From owner-freebsd-stable Mon Feb 12 06:36:38 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA02812 for stable-outgoing; Mon, 12 Feb 1996 06:36:38 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA02805 for ; Mon, 12 Feb 1996 06:36:36 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id JAA15480; Mon, 12 Feb 1996 09:36:31 -0500 Received: (from rjk@localhost) by watson.grauel.com (8.6.12/8.6.9) id JAA14442; Mon, 12 Feb 1996 09:42:47 -0500 Date: Mon, 12 Feb 1996 09:42:47 -0500 Message-Id: <199602121442.JAA14442@watson.grauel.com> From: Richard J Kuhns To: freebsd-stable@freebsd.org Subject: One problem && one question Sender: owner-stable@freebsd.org Precedence: bulk I'm now running FreeBSD-stable (I started with the sources from the 2.1 CD, and used ctm to update, as of Feb 6). Hardware: ASUS P55TP4XE motherboard, 120 MHz Pentium, 32 MB RAM, Buslogic 946C SCSI controller, 1 2GB Seagate, 1 1GB Fujitsu, SoundBlaster 16 with CD, and Diamond Stealth 64 video. The ``make world'' went without problems; it took 236 minutes. Here's the problem: I rebooted the machine a total of 7 times over the weekend; 5 times it couldn't sync the disks -- ``Syncing disks 4 4 4 .... 4 giving up'' (for example). Upon rebooting, the only thing fsck reported was ``clean flag not set in superblock''. I had never had this problem under either 2.0.5 or 2.1-RELEASE. Now my question: someone (Julian?) mentioned that there's now a wrapper to allow the 946C to be found during the PCI probe. I found the necessary file (sys/pci/bt9xx.c), but how do I convince config to use it? I have the following line in my config file: controller bt0 at isa? port "IO_BT0" bio irq ? vector bt_isa_intr Any help/comments would be greatly appreciated. Here's the dmesg output from my most recent reboot: FreeBSD 2.1-STABLE #0: Sat Feb 10 18:27:34 EST 1996 rjk@moran:/usr/2.1-STABLE/src/sys/compile/MORAN CPU: 120-MHz Pentium 735\\90 or 815\\100 (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x525 Stepping=5 Features=0x1bf real memory = 33554432 (32768K bytes) avail memory = 30654464 (29936K bytes) pcibus_setup(1): mode1res=0x80000000 (0x80000000), mode2res=0xff (0x0e) pcibus_setup(2): mode1res=0x80000000 (0x80000000) pcibus_check: device 0 is there (id=122d8086) Probing for devices on PCI bus 0: configuration mode 1 allows 32 devices. chip0 rev 2 on pci0:0 chip1 rev 2 on pci0:7 vga0 rev 0 int a irq 10 on pci0:10 mapreg[10] type=0 addr=fb000000 size=800000. pci0:12: Bus Logic, device=0x1040, class=storage (scsi) [no driver assigned] map(10): io(e400) pci0: uses 8388608 bytes of memory from fb000000 upto fb7fffff. Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> pca0 on motherboard pca0: PC speaker audio driver sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in bt0: Bt946C/ 0-(32bit) bus bt0: reading board settings, dma=5, int=11 bt0: version 4.25J, fast sync, parity, 32 mbxs, 32 ccbs bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15 bt0: targ 1 sync rate=10.00MB/s(100ns), offset=15 bt0: Using Strict Round robin scheme bt0 at 0x330 irq 11 drq 5 on isa bt0 waiting for scsi devices to settle (bt0:0:0): "FUJITSU M2694ES-512 812A" type 0 fixed SCSI 2 sd0(bt0:0:0): Direct-Access 1033MB (2117025 512 byte sectors) sd0(bt0:0:0): with 1819 cyls, 15 heads, and an average 77 sectors/track (bt0:1:0): "SEAGATE ST32550N 0011" type 0 fixed SCSI 2 sd1(bt0:1:0): Direct-Access 2047MB (4194058 512 byte sectors) sd1(bt0:1:0): with 3511 cyls, 11 heads, and an average 108 sectors/track matcd - Matsushita (Panasonic) CD-ROM Driver by FDIV, Version 1(26) 18-Oct-95 matcdc0 at 0x230-0x233 on isa matcdc0 Host interface type 0 matcd0: [CR-5630.75] sb0 at 0x220 irq 5 drq 1 on isa sb0: sbxvi0 at 0x0 drq 6 on isa sbxvo0: sbmidi0 at 0x300 on isa npx0 on motherboard npx0: INT 16 interface BIOS Geometries: 0:03ff3f20 0..1023=1024 cylinders, 0..63=64 heads, 1..32=32 sectors 1:03ff3f20 0..1023=1024 cylinders, 0..63=64 heads, 1..32=32 sectors 0 accounted for sd0s1: type 0xa5, start 735232, end = 2115583, size 1380352 : OK sd0s3: type 0x6, start 32, end = 735231, size 735200 : OK sd1s1: type 0xa5, start 32, end = 4192255, size 4192224 : OK -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Mon Feb 12 08:53:52 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA10532 for stable-outgoing; Mon, 12 Feb 1996 08:53:52 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA10522 Mon, 12 Feb 1996 08:53:49 -0800 (PST) Message-Id: <199602121653.IAA10522@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Mon, 12 Feb 1996 09:42:47 EST." <199602121442.JAA14442@watson.grauel.com> Date: Mon, 12 Feb 1996 08:53:49 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >Now my question: someone (Julian?) mentioned that there's now a wrapper to >allow the 946C to be found during the PCI probe. I found the necessary >file (sys/pci/bt9xx.c), but how do I convince config to use it? > >I have the following line in my config file: > >controller bt0 at isa? port "IO_BT0" bio irq ? vector bt_isa_intr > >Any help/comments would be greatly appreciated. The problem is that I have the PCI ID fror the Buslogic reversed in sys/pci/bt9xx.c. Change: #define PCI_DEVICE_ID_BUSLOGIC_946 0x104B1040ul To: #define PCI_DEVICE_ID_BUSLOGIC_946 0x1040104Bul I'll fix this in -current and -stable ASAP. That's what I get for writing the wrapper and not having one of these cards... :) >-- >Rich Kuhns rjk@grauel.com >PO Box 6249 >100 Sawmill Road >Lafayette, IN 47903 >(317)477-6000 x319 -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Mon Feb 12 11:35:51 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA21529 for stable-outgoing; Mon, 12 Feb 1996 11:35:51 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA21523 for ; Mon, 12 Feb 1996 11:35:49 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id OAA16722; Mon, 12 Feb 1996 14:35:47 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id OAA00256; Mon, 12 Feb 1996 14:40:40 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA15018; Mon, 12 Feb 1996 14:36:12 +0500 Date: Mon, 12 Feb 1996 14:36:12 +0500 Message-Id: <9602121936.AA15018@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602121653.IAA10522@freefall.freebsd.org> References: <199602121442.JAA14442@watson.grauel.com> <199602121653.IAA10522@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > > The problem is that I have the PCI ID fror the Buslogic reversed in > sys/pci/bt9xx.c. > > Change: > #define PCI_DEVICE_ID_BUSLOGIC_946 0x104B1040ul > To: > #define PCI_DEVICE_ID_BUSLOGIC_946 0x1040104Bul > > I'll fix this in -current and -stable ASAP. That's what I get for > writing the wrapper and not having one of these cards... :) > Thanks, Justin. I just made that change and built a new kernel. The good news is, the pci probe found the 946. The bad news is, the kernel won't boot. The message is `bt0: Invalid base address', followed almost immediately by the ``reboot in 15 seconds'' message. The only difference between the 2 kernels is the line above. Unfortunately, none of the boot messages hung around to be printed later by dmesg. Anything else I can do to make your day more enjoyable? :-) -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Mon Feb 12 12:53:52 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA28931 for stable-outgoing; Mon, 12 Feb 1996 12:53:52 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA28921 Mon, 12 Feb 1996 12:53:47 -0800 (PST) Message-Id: <199602122053.MAA28921@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Mon, 12 Feb 1996 14:36:12 +0500." <9602121936.AA15018@sparcmill.grauel.com> Date: Mon, 12 Feb 1996 12:53:46 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >Justin T. Gibbs writes: > >Thanks, Justin. > >I just made that change and built a new kernel. The good news is, the pci >probe found the 946. The bad news is, the kernel won't boot. The message >is `bt0: Invalid base address', followed almost immediately by the ``reboot >in 15 seconds'' message. The only difference between the 2 kernels is the >line above. Unfortunately, none of the boot messages hung around to be >printed later by dmesg. > >Anything else I can do to make your day more enjoyable? :-) >-- >Rich Kuhns rjk@grauel.com >PO Box 6249 >100 Sawmill Road >Lafayette, IN 47903 >(317)477-6000 x319 Can you try this patch. I'm away from my machine at the moment, so you may need to tweak it to get it to compile. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== Index: bt9xx.c =================================================================== RCS file: /home/ncvs/src/sys/pci/bt9xx.c,v retrieving revision 1.5 diff -c -r1.5 bt9xx.c *** bt9xx.c 1996/02/12 17:00:39 1.5 --- bt9xx.c 1996/02/12 20:51:26 *************** *** 35,41 **** #include /* XXX Need more device IDs */ - #define PCI_BASEADR0 PCI_MAP_REG_START #define PCI_DEVICE_ID_BUSLOGIC_946 0x1040104Bul static char* bt_pci_probe __P((pcici_t tag, pcidi_t type)); --- 35,40 ---- *************** *** 70,86 **** pcici_t config_id; int unit; { u_long io_port; unsigned opri = 0; struct bt_data *bt; ! if(!(io_port = pci_conf_read(config_id, PCI_BASEADR0))) return; - /* - * The first bit of PCI_BASEADR0 is always - * set hence we mask it off. - */ - io_port &= 0xfffffffe; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ --- 69,89 ---- pcici_t config_id; int unit; { + int reg; u_long io_port; unsigned opri = 0; struct bt_data *bt; ! for(reg = PCI_MAP_REG_START; reg <= PCI_MAP_REG_END; reg++) { ! if(!(io_port = pci_conf_read(config_id, reg))) ! return; ! if(io_port & PCI_MAP_IO) { ! ioport &= PCI_IO_MAP; ! break; ! } ! } ! if(reg > PCI_MAP_REG_END) return; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ From owner-freebsd-stable Mon Feb 12 13:29:00 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id NAA02406 for stable-outgoing; Mon, 12 Feb 1996 13:29:00 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id NAA02397 for ; Mon, 12 Feb 1996 13:28:56 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id QAA17180; Mon, 12 Feb 1996 16:28:52 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id QAA00240; Mon, 12 Feb 1996 16:33:43 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA16125; Mon, 12 Feb 1996 16:29:23 +0500 Date: Mon, 12 Feb 1996 16:29:23 +0500 Message-Id: <9602122129.AA16125@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602122053.MAA28921@freefall.freebsd.org> References: <9602121936.AA15018@sparcmill.grauel.com> <199602122053.MAA28921@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > > Can you try this patch. I'm away from my machine at the moment, > so you may need to tweak it to get it to compile. I tried; bear in mind I'm running 2.1-STABLE. I fixed what seemed to be a couple of typos in the patch; specifically ioport &= PCI_IO_MAP; to io_port &= PCI_MAP_IO; Unfortunately, it still didn't work -- same message. I tried booting with -v, but the messages don't stay on-screen long enough to read very easily - I can't believe I'm *complaining* about having a fast machine. I'm fairly sure it reported addr: 0x00006000, tho. What next, boss? -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Mon Feb 12 14:45:26 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id OAA08655 for stable-outgoing; Mon, 12 Feb 1996 14:45:26 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id OAA08648 Mon, 12 Feb 1996 14:45:24 -0800 (PST) Message-Id: <199602122245.OAA08648@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: freebsd-stable@FreeBSD.ORG Subject: Re: One problem && one question In-reply-to: Your message of "Mon, 12 Feb 1996 16:29:23 +0500." <9602122129.AA16125@sparcmill.grauel.com> Date: Mon, 12 Feb 1996 14:45:23 -0800 From: "Justin T. Gibbs" Sender: owner-stable@FreeBSD.ORG Precedence: bulk Patch round two: -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== Index: bt9xx.c =================================================================== RCS file: /home/ncvs/src/sys/pci/bt9xx.c,v retrieving revision 1.5 diff -c -r1.5 bt9xx.c *** bt9xx.c 1996/02/12 17:00:39 1.5 --- bt9xx.c 1996/02/12 22:43:20 *************** *** 35,41 **** #include /* XXX Need more device IDs */ - #define PCI_BASEADR0 PCI_MAP_REG_START #define PCI_DEVICE_ID_BUSLOGIC_946 0x1040104Bul static char* bt_pci_probe __P((pcici_t tag, pcidi_t type)); --- 35,40 ---- *************** *** 70,86 **** pcici_t config_id; int unit; { u_long io_port; unsigned opri = 0; struct bt_data *bt; ! if(!(io_port = pci_conf_read(config_id, PCI_BASEADR0))) return; - /* - * The first bit of PCI_BASEADR0 is always - * set hence we mask it off. - */ - io_port &= 0xfffffffe; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ --- 69,89 ---- pcici_t config_id; int unit; { + int reg; u_long io_port; unsigned opri = 0; struct bt_data *bt; ! for(reg = PCI_MAP_REG_START; reg < PCI_MAP_REG_END; reg+=4) { ! if(!