ntpd portmap ypbind. > > > > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem > _map > > > > too small: 6266880 total allocated > > > > > > Kernel and world out of sync. Specifically all the NFS stuff (or possibly > > > portmap). Disable NFS and you will boot fine. Then you need to build a ne > w > > > world of course. > > > > Ow. Okay. Hmm. Interesting bootstrapping problem since all of my source is NF > S > > mounted... > > Heh. That's why "conventional wisdom" is to NOT use modules while doing > kernel development except for specific drivers that you are working on and > loading/unloading manually. > > echo "NO_MODULES=too_dangerous" >> /etc/make.conf Hmm. Well- this is the second time in 2 years that this bit me. IMO, it's more valuable to run what the suckers on the street run. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 10:39:48 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id A9C3A37B42C for ; Wed, 13 Sep 2000 10:39:46 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id KAA38810; Wed, 13 Sep 2000 10:39:04 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009131739.KAA38810@pike.osd.bsdi.com> Subject: Re: interesting new one In-Reply-To: <200009131737.e8DHbUG94853@netplex.com.au> from Peter Wemm at "Sep 13, 2000 10:37:30 am" To: Peter Wemm Date: Wed, 13 Sep 2000 10:39:03 -0700 (PDT) Cc: mjacob@feral.com, alpha@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Peter Wemm wrote: > Matthew Jacob wrote: > > > > On Wed, 13 Sep 2000, John Baldwin wrote: > > > > > Matthew Jacob wrote: > > > > > > > > checking for core dump...savecore: no core dump > > > > Doing additional network setup: ntpdate ntpd portmap ypbind. > > > > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem > _map > > > > too small: 6266880 total allocated > > > > > > Kernel and world out of sync. Specifically all the NFS stuff (or possibly > > > portmap). Disable NFS and you will boot fine. Then you need to build a ne > w > > > world of course. > > > > Ow. Okay. Hmm. Interesting bootstrapping problem since all of my source is NF > S > > mounted... > > Heh. That's why "conventional wisdom" is to NOT use modules while doing > kernel development except for specific drivers that you are working on and > loading/unloading manually. This doesn't seem to have anything to do with modules though. It has to do with the interaction of the userland NFS daemons and the the kernel side of NFS. Note that it is trying to malloc approx. 300 meg. :P > echo "NO_MODULES=too_dangerous" >> /etc/make.conf > > Cheers, > -Peter > -- > Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au > "All of this is for nothing if we don't go to the stars" - JMS/B5 -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 12: 0:20 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 94FB537B440 for ; Wed, 13 Sep 2000 12:00:17 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id MAA42839 for alpha@FreeBSD.org; Wed, 13 Sep 2000 12:00:10 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009131900.MAA42839@pike.osd.bsdi.com> Subject: We have a missing Giant mtx_enter To: alpha@FreeBSD.org Date: Wed, 13 Sep 2000 12:00:10 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Well, before my last round of commits, if you compiled a kernel with INVARIANTS it didn't make it through the device probe before it died when it tried to release the sched_lock when it didn't own it (probably due to the recursion count not being properly saved & restored.) With the stuff I just committed, it can now boot into single user with INVARIANTS, but it fails on an assertion entering multi-user. We basically try to release Giant when we don't actually have it. My guess is that we aren't obtaining Giant somewhere that we are supposed to be. With my interrupt threads stuff, it is still hanging trying to probe the SCSI busses. I'm currently trying to get that working at the moment, but the missing mtx_enter on Giant is something we will need to solve eventually. My guess is it might be related to forking new processes based on where it died (when dhclient forked a child process to presumably dink with ifconfig). -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 12:26:39 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 47EE737B42C for ; Wed, 13 Sep 2000 12:26:37 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id MAA44113; Wed, 13 Sep 2000 12:25:07 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009131925.MAA44113@pike.osd.bsdi.com> Subject: Re: cvs commit: src/sys/alpha/alpha trap.c In-Reply-To: from Doug Rabson at "Sep 13, 2000 09:42:57 am" To: Doug Rabson Date: Wed, 13 Sep 2000 12:25:07 -0700 (PDT) Cc: alpha@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > On Tue, 12 Sep 2000, John Baldwin wrote: > > > Doug Rabson wrote: > > > On Tue, 12 Sep 2000, Doug Rabson wrote: > > > > > > > dfr 2000/09/12 15:47:10 PDT > > > > > > > > Modified files: > > > > sys/alpha/alpha trap.c > > > > Log: > > > > Merge changes from the i386 port to allow userret() to be called both > > > > with and without holding the Giant mutex. > > > > > > With this change, my UP test box boots and appears happy. We will see if > > > it manages to complete its 'make world'. > > > > Can you turn on the Giant enter and exit's in interrupt() now? > > I have done this already, mainly to increase the pressure on Giant. I > still think that its not the right thing to do, even for SMP > machines. Also, there is clearly a race for recursing a mutex if it can be > recursed from an irq context since the "m->mtx_recurse++" statement could > be interrupted. We should make it an atomic_add_int() then to work around that for the time being. I agree it is probably not the right thing to do at the moment. Once I get ithreads working, it probably won't be needed for device interrupts anymore, which will help some. If we can bounce the clock interrupt to a thread that will help as well. Also, I should make a note about the ithreads patch. It tries to use device_get_nameunit() when adding interrupt handlers so that ithreads can have useful names such as on the x86 that include the names of the devices that use that ithread. However, it turns out that during my testing dev isn't initialized yet, so that device_get_nameunit() ends up dereferencing NULL pointers. *sigh* In the cia pci code and the isa code I have hacked around this by just using "pci" and "isa" for the names instead. However, these hacks are only in those bus drivers atm because I don't plan to have them in the final patch, so if you are going to try out the ithreads stuff, you will have to modify your pcibus driver to use a similar hack to that of cia.c. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 13:24:46 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from finch-post-10.mail.demon.net (finch-post-10.mail.demon.net [194.217.242.38]) by hub.freebsd.org (Postfix) with ESMTP id 0305637B422 for ; Wed, 13 Sep 2000 13:24:42 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-10.mail.demon.net with esmtp (Exim 2.12 #1) id 13ZJ5D-000N61-0A; Wed, 13 Sep 2000 20:24:40 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id VAA14508; Wed, 13 Sep 2000 21:26:02 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Wed, 13 Sep 2000 21:24:58 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@FreeBSD.ORG Subject: Re: cvs commit: src/sys/alpha/alpha trap.c In-Reply-To: <200009131925.MAA44113@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 13 Sep 2000, John Baldwin wrote: > Doug Rabson wrote: > > On Tue, 12 Sep 2000, John Baldwin wrote: > > > > > Doug Rabson wrote: > > > > On Tue, 12 Sep 2000, Doug Rabson wrote: > > > > > > > > > dfr 2000/09/12 15:47:10 PDT > > > > > > > > > > Modified files: > > > > > sys/alpha/alpha trap.c > > > > > Log: > > > > > Merge changes from the i386 port to allow userret() to be called both > > > > > with and without holding the Giant mutex. > > > > > > > > With this change, my UP test box boots and appears happy. We will see if > > > > it manages to complete its 'make world'. > > > > > > Can you turn on the Giant enter and exit's in interrupt() now? > > > > I have done this already, mainly to increase the pressure on Giant. I > > still think that its not the right thing to do, even for SMP > > machines. Also, there is clearly a race for recursing a mutex if it can be > > recursed from an irq context since the "m->mtx_recurse++" statement could > > be interrupted. > > We should make it an atomic_add_int() then to work around that for the > time being. I agree it is probably not the right thing to do at the > moment. Once I get ithreads working, it probably won't be needed for > device interrupts anymore, which will help some. If we can bounce the clock > interrupt to a thread that will help as well. For device interrupts, the ithread code will handle taking Giant as appropriate for non-MP-safe drivers. For clock, we need to aim at an MP-safe hardclock with softclock handled by the SWI thread. That should remove all need to take non-spin mutexes in interrupt(). > > Also, I should make a note about the ithreads patch. It tries to use > device_get_nameunit() when adding interrupt handlers so that ithreads can > have useful names such as on the x86 that include the names of the devices > that use that ithread. However, it turns out that during my testing > dev isn't initialized yet, so that device_get_nameunit() ends up dereferencing > NULL pointers. *sigh* In the cia pci code and the isa code I have hacked > around this by just using "pci" and "isa" for the names instead. However, > these hacks are only in those bus drivers atm because I don't plan to have > them in the final patch, so if you are going to try out the ithreads stuff, > you will have to modify your pcibus driver to use a similar hack to that > of cia.c. That sounds odd. The device_get_nameunit() should work the same for both x86 and alpha. I'll investigate that and see if I can come up with something cleaner. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 14: 1:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from sasami.jurai.net (sasami.jurai.net [63.67.141.99]) by hub.freebsd.org (Postfix) with ESMTP id 4D45E37B422 for ; Wed, 13 Sep 2000 14:01:15 -0700 (PDT) Received: from localhost (winter@localhost) by sasami.jurai.net (8.9.3/8.8.7) with ESMTP id RAA64117; Wed, 13 Sep 2000 17:00:51 -0400 (EDT) Date: Wed, 13 Sep 2000 17:00:50 -0400 (EDT) From: "Matthew N. Dodd" To: Peter Wemm Cc: mjacob@feral.com, John Baldwin , alpha@FreeBSD.ORG Subject: Re: interesting new one In-Reply-To: <200009131737.e8DHbUG94853@netplex.