Date: Tue, 21 May 2002 23:19:48 +0100 From: "Jean-Mark Dupoux" <jeanm@dupx.freeserve.co.uk> To: A.J.Caines@halplant.com Cc: questions@freebsd.org Subject: re: Kernel bombs on Promise TX2 Ultra ATA/100 with disk Message-ID: <3CEAD614.2049.2788F4B@localhost>
next in thread | raw e-mail | index | archive | help
> Summary:
>
> Kernel boots with Promise ATA/100 controler and no disks, boots with disk
> on ATA/33 controler, crashes with disk on ATA/100 controler.
>
> Details:
>
> The onboard ATA/33 controler handles the boot disk (Quantum Fireball
> 6.4GB), CD-ROM and an ATA/100 disk (Maxtor 20GB).
>
> To get more out of the Maxtor, I added the Promise controler and attached
> the Maxtor to it.
>
> During boot, the BIOS picks up the Promise controler and shows it probe
> and find the Maxtor and correctly determine it to be UDMA/100 capable. The
> boot loader loads as normal, but as soon as the kernel (both 4.5-STABLE
> and recent 4.6-RC) load starts, it bombs with a console message which
> look vaguely like..
>
> epi: 0001xb4 abc: 0053dt ...
> [five more lines like that]
> BTX Halted.
>
> ..at which point it hangs hard.
>
> With the Maxtor attached to the onboard ATA/33 controler, the kernel boots
> fine and recognises the Promise controler.
>
> # dmesg | egrep '^(ata|ad)'
> atapci0: <Intel PIIX4 ATA33 controller> port 0xfc90-0xfc9f at device 7.1 on pci0
> ata0: at 0x1f0 irq 14 on atapci0
> ata1: at 0x170 irq 15 on atapci0
> atapci1: <Promise TX2 ATA100 controller> port 0xfc70-0xfc7f,0xfc8c-0xfc8f,0xfc68-0xfc6f,0xfc88-0xfc8b,0xfc60-0xfc67 mem 0xfedf8000-0xfedfbfff irq
> 9 at device 16.0 on pci0 ata2: at 0xfc60 on atapci1 ata3: at 0xfc68 on atapci1 ata0-slave: ATA identify retries exceeded ad0: 6149MB <QUANTUM
> FIREBALL ST6.4A> [13328/15/63] at ata0-master UDMA33 ad1: 19541MB <Maxtor 2B020H1> [39703/16/63] at ata1-master UDMA33
>
> When I connected an ATA/33 Zip disk to the Promise controler, it bombs in
> exactly the same fashion.
>
> Any assistance in determining why I can't have my disk on the Promise
> controler would be much appreciated.
>
> The only suspicion I have is that this may be irq related. I seem to have
> quite a few devices (all working) on irq 9:
>
> # dmesg | awk '/irq 9/{print $1,$2,$3,$4}'
> pci1: <NVidia/SGS-Thomson Riva128 graphics
> uhci0: <Intel 82371AB/EB (PIIX4)
> intpm0: <Intel 82371AB Power
> pcm0: <AudioPCI ES1370> port
> atapci1: <Promise TX2 ATA100
>
> More information on the system (dmesg, atacontrol list, kernel config,
> etc.) can be found at
>
> http://halplant.com:88/systems.html#Hardware
> http://halplant.com:88/server/config/
>
>
> -Andrew-
> --
> _______________________________________________________________________
> | -Andrew J. Caines- Unix Systems Engineer A.J.Caines@halplant.com |
> | "They that can give up essential liberty to obtain a little temporary |
> | safety deserve neither liberty nor safety" - Benjamin Franklin, 1759 |
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-questions" in the body of the message
Hello Andrew,
Your suspicions about too many devices on the same IRQ could well be right, but I
would personally start by running both the HDD's off the Promise card and then the
CD and anything else off just one of the onboard ATA33 controllers.
You could then disable the other ATA33 in the BIOS which would leave more IRQ's
free. Depending on your BIOS you may also have the option to force IRQ mapping
of PCI slots to specific values.
jeanmark
jeanmark@dupx.freeserve.co.uk
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3CEAD614.2049.2788F4B>