(io_port = pci_conf_read(config_id, reg))) ! return; ! if(io_port & PCI_MAP_IO) { ! ioport &= PCI_MAP_IO; ! break; ! } ! } ! if(reg == PCI_MAP_REG_END) return; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ From owner-freebsd-stable Tue Feb 13 06:39:27 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA14430 for stable-outgoing; Tue, 13 Feb 1996 06:39:27 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA14424 for ; Tue, 13 Feb 1996 06:39:25 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id JAA20130; Tue, 13 Feb 1996 09:39:23 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id JAA00482; Tue, 13 Feb 1996 09:44:18 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA20817; Tue, 13 Feb 1996 09:39:53 +0500 Date: Tue, 13 Feb 1996 09:39:53 +0500 Message-Id: <9602131439.AA20817@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: Richard J Kuhns , freebsd-stable@FreeBSD.ORG Subject: Re: One problem && one question In-Reply-To: <199602122245.OAA08648@freefall.freebsd.org> References: <9602122129.AA16125@sparcmill.grauel.com> <199602122245.OAA08648@freefall.freebsd.org> Sender: owner-stable@FreeBSD.ORG Precedence: bulk Justin T. Gibbs writes: > Patch round two: Sorry, still no cigar. I rebooted this machine several times this morning, trying to copy the boot messages. I guess I'd better set up a minimal configuration to test this so the bt messages will remain on-screen. At any rate, I believe it looked like: int a irq 10 pci 0:13 mapreg[10] type 1 addr 0x00006000 size ffff0004 bt0: Invalid base address I'm afraid I won't be able to test quite so often today; the machine has to do some Real Work (TM). -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Tue Feb 13 07:29:07 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id HAA17346 for stable-outgoing; Tue, 13 Feb 1996 07:29:07 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA17339 for ; Tue, 13 Feb 1996 07:28:54 -0800 (PST) Received: by Sysiphos id AA05771 (5.67b/IDA-1.5 for freebsd-stable@FreeBSD.ORG); Tue, 13 Feb 1996 16:28:44 +0100 Message-Id: <199602131528.AA05771@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Tue, 13 Feb 1996 16:28:43 +0100 In-Reply-To: Richard J Kuhns "Re: One problem && one question" (Feb 13, 9:39) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Richard J Kuhns Subject: Re: One problem && one question Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk On Feb 13, 9:39, Richard J Kuhns wrote: } Subject: Re: One problem && one question } Justin T. Gibbs writes: } > Patch round two: } } Sorry, still no cigar. I rebooted this machine several times this morning, } trying to copy the boot messages. I guess I'd better set up a minimal } configuration to test this so the bt messages will remain on-screen. } } At any rate, I believe it looked like: } } int a irq 10 pci 0:13 } mapreg[10] type 1 addr 0x00006000 size ffff0004 } bt0: Invalid base address The region size calculation seems broken! I'll check whether this is caused by some bug in the generic PCI code ... ... Ok. I'm just back from reading the sources, and it appears that the BT946C does not support a 32 bit I/O range as required by the PCI 2.0 spec ... Please try the patch I have appended to this message ... Regards, STefan Index: /sys/pci/pcireg.h =================================================================== RCS file: /usr/cvs/src/sys/pci/pcireg.h,v retrieving revision 1.7 diff -C2 -r1.7 pcireg.h *** pcireg.h 1996/01/25 18:31:59 1.7 --- pcireg.h 1996/02/13 15:24:30 *************** *** 152,156 **** #define PCI_MAP_MEMORY_ADDRESS_MASK 0xfffffff0 ! #define PCI_MAP_IO_ADDRESS_MASK 0xfffffffc /* ** PCI-PCI bridge mapping registers --- 152,156 ---- #define PCI_MAP_MEMORY_ADDRESS_MASK 0xfffffff0 ! #define PCI_MAP_IO_ADDRESS_MASK 0x0000fffc /* ** PCI-PCI bridge mapping registers -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-stable Tue Feb 13 08:10:27 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA19912 for stable-outgoing; Tue, 13 Feb 1996 08:10:27 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA19906 Tue, 13 Feb 1996 08:10:24 -0800 (PST) Message-Id: <199602131610.IAA19906@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: se@ZPR.Uni-Koeln.DE (Stefan Esser) cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Tue, 13 Feb 1996 16:28:43 +0100." <199602131528.AA05771@Sysiphos> Date: Tue, 13 Feb 1996 08:10:24 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >On Feb 13, 9:39, Richard J Kuhns wrote: >} Subject: Re: One problem && one question >} Justin T. Gibbs writes: >} > Patch round two: >} >} Sorry, still no cigar. I rebooted this machine several times this morning, >} trying to copy the boot messages. I guess I'd better set up a minimal >} configuration to test this so the bt messages will remain on-screen. >} >} At any rate, I believe it looked like: >} >} int a irq 10 pci 0:13 >} mapreg[10] type 1 addr 0x00006000 size ffff0004 >} bt0: Invalid base address > >The region size calculation seems broken! The driver only cares about the I/O port, which isn't reported above. I've seen boot -v output from bt946 cards before, and they at least report a sane I/O port address like 0x330 (when listed as a device without a driver) - never looked at the size, but it should be 4. I wouldn't bother hacking up the memory size calculation just for this card since we won't even use it. I just sent him an updated patch, that corrects some mistakes I made in the last one, but I did notice something odd in the PCI code: if (bootverbose) { for (reg=PCI_MAP_REG_START; regpb_read (tag, reg); if ((data&~7)==0) continue; switch (data&7) { case 1: case 5: printf (" map(%x): io(%lx)\n", reg, data & ~3); break; I guess that everything must be quad-word aligned so that is why you & ~3? Is this in the spec even for I/O ports? >-- > Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 > Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 > ============================================================================= >= > http://www.zpr.uni-koeln.de/~se -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Tue Feb 13 08:31:38 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA20950 for stable-outgoing; Tue, 13 Feb 1996 08:31:38 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA20941 for ; Tue, 13 Feb 1996 08:30:51 -0800 (PST) Received: by Sysiphos id AA09319 (5.67b/IDA-1.5 for freebsd-stable@freebsd.org); Tue, 13 Feb 1996 17:30:13 +0100 Message-Id: <199602131630.AA09319@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Tue, 13 Feb 1996 17:30:13 +0100 In-Reply-To: "Justin T. Gibbs" "Re: One problem && one question" (Feb 13, 8:10) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: "Justin T. Gibbs" Subject: Re: One problem && one question Cc: Richard J Kuhns , freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk On Feb 13, 8:10, "Justin T. Gibbs" wrote: } Subject: Re: One problem && one question } >On Feb 13, 9:39, Richard J Kuhns wrote: } >} Subject: Re: One problem && one question } >} Justin T. Gibbs writes: } >} > Patch round two: } >} } >} Sorry, still no cigar. I rebooted this machine several times this morning, } >} trying to copy the boot messages. I guess I'd better set up a minimal } >} configuration to test this so the bt messages will remain on-screen. } >} } >} At any rate, I believe it looked like: } >} } >} int a irq 10 pci 0:13 } >} mapreg[10] type 1 addr 0x00006000 size ffff0004 } >} bt0: Invalid base address } > } >The region size calculation seems broken! } } The driver only cares about the I/O port, which isn't reported above. I've } seen boot -v output from bt946 cards before, and they at least report a } sane I/O port address like 0x330 (when listed as a device without a driver) The BIOS is free to assign any port address, and 0x330 seems to be reasonable to emulate the ISA version of the card ... } - never looked at the size, but it should be 4. I wouldn't bother hacking } up the memory size calculation just for this card since we won't even use } it. I just sent him an updated patch, that corrects some mistakes I made The PCI code uses it as a parameter to the pci_map_io() call, but since this is not about VM, it doesn't have much of an effect ... } in the last one, but I did notice something odd in the PCI code: } } if (bootverbose) { } for (reg=PCI_MAP_REG_START; regpb_read (tag, reg); } if ((data&~7)==0) continue; } switch (data&7) { } } case 1: } case 5: } printf (" map(%x): io(%lx)\n", } reg, data & ~3); } break; } } I guess that everything must be quad-word aligned so that is why you } & ~3? Is this in the spec even for I/O ports? In PCI there are a number of map types, and those are encoded in those (assumed to be '0') bits. I.e. the type 1 encoding has a '01' in the two low order bits, which have to be masked out ... The problem with the wrong size being reported is caused by the fact, that the bt946c has only a 16 bit port base register. And for this reason, it returns 0x0000fffd if a 0xffffffff had been written ... The 'd' is a 'c' (i.e. -4) plus the mapping type 1 code. Since the PCI spec defines type 1 as a 32bit I/O port mapping (IIRC) the 946 really should return 0xfffffffd. It has no right to ignore the upper 16bits. But since ISA only supports 16 bits of port address anyway, it doesn't appear to make much of a difference to make the code always ignore those upper half for port mappings ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-stable Tue Feb 13 09:29:36 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA24935 for stable-outgoing; Tue, 13 Feb 1996 09:29:36 -0800 (PST) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA24930 for ; Tue, 13 Feb 1996 09:29:33 -0800 (PST) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id JAA02469; Tue, 13 Feb 1996 09:28:37 -0800 From: "Rodney W. Grimes" Message-Id: <199602131728.JAA02469@GndRsh.aac.dev.com> Subject: Re: One problem && one question To: se@ZPR.Uni-Koeln.DE (Stefan Esser) Date: Tue, 13 Feb 1996 09:28:36 -0800 (PST) Cc: gibbs@freefall.freebsd.org, rjk@sparcmill.grauel.com, freebsd-stable@freebsd.org In-Reply-To: <199602131630.AA09319@Sysiphos> from "Stefan Esser" at Feb 13, 96 05:30:13 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk > > On Feb 13, 8:10, "Justin T. Gibbs" wrote: > } Subject: Re: One problem && one question > } >On Feb 13, 9:39, Richard J Kuhns wrote: > } >} Subject: Re: One problem && one question > } >} Justin T. Gibbs writes: > } >} > Patch round two: > } >} > } >} Sorry, still no cigar. I rebooted this machine several times this morning, > } >} trying to copy the boot messages. I guess I'd better set up a minimal > } >} configuration to test this so the bt messages will remain on-screen. > } >} > } >} At any rate, I believe it looked like: > } >} > } >} int a irq 10 pci 0:13 > } >} mapreg[10] type 1 addr 0x00006000 size ffff0004 > } >} bt0: Invalid base address > } > > } >The region size calculation seems broken! > } > } The driver only cares about the I/O port, which isn't reported above. I've > } seen boot -v output from bt946 cards before, and they at least report a > } sane I/O port address like 0x330 (when listed as a device without a driver) > > The BIOS is free to assign any port address, and > 0x330 seems to be reasonable to emulate the ISA > version of the card ... And if you have a revision E card and turn off the ISA ``assign standard I/O port'' in the bios it will come up with something in the more normal PCI I/O address range. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-stable Tue Feb 13 09:52:28 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA26121 for stable-outgoing; Tue, 13 Feb 1996 09:52:28 -0800 (PST) Received: from shellx.best.com (root@shellx.best.com [206.86.0.11]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA26113 for ; Tue, 13 Feb 1996 09:52:26 -0800 (PST) Received: from free.emmert.com (emmert.vip.best.com [204.156.142.218]) by shellx.best.com (8.6.12/8.6.5) with SMTP id JAA08466 for ; Tue, 13 Feb 1996 09:49:30 -0800 Date: Tue, 13 Feb 1996 09:51:03 -0800 (PST) From: Steve Emmert To: freebsd-stable@freebsd.org Subject: Using -stable Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org Precedence: bulk Hi, A few questions: 1. Upgraded my src to -stable using ctm, but the latest patch src-2.1.0033 won't go in. Fails with "md5 mismatch", it seems to be out of sequence. 