lefevre a écrit 1 commentaire

  • [^] # boot on ATA normal mode fails with IT8212 RAID controller

    Posté par  . En réponse au journal Carte controlleur W-8212. Évalué à 1.

    Désolé, le message est en (mauvais anglais) mais je sonde tout azimut pour trouver des personnes susceptibles de m'aider dans la configuration de la ga7n400pro2 et surtout de son XXXX gigaraid IT8212F incorporé à la carte mère. Avis aux pros de la configuration sous linux. Merci à vous ! (mail en retour apprécié Merci.)

    Info : Ce message est également envoyé à ITE et sur les rares forum discutant de IT8212.

    Hi,

    I have some trouble with gigaraid controller on my motherboard ga7N400pro.

    My hardware is :
    ---------------
    GA 7N400pro, AMD XP 2400+
    1048 MO DDR3200
    2 ATA100 UDMA5 Maxtor, with egal capacity : 30 go

    Info +:
    Motherboard brings 4 IDE :
    - "special NFORCE" IDE for ATAPI CDROM (mine is Pioneer DVD)
    - NFORCE IDE for ATA DRIVE
    - Tw0 IT8212F (or ATARAID) IDE for raid 0, 1, 0+1, JBOD or normal IDE function

    ga7n400 bios setup :
    --------------------
    SATA : disabled
    ATARAID : enabled

    IT8212 bios setup :
    --------------------
    Setup array type as : NORMAL (No array, Normal IDE controller config)

    MOTIVATION :
    ------------
    I have 2 Hard Disk. Like my older config with asus a7v133, I want to switch windows and linux.
    I want to work with LINUX software RAID too, not hardware. Software Raid with linux is more flexible,
    in my mind and situation :]

    GENERAL PROCEDURE :
    ------------------
    Before, I make a first install with my 2 OS in order to attach later my system into ATARAID IDE.
    In this first install, I can insert ATARAID driver in my kernel 2.4.22. The driver is ITERAID.o module.

    (In a first install, with kernel 2.6.1, i encounter trouble with ITERAID.o, which don't build.)

    STEPS in my Installation :
    ------------------------

    Before, I make a first install with my 2 OS in order to insert later my system into ATARAID IDE.

    Boot manager is Lilo, with [Windows - Linux-2.4.22. - Linux-2.4.22.ITE] switch.
    Windows 2000 is on my DISK 2.
    Linux mdk 9.2 is on my DISK 1.
    DISK 1 and DISK 2 share two ext2fs partition for Linux software raid0, with same size.
    There is 2 swap file with same size on DISK 1 and DISK 2 too.

    First install is made with my onboard nforce IDE. Disk 1(linux) is master and Disk 2(w2k), on the same cable, is slave.

    I download Kernel 2.4.22.10 (like my distro mdk 9.2), setup, install NVIDIA_nforce-1.0-0261.src.rpm
    and compile the custom Kernel. Reboot, dmesg output, reboot, dmesg output. Not fine tuning, but works !.

    I build iteraid.o for mandrake, test it like procedure from Mark Lu (ITE RAIDEXPRESS 133 MDK 9.0 Installation Guide)
    and insert it in "lib/modules/"mycustomKernel"/drivers/SCSI".

    Finally, I type :

    # mkinitrd -f --preload scsi_mod --preload sd_mod --with=iteraid /boot/initrd"mycustomkernel".img "mycustomkernel"
    With this command, iteraid.o is loaded in first boot time.

    I don't have any major error with these steps.

    I don't edit any config.file, like fstab or raidtab. No-place, i specify volume like sda or sdb. I do ? No, in my mind...

    I reboot, ultim dmesg output verification (it's ok).

    Finaly, I halt. I attach my 2 ATA DISK into ATARAID IDE. DISK 1 is master on ATARAID IDE 1, DISK 2 is alone but with slave jumper ON
    into ATARAID IDE 2 (slave jumper : don't matter ? DISK 2 is alone, it will be master by default, no?)

    BOOT ISSUE :
    ---------
    I reboot. I see the controller RAIDExpress133 in my boot screen and my disks are detected like SCSI disk.

    The partition check is very long (> 120 seconds !! :( , why ? ) and issue is fatal.


    I don't know what is wrong. FAQ, FORUM support on ITERAID is very weak and i don't find good tips or
    happy experience in google search.

    Anyone can diagnose and explain why and what is wrong ? Many thanks ?

    Here is my Boot screen :

    **************************************************************
    ...
    Freeing initrd memory: 57k freed
    VFS: Mounted root (ext2 filesystem).
    Mounted devfs on /dev
    Red Hat nash version 3.4.43-mdk starting
    Loading sd_mod.o module
    Loading raid0.o module
    md: raid0 personality registered as nr 2
    Loading iteraid.o module
    Found Controller: IT8212 UDMA/ATA133 RAID Controller
    FindDevices: Device 0 is IDE
    Channel[0] BM-DMA at 0xA400-0xA407
    FindDevices: Device 3 is IDE
    Channel[1] BM-DMA at 0xA408-0xA40F
    scsi1 ! RAIDExpress133
    Vendor: ITE Model: IT8212F Rev: 1.3
    Type: Direct-Access ANSI SCSI revision: 00
    Vendor: ITE Model: IT8212F Rev: 1.3
    Type: Direct-Access ANSI SCSI revision: 00
    Attached scsi disk sda at scsi1, channel 0, id 0, lun 0
    Attached scsi disk sda at scsi1, channel 0, id 3, lun 0
    SCSI device sda: 60030432 512-byte hdwr sectors (30736 MB)
    Partition check:
    /dev/scsi/host1/bus0/target0/lun0:AtapiStartIo:Already have a request!
    AtapiResetController enter
    IT8212ResetAdapter: Reset channel 0
    IT8212ResetAdapter: Reset channel 1
    IT8212ResetAdapter Success!
    AtapiResetController exit
    AtapiResetController enter
    IT8212ResetAdapter: Reset channel 0
    IT8212ResetAdapter: Reset channel 1
    IT8212ResetAdapter Success!
    AtapiResetController exit
    scsi: device set offline - not ready or command retry failed after host reset: host 1 channel 0 id 0 lun 0
    p1 p2 < I/O error: dev 00:00, sector 45360
    I/O error: dev 00:00, sector 45360
    >
    SCSI device sdb: 58600000 512-byte hdwr sectors (30003 MB)
    /dev/scsi/host1/bus0/target3/lun0:AtapiStartIo:Already have a request!
    AtapiResetController enter
    Unable to handle kernel NULL pointer dereference at virtual address 0000020e
    printing eip:
    f88ae08a
    *pde = 00000000
    Oops: 0002
    iteraid raid0 sd_mod
    CPU: 0
    EIP: 0010:[]
    EFLAGS: 00010087
    eax: 0000000e ebx: 00009802 ecx: 00000086 edx: f7d15f08
    esi: 00009416 edi: f7d1a000 ebp: 00000002 esp: f7d15ed8
    ds: 0018 es: 0018 ss: 0018
    Process scsi_eh_1 (pid: 20, stackpage=f7d15000)
    Stack: 0000a400 000006cc f7d1a000 f88aa42f f7d1a000 f7d15f08 ...
    etc...

    Call Trace: [] [] ...
    etc...

    Code: c7 01 00 01 00 00 00 00 00 00 eb ba c7 81 88 01 00 00 00 00
    ***********************************************************************************

    -----------------------------
    jerome lefevre, New-Caledonia
    jlefevre@mls.nc