xfs
[Top] [All Lists]

Re: boot problems

To: Daniel Spratlen <spratlen@xxxxxxxxxxxxxx>
Subject: Re: boot problems
From: "D. Stimits" <stimits@xxxxxxxxxx>
Date: Mon, 11 Jun 2001 23:25:08 -0600
Cc: linux-xfs@xxxxxxxxxxx
References: <3B255A44.570F9EB8@xxxxxxxxxxxxxx>
Reply-to: stimits@xxxxxxxxxx
Sender: owner-linux-xfs@xxxxxxxxxxx
Daniel Spratlen wrote:
> 
> Hello xfsers,
>     First off, thank you for your hard and wonderful work.  Now for my
> question.  I have an IBM x220 server, and I can't get RH7.1-xfs to boot
> unless I use the boot disk that I created during the install.  System
> configuration is as follows:

One thing that has kept a couple of 2.4.x kernels from booting my system
is a bug in fs/block_dev.c, function ioctl_by_bdev, near line 596. This
is fixed in the 2.4.5-ac kernels starting around ac3, and I have hand
edited it in each version I've used that does not have the fix. You can
add the line starting with "+" (don't actually use a "+" sign):
int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long
arg)
{
        kdev_t rdev = to_kdev_t(bdev->bd_dev);
        struct inode inode_fake;
        int res;
        mm_segment_t old_fs = get_fs();

        if (!bdev->bd_op->ioctl)
                return -EINVAL;
        inode_fake.i_rdev=rdev;
+        inode_fake.i_bdev=bdev;
        init_waitqueue_head(&inode_fake.i_wait);
        set_fs(KERNEL_DS);
        res = bdev->bd_op->ioctl(&inode_fake, NULL, cmd, arg);
        set_fs(old_fs);
        return res;
}

I realize it is a long-shot that this might actually be your particular
problem, but I have the same integrated adapter which is what brought
out this particular bug. The bug fix was provided through the kernel
devel list. I'm hoping CVS here might add it in as well.