2. Ran "make world" for the first time, it asks me "overwrite .depend?" should I do this? (I stopped it there) What other questions will it ask? 3. Will "make world" make and install a new kernel? Should I put my custom kernel config in /usr/src/sys/i336/conf/? --Steve Steve Emmert steve@emmert.com ------------------------------------------------------------------------- Powered by FreeBSD -- The Daemon is FREE!!! From owner-freebsd-stable Tue Feb 13 09:55:49 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA26219 for stable-outgoing; Tue, 13 Feb 1996 09:55:49 -0800 (PST) Received: from iaehv.IAEhv.nl (root@iaehv.IAEhv.nl [192.87.208.2]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA26200 for ; Tue, 13 Feb 1996 09:55:45 -0800 (PST) Received: from oasis.IAEhv.nl by iaehv.IAEhv.nl (8.6.12/1.63) id SAA29997; Tue, 13 Feb 1996 18:55:42 +0100 X-Disclaimer: iaehv.nl is a public access UNIX system and cannot be held responsible for the opinions of its individual users. Received: by oasis (8.6.12/1.63) id SAA01160; Tue, 13 Feb 1996 18:35:10 +0100 From: volf@oasis.IAEhv.nl (Frank Volf) Message-Id: <199602131735.SAA01160@oasis> Subject: Re: Commit documentation for stable To: graichen@omega.physik.fu-berlin.de (Thomas Graichen) Date: Fri, 9 Feb 1996 22:55:23 +0100 (MET) In-Reply-To: <4fbd47$1id@prospero.physik.fu-berlin.de> from "Thomas Graichen" at Feb 7, 96 11:37:43 pm X-Mailer: ELM [version 2.4 PL25] Content-Type: text Sender: owner-stable@FreeBSD.ORG Precedence: bulk Thomas Graichen wrote: > > Frank Volf (volf@oasis.IAEhv.nl) wrote: > > : Hi, > > : Now that I have the everything setup on my system to receive the CTM patches > : by email, I would like to know what each of these patches is supposed to > : do. For FreeBSD-current you can subscribe to the cvs-* mailing lists. > : However, I was unable to find a list for FreeBSD-stable. How, can I get > : this information. > > the same lists :-) I was afraid you would say that :-(, the reason I asked is that I use a UUCP connection to the internet, so I don't want to download a lot of cvs-all mail, while I'm only interested in a few commits to the stable branch... So, it would be nice if we could have a seperate "stable branch cvs commit list" (Jordan, please, please, please) Regards, Frank ---------------------------------------------------------------------------- Frank Volf - Internet Access Eindhoven - Digitale Stad Eindhoven ---------------------------------------------------------------------------- || volf@oasis.IAEhv.nl - use for personal mail || || volf@IAEhv.nl - use for Internet Access Eindhoven related mail || || volf@dse.dse.nl - use for Digital City of Eindhoven related mail || ---------------------------------------------------------------------------- IAE Public Access Unix System - Dial +31.40.2439436 and login as new. ---------------------------------------------------------------------------- From owner-freebsd-stable Tue Feb 13 11:26:53 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA01012 for stable-outgoing; Tue, 13 Feb 1996 11:26:53 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA01006 for ; Tue, 13 Feb 1996 11:26:50 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id OAA21555; Tue, 13 Feb 1996 14:26:46 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id OAA00225; Tue, 13 Feb 1996 14:31:46 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA23929; Tue, 13 Feb 1996 14:27:19 +0500 Date: Tue, 13 Feb 1996 14:27:19 +0500 Message-Id: <9602131927.AA23929@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602131550.HAA18721@freefall.freebsd.org> References: <9602131439.AA20817@sparcmill.grauel.com> <199602131550.HAA18721@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > >Justin T. Gibbs writes: > > int a irq 10 pci 0:13 > > mapreg[10] type 1 addr 0x00006000 size ffff0004 > >bt0: Invalid base address > > boot -v with the old kernel and then use dmesg to retrieve the info. > That works fine for any ``working'' kernel; apparently none of the boot messages from a non-working kernel end up in the circular buffer dmesg uses. IE, if I boot-working then boot-nonworking then boot-working and run dmesg, I get the messages from the 2 boot-working kernels only. If that's what you were interested in, here's the most recent: FreeBSD 2.1-STABLE #2: Tue Feb 13 14:20:26 EST 1996 rjk@watson.grauel.com:/prog/2.1-STABLE/src/sys/compile/WATSON CPU: 100-MHz Pentium 815\\100 (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x525 Stepping=5 Features=0x1bf real memory = 33554432 (32768K bytes) avail memory = 30728192 (30008K bytes) pcibus_setup(1): mode1res=0x80000000 (0x80000000), mode2res=0xff (0x0e) pcibus_setup(2): mode1res=0x80000000 (0x80000000) pcibus_check: device 0 is there (id=04061039) Probing for devices on PCI bus 0: configuration mode 1 allows 32 devices. chip0 rev 0 on pci0:0 chip1 rev 0 on pci0:1 pci0:13: Bus Logic, device=0x1040, class=storage (scsi) [no driver assigned] map(10): io(6000) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 at 0x280-0x29f irq 5 maddr 0xd8000 msize 16384 on isa ed0: address 00:00:c0:15:00:ab, type SMC8216/SMC8216C (16 bit) sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface pca0 on motherboard pca0: PC speaker audio driver fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in bt0: Bt946C/ 0-(32bit) bus bt0: reading board settings, busmastering, int=10 bt0: version 4.25J, fast sync, parity, 32 mbxs, 32 ccbs bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15 bt0: targ 5 sync rate= 5.00MB/s(200ns), offset=11 bt0: Using Strict Round robin scheme bt0 at 0x330 irq 10 on isa bt0 waiting for scsi devices to settle (bt0:0:0): "MICROP 3243-19MZ Q4D HT02" type 0 fixed SCSI 2 sd0(bt0:0:0): Direct-Access 4095MB (8388315 512 byte sectors) sd0(bt0:0:0): with 3956 cyls, 19 heads, and an average 111 sectors/track (bt0:5:0): "EXABYTE EXB-82058VQANXR1 0790" type 1 removable SCSI 2 st0(bt0:5:0): Sequential-Access density code 0x0, drive empty (bt0:6:0): "Chinon CD-ROM CDS-525 V09" type 5 removable SCSI 2 cd0(bt0:6:0): CD-ROM cd0(bt0:6:0): UNIT ATTENTION asc:2a,0 Parameters changed cd present.[400000 x 2048 byte records] npx0 on motherboard npx0: INT 16 interface changing root device to sd0a BIOS Geometries: 0:03ff3f20 0..1023=1024 cylinders, 0..63=64 heads, 1..32=32 sectors 0 accounted for sd0s1: type 0xa5, start 0, end = 8388314, size 8388315 : OK > >I'm afraid I won't be able to test quite so often today; the machine has to > >do some Real Work (TM). > > When you get a chance, try this one. Now that I'm not editing over a link > with 50% packet loss, its much easier to check my work. :) > I tried it; I'm afraid it made no difference. By the way, here's one more line from the (failing) pci-probe: bt0: uses 4 bytes of I/O space from 6000 upto 6003 -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Tue Feb 13 12:22:14 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA05026 for stable-outgoing; Tue, 13 Feb 1996 12:22:14 -0800 (PST) Received: from eel.dataplex.net (EEL.DATAPLEX.NET [199.183.109.245]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA05020 for ; Tue, 13 Feb 1996 12:22:10 -0800 (PST) Received: from [199.183.109.242] (cod [199.183.109.242]) by eel.dataplex.net (8.6.11/8.6.9) with SMTP id OAA23147; Tue, 13 Feb 1996 14:21:05 -0600 X-Sender: rkw@shark.dataplex.net Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 13 Feb 1996 14:21:06 -0600 To: Steve Emmert From: rkw@dataplex.net (Richard Wackerbarth) Subject: Re: Using -stable Cc: stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk >Hi, >A few questions: > >1. Upgraded my src to -stable using ctm, but the latest patch src-2.1.0033 > won't go in. Fails with "md5 mismatch", it seems to be out of sequence. Doesn't sound good. How did you get the .0033 file? What is the content of the .ctm_status file in YOUR tree? Did you modify ANYTHING in the tree? What is the tree history? Where did you start, (eg CD-ROM, 0015C update, etc) Let's keep this off the list until I figure out what went wrong and get you back up to speed. Richard Wackerbarth rkw@dataplex.net From owner-freebsd-stable Tue Feb 13 15:14:43 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id PAA16931 for stable-outgoing; Tue, 13 Feb 1996 15:14:43 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id PAA16921 for ; Tue, 13 Feb 1996 15:14:38 -0800 (PST) Received: by Sysiphos id AA14953 (5.67b/IDA-1.5 for freebsd-stable@freebsd.org); Wed, 14 Feb 1996 00:14:00 +0100 Message-Id: <199602132314.AA14953@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Wed, 14 Feb 1996 00:14:00 +0100 In-Reply-To: Richard J Kuhns "Re: One problem && one question" (Feb 13, 14:27) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Richard J Kuhns Subject: Re: One problem && one question Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk On Feb 13, 14:27, Richard J Kuhns wrote: } Subject: Re: One problem && one question } Justin T. Gibbs writes: } > >Justin T. Gibbs writes: } > > int a irq 10 pci 0:13 } > > mapreg[10] type 1 addr 0x00006000 size ffff0004 } > >bt0: Invalid base address } > } > boot -v with the old kernel and then use dmesg to retrieve the info. } > } } That works fine for any ``working'' kernel; apparently none of the boot } messages from a non-working kernel end up in the circular buffer dmesg } uses. IE, if I boot-working then boot-nonworking then boot-working and run } dmesg, I get the messages from the 2 boot-working kernels only. } } If that's what you were interested in, here's the most recent: } pci0:13: Bus Logic, device=0x1040, class=storage (scsi) [no driver assigned] } map(10): io(6000) Well, seems the PCI code does find the BT946C, but just doesn't know that it knows about it (because a insuitable PCI ID has been specified). } bt0 at 0x330 irq 10 on isa This gives the ISA code a chance to initialize the chip, which it could not have done else :) } > >I'm afraid I won't be able to test quite so often today; the machine has to } > >do some Real Work (TM). } I tried it; I'm afraid it made no difference. By the way, here's one more } line from the (failing) pci-probe: } } bt0: uses 4 bytes of I/O space from 6000 upto 6003 That's not surprising. But that the Buslogic responds to the ISA probe at 0x330 but offers only the 0x6000 port mapping in the PCI config space registers, that's bad ... If you have the correct PCI ID in the bt9xx.c file now, then I'd expect the PCI probe to be able to initialize the card at 0x6000 (I guess it will respond there, too!). There after the ISA probe may still be able to access it at 0x330 (unless the card disables that address, if the PCI region has been accessed at least once ...). If the card actually is found twice at two different port addresses, once as ISA and once as PCI device, then bad things will happen for sure ! Could you please check the bt0 attach message in the PCI scan, and look at the ISA messages whether bt0 appears there, too. I.e. the two lines I did not remove from the quoted boot message log you sent ... Seems the chip (or its BIOS) is severely broken, if it does not announce it's ISA compatibility register range through the mapping registers. It would be hard to compensate for this, unless the ISA compatibility range was in fact disabled after a PCI announced register has been used for the first time. Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-stable Tue Feb 13 15:59:37 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id PAA18777 for stable-outgoing; Tue, 13 Feb 1996 15:59:37 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [192.216.222.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id PAA18771 for ; Tue, 13 Feb 1996 15:59:36 -0800 (PST) Received: from eel.dataplex.net (EEL.DATAPLEX.NET [199.183.109.245]) by who.cdrom.com (8.6.12/8.6.11) with ESMTP id PAA13669 for ; Tue, 13 Feb 1996 15:59:24 -0800 Received: from [199.183.109.242] (cod [199.183.109.242]) by eel.dataplex.net (8.6.11/8.6.9) with SMTP id RAA23864; Tue, 13 Feb 1996 17:42:44 -0600 X-Sender: rkw@shark.dataplex.net Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 13 Feb 1996 17:42:44 -0600 To: Steve Emmert From: rkw@dataplex.net (Richard Wackerbarth) Subject: Re: Using -stable Cc: stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk Steve Emmert writes: [After we got him to fetch the missing files] Be sure to subscribe to the ctm-src-2_1 list to get the future updates automatically. >CTM works great 8) I tried sup, but could never get the servers to stay >on line 8( (Hoists virtual martini to Richard!!) Thanks! And here's to Poul, too! >Like I hinted in my post to the list, I am a newbie to make world (but not >UN*X or making a working kernel) Will I still get those .depend questions? >Will it make a kernel? Repeating your questions for the benefit of the kernel hackers. I'm just the ctm-stable guy. If I were the "sup" advocate, would that make me a sup-er guy? :-) >>2. Ran "make world" for the first time, it asks me "overwrite .depend?" >> should I do this? (I stopped it there) What other questions will it ask? >>3. Will "make world" make and install a new kernel? Should I put my custom >> kernel config in /usr/src/sys/i336/conf/? Richard Wackerbarth rkw@dataplex.net From owner-freebsd-stable Tue Feb 13 21:22:36 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA08643 for stable-outgoing; Tue, 13 Feb 1996 21:22:36 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA08632 Tue, 13 Feb 1996 21:22:33 -0800 (PST) Message-Id: <199602140522.VAA08632@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Tue, 13 Feb 1996 14:27:19 +0500." <9602131927.AA23929@sparcmill.grauel.com> Date: Tue, 13 Feb 1996 21:22:33 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >Justin T. Gibbs writes: > > >Justin T. Gibbs writes: > > > int a irq 10 pci 0:13 > > > mapreg[10] type 1 addr 0x00006000 size ffff0004 > > >bt0: Invalid base address > > > > boot -v with the old kernel and then use dmesg to retrieve the info. > > > >That works fine for any ``working'' kernel; apparently none of the boot >messages from a non-working kernel end up in the circular buffer dmesg >uses. IE, if I boot-working then boot-nonworking then boot-working and run >dmesg, I get the messages from the 2 boot-working kernels only. How would you expect the dmesg buffer to be written to disk if the probe for your disk controller isn't successful? 8-) > > When you get a chance, try this one. Now that I'm not editing over a link > > with 50% packet loss, its much easier to check my work. :) > > > >I tried it; I'm afraid it made no difference. By the way, here's one more >line from the (failing) pci-probe: > >bt0: uses 4 bytes of I/O space from 6000 upto 6003 Ahh. It did make a difference. Unfortunately, when I was cleaning up the bt driver, I assumed that the PCI probe would always return one of the ISA compatibility addresses. As Rod pointed out, this is not always the case. This patch should do the trick >-- >Rich Kuhns rjk@grauel.com >PO Box 6249 >100 Sawmill Road >Lafayette, IN 47903 >(317)477-6000 x319 -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== Index: i386/scsi/bt.c =================================================================== RCS file: /usr/cvs/src/sys/i386/scsi/bt.c,v retrieving revision 1.5 diff -c -r1.5 bt.c *** bt.c 1996/01/25 23:03:07 1.5 --- bt.c 1996/02/14 05:03:01 *************** *** 462,471 **** } } } - if (i >= sizeof(found)/sizeof(struct bt_found)) { - printf("bt%d: Invalid base address\n", unit); - return NULL; - } bt = malloc(sizeof(struct bt_data), M_DEVBUF, M_NOWAIT); if (!bt) { --- 462,467 ---- Index: i386/pci/bt9xx.c =================================================================== RCS file: /usr/cvs/src/sys/pci/bt9xx.c,v retrieving revision 1.4 diff -c -r1.4 bt9xx.c *** bt9xx.c 1996/01/23 21:46:57 1.4 --- bt9xx.c 1996/02/13 15:46:23 *************** *** 35,42 **** #include /* XXX Need more device IDs */ ! #define PCI_BASEADR0 PCI_MAP_REG_START ! #define PCI_DEVICE_ID_BUSLOGIC_946 0x104B1040ul static char* bt_pci_probe __P((pcici_t tag, pcidi_t type)); static void bt_pci_attach __P((pcici_t config_id, int unit)); --- 35,41 ---- #include /* XXX Need more device IDs */ ! #define PCI_DEVICE_ID_BUSLOGIC_946 0x1040104Bul static char* bt_pci_probe __P((pcici_t tag, pcidi_t type)); static void bt_pci_attach __P((pcici_t config_id, int unit)); *************** *** 70,86 **** pcici_t config_id; int unit; { u_long io_port; unsigned opri = 0; struct bt_data *bt; ! if(!