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 13 Sep 2000, Peter Wemm wrote: > Heh. That's why "conventional wisdom" is to NOT use modules while > doing kernel development except for specific drivers that you are > working on and loading/unloading manually. Except that this happens on an x86 box running a kernel with NFS compiled in. It blows up when 'mountd' is run, not when any modules are loaded. -- | Matthew N. Dodd | '78 Datsun 280Z | '75 Volvo 164E | FreeBSD/NetBSD | | winter@jurai.net | 2 x '84 Volvo 245DL | ix86,sparc,pmax | | http://www.jurai.net/~winter | This Space For Rent | ISO8802.5 4ever | To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 17:59: 3 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id DA29437B422 for ; Wed, 13 Sep 2000 17:59:00 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id RAA56657; Wed, 13 Sep 2000 17:58:50 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009140058.RAA56657@pike.osd.bsdi.com> Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: from Doug Rabson at "Sep 11, 2000 00:43:49 am" To: Doug Rabson Date: Wed, 13 Sep 2000 17:58:50 -0700 (PDT) Cc: alpha@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > On first reading, it looks like a good start. I have a partially written > version which separates the interrupt source (the vector) from the > handlers which have been registered using that vector. That would simplify > the search for the ithread. In fact, there isn't much difference in what I > had between the interrupt source and the ithread itself. One idea to clean > things up would be to make the hash table entries contain the ithreads > directly, with the handlers in a list on each ithread. I'd also like to > re-do the hash function to avoid the modulus operation since that is very > expensive on alpha. > > I don't think this code will work properly without adding some code to > disable the interrupt source before returning from the low-level irq > handler. If this isn't done for at least pci interrupts, they will just > fire again immediately which tends to cause instant stack overflow. This is what is happening now. My alpha is hanging during the SCSI bus probe. When I selectively dumped the KTR output to the console, it was constantly scheduling the ithread for a PCI interrupt (I'm assuming it was my isp0 controller) because the interrupt was repeatedly firing. > One way to solve this would be to pass a couple more function pointers to > alpha_setup_intr(), one to enable a vector and one to disable it. The irq > handler would disable the vector and schedule the ithread. The ithread > would handle the interrupt and re-enable the vector before sleeping. I can go ahead and start tweaking with alpha_setup_intr() and all the places that happen to get these setup. However, what I don't know is what to stick in these functions to actually enable/disable the interrupt vector. ISA interrupts seem to be working fine, btw, so for ISA we can probably get away with just passing NULL and not calling any functions, but we definitely need something like this for PCI interrupts. I've updated the alpha.ithreads.patch on my freefall webpage to work with the latest current, btw. > -- > Doug Rabson Mail: dfr@nlsystems.com > Nonlinear Systems Ltd. Phone: +44 20 8348 3944 -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Wed Sep 13 23: 3:49 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 546FE37B424 for ; Wed, 13 Sep 2000 23:03:46 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8E63gI16106 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK) for ; Thu, 14 Sep 2000 08:03:44 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8E63gI75595 for ; Thu, 14 Sep 2000 08:03:46 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8E63gn58722 for freebsd-alpha@freebsd.org; Thu, 14 Sep 2000 08:03:42 +0200 (CEST) (envelope-from ticso) Date: Thu, 14 Sep 2000 08:03:41 +0200 From: Bernd Walter To: freebsd-alpha@freebsd.org Subject: buildworld with SMPng succeded Message-ID: <20000914080341.A58692@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org For me it was the first succesfull buildworld with a current since the SMPng merge. Sources were from yesterday. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 0: 5:23 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id DC80C37B423 for ; Thu, 14 Sep 2000 00:05:20 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8E75GL22147 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK) for ; Thu, 14 Sep 2000 09:05:19 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8E75HI75753 for ; Thu, 14 Sep 2000 09:05:17 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8E75GT58858 for freebsd-alpha@FreeBSD.ORG; Thu, 14 Sep 2000 09:05:16 +0200 (CEST) (envelope-from ticso) Date: Thu, 14 Sep 2000 09:05:16 +0200 From: Bernd Walter To: freebsd-alpha@FreeBSD.ORG Subject: Re: buildworld with SMPng succeded Message-ID: <20000914090515.A58839@cicely5.cicely.de> References: <20000914080341.A58692@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000914080341.A58692@cicely5.cicely.de>; from ticso@cicely5.cicely.de on Thu, Sep 14, 2000 at 08:03:41AM +0200 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, Sep 14, 2000 at 08:03:41AM +0200, Bernd Walter wrote: > For me it was the first succesfull buildworld with a current since the > SMPng merge. > Sources were from yesterday. To be more precise kernel source was from yesterday. After installworld the loader seems to be broken: jumping to bootstrap code Loading /boot/loader Console: SRM firmware console VMS PAL rev: 0x1000900010115 OSF PAL rev: 0x1000800020117 Switch to OSF PAL code succeeded. FreeBSD/alpha SRM disk boot, Revision 1.1 (ticso@cicely9.cicely.de, Thu Sep 14 03:37:12 CEST 2000) Memory: 262144 k - halted CPU 0 halt code = 2 kernel stack not valid halt PC = 200000000 boot failure -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 0:52:15 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 517DE37B43E for ; Thu, 14 Sep 2000 00:52:13 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8E7q7L27845 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Thu, 14 Sep 2000 09:52:09 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8E7qAI75922; Thu, 14 Sep 2000 09:52:11 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8E7qAE58938; Thu, 14 Sep 2000 09:52:10 +0200 (CEST) (envelope-from ticso) Date: Thu, 14 Sep 2000 09:52:10 +0200 From: Bernd Walter To: Matthew Jacob Cc: alpha@FreeBSD.ORG Subject: Re: so much for *that* idea..... Message-ID: <20000914095210.B58839@cicely5.cicely.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from mjacob@feral.com on Mon, Sep 11, 2000 at 10:52:30AM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Sep 11, 2000 at 10:52:30AM -0700, Matthew Jacob wrote: > > This isn't the kernel. This is the newer boot loader. Whatever was checked in > after 9/7 busted me. Replacing /boot/loader with /boot/loader.old > and symlinking /kernel -> /boot/kernel/kernel.ko and /modules -> /boot/kernel > did the right thing for me so I could boot. > > There are other breakages too (mfs is broken- EFAULT, ignored my loader.conf > to load ispfw), but at least this got me to multiuser. I run into this myself. Changing the loader to a loader.old did not help. The loader.old is the one I booted before with. Maybe this is because of the script changes. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 5:52:16 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from account.abs.net (account.abs.net [207.114.5.70]) by hub.freebsd.org (Postfix) with ESMTP id C63C737B422 for ; Thu, 14 Sep 2000 05:52:11 -0700 (PDT) Received: (from howardl@localhost) by account.abs.net (8.9.3/8.9.3+RBL+DUL+RSS) id IAA24566 for freebsd-alpha@freebsd.org; Thu, 14 Sep 2000 08:52:10 -0400 (EDT) (envelope-from howardl) From: Howard Leadmon Message-Id: <200009141252.IAA24566@account.abs.net> Subject: Make buildworld dies 4.1-STABLE.. To: freebsd-alpha@freebsd.org Date: Thu, 14 Sep 2000 08:52:10 -0400 (EDT) X-Mailer: ELM [version 2.4ME+ PL72 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Wanted to see if anyone had any ideas on what may be causing this, or has gotten broken in 4.1 STABLE. I did a cvsup on several of my machines last night (some Intel, and some Alpha), and when trying to run a make buildworld on my Intel's all went just fine. When trying to make buildworld on my Alpha's (PC164SX-533's) things blow up with the following: cc -O -pipe -DLOADER_NET_SUPPORT -I/usr/src/sys/boot/alpha/netboot/../../ficl -I/usr/src/sys/boot/alpha/netboot/../../ficl/alpha -DBOOT_FORTH -mno-fp-regs -I/usr/src/sys/boot/alpha/netboot/../../common -I/usr/src/sys/boot/alpha/netboot -I/usr/src/sys/boot/alpha/netboot/../../.. -I. -DPRIMARY_LOAD_ADDRESS=0x20000000 -DSECONDARY_LOAD_ADDRESS=0x2000c000 -Wall -I/usr/src/sys/boot/alpha/netboot/../../../../lib/libstand -I/usr/src/sys/boot/alpha/netboot/.. -I/usr/obj/usr/src/alpha/usr/include -c /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:200: conflicting types for `bootp' /usr/src/sys/boot/alpha/netboot/../../../../lib/libstand/net.h:117: previous declaration of `bootp' /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c: In function `net_getparams': /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:222: too many arguments to function `bootp' *** Error code 1 Stop in /usr/src/sys/boot/alpha/netboot. *** Error code 1 Stop in /usr/src/sys/boot/alpha. *** Error code 1 Stop in /usr/src/sys/boot. *** Error code 1 Stop in /usr/src/sys. *** Error code 1 Stop in /usr/src. *** Error code 1 Stop in /usr/src. *** Error code 1 Stop in /usr/src. Anyone have any ideas, or know if this has been fixed??? --- Howard Leadmon - howardl@abs.net - http://www.abs.net ABSnet Internet Services - Phone: 410-361-8160 - FAX: 410-361-8162 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 9:46:32 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from finch-post-12.mail.demon.net (finch-post-12.mail.demon.net [194.217.242.41]) by hub.freebsd.org (Postfix) with ESMTP id 93A1437B624 for ; Thu, 14 Sep 2000 09:46:26 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-12.mail.demon.net with esmtp (Exim 2.12 #1) id 13Zc9X-000Bgs-0C; Thu, 14 Sep 2000 16:46:24 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id RAA18539; Thu, 14 Sep 2000 17:47:57 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Thu, 14 Sep 2000 17:46:40 +0100 (BST) From: Doug Rabson To: Matthew Jacob Cc: alpha@freebsd.org Subject: Re: interesting new one In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 13 Sep 2000, Matthew Jacob wrote: > > checking for core dump...savecore: no core dump > Doing additional network setup: ntpdate ntpd portmap ypbind. > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map > too small: 6266880 total allocated > > syncing disks... I had this, something changed and mountd is passing the wrong shaped structure to mount(). The arguments are not checked very well by the kernel and it dies horribly. You can get to multiuser by simply deleting mountd and then rebuild from fresh sources. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 11: 2:36 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.swissonline.ch (mail.swissonline.ch [62.2.32.83]) by hub.freebsd.org (Postfix) with ESMTP id 422FE37B422 for ; Thu, 14 Sep 2000 11:02:27 -0700 (PDT) Received: from drowzee ([62.2.106.137]) by mail.swissonline.ch (8.9.3/8.9.3) with SMTP id UAA22221 for ; Thu, 14 Sep 2000 20:02:22 +0200 (MET DST) From: "Oliver Fehr" To: Subject: RE: Make buildworld dies 4.1-STABLE.. Date: Thu, 14 Sep 2000 20:01:47 +0200 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: <200009141252.IAA24566@account.abs.net> X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Importance: Normal Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The same happened to me 3 days ago on a DEC Personal WS 500au, the Intels did just go fine. The problem seems to be in dev_net.c with differing number of arguments: line 200: int bootp(int sock) line 222: bootp(sock,BOOTP_NONE) If I read the files correctly think there should be two arguments in line 220 according to /usr/src/lib/libstand/net.h. Cheers Oliver > -----Original Message----- > From: owner-freebsd-alpha@FreeBSD.ORG > [mailto:owner-freebsd-alpha@FreeBSD.ORG]On Behalf Of Howard Leadmon > Sent: Thursday, September 14, 2000 2:52 PM > To: freebsd-alpha@freebsd.org > Subject: Make buildworld dies 4.1-STABLE.. > > > > Wanted to see if anyone had any ideas on what may be causing this, or > has gotten broken in 4.1 STABLE. I did a cvsup on several of my machines > last night (some Intel, and some Alpha), and when trying to run a make > buildworld on my Intel's all went just fine. When trying to make > buildworld > on my Alpha's (PC164SX-533's) things blow up with the following: > > cc -O -pipe -DLOADER_NET_SUPPORT > -I/usr/src/sys/boot/alpha/netboot/../../ficl > -I/usr/src/sys/boot/alpha/netboot/../../ficl/alpha -DBOOT_FORTH > -mno-fp-regs -I/usr/src/sys/boot/alpha/netboot/../../common > -I/usr/src/sys/boot/alpha/netboot > -I/usr/src/sys/boot/alpha/netboot/../../.. -I. > -DPRIMARY_LOAD_ADDRESS=0x20000000 > -DSECONDARY_LOAD_ADDRESS=0x2000c000 -Wall > -I/usr/src/sys/boot/alpha/netboot/../../../../lib/libstand > -I/usr/src/sys/boot/alpha/netboot/.. > -I/usr/obj/usr/src/alpha/usr/include -c > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:200: > conflicting types for `bootp' > /usr/src/sys/boot/alpha/netboot/../../../../lib/libstand/net.h:117 > : previous declaration of `bootp' > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c: In > function `net_getparams': > /usr/src/sys/boot/alpha/netboot/../../common/dev_net.c:222: too > many arguments to function `bootp' > *** Error code 1 > > Stop in /usr/src/sys/boot/alpha/netboot. > *** Error code 1 > > Stop in /usr/src/sys/boot/alpha. > *** Error code 1 > > Stop in /usr/src/sys/boot. > *** Error code 1 > > Stop in /usr/src/sys. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > Stop in /usr/src. > > > > Anyone have any ideas, or know if this has been fixed??? > > > --- > Howard Leadmon - howardl@abs.net - http://www.abs.net > ABSnet Internet Services - Phone: 410-361-8160 - FAX: 410-361-8162 > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 12:57:37 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mass.osd.bsdi.com (adsl-63-202-177-115.dsl.snfc21.pacbell.net [63.202.177.115]) by hub.freebsd.org (Postfix) with ESMTP id 9FDE137B422 for ; Thu, 14 Sep 2000 12:57:35 -0700 (PDT) Received: from mass.osd.bsdi.com (localhost [127.0.0.1]) by mass.osd.bsdi.com (8.9.3/8.9.3) with ESMTP id MAA00889; Thu, 14 Sep 2000 12:57:13 -0700 (PDT) (envelope-from msmith@mass.osd.bsdi.com) Message-Id: <200009141957.MAA00889@mass.osd.bsdi.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: Doug Rabson Cc: Matthew Jacob , alpha@freebsd.org Subject: Re: interesting new one In-reply-to: Your message of "Thu, 14 Sep 2000 17:46:40 BST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 14 Sep 2000 12:57:13 -0700 From: Mike Smith Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On Wed, 13 Sep 2000, Matthew Jacob wrote: > > > > > checking for core dump...savecore: no core dump > > Doing additional network setup: ntpdate ntpd portmap ypbind. > > Starting final network daemons: mountdpanic: kmem_malloc(301973504): kmem_map > > too small: 6266880 total allocated > > > > syncing disks... > > I had this, something changed and mountd is passing the wrong shaped > structure to mount(). The arguments are not checked very well by the > kernel and it dies horribly. You can get to multiuser by simply deleting > mountd and then rebuild from fresh sources. I think it was the ucred structure growing an extra pointer. This worried me because I thought it might affect other kernel interfaces as well; I think the 'right' solution here would have been to split the kernel and userland versions of the struct... -- ... every activity meets with opposition, everyone who acts has his rivals and unfortunately opponents also. But not because people want to be opponents, rather because the tasks and relationships force people to take different points of view. [Dr. Fritz Todt] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 15:55:10 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 286A637B422; Thu, 14 Sep 2000 15:55:07 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id PAA95214; Thu, 14 Sep 2000 15:54:58 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009142254.PAA95214@pike.osd.bsdi.com> Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: from Doug Rabson at "Sep 14, 2000 03:55:04 pm" To: Doug Rabson Date: Thu, 14 Sep 2000 15:54:58 -0700 (PDT) Cc: alpha@FreeBSD.org, smp@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > On Wed, 13 Sep 2000, John Baldwin wrote: > > I've updated the alpha.ithreads.patch on my freefall webpage to work with > > the latest current, btw. > > I've been working on this today and I have a kernel which boots on my > miata and is happily chewing its way through a buildworld. I fixed the > device_get_nameunit() problem - it was just going wrong for ATA interrupts > which are handled quite strangely on alpha. I also changed the p_comm > string to include the whole vector number in hex since on some platforms > all bits are relavent, not just (vec-0x900)>>4. > > I've attached my modified version of your patch to this message. Its not > quite ready to commit yet since I didn't write the enable/disable hooks > for mcpcia or dwlpx. I have an mcpcia here (the AS4100) so I'll be able to > test that soon. After that last fix to trap.c to add an acquire/release of sched_lock, I am now happily running interrupt threads on my miata here with your patch. I've gone ahead and replaced my patch on freefall with yours, but hopefully we can commit this very soon when you get the other PCI chipsets finished. Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it is running fine so far. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 19:18:15 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id D72A437B422 for ; Thu, 14 Sep 2000 19:18:13 -0700 (PDT) Received: from zeppo.feral.com (IDENT:mjacob@zeppo [192.67.166.71]) by feral.com (8.9.3/8.9.3) with ESMTP id TAA26989; Thu, 14 Sep 2000 19:17:56 -0700 Date: Thu, 14 Sep 2000 19:17:54 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Peter Wemm Cc: alpha@freebsd.org Subject: PRE_SMPNG question.... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org If I want to make sure that PC164's boot and run in the source tree that was tagged PRE_SMPNG, is it acceptable to MFC, as it were, the fix? I've had to, hah hah (coz I wasn't as careful as I should) spend most of the day reinstalling 4.1 so I could get back to having some stable disks I can actually boot from prior to moving ahead back into -current for alpha. It strikes me that having the PRE_SMPNG stuff be stable would be good. But I'll admit that I'm a bit hazy as to how branches work vs. tags. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Thu Sep 14 22:58:22 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id CD56537B42C; Thu, 14 Sep 2000 22:58:18 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8F5wCV08120 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Fri, 15 Sep 2000 07:58:14 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8F5wGI79635; Fri, 15 Sep 2000 07:58:17 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8F5wCG60394; Fri, 15 Sep 2000 07:58:12 +0200 (CEST) (envelope-from ticso) Date: Fri, 15 Sep 2000 07:58:12 +0200 From: Bernd Walter To: John Baldwin Cc: Doug Rabson , alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000915075812.B60348@cicely5.cicely.de> References: <200009142254.PAA95214@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009142254.PAA95214@pike.osd.bsdi.com>; from jhb@pike.osd.bsdi.com on Thu, Sep 14, 2000 at 03:54:58PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, Sep 14, 2000 at 03:54:58PM -0700, John Baldwin wrote: > After that last fix to trap.c to add an acquire/release of sched_lock, I > am now happily running interrupt threads on my miata here with your patch. > I've gone ahead and replaced my patch on freefall with yours, but hopefully > we can commit this very soon when you get the other PCI chipsets finished. > Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it > is running fine so far. miata means that it should be testable on a PC164 - right? -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 9:18:40 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from dragon.nuxi.com (trang.nuxi.com [209.152.133.57]) by hub.freebsd.org (Postfix) with ESMTP id 100CC37B422; Fri, 15 Sep 2000 09:18:37 -0700 (PDT) Received: (from obrien@localhost) by dragon.nuxi.com (8.9.3/8.9.1) id JAA72911; Fri, 15 Sep 2000 09:18:26 -0700 (PDT) (envelope-from obrien) Date: Fri, 15 Sep 2000 09:18:26 -0700 From: "David O'Brien" To: Bernd Walter Cc: alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000915091826.A72881@dragon.nuxi.com> Reply-To: obrien@FreeBSD.ORG References: <200009142254.PAA95214@pike.osd.bsdi.com> <20000915075812.B60348@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20000915075812.B60348@cicely5.cicely.de>; from ticso@cicely5.cicely.de on Fri, Sep 15, 2000 at 07:58:12AM +0200 X-Operating-System: FreeBSD 5.0-CURRENT Organization: The NUXI BSD group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 07:58:12AM +0200, Bernd Walter wrote: > > miata means that it should be testable on a PC164 - right? No. There are some differences in the chipsets used. -- -- David (obrien@FreeBSD.org) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 9:26:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from dragon.nuxi.com (trang.nuxi.com [209.152.133.57]) by hub.freebsd.org (Postfix) with ESMTP id 8C40237B423; Fri, 15 Sep 2000 09:26:14 -0700 (PDT) Received: (from obrien@localhost) by dragon.nuxi.com (8.9.3/8.9.1) id JAA72956; Fri, 15 Sep 2000 09:26:09 -0700 (PDT) (envelope-from obrien) Date: Fri, 15 Sep 2000 09:26:09 -0700 From: "David O'Brien" To: Bernd Walter Cc: alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000915092609.B72881@dragon.nuxi.com> Reply-To: obrien@FreeBSD.ORG References: <200009142254.PAA95214@pike.osd.bsdi.com> <20000915075812.B60348@cicely5.cicely.de> <20000915091826.A72881@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20000915091826.A72881@dragon.nuxi.com>; from obrien@FreeBSD.ORG on Fri, Sep 15, 2000 at 09:18:26AM -0700 X-Operating-System: FreeBSD 5.0-CURRENT Organization: The NUXI BSD group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 09:18:26AM -0700, David O'Brien wrote: > On Fri, Sep 15, 2000 at 07:58:12AM +0200, Bernd Walter wrote: > > > > miata means that it should be testable on a PC164 - right? > > No. There are some differences in the chipsets used. Let me expand on this answer -- just because something worked on a Miata does not mean it will work on a PC164. Of course something working (or not working) does mean it is worth testing to see if it works on a PC164. -- -- David (obrien@FreeBSD.org) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 10:32:43 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id AA17037B423; Fri, 15 Sep 2000 10:32:38 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8FHWZB22377 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Fri, 15 Sep 2000 19:32:36 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8FHWeI81531; Fri, 15 Sep 2000 19:32:40 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8FHWaY61100; Fri, 15 Sep 2000 19:32:36 +0200 (CEST) (envelope-from ticso) Date: Fri, 15 Sep 2000 19:32:36 +0200 From: Bernd Walter To: "David O'Brien" Cc: alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000915193236.B61065@cicely5.cicely.de> References: <200009142254.PAA95214@pike.osd.bsdi.com> <20000915075812.B60348@cicely5.cicely.de> <20000915091826.A72881@dragon.nuxi.com> <20000915092609.B72881@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000915092609.B72881@dragon.nuxi.com>; from obrien@FreeBSD.ORG on Fri, Sep 15, 2000 at 09:26:09AM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 09:26:09AM -0700, David O'Brien wrote: > On Fri, Sep 15, 2000 at 09:18:26AM -0700, David O'Brien wrote: > > On Fri, Sep 15, 2000 at 07:58:12AM +0200, Bernd Walter wrote: > > > > > > miata means that it should be testable on a PC164 - right? > > > > No. There are some differences in the chipsets used. > > Let me expand on this answer -- just because something worked on a Miata > does not mean it will work on a PC164. Of course something working (or > not working) does mean it is worth testing to see if it works on a PC164. I wanted to know if there are known show stoppers. "worth testing" is exactly what I tried to say with testable. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 11: 3:47 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id D73DE37B424; Fri, 15 Sep 2000 11:03:44 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id LAA24561; Fri, 15 Sep 2000 11:03:31 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009151803.LAA24561@pike.osd.bsdi.com> Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: <20000915193236.B61065@cicely5.cicely.de> from Bernd Walter at "Sep 15, 2000 07:32:36 pm" To: Bernd Walter Date: Fri, 15 Sep 2000 11:03:31 -0700 (PDT) Cc: "David O'Brien" , alpha@FreeBSD.ORG, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Fri, Sep 15, 2000 at 09:26:09AM -0700, David O'Brien wrote: > > On Fri, Sep 15, 2000 at 09:18:26AM -0700, David O'Brien wrote: > > > On Fri, Sep 15, 2000 at 07:58:12AM +0200, Bernd Walter wrote: > > > > > > > > miata means that it should be testable on a PC164 - right? > > > > > > No. There are some differences in the chipsets used. > > > > Let me expand on this answer -- just because something worked on a Miata > > does not mean it will work on a PC164. Of course something working (or > > not working) does mean it is worth testing to see if it works on a PC164. > > I wanted to know if there are known show stoppers. > "worth testing" is exactly what I tried to say with testable. I believe it should be worth testing as I know that the cia PCI bus driver is ok. In fact, most of the PCI bus drivers are fixed, so it should work on most alphas, AFAIK. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 12: 4:11 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id 692D637B42C for ; Fri, 15 Sep 2000 12:04:09 -0700 (PDT) Received: (qmail 52219 invoked by uid 1142); 15 Sep 2000 19:04:08 -0000 Date: 15 Sep 2000 12:04:08 -0700 Date: Fri, 15 Sep 2000 11:52:39 -0700 From: Jason Evans To: Matthew Jacob Cc: alpha@freebsd.org Subject: Re: PRE_SMPNG question.... Message-ID: <20000915115239.E7247@blitz.canonware.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from mjacob@feral.com on Thu, Sep 14, 2000 at 07:17:54PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, Sep 14, 2000 at 07:17:54PM -0700, Matthew Jacob wrote: > > If I want to make sure that PC164's boot and run in the source tree > that was tagged PRE_SMPNG, is it acceptable to MFC, as it were, the fix? > > I've had to, hah hah (coz I wasn't as careful as I should) spend most of the > day reinstalling 4.1 so I could get back to having some stable disks I can > actually boot from prior to moving ahead back into -current for alpha. It > strikes me that having the PRE_SMPNG stuff be stable would be good. > > But I'll admit that I'm a bit hazy as to how branches work vs. tags. PRE_SMPNG is a static tag. That means that the only way to modify the PRE_SMPNG version of a file is to move the tag to a different version of a file. So, unless the file hasn't changed since the PRE_SMPNG tag was created, moving the tag will prove very difficult. Indeed, the purpose of the PRE_SMPNG tag was not to provide a branch point; it was merely meant to mark the last moment before the SMP code destabilized the tree. Jason To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 14:36:38 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 3149B37B422; Fri, 15 Sep 2000 14:36:28 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8FLaLh09023 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Fri, 15 Sep 2000 23:36:23 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8FLaRI82145; Fri, 15 Sep 2000 23:36:27 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8FLaMv61464; Fri, 15 Sep 2000 23:36:22 +0200 (CEST) (envelope-from ticso) Date: Fri, 15 Sep 2000 23:36:22 +0200 From: Bernd Walter To: John Baldwin Cc: "David O'Brien" , alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000915233621.A61424@cicely5.cicely.de> References: <20000915193236.B61065@cicely5.cicely.de> <200009151803.LAA24561@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009151803.LAA24561@pike.osd.bsdi.com>; from jhb@pike.osd.bsdi.com on Fri, Sep 15, 2000 at 11:03:31AM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 11:03:31AM -0700, John Baldwin wrote: > I believe it should be worth testing as I know that the cia PCI bus driver > is ok. In fact, most of the PCI bus drivers are fixed, so it should work > on most alphas, AFAIK. OK. I took the Version from http://people.FreeBSD.org/~jhb/patches/alpha.ithreads.patch With boot -v it hangs here: [...] Creating DISK cd0 Creating DISK da0 Creating DISK da1 A none -v boot get's sometimes up to printing some or all SCSI-devices but never gets to mounting /. I'm using seriel console. The kernel is compiled with INVARIANTS. A working boot looks like this: Copyright (c) 1992-2000 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 5.0-CURRENT #0: Thu Sep 14 08:19:11 CEST 2000 ticso@cicely9.cicely.de:/var/d7/src-2000-09-13/src/sys/compile/CICELY9 EB164 Digital AlphaPC 164 500 MHz, 500MHz 8192 byte page size, 1 processor. CPU: EV56 (21164A) major=7 minor=2 extensions=0x1 OSF PAL rev: 0x1000800020117 real memory = 265904128 (259672K bytes) avail memory = 253460480 (247520K bytes) Preloaded elf kernel "kernel.ko" at 0xfffffc000061e000. cia0: <2117x Core Logic chipset> cia0: ALCOR/ALCOR2, pass 3 cia0: extended capabilities: 21 pcib0: <2117x PCI host bus adapter> on cia0 pci0: on pcib0 sym0: <895> port 0x10200-0x102ff mem 0x82030000-0x82030fff,0x82031200-0x820312ff irq 2 at device 5.0 on pci0 sym0: Symbios NVRAM, ID 7, Fast-40, LVD, parity checking sym0: open drain IRQ line driver, using on-chip SRAM sym0: using LOAD/STORE-based firmware. sym0: interrupting at CIA irq 2 sym1: <810a> port 0x10100-0x101ff mem 0x82031100-0x820311ff irq 0 at device 6.0 on pci0 sym1: No NVRAM, ID 7, Fast-10, SE, parity checking sym1: interrupting at CIA irq 0 sym2: <810a> port 0x10000-0x100ff mem 0x82031000-0x820310ff irq 1 at device 7.0 on pci0 sym2: No NVRAM, ID 7, Fast-10, SE, parity checking sym2: interrupting at CIA irq 1 isab0: at device 8.0 on pci0 isa0: on isab0 xl0: <3Com 3c905B-TX Fast Etherlink XL> port 0x10300-0x1037f mem 0x82031300-0x8203137f irq 3 at device 9.0 on pci0 xl0: interrupting at CIA irq 3 xl0: Ethernet address: 00:10:5a:30:1c:1a miibus0: on xl0 xlphy0: <3Com internal media interface> on miibus0 xlphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto atapci0: port 0x10380-0x1038f irq 5 at device 11.0 on pci0 ata0: at 0x1f0 irq 14 on atapci0 fdc0: at port 0x3f0-0x3f5,0x3f7 irq 6 drq 2 on isa0 fdc0: interrupting at ISA irq 6 fdc0: FIFO enabled, 8 bytes threshold fd0: <1440-KB 3.5" drive> on fdc0 drive 0 mcclock0: at port 0x70-0x71 on isa0 sio0 at port 0x3f8-0x3ff irq 4 on isa0 sio0: type 16550A, console sio0: interrupting at ISA irq 4 sio1: reserved for low-level i/o Timecounter "i8254" frequency 1193182 Hz Timecounter "alpha" frequency 500006831 Hz ad0: 1219MB [2477/16/63] at ata0-master using WDMA2 Waiting 15 seconds for SCSI devices to settle (noperiph:sym0:0:-1:-1): SCSI BUS reset delivered. da0 at sym0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: 10.000MB/s transfers (10.000MHz, offset 15), Tagged Queueing Enabled da0: 1041MB (2131992 512 byte sectors: 255H 63S/T 132C) cd0 at sym2 bus 0 target 4 lun 0 cd0: Removable CD-ROM SCSI-2 device cd0: 4.237MB/s transfers (4.237MHz, offset 8) cd0: cd present [314602 x 2048 byte records] da1 at sym0 bus 0 target 1 lun 0 da1: Fixed Direct Access SCSI-CCS device da1: 3.300MB/s transfers da1: 638MB (1308087 512 byte sectors: 64H 32S/T 638C) da6 at sym1 bus 0 target 0 lun 0 da6: Fixed Direct Access SCSI-2 device da6: 5.000MB/s transfers (5.000MHz, offset 7) da6: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da2 at sym0 bus 0 target 2 lun 0 da2: Fixed Direct Access SCSI-2 device da2: 5.000MB/s transfers (5.000MHz, offset 7) da2: 103MB (210944 512 byte sectors: 64H 32S/T 103C) da13 at sym2 bus 0 target 2 lun 0 da13: Fixed Direct Access SCSI-CCS device da13: 3.300MB/s transfers da13: 316MB (649040 512 byte sectors: 64H 32S/T 316C) da7 at sym1 bus 0 target 1 lun 0 da7: Fixed Direct Access SCSI-2 device da7: 5.000MB/s transfers (5.000MHz, offset 7) da7: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da15 at sym2 bus 0 target 5 lun 0 da15: Removable Direct Access SCSI-2 device da15: 10.000MB/s transfers (10.000MHz, offset 8) da15: 1243MB (1273011 1024 byte sectors: 255H 63S/T 79C) da3 at sym0 bus 0 target 4 lun 0 da3: Fixed Direct Access SCSI-CCS device da3: 3.300MB/s transfers da3: 405MB (830340 512 byte sectors: 64H 32S/T 405C) da4 at sym0 bus 0 target 9 lun 0 da4: Fixed Direct Access SCSI-2 device da4: 80.000MB/s transfers (40.000MHz, offset 15, 16bit), Tagged Queueing Enabled da4: 17366MB (35566480 512 byte sectors: 255H 63S/T 2213C) da5 at sym0 bus 0 target 10 lun 0 da5: Fixed Direct Access SCSI-2 device da5: 80.000MB/s transfers (40.000MHz, offset 15, 16bit), Tagged Queueing Enabled da5: 17366MB (35566480 512 byte sectors: 255H 63S/T 2213C) da8 at sym1 bus 0 target 2 lun 0 da8: Fixed Direct Access SCSI-2 device da8: 5.000MB/s transfers (5.000MHz, offset 7) da8: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da9 at sym1 bus 0 target 3 lun 0 da9: Fixed Direct Access SCSI-2 device da9: 5.000MB/s transfers (5.000MHz, offset 7) da9: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da10 at sym1 bus 0 target 4 lun 0 da10: Fixed Direct Access SCSI-2 device da10: 5.000MB/s transfers (5.000MHz, offset 7) da10: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da11 at sym1 bus 0 target 5 lun 0 da11: Fixed Direct Access SCSI-2 device da11: 5.000MB/s transfers (5.000MHz, offset 7) da11: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da12 at sym1 bus 0 target 6 lun 0 da12: Fixed Direct Access SCSI-2 device da12: 5.000MB/s transfers (5.000MHz, offset 7) da12: 206MB (422912 512 byte sectors: 64H 32S/T 206C) da14 at sym2 bus 0 target 3 lun 0 da14: Fixed Direct Access SCSI-CCS device da14: 3.300MB/s transfers da14: 170MB (349770 512 byte sectors: 64H 32S/T 170C) Mounting root from ufs:/dev/da0a [...] -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 15:26:34 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id CE9F237B42C; Fri, 15 Sep 2000 15:26:29 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8FMQNc12097 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sat, 16 Sep 2000 00:26:25 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8FMQPI82345; Sat, 16 Sep 2000 00:26:29 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8FMQPl61567; Sat, 16 Sep 2000 00:26:25 +0200 (CEST) (envelope-from ticso) Date: Sat, 16 Sep 2000 00:26:24 +0200 From: Bernd Walter To: John Baldwin Cc: "David O'Brien" , alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000916002624.B61424@cicely5.cicely.de> References: <20000915193236.B61065@cicely5.cicely.de> <200009151803.LAA24561@pike.osd.bsdi.com> <20000915233621.A61424@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000915233621.A61424@cicely5.cicely.de>; from ticso@cicely5.cicely.de on Fri, Sep 15, 2000 at 11:36:22PM +0200 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 11:36:22PM +0200, Bernd Walter wrote: > On Fri, Sep 15, 2000 at 11:03:31AM -0700, John Baldwin wrote: > > I believe it should be worth testing as I know that the cia PCI bus driver > > is ok. In fact, most of the PCI bus drivers are fixed, so it should work > > on most alphas, AFAIK. > > OK. I took the Version from > http://people.FreeBSD.org/~jhb/patches/alpha.ithreads.patch > > With boot -v it hangs here: > [...] > Creating DISK cd0 > Creating DISK da0 > Creating DISK da1 Sorry - I forgot to cvs update so I had an older trap.c. I will retry with the propper source. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 16:59:43 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 0A85837B423 for ; Fri, 15 Sep 2000 16:59:41 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id QAA30724; Fri, 15 Sep 2000 16:23:28 -0700 Date: Fri, 15 Sep 2000 16:23:04 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Jason Evans Cc: alpha@freebsd.org Subject: Re: PRE_SMPNG question.... In-Reply-To: <20000915115239.E7247@blitz.canonware.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org If it's ok with the CVS meisters, I'd like the tag shifted one for sys/alpha/pci/cia.c then- the tree wasn't quite stable at the point it was tagged. There should be other required changes. On 15 Sep 2000, Jason Evans wrote: > On Thu, Sep 14, 2000 at 07:17:54PM -0700, Matthew Jacob wrote: > > > > If I want to make sure that PC164's boot and run in the source tree > > that was tagged PRE_SMPNG, is it acceptable to MFC, as it were, the fix? > > > > I've had to, hah hah (coz I wasn't as careful as I should) spend most of the > > day reinstalling 4.1 so I could get back to having some stable disks I can > > actually boot from prior to moving ahead back into -current for alpha. It > > strikes me that having the PRE_SMPNG stuff be stable would be good. > > > > But I'll admit that I'm a bit hazy as to how branches work vs. tags. > > PRE_SMPNG is a static tag. That means that the only way to modify the > PRE_SMPNG version of a file is to move the tag to a different version of a > file. So, unless the file hasn't changed since the PRE_SMPNG tag was > created, moving the tag will prove very difficult. Indeed, the purpose of > the PRE_SMPNG tag was not to provide a branch point; it was merely meant to > mark the last moment before the SMP code destabilized the tree. > > Jason > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 17:36:35 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id 79F5437B422 for ; Fri, 15 Sep 2000 17:36:33 -0700 (PDT) Received: (qmail 65185 invoked by uid 1142); 16 Sep 2000 00:36:32 -0000 Date: 15 Sep 2000 17:36:32 -0700 Date: Fri, 15 Sep 2000 17:36:25 -0700 From: Jason Evans To: Matthew Jacob Cc: alpha@freebsd.org Subject: Re: PRE_SMPNG question.... Message-ID: <20000915173625.G7247@blitz.canonware.com> References: <20000915115239.E7247@blitz.canonware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from mjacob@feral.com on Fri, Sep 15, 2000 at 04:23:04PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 04:23:04PM -0700, Matthew Jacob wrote: > > If it's ok with the CVS meisters, I'd like the tag shifted one for > sys/alpha/pci/cia.c then- the tree wasn't quite stable at the point it was > tagged. There should be other required changes. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ I don't understand that sentence. Can you clarify? Thanks, Jason To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 18:20:58 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 7EF6937B43E; Fri, 15 Sep 2000 18:20:57 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id SAA45920; Fri, 15 Sep 2000 18:20:54 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009160120.SAA45920@pike.osd.bsdi.com> Subject: Loader badly broken on the alpha To: alpha@FreeBSD.org, dcs@FreeBSD.org Date: Fri, 15 Sep 2000 18:20:53 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Well, after a rather long and frustrating afternoon, it seems that revision of the loader after Fri Sep 8 16:30 2000 UTC is broken for the alpha arch. I've tried to narrow things down to see if it was just the 4th changes or the pnp changes, but have been unable to get anything to boot except for a binary and 4th files from that date. I've even tried removing all of the 4th stuff and /boot/loader.rc and the newer loader binaries still break. Every time they break, they trigger a kernel stack not valid fault. My guess is that ficl is recursing in an infinite loop somewhere, that it is recursing too deeply and using up all available memory, or that something is failing to handle error conditions and is dereferencing a NULL pointer. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 18:50:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 4985037B42C for ; Fri, 15 Sep 2000 18:50:15 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id SAA30982; Fri, 15 Sep 2000 18:50:13 -0700 Date: Fri, 15 Sep 2000 18:49:49 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Jason Evans Cc: alpha@freebsd.org Subject: Re: PRE_SMPNG question.... In-Reply-To: <20000915173625.G7247@blitz.canonware.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On 15 Sep 2000, Jason Evans wrote: > On Fri, Sep 15, 2000 at 04:23:04PM -0700, Matthew Jacob wrote: > > > > If it's ok with the CVS meisters, I'd like the tag shifted one for > > sys/alpha/pci/cia.c then- the tree wasn't quite stable at the point it was > > tagged. There should be other required changes. > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > I don't understand that sentence. Can you clarify? Fatigue. There should be "no" other changes required. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 18:51:33 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 63F8137B423; Fri, 15 Sep 2000 18:51:31 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id SAA30991; Fri, 15 Sep 2000 18:51:30 -0700 Date: Fri, 15 Sep 2000 18:51:06 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: John Baldwin Cc: alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha In-Reply-To: <200009160120.SAA45920@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org And you just noticed? Hello? What the hell do you think I've been saying... On Fri, 15 Sep 2000, John Baldwin wrote: > Well, after a rather long and frustrating afternoon, it seems that > revision of the loader after Fri Sep 8 16:30 2000 UTC is broken for > the alpha arch. I've tried to narrow things down to see if it was just > the 4th changes or the pnp changes, but have been unable to get anything > to boot except for a binary and 4th files from that date. I've even > tried removing all of the 4th stuff and /boot/loader.rc and the newer > loader binaries still break. Every time they break, they trigger a > kernel stack not valid fault. My guess is that ficl is recursing in an > infinite loop somewhere, that it is recursing too deeply and using up > all available memory, or that something is failing to handle error conditions > and is dereferencing a NULL pointer. > > -- > > John Baldwin -- http://www.FreeBSD.org/~jhb/ > "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 19: 9:35 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from peach.ocn.ne.jp (peach.ocn.ne.jp [210.145.254.87]) by hub.freebsd.org (Postfix) with ESMTP id B85B237B422; Fri, 15 Sep 2000 19:09:31 -0700 (PDT) Received: from newsguy.com (p59-dn01kiryunisiki.gunma.ocn.ne.jp [211.0.245.60]) by peach.ocn.ne.jp (8.9.1a/OCN/) with ESMTP id LAA08512; Sat, 16 Sep 2000 11:09:20 +0900 (JST) Message-ID: <39C2D621.A7BDD4DE@newsguy.com> Date: Sat, 16 Sep 2000 11:08:33 +0900 From: "Daniel C. Sobral" X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en,pt-BR MIME-Version: 1.0 To: mjacob@feral.com Cc: John Baldwin , alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > And you just noticed? Hello? What the hell do you think I've been saying... Well, this is the first *I* heard of it! Let me get this straight... it's broken even after the pnp fixes went in? Alas, the pnp fixes should have screwed linking, not booting... The 8th, you say? Anyone was able to pinpoint the revision? -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org capo@the.secret.bsdconspiracy.net "I demand that my picture show a handsome face, even if it doesn't look like me." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 19:20:15 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id DC30C37B422; Fri, 15 Sep 2000 19:20:13 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id TAA42547; Fri, 15 Sep 2000 19:20:09 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009160220.TAA42547@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: from Matthew Jacob at "Sep 15, 2000 06:51:06 pm" To: mjacob@feral.com Date: Fri, 15 Sep 2000 19:20:09 -0700 (PDT) Cc: alpha@FreeBSD.ORG, dcs@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > And you just noticed? Hello? What the hell do you think I've been saying... I heard you, which is why I wasn't surprised when it died on me, but I hadn't seen it myself yet. Also, I tried to pinpoint what revision broke it but failed in my attempts to do so and wanted to try to communicate that to hopefully make it easier to pinpoint when it did actually break. If that makes any sense. > On Fri, 15 Sep 2000, John Baldwin wrote: > > > Well, after a rather long and frustrating afternoon, it seems that > > revision of the loader after Fri Sep 8 16:30 2000 UTC is broken for > > the alpha arch. I've tried to narrow things down to see if it was just > > the 4th changes or the pnp changes, but have been unable to get anything > > to boot except for a binary and 4th files from that date. I've even > > tried removing all of the 4th stuff and /boot/loader.rc and the newer > > loader binaries still break. Every time they break, they trigger a > > kernel stack not valid fault. My guess is that ficl is recursing in an > > infinite loop somewhere, that it is recursing too deeply and using up > > all available memory, or that something is failing to handle error conditions > > and is dereferencing a NULL pointer. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 19:21:10 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 32EBA37B43F; Fri, 15 Sep 2000 19:21:09 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id TAA42952; Fri, 15 Sep 2000 19:20:34 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009160220.TAA42952@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: <39C2D621.A7BDD4DE@newsguy.com> from "Daniel C. Sobral" at "Sep 16, 2000 11:08:33 am" To: "Daniel C. Sobral" Date: Fri, 15 Sep 2000 19:20:34 -0700 (PDT) Cc: mjacob@feral.com, alpha@FreeBSD.ORG, dcs@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Daniel C. Sobral wrote: > Matthew Jacob wrote: > > > > And you just noticed? Hello? What the hell do you think I've been saying... > > Well, this is the first *I* heard of it! > > Let me get this straight... it's broken even after the pnp fixes went > in? Alas, the pnp fixes should have screwed linking, not booting... Yes. > The 8th, you say? Anyone was able to pinpoint the revision? Not so far, no. :( -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 19:27:37 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from peach.ocn.ne.jp (peach.ocn.ne.jp [210.145.254.87]) by hub.freebsd.org (Postfix) with ESMTP id 3791F37B43C; Fri, 15 Sep 2000 19:27:35 -0700 (PDT) Received: from newsguy.com (p59-dn01kiryunisiki.gunma.ocn.ne.jp [211.0.245.60]) by peach.ocn.ne.jp (8.9.1a/OCN/) with ESMTP id LAA11201; Sat, 16 Sep 2000 11:27:27 +0900 (JST) Message-ID: <39C2DA60.93005FFC@newsguy.com> Date: Sat, 16 Sep 2000 11:26:40 +0900 From: "Daniel C. Sobral" X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en,pt-BR MIME-Version: 1.0 To: John Baldwin Cc: mjacob@feral.com, alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha References: <200009160220.TAA42952@pike.osd.bsdi.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Ok, first thing I need to know is whether the problem is with .4th files or .c files. The changes to both are independent of each other, so: 1) Try the last working loader with the most recent forth/* files. 2) Try the most recent loader with the last working forth/* files. Hopefully, only one is broken. -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org capo@the.secret.bsdconspiracy.net "I demand that my picture show a handsome face, even if it doesn't look like me." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 19:38:19 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from peach.ocn.ne.jp (peach.ocn.ne.jp [210.145.254.87]) by hub.freebsd.org (Postfix) with ESMTP id 1016237B43E for ; Fri, 15 Sep 2000 19:38:17 -0700 (PDT) Received: from newsguy.com (p59-dn01kiryunisiki.gunma.ocn.ne.jp [211.0.245.60]) by peach.ocn.ne.jp (8.9.1a/OCN/) with ESMTP id LAA12739; Sat, 16 Sep 2000 11:37:45 +0900 (JST) Message-ID: <39C2DCCA.D566C1D2@newsguy.com> Date: Sat, 16 Sep 2000 11:36:58 +0900 From: "Daniel C. Sobral" X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en,pt-BR MIME-Version: 1.0 To: John Baldwin Cc: mjacob@feral.com, alpha@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha References: <200009160220.TAA42952@pike.osd.bsdi.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org John Baldwin wrote: > > > The 8th, you say? Anyone was able to pinpoint the revision? > > Not so far, no. :( Funny, it should be easy. The changes around that date were pretty much ortoghonal. 1) ficl/*, common/bootstrap.h and pnp.c. 2) common/boot.c and module.c 3) forth/* 4) alpha/loader/version (now, *THAT* would be funny) Unless I'm missing something, it must be *one* of these. If the problem are with the .4th files, I'm very screwed. Otherwise, it should be easy to find. -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org capo@the.secret.bsdconspiracy.net "I demand that my picture show a handsome face, even if it doesn't look like me." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 20: 2:45 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id DE48437B43E; Fri, 15 Sep 2000 20:02:42 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id UAA31225; Fri, 15 Sep 2000 20:02:39 -0700 Date: Fri, 15 Sep 2000 20:02:15 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: "Daniel C. Sobral" Cc: John Baldwin , alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha In-Reply-To: <39C2D621.A7BDD4DE@newsguy.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 16 Sep 2000, Daniel C. Sobral wrote: > Matthew Jacob wrote: > > > > And you just noticed? Hello? What the hell do you think I've been saying... > Sorry- I shouldn't send mail with a 100 degree fever... > Well, this is the first *I* heard of it! I think I said "bad kernel stack??- not kernel- it's the loader changes"" This bit me twice, as an installworld nuked me bad enough (oops, my mistake) so that I had to reinstall from 4.1 . > > Let me get this straight... it's broken even after the pnp fixes went > in? Alas, the pnp fixes should have screwed linking, not booting... > > The 8th, you say? Anyone was able to pinpoint the revision? John might be able to. I'm ill and can't get back to the office probably until Monday. I was waiting for the drive by snipe mail to stop and the final resting place for to be settled on before trying to fool around with this whole mess again. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 20: 3:25 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 50A3237B43C; Fri, 15 Sep 2000 20:03:24 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id UAA31235; Fri, 15 Sep 2000 20:03:23 -0700 Date: Fri, 15 Sep 2000 20:02:59 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: John Baldwin Cc: alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha In-Reply-To: <200009160220.TAA42547@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 15 Sep 2000, John Baldwin wrote: > Matthew Jacob wrote: > > > > And you just noticed? Hello? What the hell do you think I've been saying... > > I heard you, which is why I wasn't surprised when it died on me, but > I hadn't seen it myself yet. Also, I tried to pinpoint what revision > broke it but failed in my attempts to do so and wanted to try to communicate > that to hopefully make it easier to pinpoint when it did actually break. > If that makes any sense. Yes. Like I said, I shouldn't have sent the above mail. Sorry. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Fri Sep 15 20: 5: 6 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 027C837B43C; Fri, 15 Sep 2000 20:05:05 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id UAA31253; Fri, 15 Sep 2000 20:05:02 -0700 Date: Fri, 15 Sep 2000 20:04:38 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: "Daniel C. Sobral" Cc: John Baldwin , alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha In-Reply-To: <39C2DA60.93005FFC@newsguy.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I think I had determined (I thought) that it was just loader itself, but I could be wrong. On Sat, 16 Sep 2000, Daniel C. Sobral wrote: > Ok, first thing I need to know is whether the problem is with .4th files > or .c files. The changes to both are independent of each other, so: > > 1) Try the last working loader with the most recent forth/* files. > 2) Try the most recent loader with the last working forth/* files. > > Hopefully, only one is broken. > > -- > Daniel C. Sobral (8-DCS) > dcs@newsguy.com > dcs@freebsd.org > capo@the.secret.bsdconspiracy.net > > "I demand that my picture show a handsome face, even if it doesn't look > like me." > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 0:24: 7 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 7483837B422; Sat, 16 Sep 2000 00:24:04 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8G7Nsp05632 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sat, 16 Sep 2000 09:23:56 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8G7NxI86180; Sat, 16 Sep 2000 09:24:00 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8G7NsR65787; Sat, 16 Sep 2000 09:23:54 +0200 (CEST) (envelope-from ticso) Date: Sat, 16 Sep 2000 09:23:54 +0200 From: Bernd Walter To: Matthew Jacob Cc: John Baldwin , alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha Message-ID: <20000916092354.A65727@cicely5.cicely.de> References: <200009160220.TAA42547@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from mjacob@feral.com on Fri, Sep 15, 2000 at 08:02:59PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 15, 2000 at 08:02:59PM -0700, Matthew Jacob wrote: > > On Fri, 15 Sep 2000, John Baldwin wrote: > > > Matthew Jacob wrote: > > > > > > And you just noticed? Hello? What the hell do you think I've been saying... > > > > I heard you, which is why I wasn't surprised when it died on me, but > > I hadn't seen it myself yet. Also, I tried to pinpoint what revision > > broke it but failed in my attempts to do so and wanted to try to communicate > > that to hopefully make it easier to pinpoint when it did actually break. > > If that makes any sense. > > Yes. Like I said, I shouldn't have sent the above mail. Sorry. I had a world from the 8th septempber running. After updating to the 13th it broke for me. Using the older loader did not help. Removing all 4th files and references work - beside that I had to set device.hints in loader.rc. With the kernel source from yesterday (World still from 13th) the kernel now gets installed as /boot/kernel/kernel instead of kernel.ko and the loader paniced similar because it was not able to load the kernel! I thought that we might have a similar unavailable file reference now with the 4th files. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 0:26:33 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from netplex.com.au (adsl-63-207-30-186.dsl.snfc21.pacbell.net [63.207.30.186]) by hub.freebsd.org (Postfix) with ESMTP id 6FF2E37B424; Sat, 16 Sep 2000 00:26:30 -0700 (PDT) Received: from netplex.com.au (peter@localhost [127.0.0.1]) by netplex.com.au (8.11.0/8.9.3) with ESMTP id e8G7QPG10931; Sat, 16 Sep 2000 00:26:25 -0700 (PDT) (envelope-from peter@netplex.com.au) Message-Id: <200009160726.e8G7QPG10931@netplex.com.au> X-Mailer: exmh version 2.1.1 10/15/1999 To: mjacob@feral.com Cc: "Daniel C. Sobral" , John Baldwin , alpha@FreeBSD.ORG, dcs@FreeBSD.ORG Subject: Re: Loader badly broken on the alpha In-Reply-To: Date: Sat, 16 Sep 2000 00:26:25 -0700 From: Peter Wemm Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matthew Jacob wrote: > > I think I had determined (I thought) that it was just loader itself, but I > could be wrong. The problem is that the loader is now too large, it cannot fit inside the ~208K window that boot1 has got available to it. peter@ashburton[12:14am]/home/obj/home/src/sys/boot/alpha/loader-130# nm loader.sym | sort | tail -1 000000002003e190 A end The VM space ends at 2004000. I can't find where, but I was pretty sure that the last 8K page was already spoken for and we've clobbered it. Was the pnp stuff added to the Alpha loader images? That is what probably tipped it over the edge. > On Sat, 16 Sep 2000, Daniel C. Sobral wrote: > > > Ok, first thing I need to know is whether the problem is with .4th files > > or .c files. The changes to both are independent of each other, so: > > > > 1) Try the last working loader with the most recent forth/* files. > > 2) Try the most recent loader with the last working forth/* files. > > > > Hopefully, only one is broken. > > > > -- > > Daniel C. Sobral (8-DCS) > > dcs@newsguy.com > > dcs@freebsd.org > > capo@the.secret.bsdconspiracy.net > > > > "I demand that my picture show a handsome face, even if it doesn't look > > like me." > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-alpha" in the body of the message > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 2:26:57 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from post.mail.nl.demon.net (post-11.mail.nl.demon.net [194.159.73.21]) by hub.freebsd.org (Postfix) with ESMTP id 1EB4437B424; Sat, 16 Sep 2000 02:26:54 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #4) id 13aEFH-0007RS-00; Sat, 16 Sep 2000 09:26:51 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e8G9Rx101391; Sat, 16 Sep 2000 11:27:59 +0200 (CEST) (envelope-from wkb) Date: Sat, 16 Sep 2000 11:27:59 +0200 From: Wilko Bulte To: Bernd Walter Cc: Matthew Jacob , John Baldwin , alpha@freebsd.org, dcs@freebsd.org Subject: Re: Loader badly broken on the alpha Message-ID: <20000916112759.B427@freebie.demon.nl> References: <200009160220.TAA42547@pike.osd.bsdi.com> <20000916092354.A65727@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20000916092354.A65727@cicely5.cicely.de>; from ticso@cicely5.cicely.de on Sat, Sep 16, 2000 at 09:23:54AM +0200 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 16, 2000 at 09:23:54AM +0200, Bernd Walter wrote: > On Fri, Sep 15, 2000 at 08:02:59PM -0700, Matthew Jacob wrote: > > > > On Fri, 15 Sep 2000, John Baldwin wrote: > > > I heard you, which is why I wasn't surprised when it died on me, but > > > I hadn't seen it myself yet. Also, I tried to pinpoint what revision > > > broke it but failed in my attempts to do so and wanted to try to communicate > > > that to hopefully make it easier to pinpoint when it did actually break. > > > If that makes any sense. > > > > Yes. Like I said, I shouldn't have sent the above mail. Sorry. > > I had a world from the 8th septempber running. > After updating to the 13th it broke for me. > Using the older loader did not help. > Removing all 4th files and references work - beside that I had to > set device.hints in loader.rc. > With the kernel source from yesterday (World still from 13th) the kernel > now gets installed as /boot/kernel/kernel instead of kernel.ko and the > loader paniced similar because it was not able to load the kernel! > I thought that we might have a similar unavailable file reference now with > the 4th files. Great :-( I nuked my -current box yesterday with the same loader problems. I promised billf to test the tx driver but now I'm screwed it seems. Any good suggestions to fix this loader thing short of reinstalling 4.1 first? -- Wilko Bulte wilko@freebsd.org Arnhem, the Netherlands