Re: Adding SCSI Controller, SCSI Tape, and SCSI Disk to 2.2.20-idepci system

["Doug MacFarlane" <madmac@covad.net>]

On 13 Jan 2003, 15:30:52, nate wrote:
> Doug MacFarlane said:
> > So I apt-got a 2.4.19-686 kernel image, and when I rebooted, I had to use
> > modconf to enable the 3c59x driver, and when I rebooted, it runs like
> > molasses . . . during boot, it takes minutes to complete the calculate
> > module dependencies step.  This hadn't happened before, and I don't have
> > time to figure it out yet . . .
> 
> this is odd, but does it boot eventually? does this happen everytime
> it boots? after it boots can you delete /lib/modules/`uname -r`/modules.dep
> and run depmod -a ..does it take the same amount of time?

Yes, it boots but runs very slowly.

If I delete modules.dep and then run depmod -a, the depmod completes in less
than a second.  But the Calculating Module Dependencies during boot takes
3 minutes . . .

I think I have a problem with initrd - I'm pasting in a piece of dmesg below:

PIIX4: IDE controller on PCI bus 00 dev a1
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0x14c0-0x14c7, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0x14c8-0x14cf, BIOS settings: hdc:DMA, hdd:pio
hda: ST34321A, ATA DISK drive
hdb: ST38641A, ATA DISK drive
hdc: TOSHIBA CD-ROM XM-5602B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
blk: queue cc84eea4, I/O limit 4095Mb (mask 0xffffffff)
hda: 8421840 sectors (4312 MB) w/128KiB Cache, CHS=8912/15/63, UDMA(25)
blk: queue cc84eff0, I/O limit 4095Mb (mask 0xffffffff)
hdb: 16809660 sectors (8607 MB) w/128KiB Cache, CHS=16676/16/63, UDMA(25)
Partition check:
 /dev/ide/host0/bus0/target0/lun0: [PTBL] [557/240/63] p1 p2 < p5 >
 /dev/ide/host0/bus0/target1/lun0:hdb: set_multmode: status=0x51 { DriveReady SeekComplete Error }
hdb: set_multmode: error=0x04 { DriveStatusError }
 [PTBL] [1111/240/63] p1 < p5 p6 p7 p8 p9 > p2
ext3: No journal on filesystem on ide0(3,69)

I have no idea why it's trying to use ext3 and jbd  . . . 

When I boot 2.2.20, it, ofcourse, is NOT an initrd kernel, and it runs fine,
and I don't get any of these messages in dmesg . . . 

So my workaround will be to recompile the 2.4.29 or 2.4.20 kernel without
changing the config, and leaving off the --initrd make-kpkg option, and editing
lilo.conf to remove the initrd option, and see if that fixes it . .  . 

Any thoughts?

As always, thanks for your help.

madmac