(io_port = pci_conf_read(config_id, PCI_BASEADR0))) return; - /* - * The first bit of PCI_BASEADR0 is always - * set hence we mask it off. - */ - io_port &= 0xfffffffe; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ --- 69,89 ---- pcici_t config_id; int unit; { + u_char reg; u_long io_port; unsigned opri = 0; struct bt_data *bt; ! for(reg = PCI_MAP_REG_START; reg < PCI_MAP_REG_END; reg+=4) { ! io_port = pci_conf_read(config_id, reg); ! if ((io_port&~7)==0) continue; ! if(io_port & PCI_MAP_IO) { ! io_port &= ~PCI_MAP_IO; ! break; ! } ! } ! if(reg == PCI_MAP_REG_END) return; if(!(bt = bt_alloc(unit, io_port))) return; /* XXX PCI code should take return status */ From owner-freebsd-stable Tue Feb 13 21:33:05 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA09776 for stable-outgoing; Tue, 13 Feb 1996 21:33:05 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA09769 Tue, 13 Feb 1996 21:33:02 -0800 (PST) Message-Id: <199602140533.VAA09769@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: se@zpr.uni-koeln.de (Stefan Esser) cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Tue, 13 Feb 1996 17:30:13 +0100." <199602131630.AA09319@Sysiphos> Date: Tue, 13 Feb 1996 21:33:01 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >} switch (data&7) { >} >} case 1: >} case 5: >} printf (" map(%x): io(%lx)\n", >} reg, data & ~3); >} break; >} >} I guess that everything must be quad-word aligned so that is why you >} & ~3? Is this in the spec even for I/O ports? > >In PCI there are a number of map types, and those are >encoded in those (assumed to be '0') bits. I.e. the type >1 encoding has a '01' in the two low order bits, which >have to be masked out ... Actually, your case statement ensures that the second bit is already 0. Perhaps ~PCI_MAP_IO would make the code clearer? >The problem with the wrong size being reported is caused >by the fact, that the bt946c has only a 16 bit port base >register. And for this reason, it returns 0x0000fffd if >a 0xffffffff had been written ... > >The 'd' is a 'c' (i.e. -4) plus the mapping type 1 code. >Since the PCI spec defines type 1 as a 32bit I/O port >mapping (IIRC) the 946 really should return 0xfffffffd. >It has no right to ignore the upper 16bits. > >But since ISA only supports 16 bits of port address >anyway, it doesn't appear to make much of a difference >to make the code always ignore those upper half for >port mappings ... Its not bits of port address, but bits of port address size right? I think that the Adaptec cards map their I/O space much higer than that. Regardless, I'd hate to see us do the "wrong" thing in our PCI code just because of one broken card. We should special case the Buslogic as a 'rogue' PCI card and deal with it that way since the PCI ID makes it easy to do this. >Regards, STefan >-- > Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 > Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 > ============================================================================= >= > http://www.zpr.uni-koeln.de/~se -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Tue Feb 13 21:34:17 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA09916 for stable-outgoing; Tue, 13 Feb 1996 21:34:17 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA09907 Tue, 13 Feb 1996 21:34:16 -0800 (PST) Message-Id: <199602140534.VAA09907@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: "Rodney W. Grimes" cc: se@ZPR.Uni-Koeln.DE (Stefan Esser), rjk@sparcmill.grauel.com, freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Tue, 13 Feb 1996 09:28:36 PST." <199602131728.JAA02469@GndRsh.aac.dev.com> Date: Tue, 13 Feb 1996 21:34:15 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >And if you have a revision E card and turn off the ISA ``assign standard I/O >port'' in the bios it will come up with something in the more normal PCI >I/O address range. > Then the generic bt driver is too strict about its I/O address which is the last of the problems with the PCI probe(I hope). >-- >Rod Grimes rgrimes@gndrsh.aac.dev.com >Accurate Automation Company Reliable computers for FreeBSD -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Wed Feb 14 07:56:12 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id HAA20354 for stable-outgoing; Wed, 14 Feb 1996 07:56:12 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA20346 for ; Wed, 14 Feb 1996 07:56:08 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id KAA25148; Wed, 14 Feb 1996 10:56:00 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id LAA02938; Wed, 14 Feb 1996 11:01:23 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA00009; Wed, 14 Feb 1996 10:56:22 +0500 Date: Wed, 14 Feb 1996 10:56:22 +0500 Message-Id: <9602141556.AA00009@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602140522.VAA08632@freefall.freebsd.org> References: <9602131927.AA23929@sparcmill.grauel.com> <199602140522.VAA08632@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > >That works fine for any ``working'' kernel; apparently none of the boot > >messages from a non-working kernel end up in the circular buffer dmesg > >uses. IE, if I boot-working then boot-nonworking then boot-working and run > >dmesg, I get the messages from the 2 boot-working kernels only. > > How would you expect the dmesg buffer to be written to disk if the > probe for your disk controller isn't successful? 8-) > I didn't think the dmesg buffer was ever written to disk -- I thought it was just a circular in-core deal. I don't recall ever seeing messages from an earlier boot after a power cycle, anyhow. > > > When you get a chance, try this one. Now that I'm not editing over a link > > > with 50% packet loss, its much easier to check my work. :) > > > > > > >I tried it; I'm afraid it made no difference. By the way, here's one more > >line from the (failing) pci-probe: > > > >bt0: uses 4 bytes of I/O space from 6000 upto 6003 > > Ahh. It did make a difference. Unfortunately, when I was cleaning > up the bt driver, I assumed that the PCI probe would always return > one of the ISA compatibility addresses. As Rod pointed out, this is > not always the case. This patch should do the trick > Thanks... I'll try it over lunch and let you know how it goes down... -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Wed Feb 14 08:11:01 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA21608 for stable-outgoing; Wed, 14 Feb 1996 08:11:01 -0800 (PST) Received: from shellx.best.com (root@shellx.best.com [206.86.0.11]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA21603 Wed, 14 Feb 1996 08:10:56 -0800 (PST) Received: from free.emmert.com (emmert.vip.best.com [204.156.142.218]) by shellx.best.com (8.6.12/8.6.5) with SMTP id IAA24846; Wed, 14 Feb 1996 08:04:13 -0800 Date: Wed, 14 Feb 1996 00:05:22 -0800 (PST) From: Steve Emmert To: "Frank E. Terhaar-Yonkers" cc: freebsd-current@FreeBSD.org, freebsd-stable@FreeBSD.org Subject: Re: bogon in -stable In-Reply-To: <199602141346.IAA18414@robin.mcnc.org.mcnc.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org Precedence: bulk On Wed, 14 Feb 1996, Frank E. Terhaar-Yonkers wrote: > As of Monday night the kernel in -stable won't compile. lpt.c has changed > a couple of times since the weekend. > > - Frank > > cc -c -O -pipe -m486 -W -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit > -nostdinc -I. -I../.. -I../../sys -I../../../include -DLOCO -DI486_CPU > -DI386_CPU -DXSERVER -DDODUMP -DDDB -DSYSVMSG -DSYSVSEM -DSYSVSHM -DNCONS=4 > -DUCONSOLE -DTUNE_1542 -DSCSI_REPORT_GEOMETRY -DSCSI_DELAY=10 -DCOMPAT_43 > -DKTRACE -DIPACCT -DARP_PROXYALL -DGATEWAY -DFFS -DINET -DKERNEL -Di386 > -DLOAD_ADDRESS=0xF0100000 ../../i386/isa/lpt.c > > ../../i386/isa/lpt.c: In function `lpattach': > ../../i386/isa/lpt.c:892: warning: passing arg 1 of `bpfattach' from incompatible pointer type > ../../i386/isa/lpt.c:892: too few arguments to function `bpfattach' > ../../i386/isa/lpt.c: In function `lpintr': > ../../i386/isa/lpt.c:1159: warning: passing arg 1 of `bpf_tap' from incompatible pointer type > ../../i386/isa/lpt.c: In function `lpoutput': > ../../i386/isa/lpt.c:1361: warning: passing arg 1 of `bpf_mtap' from incompatible pointer type > *** Error code 1 > > Stop. > > \\\\////\\\\////\\\\\////\\\\\////\\\\////\\\\////\\\\////\\\\////\\\\////\\\\ > Frank Terhaar-Yonkers > High Performance Computing and Communications Research > MCNC > PO Box 12889 3021 Cornwallis Road > Research Triangle Park, North Carolina 27709-2889 > fty@mcnc.org voice (919)248-1417 FAX (919)248-1455 > > http://www.mcnc.org/hpcc.html I had the same problem, I removed the bpf line in my kernel config: pseudo-device bpfilter 4 #Berkeley packet filter This allowed the kernel to compile --Steve Steve Emmert steve@emmert.com ------------------------------------------------------------------------- Powered by FreeBSD -- The Daemon is FREE!!! From owner-freebsd-stable Wed Feb 14 08:33:00 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA25932 for stable-outgoing; Wed, 14 Feb 1996 08:33:00 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA25909 for ; Wed, 14 Feb 1996 08:32:56 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id LAA25351; Wed, 14 Feb 1996 11:32:46 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id LAA00233; Wed, 14 Feb 1996 11:37:51 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA00264; Wed, 14 Feb 1996 11:33:20 +0500 Date: Wed, 14 Feb 1996 11:33:20 +0500 Message-Id: <9602141633.AA00264@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602140522.VAA08632@freefall.freebsd.org> References: <9602131927.AA23929@sparcmill.grauel.com> <199602140522.VAA08632@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > Ahh. It did make a difference. Unfortunately, when I was cleaning > up the bt driver, I assumed that the PCI probe would always return > one of the ISA compatibility addresses. As Rod pointed out, this is > not always the case. This patch should do the trick > ... and there was great rejoicing. Here's the dmesg output; I find it interesting that the isa probe code apparently still tries to look for a bt card. Thanks a lot. -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 == FreeBSD 2.1-STABLE #0: Wed Feb 14 11:25:54 EST 1996 rjk@watson.grauel.com:/prog/2.1-STABLE/src/sys/compile/BTTEST CPU: 100-MHz Pentium 815\\100 (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x525 Stepping=5 Features=0x1bf real memory = 33554432 (32768K bytes) avail memory = 30728192 (30008K bytes) pcibus_setup(1): mode1res=0x80000000 (0x80000000), mode2res=0xff (0x0e) pcibus_setup(2): mode1res=0x80000000 (0x80000000) pcibus_check: device 0 is there (id=04061039) Probing for devices on PCI bus 0: configuration mode 1 allows 32 devices. chip0 rev 0 on pci0:0 chip1 rev 0 on pci0:1 bt0 rev 0 int a irq 10 on pci0:13 mapreg[10] type=1 addr=00006000 size=ffff0004. bt0: Bt946C/ 0-(32bit) bus bt0: reading board settings, busmastering, int=10 bt0: version 4.25J, fast sync, parity, 32 mbxs, 32 ccbs bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15 bt0: targ 5 sync rate= 5.00MB/s(200ns), offset=11 bt0: Using Strict Round robin scheme bt0 waiting for scsi devices to settle (bt0:0:0): "MICROP 3243-19MZ Q4D HT02" type 0 fixed SCSI 2 sd0(bt0:0:0): Direct-Access 4095MB (8388315 512 byte sectors) sd0(bt0:0:0): with 3956 cyls, 19 heads, and an average 111 sectors/track (bt0:5:0): "EXABYTE EXB-82058VQANXR1 0790" type 1 removable SCSI 2 st0(bt0:5:0): Sequential-Access density code 0x0, drive empty (bt0:6:0): "Chinon CD-ROM CDS-525 V09" type 5 removable SCSI 2 cd0(bt0:6:0): CD-ROM cd0(bt0:6:0): UNIT ATTENTION asc:2a,0 Parameters changed cd present.[400000 x 2048 byte records] pci0: uses 4 bytes of I/O space from 6000 upto 6003. Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 at 0x280-0x29f irq 5 maddr 0xd8000 msize 16384 on isa ed0: address 00:00:c0:15:00:ab, type SMC8216/SMC8216C (16 bit) sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface pca0 on motherboard pca0: PC speaker audio driver fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in bt: unit number (1) too high bt0 not found at 0x330 npx0 on motherboard npx0: INT 16 interface changing root device to sd0a BIOS Geometries: 0:03ff3f20 0..1023=1024 cylinders, 0..63=64 heads, 1..32=32 sectors 0 accounted for sd0s1: type 0xa5, start 0, end = 8388314, size 8388315 : OK From owner-freebsd-stable Wed Feb 14 11:14:14 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA07561 for stable-outgoing; Wed, 14 Feb 1996 11:14:14 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA07554 Wed, 14 Feb 1996 11:14:11 -0800 (PST) Message-Id: <199602141914.LAA07554@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Wed, 14 Feb 1996 11:33:20 +0500." <9602141633.AA00264@sparcmill.grauel.com> Date: Wed, 14 Feb 1996 11:14:11 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >Justin T. Gibbs writes: > > > Ahh. It did make a difference. Unfortunately, when I was cleaning > > up the bt driver, I assumed that the PCI probe would always return > > one of the ISA compatibility addresses. As Rod pointed out, this is > > not always the case. This patch should do the trick > > > >... and there was great rejoicing. Here's the dmesg output; I find it >interesting that the isa probe code apparently still tries to look for a bt >card. Having a line in your config file like: controller bt0 Will give you a pci and eisa probe so long as the pci and eisa busses are defined in your config file. An entry like: controller bt0 at isa? ... Gives you an ISA/EISA and PCI probe and so its what is in our GENERIC kernel config file. You can prevent the ISA probe by using a config line without the "at isa?" part. One thing that would be interesting to know is if your card can be probed twice (ie found by the ISA probe after a successfull PCI probe). Can you add a "controller bt1 at isa? ..." entry into your config file in addition to the one you have and see what happens? The probe won't even attempt to look at the card if its out of unit numbers, so the ouput you give below isn't enough to prove that we can't double probe the card. >Thanks a lot. Sorry it took so many itterations. >-- >Rich Kuhns rjk@grauel.com >PO Box 6249 >100 Sawmill Road >Lafayette, IN 47903 >(317)477-6000 x319 > >== >FreeBSD 2.1-STABLE #0: Wed Feb 14 11:25:54 EST 1996 > rjk@watson.grauel.com:/prog/2.1-STABLE/src/sys/compile/BTTEST >CPU: 100-MHz Pentium 815\\100 (Pentium-class CPU) > Origin = "GenuineIntel" Id = 0x525 Stepping=5 > Features=0x1bf >real memory = 33554432 (32768K bytes) >avail memory = 30728192 (30008K bytes) >pcibus_setup(1): mode1res=0x80000000 (0x80000000), mode2res=0xff (0x0e) >pcibus_setup(2): mode1res=0x80000000 (0x80000000) >pcibus_check: device 0 is there (id=04061039) >Probing for devices on PCI bus 0: > configuration mode 1 allows 32 devices. >chip0 rev 0 on pci0:0 >chip1 rev 0 on pci0:1 >bt0 rev 0 int a irq 10 on pci0:13 > mapreg[10] type=1 addr=00006000 size=ffff0004. >bt0: Bt946C/ 0-(32bit) bus >bt0: reading board settings, busmastering, int=10 >bt0: version 4.25J, fast sync, parity, 32 mbxs, 32 ccbs >bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15 >bt0: targ 5 sync rate= 5.00MB/s(200ns), offset=11 >bt0: Using Strict Round robin scheme >bt0 waiting for scsi devices to settle >(bt0:0:0): "MICROP 3243-19MZ Q4D HT02" type 0 fixed SCSI 2 >sd0(bt0:0:0): Direct-Access 4095MB (8388315 512 byte sectors) >sd0(bt0:0:0): with 3956 cyls, 19 heads, and an average 111 sectors/track >(bt0:5:0): "EXABYTE EXB-82058VQANXR1 0790" type 1 removable SCSI 2 >st0(bt0:5:0): Sequential-Access density code 0x0, drive empty >(bt0:6:0): "Chinon CD-ROM CDS-525 V09" type 5 removable SCSI 2 >cd0(bt0:6:0): CD-ROM >cd0(bt0:6:0): UNIT ATTENTION asc:2a,0 Parameters changed >cd present.[400000 x 2048 byte records] >pci0: uses 4 bytes of I/O space from 6000 upto 6003. >Probing for devices on the ISA bus: >sc0 at 0x60-0x6f irq 1 on motherboard >sc0: VGA color <16 virtual consoles, flags=0x0> >ed0 at 0x280-0x29f irq 5 maddr 0xd8000 msize 16384 on isa >ed0: address 00:00:c0:15:00:ab, type SMC8216/SMC8216C (16 bit) >sio0 at 0x3f8-0x3ff irq 4 on isa >sio0: type 16550A >sio1 at 0x2f8-0x2ff irq 3 on isa >sio1: type 16550A >lpt0 at 0x378-0x37f irq 7 on isa >lpt0: Interrupt-driven port >lp0: TCP/IP capable interface >pca0 on motherboard >pca0: PC speaker audio driver >fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa >fdc0: NEC 72065B >fd0: 1.44MB 3.5in >bt: unit number (1) too high >bt0 not found at 0x330 >npx0 on motherboard >npx0: INT 16 interface >changing root device to sd0a >BIOS Geometries: > 0:03ff3f20 0..1023=1024 cylinders, 0..63=64 heads, 1..32=32 sectors > 0 accounted for >sd0s1: type 0xa5, start 0, end = 8388314, size 8388315 : OK -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Wed Feb 14 19:34:58 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA06466 for stable-outgoing; Wed, 14 Feb 1996 19:34:58 -0800 (PST) Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id TAA06457 for ; Wed, 14 Feb 1996 19:34:52 -0800 (PST) Received: from tahoma.cwu.edu (skynyrd@tahoma.cwu.edu [198.104.67.25]) by pahtoh.cwu.edu (8.6.12/8.6.9) with ESMTP id TAA11912; Wed, 14 Feb 1996 19:34:42 -0800 Received: (from skynyrd@localhost) by tahoma.cwu.edu (8.6.12/8.6.9) id TAA02470; Wed, 14 Feb 1996 19:34:41 -0800 Date: Wed, 14 Feb 1996 19:34:40 -0800 (PST) From: Chris Timmons To: Steve Emmert cc: "Frank E. Terhaar-Yonkers" , freebsd-stable@FreeBSD.org Subject: Re: bogon in -stable In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org Precedence: bulk I ran into this and discovered that if you do not say 'make depend' ahead of your 'make' in the kernel compile directory, it will build. When I was flailing I looked in the handbook section about kernel compiling and it didn't have the 'make depend' in there, so I tried it that way and I've been running fine... Except that I've encountered the ld.so failed: Undefined symbol "_encrypt_debug_mode" in telnetd:telnetd problem now... I am finding stuff in the archives about it but no cure so far. Any hints? -Chris On Wed, 14 Feb 1996, Steve Emmert wrote: > On Wed, 14 Feb 1996, Frank E. Terhaar-Yonkers wrote: > > > As of Monday night the kernel in -stable won't compile. lpt.c has changed > > a couple of times since the weekend. > > > > - Frank > > > > cc -c -O -pipe -m486 -W -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit > > -nostdinc -I. -I../.. -I../../sys -I../../../include -DLOCO -DI486_CPU > > -DI386_CPU -DXSERVER -DDODUMP -DDDB -DSYSVMSG -DSYSVSEM -DSYSVSHM -DNCONS=4 > > -DUCONSOLE -DTUNE_1542 -DSCSI_REPORT_GEOMETRY -DSCSI_DELAY=10 -DCOMPAT_43 > > -DKTRACE -DIPACCT -DARP_PROXYALL -DGATEWAY -DFFS -DINET -DKERNEL -Di386 > > -DLOAD_ADDRESS=0xF0100000 ../../i386/isa/lpt.c > > > > ../../i386/isa/lpt.c: In function `lpattach': > > ../../i386/isa/lpt.c:892: warning: passing arg 1 of `bpfattach' from incompatible pointer type > > ../../i386/isa/lpt.c:892: too few arguments to function `bpfattach' > > ../../i386/isa/lpt.c: In function `lpintr': > > ../../i386/isa/lpt.c:1159: warning: passing arg 1 of `bpf_tap' from incompatible pointer type > > ../../i386/isa/lpt.c: In function `lpoutput': > > ../../i386/isa/lpt.c:1361: warning: passing arg 1 of `bpf_mtap' from incompatible pointer type > > *** Error code 1 > > > > Stop. > > > > \\\\////\\\\////\\\\\////\\\\\////\\\\////\\\\////\\\\////\\\\////\\\\////\\\\ > > Frank Terhaar-Yonkers > > High Performance Computing and Communications Research > > MCNC > > PO Box 12889 3021 Cornwallis Road > > Research Triangle Park, North Carolina 27709-2889 > > fty@mcnc.org voice (919)248-1417 FAX (919)248-1455 > > > > http://www.mcnc.org/hpcc.html > > I had the same problem, I removed the bpf line in my kernel config: > > pseudo-device bpfilter 4 #Berkeley packet filter > > This allowed the kernel to compile > > --Steve > > Steve Emmert > steve@emmert.com > ------------------------------------------------------------------------- > Powered by FreeBSD -- The Daemon is FREE!!! > From owner-freebsd-stable Wed Feb 14 19:54:00 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA07945 for stable-outgoing; Wed, 14 Feb 1996 19:54:00 -0800 (PST) Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id TAA07937 for ; Wed, 14 Feb 1996 19:53:55 -0800 (PST) Received: by haven.uniserve.com id <30957-2926>; Wed, 14 Feb 1996 19:56:30 -0800 Date: Wed, 14 Feb 1996 19:56:24 -0800 (PST) From: Tom Samplonius To: Chris Timmons cc: Steve Emmert , "Frank E. Terhaar-Yonkers" , freebsd-stable@FreeBSD.org Subject: Re: bogon in -stable In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org Precedence: bulk On Wed, 14 Feb 1996, Chris Timmons wrote: > > I ran into this and discovered that if you do not say 'make depend' ahead > of your 'make' in the kernel compile directory, it will build. When I > was flailing I looked in the handbook section about kernel compiling and > it didn't have the 'make depend' in there, so I tried it that way and > I've been running fine... > > Except that I've encountered the > > ld.so failed: Undefined symbol "_encrypt_debug_mode" in telnetd:telnetd > > problem now... I am finding stuff in the archives about it but no cure so > far. Any hints? That should only happen during a "make world" while shared libraries are being built. It should be fixed before "make world" ends. Tom From owner-freebsd-stable Thu Feb 15 05:43:03 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id FAA00664 for stable-outgoing; Thu, 15 Feb 1996 05:43:03 -0800 (PST) Received: from moonpie.w8hd.org (moonpie.w8hd.org [198.252.159.14]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id FAA00659 for ; Thu, 15 Feb 1996 05:43:01 -0800 (PST) Received: (from kimc@localhost) by moonpie.w8hd.org (8.6.12/8.6.12) id IAA19028; Thu, 15 Feb 1996 08:42:43 -0500 Date: Thu, 15 Feb 1996 08:42:43 -0500 (EST) From: Kim Culhan To: stable@freebsd.org Subject: latest stable kernel build failure Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org Precedence: bulk Fails with this: EL -Di386 -DLOAD_ADDRESS=0xF0100000 ../../i386/isa/lpt.c ../../i386/isa/lpt.c: In function `lpattach': ../../i386/isa/lpt.c:892: warning: passing arg 1 of `bpfattach' from incompatible pointer type ../../i386/isa/lpt.c:892: too few arguments to function `bpfattach' ../../i386/isa/lpt.c: In function `lpintr': ../../i386/isa/lpt.c:1159: warning: passing arg 1 of `bpf_tap' from incompatible pointer type ../../i386/isa/lpt.c: In function `lpoutput': ../../i386/isa/lpt.c:1361: warning: passing arg 1 of `bpf_mtap' from incompatible pointer type lpt.c is: -rw-r--r-- 1 root wheel 34524 Feb 12 16:30 lpt.c regards kim -- kimc@w8hd.org From owner-freebsd-stable Thu Feb 15 06:58:35 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA02801 for stable-outgoing; Thu, 15 Feb 1996 06:58:35 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA02795 for ; Thu, 15 Feb 1996 06:58:32 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id JAA29907; Thu, 15 Feb 1996 09:58:29 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id KAA00270; Thu, 15 Feb 1996 10:03:40 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA08224; Thu, 15 Feb 1996 09:58:36 +0500 Date: Thu, 15 Feb 1996 09:58:36 +0500 Message-Id: <9602151458.AA08224@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: Richard J Kuhns , freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602141914.LAA07554@freefall.freebsd.org> References: <9602141633.AA00264@sparcmill.grauel.com> <199602141914.LAA07554@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > > controller bt0 > > Will give you a pci and eisa probe so long as the pci and eisa busses > are defined in your config file. An entry like: > > controller bt0 at isa? ... > > Gives you an ISA/EISA and PCI probe and so its what is in our GENERIC > kernel config file. You can prevent the ISA probe by using a config > line without the "at isa?" part. > Thanks -- that worked. > One thing that would be interesting to know is if your card can > be probed twice (ie found by the ISA probe after a successfull > PCI probe). Can you add a "controller bt1 at isa? ..." entry into > your config file in addition to the one you have and see what happens? > The probe won't even attempt to look at the card if its out of unit numbers, > so the ouput you give below isn't enough to prove that we can't > double probe the card. > I'm afraid the card IS double probed. I built and booted a kernel with the following in the config file: controller bt0 controller bt1 at isa? port "IO_BT0" bio irq ? vector bt_isa_intr The PCI probe found the card first as bt0; the ISA probe found it later as bt1. In the interest of minimizing possible damage to my file systems, I'm afraid I hit the reset button while the bt1 portion was ``waiting for scsi devices to settle'. -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319 From owner-freebsd-stable Thu Feb 15 07:23:19 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id HAA03887 for stable-outgoing; Thu, 15 Feb 1996 07:23:19 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA03872 for ; Thu, 15 Feb 1996 07:23:06 -0800 (PST) Received: by Sysiphos id AA02983 (5.67b/IDA-1.5 for stable@freebsd.org); Thu, 15 Feb 1996 16:22:46 +0100 Message-Id: <199602151522.AA02983@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Thu, 15 Feb 1996 16:22:45 +0100 In-Reply-To: Richard J Kuhns "Re: One problem && one question" (Feb 15, 9:58) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Richard J Kuhns Subject: Re: One problem && one question Cc: stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk On Feb 15, 9:58, Richard J Kuhns wrote: } Subject: Re: One problem && one question } Justin T. Gibbs writes: } I'm afraid the card IS double probed. } I built and booted a kernel with the following in the config file: } } controller bt0 } controller bt1 at isa? port "IO_BT0" bio irq ? vector bt_isa_intr } } The PCI probe found the card first as bt0; the ISA probe found it later as } bt1. In the interest of minimizing possible damage to my file systems, I'm } afraid I hit the reset button while the bt1 portion was ``waiting for scsi } devices to settle'. We definitely need conflict check code that covers all bus types ... (And I have part of that ready.) But the BusLogic would still be double