To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 5:35: 7 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-32.mail.demon.net (anchor-post-32.mail.demon.net [194.217.242.90]) by hub.freebsd.org (Postfix) with ESMTP id EEF4737B424; Sat, 16 Sep 2000 05:35:02 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-32.mail.demon.net with esmtp (Exim 2.12 #1) id 13aHBM-0007hE-0W; Sat, 16 Sep 2000 13:35:01 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id NAA27694; Sat, 16 Sep 2000 13:12:04 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Sat, 16 Sep 2000 13:10:18 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: alpha@FreeBSD.org, smp@FreeBSD.org Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: <200009142254.PAA95214@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 14 Sep 2000, John Baldwin wrote: > Doug Rabson wrote: > > On Wed, 13 Sep 2000, John Baldwin wrote: > > > I've updated the alpha.ithreads.patch on my freefall webpage to work with > > > the latest current, btw. > > > > I've been working on this today and I have a kernel which boots on my > > miata and is happily chewing its way through a buildworld. I fixed the > > device_get_nameunit() problem - it was just going wrong for ATA interrupts > > which are handled quite strangely on alpha. I also changed the p_comm > > string to include the whole vector number in hex since on some platforms > > all bits are relavent, not just (vec-0x900)>>4. > > > > I've attached my modified version of your patch to this message. Its not > > quite ready to commit yet since I didn't write the enable/disable hooks > > for mcpcia or dwlpx. I have an mcpcia here (the AS4100) so I'll be able to > > test that soon. > > After that last fix to trap.c to add an acquire/release of sched_lock, I > am now happily running interrupt threads on my miata here with your patch. > I've gone ahead and replaced my patch on freefall with yours, but hopefully > we can commit this very soon when you get the other PCI chipsets finished. > Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it > is running fine so far. Good. I'll try to deal with mcpcia and dwlpx today or tomorrow and we can get this thing committed. After that, I can get back to trying to get a secondary CPU into the scheduler without hard-locking the machine :-). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 5:35:10 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from anchor-post-32.mail.demon.net (anchor-post-32.mail.demon.net [194.217.242.90]) by hub.freebsd.org (Postfix) with ESMTP id C330A37B422; Sat, 16 Sep 2000 05:35:04 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-32.mail.demon.net with esmtp (Exim 2.12 #1) id 13aHBP-0007hE-0W; Sat, 16 Sep 2000 13:35:03 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id NAA27717; Sat, 16 Sep 2000 13:27:21 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Sat, 16 Sep 2000 13:25:36 +0100 (BST) From: Doug Rabson To: Bernd Walter Cc: John Baldwin , alpha@FreeBSD.org, smp@FreeBSD.org Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: <20000915075812.B60348@cicely5.cicely.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 15 Sep 2000, Bernd Walter wrote: > On Thu, Sep 14, 2000 at 03:54:58PM -0700, John Baldwin wrote: > > After that last fix to trap.c to add an acquire/release of sched_lock, I > > am now happily running interrupt threads on my miata here with your patch. > > I've gone ahead and replaced my patch on freefall with yours, but hopefully > > we can commit this very soon when you get the other PCI chipsets finished. > > Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it > > is running fine so far. > > miata means that it should be testable on a PC164 - right? The patch should work on all except AS4100 and AS8200. I would like to get some testing on tsunami, apecs and lca based machines for a sanity check but it ought to work (crossed fingers). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 9:24:59 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 9C17837B422 for ; Sat, 16 Sep 2000 09:24:57 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id JAA19314; Sat, 16 Sep 2000 09:24:40 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009161624.JAA19314@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: <20000916112759.B427@freebie.demon.nl> from Wilko Bulte at "Sep 16, 2000 11:27:59 am" To: Wilko Bulte Date: Sat, 16 Sep 2000 09:24:40 -0700 (PDT) Cc: Bernd Walter , Matthew Jacob , alpha@FreeBSD.ORG, dcs@FreeBSD.ORGG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Wilko Bulte wrote: > On Sat, Sep 16, 2000 at 09:23:54AM +0200, Bernd Walter wrote: > > On Fri, Sep 15, 2000 at 08:02:59PM -0700, Matthew Jacob wrote: > > > > > > On Fri, 15 Sep 2000, John Baldwin wrote: > > > > > I heard you, which is why I wasn't surprised when it died on me, but > > > > I hadn't seen it myself yet. Also, I tried to pinpoint what revision > > > > broke it but failed in my attempts to do so and wanted to try to communicate > > > > that to hopefully make it easier to pinpoint when it did actually break. > > > > If that makes any sense. > > > > > > Yes. Like I said, I shouldn't have sent the above mail. Sorry. > > > > I had a world from the 8th septempber running. > > After updating to the 13th it broke for me. > > Using the older loader did not help. > > Removing all 4th files and references work - beside that I had to > > set device.hints in loader.rc. > > With the kernel source from yesterday (World still from 13th) the kernel > > now gets installed as /boot/kernel/kernel instead of kernel.ko and the > > loader paniced similar because it was not able to load the kernel! > > I thought that we might have a similar unavailable file reference now with > > the 4th files. > > Great :-( I nuked my -current box yesterday with the same loader problems. > I promised billf to test the tx driver but now I'm screwed it seems. > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > first? Boot teh 4.0 CD's. Use the live filesystem CD fixit option. Then do this: # mount /dist/dev/da0a /mnt # cp /dist/boot/loader /mnt/boot/loader # cd /mnt/boot # /dist/bin/sed -e 's/^/set /' device.hints > hints Reboot. The 4.0 loader doesn't have 4th support, so it will ignore all the 4th files. Break into a loader prompt, type 'include hints' to set all your hints. Then 'boot /boot/mykernelname/kernel'. > -- > Wilko Bulte wilko@freebsd.org -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 9:27:16 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id BD70337B422 for ; Sat, 16 Sep 2000 09:27:14 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id JAA19366; Sat, 16 Sep 2000 09:26:04 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009161626.JAA19366@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: <200009160726.e8G7QPG10931@netplex.com.au> from Peter Wemm at "Sep 16, 2000 00:26:25 am" To: Peter Wemm Date: Sat, 16 Sep 2000 09:26:04 -0700 (PDT) Cc: mjacob@feral.com, "Daniel C. Sobral" , alpha@FreeBSD.ORG, dcs@FreeBSD.ORGG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Peter Wemm wrote: > Matthew Jacob wrote: > > > > I think I had determined (I thought) that it was just loader itself, but I > > could be wrong. > > The problem is that the loader is now too large, it cannot fit inside the > ~208K window that boot1 has got available to it. Isn't there a better solution where we can change the way in which we do memory setup in boot1 so that we can get rid of this limit? Or at least make it significantly larger? I thought you mentioned something along those lines at Usenix.. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 10:53:32 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from post.mail.nl.demon.net (post-10.mail.nl.demon.net [194.159.73.20]) by hub.freebsd.org (Postfix) with ESMTP id 1B3C837B423 for ; Sat, 16 Sep 2000 10:53:30 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #2) id 13aM9T-0001nQ-00; Sat, 16 Sep 2000 17:53:23 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e8GHpCt41803; Sat, 16 Sep 2000 19:51:12 +0200 (CEST) (envelope-from wkb) Date: Sat, 16 Sep 2000 19:51:12 +0200 From: Wilko Bulte To: John Baldwin Cc: Bernd Walter , Matthew Jacob , alpha@freebsd.org, dcs@FreeBSD.ORGG Subject: Re: Loader badly broken on the alpha Message-ID: <20000916195112.A41783@freebie.demon.nl> References: <20000916112759.B427@freebie.demon.nl> <200009161624.JAA19314@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <200009161624.JAA19314@pike.osd.bsdi.com>; from jhb@pike.osd.bsdi.com on Sat, Sep 16, 2000 at 09:24:40AM -0700 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 16, 2000 at 09:24:40AM -0700, John Baldwin wrote: > Wilko Bulte wrote: > > On Sat, Sep 16, 2000 at 09:23:54AM +0200, Bernd Walter wrote: > > > On Fri, Sep 15, 2000 at 08:02:59PM -0700, Matthew Jacob wrote: > > > > > > > > On Fri, 15 Sep 2000, John Baldwin wrote: ... > > > With the kernel source from yesterday (World still from 13th) the kernel > > > now gets installed as /boot/kernel/kernel instead of kernel.ko and the > > > loader paniced similar because it was not able to load the kernel! > > > I thought that we might have a similar unavailable file reference now with > > > the 4th files. > > > > Great :-( I nuked my -current box yesterday with the same loader problems. > > I promised billf to test the tx driver but now I'm screwed it seems. > > > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > > first? > > Boot teh 4.0 CD's. Use the live filesystem CD fixit option. Then do this: > > # mount /dist/dev/da0a /mnt > # cp /dist/boot/loader /mnt/boot/loader > # cd /mnt/boot > # /dist/bin/sed -e 's/^/set /' device.hints > hints > > Reboot. The 4.0 loader doesn't have 4th support, so it will ignore all > the 4th files. Break into a loader prompt, type 'include hints' to set > all your hints. Then 'boot /boot/mykernelname/kernel'. Thanks John. Now the only thing that I lack is the live fs CD as I pulled the #1 cd as an .iso off the net :-) Time to get a full distribution set.. -- Wilko Bulte wilko@freebsd.org Arnhem, the Netherlands To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 11: 7: 7 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 4107B37B422; Sat, 16 Sep 2000 11:07:04 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8GI6tQ05918 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sat, 16 Sep 2000 20:06:57 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8GI72I90675; Sat, 16 Sep 2000 20:07:02 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8GI6w066498; Sat, 16 Sep 2000 20:06:58 +0200 (CEST) (envelope-from ticso) Date: Sat, 16 Sep 2000 20:06:58 +0200 From: Bernd Walter To: Wilko Bulte Cc: Matthew Jacob , John Baldwin , alpha@freebsd.org, dcs@freebsd.org Subject: Re: Loader badly broken on the alpha Message-ID: <20000916200658.B66455@cicely5.cicely.de> References: <200009160220.TAA42547@pike.osd.bsdi.com> <20000916092354.A65727@cicely5.cicely.de> <20000916112759.B427@freebie.demon.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000916112759.B427@freebie.demon.nl>; from wkb@freebie.demon.nl on Sat, Sep 16, 2000 at 11:27:59AM +0200 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 16, 2000 at 11:27:59AM +0200, Wilko Bulte wrote: > Great :-( I nuked my -current box yesterday with the same loader problems. > I promised billf to test the tx driver but now I'm screwed it seems. > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > first? I had the real luck to have an ancient loader from April on the disk. It did not understood some configfiles and stopped with an prompt. Then I had to do the set commands for the device hints. You should try offering a loader from a disk and then switching disks on the loader prompt. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 11:21:26 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 4AEB737B424; Sat, 16 Sep 2000 11:21:21 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8GILEQ06813 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sat, 16 Sep 2000 20:21:16 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8GILLI90692; Sat, 16 Sep 2000 20:21:21 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8GILHk66530; Sat, 16 Sep 2000 20:21:17 +0200 (CEST) (envelope-from ticso) Date: Sat, 16 Sep 2000 20:21:17 +0200 From: Bernd Walter To: Doug Rabson Cc: John Baldwin , alpha@FreeBSD.org, smp@FreeBSD.org Subject: Re: Prelimiary interrupt thread patches for alpha Message-ID: <20000916202117.C66455@cicely5.cicely.de> References: <20000915075812.B60348@cicely5.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from dfr@nlsystems.com on Sat, Sep 16, 2000 at 01:25:36PM +0100 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 16, 2000 at 01:25:36PM +0100, Doug Rabson wrote: > On Fri, 15 Sep 2000, Bernd Walter wrote: > > > On Thu, Sep 14, 2000 at 03:54:58PM -0700, John Baldwin wrote: > > > After that last fix to trap.c to add an acquire/release of sched_lock, I > > > am now happily running interrupt threads on my miata here with your patch. > > > I've gone ahead and replaced my patch on freefall with yours, but hopefully > > > we can commit this very soon when you get the other PCI chipsets finished. > > > Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it > > > is running fine so far. > > > > miata means that it should be testable on a PC164 - right? > > The patch should work on all except AS4100 and AS8200. I would like to get > some testing on tsunami, apecs and lca based machines for a sanity check > but it ought to work (crossed fingers). I have an AxpPCI running 4.1-RELEASE now which I want to update to current. But at this moment I'm more interested to get it running on my PC164 system. I updated the source yesterday and it still hangs while printing the SCSI devices when using the ithread patches :( -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 11:31:45 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id F298D37B422 for ; Sat, 16 Sep 2000 11:31:43 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id LAA21933; Sat, 16 Sep 2000 11:31:27 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009161831.LAA21933@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: <20000916200658.B66455@cicely5.cicely.de> from Bernd Walter at "Sep 16, 2000 08:06:58 pm" To: Bernd Walter Date: Sat, 16 Sep 2000 11:31:27 -0700 (PDT) Cc: Wilko Bulte , Matthew Jacob , alpha@freebsd.org, dcs@freebsd.orgg X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Sat, Sep 16, 2000 at 11:27:59AM +0200, Wilko Bulte wrote: > > Great :-( I nuked my -current box yesterday with the same loader problems. > > I promised billf to test the tx driver but now I'm screwed it seems. > > > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > > first? > > I had the real luck to have an ancient loader from April on the disk. > It did not understood some configfiles and stopped with an prompt. > Then I had to do the set commands for the device hints. > > You should try offering a loader from a disk and then switching disks > on the loader prompt. You can't do that with SRM. :( SRM only tells you about your boot device, unlike the x86 BIOS which lets you talk to all the devices it knows about. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 11:53:41 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 8567E37B424; Sat, 16 Sep 2000 11:53:35 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id LAA00866; Sat, 16 Sep 2000 11:53:28 -0700 Date: Sat, 16 Sep 2000 11:53:03 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: Doug Rabson Cc: Bernd Walter , John Baldwin , alpha@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Prelimiary interrupt thread patches for alpha In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I can look at Rawhide and TurboLaser next week when my temperature comes down (flu). Doug- you have a rawhide. I have the turbolaseers. Why don't y'all check the patch in so we have something same to work with? On Sat, 16 Sep 2000, Doug Rabson wrote: > On Fri, 15 Sep 2000, Bernd Walter wrote: > > > On Thu, Sep 14, 2000 at 03:54:58PM -0700, John Baldwin wrote: > > > After that last fix to trap.c to add an acquire/release of sched_lock, I > > > am now happily running interrupt threads on my miata here with your patch. > > > I've gone ahead and replaced my patch on freefall with yours, but hopefully > > > we can commit this very soon when you get the other PCI chipsets finished. > > > Thanks. :) I know have a buildworld going on my INVARIANTS kernel and it > > > is running fine so far. > > > > miata means that it should be testable on a PC164 - right? > > The patch should work on all except AS4100 and AS8200. I would like to get > some testing on tsunami, apecs and lca based machines for a sanity check > but it ought to work (crossed fingers). > > -- > Doug Rabson Mail: dfr@nlsystems.com > Nonlinear Systems Ltd. Phone: +44 20 8348 3944 > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-alpha" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 11:55:40 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id 5E66B37B422 for ; Sat, 16 Sep 2000 11:55:37 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8GItSQ08697 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sat, 16 Sep 2000 20:55:30 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8GItZI90890; Sat, 16 Sep 2000 20:55:35 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8GItVb66634; Sat, 16 Sep 2000 20:55:31 +0200 (CEST) (envelope-from ticso) Date: Sat, 16 Sep 2000 20:55:31 +0200 From: Bernd Walter To: John Baldwin Cc: Wilko Bulte , Matthew Jacob , alpha@freebsd.org, dcs@freebsd.orgg Subject: Re: Loader badly broken on the alpha Message-ID: <20000916205531.B66584@cicely5.cicely.de> References: <20000916200658.B66455@cicely5.cicely.de> <200009161831.LAA21933@pike.osd.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <200009161831.LAA21933@pike.osd.bsdi.com>; from jhb@pike.osd.bsdi.com on Sat, Sep 16, 2000 at 11:31:27AM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, Sep 16, 2000 at 11:31:27AM -0700, John Baldwin wrote: > Bernd Walter wrote: > > On Sat, Sep 16, 2000 at 11:27:59AM +0200, Wilko Bulte wrote: > > > Great :-( I nuked my -current box yesterday with the same loader problems. > > > I promised billf to test the tx driver but now I'm screwed it seems. > > > > > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > > > first? > > > > I had the real luck to have an ancient loader from April on the disk. > > It did not understood some configfiles and stopped with an prompt. > > Then I had to do the set commands for the device hints. > > > > You should try offering a loader from a disk and then switching disks > > on the loader prompt. > > You can't do that with SRM. :( SRM only tells you about your boot > device, unlike the x86 BIOS which lets you talk to all the devices it > knows about. Mmmh - Bad. Then the first place is the kernel root mount where we can change drives so the kernel need to be on the disk too. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 12:29:41 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 5A88A37B422 for ; Sat, 16 Sep 2000 12:29:39 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id MAA23231; Sat, 16 Sep 2000 12:29:22 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009161929.MAA23231@pike.osd.bsdi.com> Subject: Re: Loader badly broken on the alpha In-Reply-To: <20000916205531.B66584@cicely5.cicely.de> from Bernd Walter at "Sep 16, 2000 08:55:31 pm" To: Bernd Walter Date: Sat, 16 Sep 2000 12:29:22 -0700 (PDT) Cc: Wilko Bulte , Matthew Jacob , alpha@freebsd.org, dcs@freebsd.orggg X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Bernd Walter wrote: > On Sat, Sep 16, 2000 at 11:31:27AM -0700, John Baldwin wrote: > > Bernd Walter wrote: > > > On Sat, Sep 16, 2000 at 11:27:59AM +0200, Wilko Bulte wrote: > > > > Great :-( I nuked my -current box yesterday with the same loader problems. > > > > I promised billf to test the tx driver but now I'm screwed it seems. > > > > > > > > Any good suggestions to fix this loader thing short of reinstalling 4.1 > > > > first? > > > > > > I had the real luck to have an ancient loader from April on the disk. > > > It did not understood some configfiles and stopped with an prompt. > > > Then I had to do the set commands for the device hints. > > > > > > You should try offering a loader from a disk and then switching disks > > > on the loader prompt. > > > > You can't do that with SRM. :( SRM only tells you about your boot > > device, unlike the x86 BIOS which lets you talk to all the devices it > > knows about. > > Mmmh - Bad. > Then the first place is the kernel root mount where we can change drives > so the kernel need to be on the disk too. Yep, and in this case, the loader is dying before it even gets to that point. :-( -- John Baldwin -- http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 18:40:49 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.inka.de (quechua.inka.de [212.227.14.2]) by hub.freebsd.org (Postfix) with ESMTP id 21FAF37B422 for ; Sat, 16 Sep 2000 18:40:47 -0700 (PDT) Received: from ganerc.mips.inka.de (uucp@) by mail.inka.de with local-bsmtp id 13aTRl-0008Hq-00; Sun, 17 Sep 2000 03:40:45 +0200 Received: (from daemon@localhost) by ganerc.mips.inka.de (8.11.0/8.11.0) id e8H1Zwa72814 for freebsd-alpha@freebsd.org; Sun, 17 Sep 2000 03:35:58 +0200 (CEST) (envelope-from daemon) From: naddy@mips.inka.de (Christian Weisgerber) Subject: rsh? Date: 17 Sep 2000 03:35:58 +0200 Message-ID: <8q175u$2735$1@ganerc.mips.inka.de> To: freebsd-alpha@freebsd.org Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org FreeBSD PRE_SMPNG/alpha. rhosts authentication doesn't seem to work. I can't rsh into the box and rlogin asks for a password. No such problem on i386. Is this a silly configuration problem on my side, or does anybody else see this too? -- Christian "naddy" Weisgerber naddy@mips.inka.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message From owner-freebsd-alpha Sat Sep 16 23:27:17 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id B916437B423 for ; Sat, 16 Sep 2000 23:27:14 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8H6Qup03337 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Sun, 17 Sep 2000 08:27:08 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8H6R0I92218; Sun, 17 Sep 2000 08:27:01 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8H6Qwj67504; Sun, 17 Sep 2000 08:26:58 +0200 (CEST) (envelope-from ticso) Date: Sun, 17 Sep 2000 08:26:58 +0200 From: Bernd Walter To: Christian Weisgerber Cc: freebsd-alpha@FreeBSD.ORG Subject: Re: rsh? Message-ID: <20000917082658.A67471@cicely5.cicely.de> References: <8q175u$2735$1@ganerc.mips.inka.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <8q175u$2735$1@ganerc.mips.inka.de>; from naddy@mips.inka.de on Sun, Sep 17, 2000 at 03:35:58AM +0200 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Sep 17, 2000 at 03:35:58AM +0200, Christian Weisgerber wrote: > FreeBSD PRE_SMPNG/alpha. > > rhosts authentication doesn't seem to work. I can't rsh into the > box and rlogin asks for a password. No such problem on i386. Is > this a silly configuration problem on my side, or does anybody else > see this too? bash-2.03$ rsh c9 uname -a FreeBSD cicely9.cicely.de 5.0-CURRENT FreeBSD 5.0-CURRENT #0: Thu Sep 14 08:19:11 CEST 2000 ticso@cicely9.cicely.de:/var/d7/src-2000-09-13/src/sys/compile/CICELY9 alpha It's done via .rhost The system is newer but I never had problems with at least rlogin. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message