D. Stimits, stimits@xxxxxxxxxx

> 
> # df
> Filesystem           1k-blocks      Used Available Use% Mounted on
> /dev/scsi/host0/bus0/target5/lun0/part6
>                       16606376    463144  16143232   3% /
> /dev/scsi/host0/bus0/target5/lun0/part1
>                          99588      5376     94212   6% /boot
> /dev/scsi/host0/bus0/target6/lun0/part5
>                        7521620       144   7521476   1% /opt
> /dev/scsi/host0/bus0/target6/lun0/part1
>                       10236604     29812  10206792   1% /var
> 
> There are two scsi disks in this system which both come off of a Adaptec
> AIC-7892 scsi adapter.  Here is /proc/scsi/scsi:
> 
> # more scsi
> Attached devices:
> Host: scsi0 Channel: 00 Id: 00 Lun: 00
>   Vendor: Seagate  Model: STT20000N        Rev: 7A61
>   Type:   Sequential-Access                ANSI SCSI revision: 02
> Host: scsi0 Channel: 00 Id: 05 Lun: 00
>   Vendor: IBM-PSG  Model: ST318436LW    !# Rev: 3281
>   Type:   Direct-Access                    ANSI SCSI revision: 03
> Host: scsi0 Channel: 00 Id: 06 Lun: 00
>   Vendor: IBM-PSG  Model: DPSS-318350N  M  Rev: S9AA
>   Type:   Direct-Access                    ANSI SCSI revision: 03
> 
> I
> 
> n the scsi bios, I set the disk with the Id of 6 to be the boot disk.
> Here is /proc/scsi/aicxxxx/0:
> 
> # more 0
> Adaptec AIC7xxx driver version: 5.2.4/5.2.0
> Compile Options:
>   TCQ Enabled By Default : Enabled
>   AIC7XXX_PROC_STATS     : Enabled
> 
> Adapter Configuration:
>            SCSI Adapter: Adaptec AIC-7892 Ultra 160/m SCSI host adapter
>                            Ultra-160/m LVD/SE Wide Controller at PCI
> 1/3/0
>     PCI MMAPed I/O Base: 0xeffff000
>  Adapter SEEPROM Config: SEEPROM found and used.
>       Adaptec SCSI BIOS: Enabled
>                     IRQ: 9
>                    SCBs: Active 0, Max Active 64,
>                          Allocated 93, HW 32, Page 255
>              Interrupts: 26199
>       BIOS Control Word: 0x58d4
>    Adapter Control Word: 0x5c5e
>    Extended Translation: Enabled
> Disconnect Enable Flags: 0xffff
>      Ultra Enable Flags: 0x0000
>  Tag Queue Enable Flags: 0x0060
> Ordered Queue Tag Flags: 0x0060
> Default Tag Queue Depth: 32
>     Tagged Queue By Device array for aic7xxx host instance 0:
>       {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}
>     Actual queue depth per device for aic7xxx host instance 0:
>       {1,1,1,1,1,32,32,1,1,1,1,1,1,1,1,1}
> 
> Statistics:
> 
> (scsi0:0:0:0)
>   Device using Narrow/Async transfers.
>   Transinfo settings: current(0/0/0/3), goal(10/127/0/0),
> user(9/127/1/2)
>   Total transfers 0 (0 reads and 0 writes)
>              < 2K      2K+     4K+     8K+    16K+    32K+    64K+
> 128K+
>    Reads:       0       0       0       0       0       0       0
> 0
>   Writes:       0       0       0       0       0       0       0
> 0
> 
> (scsi0:0:5:0)
>   Device using Wide/Sync transfers at 40.0 MByte/sec, offset 31
>   Transinfo settings: current(12/31/1/0), goal(12/31/1/0),
> user(9/127/1/2)
>   Total transfers 17814 (7090 reads and 10724 writes)
>              < 2K      2K+     4K+     8K+    16K+    32K+    64K+
> 128K+
>    Reads:     129       0    4633    1280     264     784       0
> 0
>   Writes:    1042     215     361    6227    1185    1694       0
> 0
> 
> (scsi0:0:6:0)
>   Device using Wide/Sync transfers at 40.0 MByte/sec, offset 63
>   Transinfo settings: current(12/63/1/0), goal(9/127/1/2),
> user(9/127/1/2)
>   Total transfers 8276 (627 reads and 7649 writes)
>              < 2K      2K+     4K+     8K+    16K+    32K+    64K+
> 128K+
>    Reads:      43       0     102     146     146     190       0
> 0
>   Writes:    2856     262    1506    1775     605     645       0
> 0
> 
> And also included, here is my lilo.conf file:
> 
> # more lilo.conf
> boot=/dev/sda
> map=/boot/map
> install=/boot/boot.b
> prompt
> timeout=50
> message=/boot/message
> linear
> default=linux
> 
> image=/boot/vmlinuz-2.4.2-SGI_XFS_1.0
>         label=linux
>         initrd=/boot/initrd-2.4.2-SGI_XFS_1.0.img
>         read-only
>         root=/dev/sda6
>         append="ramdisk_size=2500"
> 
> I know that this email is long, and has a lot of files attached, and I
> apologize for the length, but I thought that I'd provide as much info as
> possible.  As I said, I can only boot if I use the boot disk that I
> created during the install process.  I made sure that I installed lilo
> on the MBA as per the FAQ.  I have the most updated bios available from
> IBM.  If I boot with the boot disk, everything works wonderfully, but
> I'd like to be able to boot from sda.  The only thing that I can think
> of at this moment, is that hda on this box is a cdrom, but I don't think
> that this would be a problem.  Upon bootup, the machine looks on the
> media as per the boot order in the bios, It checks the scsi drives
> first, and then looks on the cdrom, and then the floppy.   This is the
> order that the drive lights come on, and also the order that they are
> supposed to boot from according to the bios.
> 
> TIA,
> Daniel Spratlen
> spratlen@xxxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>