probed, I suppose, since it responds to 0x330 ISA probes but doesn't announce that port range as active to the PCI driver. This is a very bad thing, and the card seems to severely violate the PCI specs ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-stable Thu Feb 15 08:17:24 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA06578 for stable-outgoing; Thu, 15 Feb 1996 08:17:24 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA06573 for ; Thu, 15 Feb 1996 08:17:20 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id DAA07375 for stable@freebsd.org; Fri, 16 Feb 1996 03:17:16 +1100 From: michael butler Message-Id: <199602151617.DAA07375@asstdc.scgt.oz.au> Subject: arp and async problems To: stable@freebsd.org Date: Fri, 16 Feb 1996 03:17:15 +1100 (EST) X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk I have two problems with -stable as it is today .. i) with whatever recently changed in the arp code, I now get a stream of error messages every 30 seconds when gated wants to advertise routes. Each one relates to an alias on the same ethernet card, e.g. .. arpresolve: can't allocate llinfo arplookup 202.12.127. failed: could not allocate llinfo .. considering how many aliases this machine has, it'll fill the disk before morning :-( ii) when running with the root in 'async' mode, I happened to start elm while the machine was otherwise idle .. it blocked in "ufslk2" and stayed there until I used "mount -u -o noasync /". This is the same machine that I have had repeated arguments with over this 2842 (with and without tagged queueing and "QUEUE_FULL_SUPPORT") .. maybe there's some interaction there ? michael From owner-freebsd-stable Thu Feb 15 08:41:47 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA08056 for stable-outgoing; Thu, 15 Feb 1996 08:41:47 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA08051 Thu, 15 Feb 1996 08:41:45 -0800 (PST) Message-Id: <199602151641.IAA08051@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: se@ZPR.Uni-Koeln.DE (Stefan Esser) cc: Richard J Kuhns , stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Thu, 15 Feb 1996 16:22:45 +0100." <199602151522.AA02983@Sysiphos> Date: Thu, 15 Feb 1996 08:41:45 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >We definitely need conflict check code >that covers all bus types ... >(And I have part of that ready.) Yeah!!! >But the BusLogic would still be double >probed, I suppose, since it responds to >0x330 ISA probes but doesn't announce >that port range as active to the PCI >driver. This is a very bad thing, and >the card seems to severely violate the >PCI specs ... I've come up with another solution. All probes call bt_init which can determine the board model. I've now changed bt_init to take a bus type argument and only succeed for cards of the same bus type were looking for. Test patch in a little bit... >Regards, STefan >-- > Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 > Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 > ============================================================================= >= > http://www.zpr.uni-koeln.de/~se -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Thu Feb 15 09:12:02 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA09468 for stable-outgoing; Thu, 15 Feb 1996 09:12:02 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA09439 for ; Thu, 15 Feb 1996 09:11:51 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id KAA29800; Thu, 15 Feb 1996 10:14:22 -0700 Date: Thu, 15 Feb 1996 10:14:22 -0700 From: Nate Williams Message-Id: <199602151714.KAA29800@rocky.sri.MT.net> To: michael butler Cc: stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151617.DAA07375@asstdc.scgt.oz.au> References: <199602151617.DAA07375@asstdc.scgt.oz.au> Sender: owner-stable@freebsd.org Precedence: bulk > i) with whatever recently changed in the arp code, I now get a stream of > error messages every 30 seconds when gated wants to advertise routes. > > Each one relates to an alias on the same ethernet card, e.g. .. > > arpresolve: can't allocate llinfo > arplookup 202.12.127. failed: could not allocate llinfo > > .. considering how many aliases this machine has, it'll fill the disk > before morning :-( I don't think the arp changes Bill did could cause this, since it only affect host routes, which alias aren't. Nate From owner-freebsd-stable Thu Feb 15 09:23:53 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA10158 for stable-outgoing; Thu, 15 Feb 1996 09:23:53 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA10153 for ; Thu, 15 Feb 1996 09:23:48 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id EAA08635; Fri, 16 Feb 1996 04:23:31 +1100 From: michael butler Message-Id: <199602151723.EAA08635@asstdc.scgt.oz.au> Subject: Re: arp and async problems To: nate@sri.MT.net (Nate Williams) Date: Fri, 16 Feb 1996 04:23:28 +1100 (EST) Cc: stable@freebsd.org In-Reply-To: <199602151714.KAA29800@rocky.sri.MT.net> from "Nate Williams" at Feb 15, 96 10:14:22 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk Nate Williams writes: > > i) with whatever recently changed in the arp code, I now get a stream of > > error messages every 30 seconds when gated wants to advertise routes. > > Each one relates to an alias on the same ethernet card, e.g. .. > I don't think the arp changes Bill did could cause this, since it only > affect host routes, which alias aren't. The only other relevant info is: i) it runs an Etherlink III (stock -stable driver as of 2 hours ago, no added patches) ii) the errors do not occur with 2.1-release (to which I've now reverted), iii) the aliases are added in /etc/netstart immediately after the initial ifconfig with .. ifconfig ep0 inet 202.12.127. alias netmask 255.255.255.255 .. none of them fall within the primary network's subnet (i.e. it primary address is .70, it has aliases from 225 through 238 of which only the first six are accessible to the outside world). iv) attempts at installing a proxy arp ("arp -s
pub") fails with "not an 802 interface" or something similar. Leaving off the "pub" yields something like "cannot set arp for
"), michael From owner-freebsd-stable Thu Feb 15 09:31:59 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA10612 for stable-outgoing; Thu, 15 Feb 1996 09:31:59 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA10606 for ; Thu, 15 Feb 1996 09:31:56 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id KAA29970; Thu, 15 Feb 1996 10:34:34 -0700 Date: Thu, 15 Feb 1996 10:34:34 -0700 From: Nate Williams Message-Id: <199602151734.KAA29970@rocky.sri.MT.net> To: michael butler Cc: nate@sri.MT.net (Nate Williams), stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151723.EAA08635@asstdc.scgt.oz.au> References: <199602151714.KAA29800@rocky.sri.MT.net> <199602151723.EAA08635@asstdc.scgt.oz.au> Sender: owner-stable@freebsd.org Precedence: bulk > > > i) with whatever recently changed in the arp code, I now get a stream of > > > error messages every 30 seconds when gated wants to advertise routes. Garrett usually responds to folks having problems with this, but... > iii) the aliases are added in /etc/netstart immediately after the initial > ifconfig with .. > > ifconfig ep0 inet 202.12.127. alias netmask 255.255.255.255 > > .. none of them fall within the primary network's subnet (i.e. it primary > address is .70, it has aliases from 225 through 238 of which only the first > six are accessible to the outside world). Um, is this legal? I don't think so. > iv) attempts at installing a proxy arp ("arp -s
pub") fails > with "not an 802 interface" or something similar. Leaving off the "pub" > yields something like "cannot set arp for
"), Which is because those addresses don't have a subnet which arp can publish the entries on. I think you are doing something that worked in 2.1, but was illegal and is no longer allowed. This comment from /sys/netinet/in.c might be of interest: revision 1.14.4.2 date: 1996/01/30 21:41:34; author: davidg; state: Exp; lines: +8 -2 Brought in change from rev 1.21: fix for aliases & all-ones netmasks. Nate From owner-freebsd-stable Thu Feb 15 09:57:59 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA12077 for stable-outgoing; Thu, 15 Feb 1996 09:57:59 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA12069 for ; Thu, 15 Feb 1996 09:57:48 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id EAA09376; Fri, 16 Feb 1996 04:57:33 +1100 From: michael butler Message-Id: <199602151757.EAA09376@asstdc.scgt.oz.au> Subject: Re: arp and async problems To: nate@sri.MT.net (Nate Williams) Date: Fri, 16 Feb 1996 04:57:31 +1100 (EST) Cc: nate@sri.MT.net, stable@freebsd.org In-Reply-To: <199602151734.KAA29970@rocky.sri.MT.net> from "Nate Williams" at Feb 15, 96 10:34:34 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk Nate Williams writes: > I think you are doing something that worked in 2.1, but was illegal and > is no longer allowed. Oh, terrific, dandy, *&#@$%* .. > This comment from /sys/netinet/in.c might be of interest: > revision 1.14.4.2 > date: 1996/01/30 21:41:34; author: davidg; state: Exp; lines: +8 -2 > Brought in change from rev 1.21: fix for aliases & all-ones netmasks. I'd be interested in the impact on the functionality of named, if any. Some of these aliases are primary DNS server addresses (not to mention WWW "virtual hosts") which is going to make "interesting" to change .. (read as "politically unwise" to even attempt). There is no way I can afford to split out 14 subnets with only one address in each to support a DNS and WWW server nor is it practical to stick that many ethernet cards into the box. If this is to be a permanent feature that's one less box with which I can continue to test the stuff. It will just have to stay at it's current release level .. michael From owner-freebsd-stable Thu Feb 15 10:08:51 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA12809 for stable-outgoing; Thu, 15 Feb 1996 10:08:51 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA12801 for ; Thu, 15 Feb 1996 10:08:45 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id LAA00327; Thu, 15 Feb 1996 11:11:21 -0700 Date: Thu, 15 Feb 1996 11:11:21 -0700 From: Nate Williams Message-Id: <199602151811.LAA00327@rocky.sri.MT.net> To: michael butler Cc: nate@sri.MT.net (Nate Williams), stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151757.EAA09376@asstdc.scgt.oz.au> References: <199602151734.KAA29970@rocky.sri.MT.net> <199602151757.EAA09376@asstdc.scgt.oz.au> Sender: owner-stable@freebsd.org Precedence: bulk michael butler writes: > Nate Williams writes: > > > I think you are doing something that worked in 2.1, but was illegal and > > is no longer allowed. > > Oh, terrific, dandy, *&#@$%* .. > > > This comment from /sys/netinet/in.c might be of interest: > > revision 1.14.4.2 > > date: 1996/01/30 21:41:34; author: davidg; state: Exp; lines: +8 -2 > > Brought in change from rev 1.21: fix for aliases & all-ones netmasks. > > I'd be interested in the impact on the functionality of named, if any. Some > of these aliases are primary DNS server addresses (not to mention WWW > "virtual hosts") which is going to make "interesting" to change .. (read as > "politically unwise" to even attempt). > > There is no way I can afford to split out 14 subnets with only one address > in each to support a DNS and WWW server nor is it practical to stick that > many ethernet cards into the box. Why split it out? Why not stick all of the alias' on the same network, so that the IP address are relevant on your primary network? > ifconfig ep0 inet 202.12.127. alias netmask 255.255.255.255 > .. none of them fall within the primary network's subnet (i.e. it > primary address is .70, it has aliases from 225 through 238 of which > only the first six are accessible to the outside world). This is illegal. > If this is to be a permanent feature that's one less box with which I can > continue to test the stuff. It will just have to stay at it's current > release level .. Or else you could setup the networking correctly. Again, I'm not sure I'm hearing you correctly, but it sounds like you are trying to use aliases which have no matching network address. Nate From owner-freebsd-stable Thu Feb 15 10:15:23 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA13229 for stable-outgoing; Thu, 15 Feb 1996 10:15:23 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA13214 for ; Thu, 15 Feb 1996 10:15:00 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id FAA09626; Fri, 16 Feb 1996 05:14:41 +1100 From: michael butler Message-Id: <199602151814.FAA09626@asstdc.scgt.oz.au> Subject: Re: arp and async problems To: nate@sri.MT.net (Nate Williams) Date: Fri, 16 Feb 1996 05:14:39 +1100 (EST) Cc: nate@sri.MT.net, stable@freebsd.org In-Reply-To: <199602151811.LAA00327@rocky.sri.MT.net> from "Nate Williams" at Feb 15, 96 11:11:21 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk Nate Williams writes: > > There is no way I can afford to split out 14 subnets with only one address > > in each to support a DNS and WWW server nor is it practical to stick that > > many ethernet cards into the box. > Why split it out? Why not stick all of the alias' on the same network, > so that the IP address are relevant on your primary network? So you're telling me that I can now use .. ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 ifconfig ep0 inet 202.12.127.226 alias netmask 255.255.255.240 ifconfig ep0 inet 202.12.127.227 alias netmask 255.255.255.240 : : .. this didn't work before (I tried it), does it now ? michael From owner-freebsd-stable Thu Feb 15 10:19:30 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA13453 for stable-outgoing; Thu, 15 Feb 1996 10:19:30 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA13447 for ; Thu, 15 Feb 1996 10:19:27 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id LAA00383; Thu, 15 Feb 1996 11:21:43 -0700 Date: Thu, 15 Feb 1996 11:21:43 -0700 From: Nate Williams Message-Id: <199602151821.LAA00383@rocky.sri.MT.net> To: michael butler Cc: nate@sri.MT.net (Nate Williams), stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151814.FAA09626@asstdc.scgt.oz.au> References: <199602151811.LAA00327@rocky.sri.MT.net> <199602151814.FAA09626@asstdc.scgt.oz.au> Sender: owner-stable@freebsd.org Precedence: bulk > > > There is no way I can afford to split out 14 subnets with only one address > > > in each to support a DNS and WWW server nor is it practical to stick that > > > many ethernet cards into the box. > > > Why split it out? Why not stick all of the alias' on the same network, > > so that the IP address are relevant on your primary network? > > So you're telling me that I can now use .. > > ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 > ifconfig ep0 inet 202.12.127.226 alias netmask 255.255.255.240 > ifconfig ep0 inet 202.12.127.227 alias netmask 255.255.255.240 > : > : > > .. this didn't work before (I tried it), does it now ? I don't believe so. What does 'ifconfig -a' return, and what alias's are you trying to use? nate From owner-freebsd-stable Thu Feb 15 10:53:20 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA15124 for stable-outgoing; Thu, 15 Feb 1996 10:53:20 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA15119 for ; Thu, 15 Feb 1996 10:53:15 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id FAA22591; Fri, 16 Feb 1996 05:53:05 +1100 From: michael butler Message-Id: <199602151853.FAA22591@asstdc.scgt.oz.au> Subject: Re: arp and async problems To: nate@sri.MT.net (Nate Williams) Date: Fri, 16 Feb 1996 05:53:05 +1100 (EST) Cc: nate@sri.MT.net, stable@freebsd.org In-Reply-To: <199602151821.LAA00383@rocky.sri.MT.net> from "Nate Williams" at Feb 15, 96 11:21:43 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk > > So you're telling me that I can now use .. > > ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 > > ifconfig ep0 inet 202.12.127.226 alias netmask 255.255.255.240 > > ifconfig ep0 inet 202.12.127.227 alias netmask 255.255.255.240 > > : > > : > > .. this didn't work before (I tried it), does it now ? > I don't believe so. Then there are few remaining options .. if the above doesn't work and a netmask of 255.255.255.255 doesn't work yet you say that I may not need to split them out .. sorry, now I am confused .. what _does_ work ? > What does 'ifconfig -a' return, and what alias's are you trying to use? Currently it's primary address is 202.12.127.70 with no aliases in that subnet (I simply don't have enough "spare" in this one). Under 2.1R (generic kernel), it says .. pcm:~ % /sbin/ifconfig -a ep0: flags=863 mtu 1500 inet 202.12.127.70 netmask 0xfffffff0 broadcast 202.12.127.79 lo0: flags=8009 mtu 16384 inet 127.0.0.1 netmask 0xff000000 sl0: flags=c010 mtu 552 tun0: flags=8010 mtu 1500 It is (naturally) the ep0 interface which presently carries all the other aliases. However, I could accomodate another ethernet card to carry the WWW and DNS stuff for the other aliases (currently, .225 through .238). What I (obviously) need to know is how this is to be constructed since the method that used to work no longer does. Am I really the only one who uses aliases this extensively ? michael From owner-freebsd-stable Thu Feb 15 11:06:39 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA15828 for stable-outgoing; Thu, 15 Feb 1996 11:06:39 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA15822 for ; Thu, 15 Feb 1996 11:06:28 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id MAA00761; Thu, 15 Feb 1996 12:08:29 -0700 Date: Thu, 15 Feb 1996 12:08:29 -0700 From: Nate Williams Message-Id: <199602151908.MAA00761@rocky.sri.MT.net> To: michael butler Cc: nate@sri.MT.net (Nate Williams), stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151853.FAA22591@asstdc.scgt.oz.au> References: <199602151821.LAA00383@rocky.sri.MT.net> <199602151853.FAA22591@asstdc.scgt.oz.au> Sender: owner-stable@freebsd.org Precedence: bulk > > What does 'ifconfig -a' return, and what alias's are you trying to use? > > Currently it's primary address is 202.12.127.70 with no aliases in that > subnet (I simply don't have enough "spare" in this one). Under 2.1R (generic > kernel), it says .. > > pcm:~ % /sbin/ifconfig -a > ep0: flags=863 mtu 1500 > inet 202.12.127.70 netmask 0xfffffff0 broadcast 202.12.127.79 You're limiting your subnet to machines in the 202.12.127.79 range, but yet you also want to use hosts with IP address like 227 and such. Why are you sub-netting when you still wants machines in the entire class C to use the same wire? Nate From owner-freebsd-stable Thu Feb 15 11:08:58 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA16009 for stable-outgoing; Thu, 15 Feb 1996 11:08:58 -0800 (PST) Received: from alpha.xerox.com (alpha.Xerox.COM [13.1.64.93]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA16004 for ; Thu, 15 Feb 1996 11:08:56 -0800 (PST) Received: from crevenia.parc.xerox.com ([13.2.116.11]) by alpha.xerox.com with SMTP id <15539(15)>; Thu, 15 Feb 1996 11:08:16 PST Received: from localhost ([127.0.0.1]) by crevenia.parc.xerox.com with SMTP id <177478>; Thu, 15 Feb 1996 11:08:03 -0800 X-Mailer: exmh version 1.6.4 10/10/95 To: Nate Williams cc: michael butler , stable@freebsd.org Subject: Re: arp and async problems In-reply-to: Your message of "Thu, 15 Feb 1996 09:14:22 PST." <199602151714.KAA29800@rocky.sri.MT.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 15 Feb 1996 11:07:59 PST From: Bill Fenner Message-Id: <96Feb15.110803pst.177478@crevenia.parc.xerox.com> Sender: owner-stable@freebsd.org Precedence: bulk In message <199602151714.KAA29800@rocky.sri.MT.net>you write: >I don't think the arp changes Bill did could cause this, since it only >affect host routes, which alias aren't. Agreed. I wonder, Michael, if you could drop back to the old revision that you were running, and see if gated sends out multiple copies of the same data? This seems almost like a bug in gated, or a bug in alias processing that was uncovered by other recent changes. Bill From owner-freebsd-stable Thu Feb 15 11:09:05 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA16028 for stable-outgoing; Thu, 15 Feb 1996 11:09:05 -0800 (PST) Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA16023 Thu, 15 Feb 1996 11:09:03 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.6.12/8.6.5) with SMTP id LAA07135; Thu, 15 Feb 1996 11:08:33 -0800 Message-Id: <199602151908.LAA07135@Root.COM> X-Authentication-Warning: implode.Root.COM: Host localhost didn't use HELO protocol To: michael butler cc: nate@sri.MT.net (Nate Williams), stable@freebsd.org, wollman@freebsd.org Subject: Re: arp and async problems In-reply-to: Your message of "Fri, 16 Feb 1996 04:57:31 +1100." <199602151757.EAA09376@asstdc.scgt.oz.au> From: David Greenman Reply-To: davidg@Root.COM Date: Thu, 15 Feb 1996 11:08:33 -0800 Sender: owner-stable@freebsd.org Precedence: bulk >Nate Williams writes: > >> I think you are doing something that worked in 2.1, but was illegal and >> is no longer allowed. > >Oh, terrific, dandy, *&#@$%* .. > >> This comment from /sys/netinet/in.c might be of interest: >> revision 1.14.4.2 >> date: 1996/01/30 21:41:34; author: davidg; state: Exp; lines: +8 -2 >> Brought in change from rev 1.21: fix for aliases & all-ones netmasks. > >I'd be interested in the impact on the functionality of named, if any. Some >of these aliases are primary DNS server addresses (not to mention WWW >"virtual hosts") which is going to make "interesting" to change .. (read as >"politically unwise" to even attempt). > >There is no way I can afford to split out 14 subnets with only one address >in each to support a DNS and WWW server nor is it practical to stick that >many ethernet cards into the box. > >If this is to be a permanent feature that's one less box with which I can >continue to test the stuff. It will just have to stay at it's current >release level .. Actually, I think you're either seeing a bug or you're doing something slightly wrong in the ifconfigs. It *should* be possible to have multiple IP aliases in one subnet. Garrett might be able to say more. -DG David Greenman Core Team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Thu Feb 15 11:12:56 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA16275 for stable-outgoing; Thu, 15 Feb 1996 11:12:56 -0800 (PST) Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA16262 Thu, 15 Feb 1996 11:12:48 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.6.12/8.6.5) with SMTP id LAA07155; Thu, 15 Feb 1996 11:12:53 -0800 Message-Id: <199602151912.LAA07155@Root.COM> X-Authentication-Warning: implode.Root.COM: Host localhost didn't use HELO protocol To: Nate Williams cc: michael butler , stable@freebsd.org, wollman@freebsd.org Subject: Re: arp and async problems In-reply-to: Your message of "Thu, 15 Feb 1996 11:21:43 MST." <199602151821.LAA00383@rocky.sri.MT.net> From: David Greenman Reply-To: davidg@Root.COM Date: Thu, 15 Feb 1996 11:12:52 -0800 Sender: owner-stable@freebsd.org Precedence: bulk >> > > There is no way I can afford to split out 14 subnets with only one address >> > > in each to support a DNS and WWW server nor is it practical to stick that >> > > many ethernet cards into the box. >> >> > Why split it out? Why not stick all of the alias' on the same network, >> > so that the IP address are relevant on your primary network? >> >> So you're telling me that I can now use .. >> >> ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 >> ifconfig ep0 inet 202.12.127.226 alias netmask 255.255.255.240 >> ifconfig ep0 inet 202.12.127.227 alias netmask 255.255.255.240 >> : >> : >> >> .. this didn't work before (I tried it), does it now ? I think the above should be: ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 ifconfig ep0 inet 202.12.127.226 netmask 255.255.255.255 alias ifconfig ep0 inet 202.12.127.227 netmask 255.255.255.255 alias -DG David Greenman Core Team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Thu Feb 15 11:12:57 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA16284 for stable-outgoing; Thu, 15 Feb 1996 11:12:57 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA16264 for ; Thu, 15 Feb 1996 11:12:51 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id GAA03067; Fri, 16 Feb 1996 06:12:23 +1100 From: michael butler Message-Id: <199602151912.GAA03067@asstdc.scgt.oz.au> Subject: Re: arp and async problems To: fenner@parc.xerox.com (Bill Fenner) Date: Fri, 16 Feb 1996 06:12:22 +1100 (EST) Cc: nate@sri.mt.net, stable@freebsd.org In-Reply-To: <96Feb15.110803pst.177478@crevenia.parc.xerox.com> from "Bill Fenner" at Feb 15, 96 11:07:59 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk > >I don't think the arp changes Bill did could cause this, since it only > >affect host routes, which alias aren't. > Agreed. I wonder, Michael, if you could drop back to the old revision > that you were running, and see if gated sends out multiple copies of the > same data? No, it seems to advertise (via RIP II) routes to each (primary and virtual) interface with the correct subnet mask. > This seems almost like a bug in gated, or a bug in alias processing that > was uncovered by other recent changes. routed triggers the same sequence of errors every 30 seconds even though it has no capability to advertise the VLSMs in RIP I, michael From owner-freebsd-stable Thu Feb 15 11:16:58 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA16416 for stable-outgoing; Thu, 15 Feb 1996 11:16:58 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA16410 for ; Thu, 15 Feb 1996 11:16:55 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id MAA00870; Thu, 15 Feb 1996 12:19:37 -0700 Date: Thu, 15 Feb 1996 12:19:37 -0700 From: Nate Williams Message-Id: <199602151919.MAA00870@rocky.sri.MT.net> To: davidg@Root.COM Cc: Nate Williams , michael butler , stable@freebsd.org Subject: Re: arp and async problems In-Reply-To: <199602151912.LAA07155@Root.COM> References: <199602151821.LAA00383@rocky.sri.MT.net> <199602151912.LAA07155@Root.COM> Sender: owner-stable@freebsd.org Precedence: bulk > ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 > ifconfig ep0 inet 202.12.127.226 netmask 255.255.255.255 alias > ifconfig ep0 inet 202.12.127.227 netmask 255.255.255.255 alias Since ep0 is also configured as follows: pcm:~ % /sbin/ifconfig -a ep0: flags=863 mtu 1500 inet 202.12.127.70 netmask 0xfffffff0 broadcast 202.12.127.79 Shouldn't the first line be: ifconfig ep0 inet 202.12.127.225 netmask 255.255.255.240 alias It seems that Michael is trying to use two 'virtual' networks with one ethernet card. Nate From owner-freebsd-stable Thu Feb 15 11:36:16 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA17367 for stable-outgoing; Thu, 15 Feb 1996 11:36:16 -0800 (PST) Received: from ormail.intel.com (ormail.intel.com [134.134.192.