From owner-freebsd-current Wed Jan 27 09:14:39 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA11402 for freebsd-current-outgoing; Wed, 27 Jan 1999 09:14:39 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from megaweapon.zigg.com (megaweapon.zigg.com [206.114.60.8]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA11397 for ; Wed, 27 Jan 1999 09:14:33 -0800 (PST) (envelope-from matt@zigg.com) Received: from localhost (matt@localhost) by megaweapon.zigg.com (8.9.2/8.9.2) with ESMTP id MAA00345; Wed, 27 Jan 1999 12:08:08 -0500 (EST) (envelope-from matt@zigg.com) Date: Wed, 27 Jan 1999 12:07:52 -0500 (EST) From: Matt Behrens To: Gregory Bond cc: Matthew Dillon , Mike Smith , current@FreeBSD.ORG Subject: Re: WD/ide_pci bug! [was Re: SOFTUPDATES hangs keyboard ] In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Never mind. Apparently it went south on me while I was having lunch. I got this curious error when I paniced the debugger (paraphrased): HELP! busy_count is less then 0 (-1) Is this something of a clue? I've never been able to sync the drives at all by panicing, it can't communicate with the IDE controller for some reason. Lots of status 0x58's. On Wed, 27 Jan 1999, Matt Behrens wrote: : Greg et al, : : I added "flags 0x80ff" to my wd0 and wd1 lines (I don't have PCI : IDE, but rather VLB, so the 0x4000 bit doesn't apply) :), was able : to successfully get 32-bit mode and multiblock-16 out of both of : my drives, and have been able to enable softupdates all around : again without seeing any of my problems either. (I did re-cvsup : -STABLE just before the kernel rebuild and did config -r to make : sure everything was cleared out -- my old kernel was from the 20th, : I believe.) : : My previous test was simple -- tar zxvf samba-2.0.0.tar.gz. :) : Just to see if maybe it was just because it was being faster, I : also concurrently ran a chflags -R noschg /usr/obj;rm -rf /usr/obj : on the same partition. I did see somewhat "pausy" activity, but : it didn't actually hang. : : Now, we just have to figure out why 32-bit-mode is so special. :) : I have access to a few other boxes; I'll get those set up and see : what we can see. : : On Wed, 27 Jan 1999, Gregory Bond wrote: : : : Well stone the bloody crows. : : : : System: HP Kayak XA 440BX M/b, P-II, Quantum Fireball ex4.3a : : : : For months I've been running with no flags on either wdc0 or wd0. : : Having followed the recent discussions about DMA mode etc I looked at : : LINT and added "flags 0xa0ff" to wd0. Rebuilt the system using the : : latest 3-Stable code, including Matt D.s VM fixes, and tried yet again : : the test that has quite reliably frozen my system (softupdates enabled : : on /usr/src, cd /usr/src/games, make) and, lo and behold, it worked. I : : thought it was probably Matt's VM patches but (just to make sure) I : : built a kernel without the wd flags and tried again. It failed. : : : : This seems to be saying the problem is the WD driver: When I have flags : : on my disk, it all works; when I use the default, it fails (but only : : on filesystems with softupdates activity...) : : : : Any ide_pci experts like me to try some more tests? : : : : Matt: I'm sorry I ever doubted the VM code or the softupdates! : : : : Probing for devices on PCI bus 0: : : chip0: rev 0x02 on pci0.0.0 : : chip1: rev 0x02 on pci0.1.0 : : chip2: rev 0x02 on pci0.7.0 : : ide_pci0: rev 0x01 on pci0.7.1 : : chip3: rev 0x02 on pci0.7.3 : : [...] : : wdc0 at 0x1f0-0x1f7 irq 14 on isa : : wdc0: unit 0 (wd0): , DMA, 32-bit, multi-block-16 : : wd0: 4104MB (8405775 sectors), 8895 cyls, 15 heads, 63 S/T, 512 B/S : : wdc1 at 0x170-0x177 irq 15 on isa : : wdc1: unit 0 (atapi): , removable, accel, ovlap, dma, iordis : : acd0: drive speed 5512KB/sec, 128KB cache : : acd0: supported read types: CD-R, CD-RW, CD-DA, packet track : : acd0: Audio: play, 16 volume levels : : acd0: Mechanism: ejectable tray : : acd0: Medium: CD-ROM 120mm data disc loaded, unlocked : : vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa : : : : Greg. : : : : - Matt Behrens : Network Administrator, zigg.com : Engineer, Nameless IRC Network : : - Matt Behrens Network Administrator, zigg.com Engineer, Nameless IRC Network To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message