booting lfs - ongoing VFS error msg <- BIOS Disk in LILO?

Barry Fitzgerald Barry at hartford.uconn.edu
Fri Jan 12 10:18:51 PST 2001


On 12 Jan 2001, at 3:14, Bill Suit wrote:

> On Thursday 11 January 2001 06:08 pm, Bryan wrote:
> <snip>
> > BIOS DISK != OS DISK
> >
> > The BIOS comes in ignorant and just labels disk drives 0x80, 0x81,
> > etc... in order of the BIOSes loaded.  Of course your off-chipset
> > BIOS loads first and sets /dev/hde as BIOS disk 0x80, when LILO
> > assumes /dev/hda is BIOS disk 0x80.  Although Windows 9x could care
> > less about what it thinks the disk is (it'll just die if a drive
> > letter changes ;-), both Linux and (in some cases) NT will get into
> > issues (because they set drive devices, etc...).
> <snip>
> 
> Thanks Bryan. On another post approx 30mins ago to lfs I included my 
> lilo.conf file. I have cut, what may help you or someone else trying to 
> figure out my problem, from my last Mdk boot dmsg the following:
> ...
> Uniform Multi-Platform E-IDE driver Revision: 6.30
> ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
> PIIX4: IDE controller on PCI bus 00 dev 39
> PIIX4: chipset revision 1
> PIIX4: not 100% native mode: will probe irqs later
>     ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
> PDC20262: IDE controller on PCI bus 00 dev 58
> PDC20262: chipset revision 1
> PDC20262: not 100% native mode: will probe irqs later
> PDC20262: ROM enabled at 0xe8000000
> PDC20262: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode.
> PDC20262: FORCING PRIMARY MODE BIT 0x00 -> 0x01 MASTER
> PDC20262: FORCING SECONDARY MODE BIT 0x00 -> 0x01 MASTER
>     ide2: BM-DMA at 0xc000-0xc007, BIOS settings: hde:DMA, hdf:DMA
>     ide3: BM-DMA at 0xc008-0xc00f, BIOS settings: hdg:DMA, hdh:DMA
> hdc: NEC CD-ROM DRIVE:282, ATAPI CDROM drive
> hdd: YAMAHA CRW2100E, ATAPI CDROM drive
> hde: WDC AC34300L, ATA DISK drive
> hdf: Maxtor 54098U8, ATA DISK drive
> ide1 at 0x170-0x177,0x376 on irq 15
> ide2 at 0xb000-0xb007,0xb402 on irq 10
> hde: WDC AC34300L, 4104MB w/256kB Cache, CHS=8896/15/63, UDMA(33)
> hdf: Maxtor 54098U8, 39082MB w/2048kB Cache, CHS=79406/16/63, UDMA(66)
> hdc: ATAPI 32X CD-ROM drive, 256kB Cache
> Uniform CD-ROM driver Revision: 3.11
> hdd: ATAPI 40X CD-ROM CD-R/RW drive, 8192kB Cache
> Floppy drive(s): fd0 is 1.44M
> FDC 0 is a post-1991 82077
> md driver 0.90.0 MAX_MD_DEVS=256, MAX_REAL=12
> ...
> 
> I have a good boot floppy. Ready to try ideas. Bill
> 

what version kernel are you using in your LFS system?

I think someone already mentioned Andre Hedrick's IDE kernel 
patches available at 
http://www.kernel.org/pub/linux/kernel/people/hedrick/

if you're using a 2.2.x kernel, there is no support for the pdc20262 
ata66 card as an offboard chipset.  It is available in these patches.  
Note, support is now available in the stable 2.4.0 line.  However, I 
have to warn you.  I had almost the same IDE configuration as you 
do, and while I got the patches to work in 2.2.16... every 
subsequent patch failed to produce an operational kernel on my 
system.  Then again, I was also using the rip-off I/O Magic card.  It 
was basically a repackaged promise card, and I'm fairly certain that 
it was the same exact card, but I'm not willing to bet my life on it... 
You may have better luck than I did.  I'm now running off my PIIX4 
IDE bus, and am none the worse for wear running kernel 2.4 ...

when you do recompile, make sure that "boot offboard chipsets 
first" is set to "no" ... This is the way it works on your mandrake 
system, and setting this to yes in the kernel config will make the 
system try to load your pdc20262 card as ide0/ide1 and the PIIX4 
as ide2/ide3 (the exact opposite of your dmesg paste)...

this alone would cause the problem that you're having, if the 
support existed in the kernel.  The rationale: ide0 maps to /dev/hda 
and /dev/hdb .. ide1 to /dev/hdc /dev/hdd ...etc... Hence doing this 
would make /dev/hde not have a drive connected to it because it's 
trying to boot from the PIIX4 bus...

personally, I'd just try 2.4 and see if it works.  Don't forget to enable 
support for the chipset in the kernel and leave that offboard chipset 
boot option at 'no'... this will most likely fix your problem...

also, one suggestion.  append 'idebus=66' to your kernel command 
line.  The kernel is safedefaulting to a 33MHz bus... adding this line 
will make the kernel default to 66MHz and it will most likely speed 
up your system, even with burst bit enabled in the pdc20262...

and never, and I mean NEVER, enable hdparm -X66 no matter how 
tempting it looks.  This option is marked as experimental for good 
reason.  I have a VERY stable hardware configuration at home and 
enabling this spead up my IDE bus transfer rate to a loss.  It was 
like alcohol for the system.  It was very nice at first, but the next 
morning everything was bad.

I hosed an entire installation due to enabling that... I have never 
seen more data corruption on a stable system in my life...

good luck :) 


-- 
Unsubscribe: send email to lfs-apps-request at linuxfromscratch.org
and put unsubscribe in the subject header of the message




More information about the blfs-support mailing list