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA17358 for ; Thu, 15 Feb 1996 11:36:06 -0800 (PST) Received: from ichips.intel.com (ichips.intel.com [134.134.50.200]) by ormail.intel.com (8.6.12/8.6.12) with ESMTP id LAA03039; Thu, 15 Feb 1996 11:36:05 -0800 Received: from pdx202 by ichips.intel.com (8.7.1/jIII); Thu, 15 Feb 1996 11:36:01 -0800 (PST) Message-Id: <199602151936.LAA07562@ichips.intel.com> To: michael butler cc: stable@freebsd.org Subject: Re: arp and async problems In-reply-to: Your message of "Fri, 16 Feb 1996 03:17:15 PST." <199602151617.DAA07375@asstdc.scgt.oz.au> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Date: Thu, 15 Feb 1996 11:35:46 -0800 From: Wayne Scott Sender: owner-stable@freebsd.org Precedence: bulk > > I have two problems with -stable as it is today .. > > i) with whatever recently changed in the arp code, I now get a stream of > error messages every 30 seconds when gated wants to advertise routes. > > Each one relates to an alias on the same ethernet card, e.g. .. > > arpresolve: can't allocate llinfo > arplookup 202.12.127. failed: could not allocate llinfo > > .. considering how many aliases this machine has, it'll fill the disk > before morning :-( I have also seen a very similar problem, but I am not doing anything strange with aliases. I get the "arpresolve: can't allocate llinfo" whenever I try to mount a NFS disk. This is with a plain system running the -stable release from yesterday. Reverting to 2.1 fixes the problem. System: Pentium Pro 200/256k DEC 21040 based ethernet card (Cogent) Wayne Scott P6 Architecture wscott@ichips.intel.com Work #: (503) 264-4165 Disclaimer: All views expressed are my own opinions, and not necessarily those of Intel Corporation. From owner-freebsd-stable Thu Feb 15 13:12:33 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id NAA23374 for stable-outgoing; Thu, 15 Feb 1996 13:12:33 -0800 (PST) Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id NAA23367 for ; Thu, 15 Feb 1996 13:12:27 -0800 (PST) Received: from tahoma.cwu.edu (skynyrd@tahoma.cwu.edu [198.104.67.25]) by pahtoh.cwu.edu (8.6.12/8.6.9) with ESMTP id NAA15494 for ; Thu, 15 Feb 1996 13:11:21 -0800 Received: (from skynyrd@localhost) by tahoma.cwu.edu (8.6.12/8.6.9) id NAA04502; Thu, 15 Feb 1996 13:11:19 -0800 Date: Thu, 15 Feb 1996 13:11:18 -0800 (PST) From: Chris Timmons To: freebsd-stable@freebsd.org Subject: kernel build & lpt.c problem Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org Precedence: bulk In my earlier note I mentioned bypassing this problem by not doing a 'make depend' in the kernel build directory. I've since realized that I had pseudo-device bpf 4 instead of pseudo-device bpfilter 4 in my config file. by not doing the 'make depend' I wasn't getting the bpf support... just noticed that when I tried to use tcpdump for the first time since building -STABLE. changing to 'bpfilter' in the config file once again brings back the lpt.c problem, even w/o doing 'make depend' ../../i386/isa/lpt.c: In function `lpattach': ../../i386/isa/lpt.c:892: warning: passing arg 1 of `bpfattach' from incompatible pointer type ../../i386/isa/lpt.c:892: too few arguments to function `bpfattach' ../../i386/isa/lpt.c: In function `lpintr': ../../i386/isa/lpt.c:1159: warning: passing arg 1 of `bpf_tap' from incompatible pointer type ../../i386/isa/lpt.c: In function `lpoutput': ../../i386/isa/lpt.c:1361: warning: passing arg 1 of `bpf_mtap' from incompatible pointer type *** Error code 1 Stop. -Chris From owner-freebsd-stable Thu Feb 15 17:14:23 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id RAA09938 for stable-outgoing; Thu, 15 Feb 1996 17:14:23 -0800 (PST) Received: (from fenner@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id RAA09932 Thu, 15 Feb 1996 17:14:22 -0800 (PST) Date: Thu, 15 Feb 1996 17:14:22 -0800 (PST) From: Bill Fenner Message-Id: <199602160114.RAA09932@freefall.freebsd.org> To: stable@freebsd.org, wscott@ichips.intel.com Subject: Re: "arpresolve: can't allocate llinfo" Sender: owner-stable@freebsd.org Precedence: bulk Could anyone having trouble with seeing "arpresolve: can't allocate llinfo" please apply the following patch, to make the message at least a little more useful? The next step is to do a "route get x.x.x.x" on the address that gets printed and see what it says. Thanks, Bill cvs diff: Diffing . Index: if_ether.c =================================================================== RCS file: /home/ncvs/src/sys/netinet/if_ether.c,v retrieving revision 1.17.4.1 diff -c -r1.17.4.1 if_ether.c *** if_ether.c 1995/07/23 05:26:13 1.17.4.1 --- if_ether.c 1996/02/10 20:35:57 *************** *** 314,320 **** rt = la->la_rt; } if (la == 0 || rt == 0) { ! log(LOG_DEBUG, "arpresolve: can't allocate llinfo\n"); m_freem(m); return (0); } --- 314,321 ---- rt = la->la_rt; } if (la == 0 || rt == 0) { ! log(LOG_DEBUG, "arpresolve: can't allocate llinfo for %s\n", ! inet_ntoa(SIN(dst)->sin_addr)); m_freem(m); return (0); } From owner-freebsd-stable Fri Feb 16 00:31:43 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id AAA19332 for stable-outgoing; Fri, 16 Feb 1996 00:31:43 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id AAA19323 Fri, 16 Feb 1996 00:31:40 -0800 (PST) Message-Id: <199602160831.AAA19323@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Richard J Kuhns cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-reply-to: Your message of "Thu, 15 Feb 1996 09:58:36 +0500." <9602151458.AA08224@sparcmill.grauel.com> Date: Fri, 16 Feb 1996 00:31:40 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk > > One thing that would be interesting to know is if your card can > > be probed twice (ie found by the ISA probe after a successfull > > PCI probe). Can you add a "controller bt1 at isa? ..." entry into > > your config file in addition to the one you have and see what happens? > > The probe won't even attempt to look at the card if its out of unit numbers >, > > so the ouput you give below isn't enough to prove that we can't > > double probe the card. > > > >I'm afraid the card IS double probed. >I built and booted a kernel with the following in the config file: What a pooorly designed piece of hardware. Can you disable ISA compatibility in the BIOS? I attempted a different approach that would filter out cards based on the model number and the type of probe, but since the probe uses reseting the card as its primary detection method, it just won't work (at least not with the bt747 I have here to test with). Even attempting the bt_cmd to get the board type for a non-existant board, just hangs up the machine or simply takes too long to be practical (I waited 5 minutes before hitting the reset button. I think that we can document this enough for it to not be a problem especially since the GENERIC kernel only has one bt unit. What we do once we have a configuration manager is another story. Perhps the card has a second I/O register mapped in the PCI configuration space? Stephan is this allowed? Could it be that I should just search through them all and try to find and "ISA" port? >-- >Rich Kuhns rjk@grauel.com >PO Box 6249 >100 Sawmill Road >Lafayette, IN 47903 >(317)477-6000 x319 -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Fri Feb 16 06:40:12 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA07737 for stable-outgoing; Fri, 16 Feb 1996 06:40:12 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA07652 for ; Fri, 16 Feb 1996 06:40:01 -0800 (PST) Received: by Sysiphos id AA03076 (5.67b/IDA-1.5 for stable@freebsd.org); Fri, 16 Feb 1996 15:37:58 +0100 Message-Id: <199602161437.AA03076@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Fri, 16 Feb 1996 15:37:57 +0100 In-Reply-To: "Justin T. Gibbs" "Re: One problem && one question" (Feb 16, 0:31) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: "Justin T. Gibbs" Subject: Re: One problem && one question Cc: stable@freebsd.org Sender: owner-stable@freebsd.org Precedence: bulk On Feb 16, 0:31, "Justin T. Gibbs" wrote: } Subject: Re: One problem && one question } >I'm afraid the card IS double probed. } >I built and booted a kernel with the following in the config file: } } Perhps the card has a second I/O register mapped in the PCI configuration } space? Stephan is this allowed? Could it be that I should just search } through them all and try to find and "ISA" port? If the card used the standard PCI mechanism to declare all the resources claimed, it would be trivial to solve. We need a more general conflicts check code anyway (i.e. check for several non-overlapping mem/port regions). There are several mamping registers in the PCI configuration space, and each one can either specify a memory or port range. A verbose boot lists all those registers, unless they are set to '0', i.e. unused: bt0 rev 0 int a irq 10 on pci0:13 mapreg[10] type=1 addr=00006000 size=ffff0004. This is from one of the probe messages (with the misleading port range) and there was NO other map register listed, i.e. the card violates the PCI specs by not telling about the port range 0x330-333 that it decodes. But since those mapping registers are initialized by the PCI BIOS, this is not really surprising. I'm not sure whether the PCI spec would tolerate a map register hardcoded to some value for compatibility reasons ... I had been hoping that the first access to the PCI port range (0x6000-6003, but depends on the systems PCI BIOS) would disable the ports at 0x330, but it doesn't work that way. Now, the only solution appears to be to have conflicts check code that can deal with multiple port ranges, and enter both ranges when the PCI Buslogic card is found. But that doesn't seem right, neither, since there might be two BT946C, and I hope only one of them will respond to port 0x330 accesses. I.e. the cards BIOS must currently be able to determine either the existence of another such card, or there must be a jumper to disable ISA port accesses (or move the port to an alternate address). Since we can assume that PCI resources are assigned by a single authority (the PCI BIOS) there should never be conflicts between PCI devices. It might suffice to register the port range 0x330 for the PCI version, even if 0x6000 is actually used by the driver. But this could prevent another BT946C from being attached, since there would be a conflict (though that port range is never used in case of a PCI card). It's an ugly behaviour, but we'll see it increasingly often with the emerging crowd of ISA compatible low cost PCI Ethernet cards ... :( Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-stable Fri Feb 16 07:02:04 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id HAA08688 for stable-outgoing; Fri, 16 Feb 1996 07:02:04 -0800 (PST) Received: from public.wintek.com (public.wintek.com [199.233.104.88]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA08680 for ; Fri, 16 Feb 1996 07:02:00 -0800 (PST) Received: from watson.grauel.com (watson.grauel.com [199.233.104.36]) by public.wintek.com (8.6.12/1.17wintek(3.6davy)) with ESMTP id KAA02068; Fri, 16 Feb 1996 10:01:56 -0500 Received: from sparcmill.grauel.com (sparcmill.grauel.com [199.233.104.34]) by watson.grauel.com (8.6.12/8.6.9) with SMTP id KAA00245; Fri, 16 Feb 1996 10:07:10 -0500 Received: by sparcmill.grauel.com (5.0/SMI-SVR4) id AA17836; Fri, 16 Feb 1996 10:02:25 +0500 Date: Fri, 16 Feb 1996 10:02:25 +0500 Message-Id: <9602161502.AA17836@sparcmill.grauel.com> From: Richard J Kuhns To: "Justin T. Gibbs" Cc: freebsd-stable@freebsd.org Subject: Re: One problem && one question In-Reply-To: <199602160831.AAA19323@freefall.freebsd.org> References: <9602151458.AA08224@sparcmill.grauel.com> <199602160831.AAA19323@freefall.freebsd.org> Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > >I'm afraid the card IS double probed. > >I built and booted a kernel with the following in the config file: > > What a pooorly designed piece of hardware. Can you disable ISA > compatibility in the BIOS? I attempted a different approach that would > filter out cards based on the model number and the type of probe, but since > the probe uses reseting the card as its primary detection method, it just > won't work (at least not with the bt747 I have here to test with). Even > attempting the bt_cmd to get the board type for a non-existant board, > just hangs up the machine or simply takes too long to be practical (I > waited 5 minutes before hitting the reset button. I think that we can > document this enough for it to not be a problem especially since the > GENERIC kernel only has one bt unit. What we do once we have a > configuration manager is another story. > > Perhps the card has a second I/O register mapped in the PCI configuration > space? Stephan is this allowed? Could it be that I should just search > through them all and try to find and "ISA" port? > OK, I broke out the installation manual and started going through it. Under the `Advanced Options' menu, (I'm going to quote from the book) -Option: Set ISA Compatible IO Port. This option allows you to specify -an ISA backward-compatible I/O Port. The ISA compatible I/O Port is in -addition to the PCI compliant I/O port assigned by the PCI system BIOS. -The factory default (Primary) Allows the onboard BIOS to assign the next -available I/O port to the adapter starting from 330H. I just rebuilt the kernel that does the PCI probe for bt0 and the ISA probe for bt1, rebooted, and set the `ISA Compatible IO Port' to Disable. Everything worked as it should; the PCI probe found the card as bt0, but the ISA probe didn't find anything. The particular machine I've been working with is FreeBSD-only (dangerously dedicated, but I like it ;-). I don't know how disabling the ISA Port will affect DOS, but I hope to find out soon on a different machine -- I'm enjoying playing Duke Nukem 3D... -- Rich Kuhns rjk@grauel.com PO Box 6249 100 Sawmill Road Lafayette, IN 47903 (317)477-6000 x319