From: Hugo van.der.Kooij (hugo@rusthoff.hacktic.nl)
Date: 08/02/93


From: hugo@rusthoff.hacktic.nl (Hugo van.der.Kooij)
Subject: 486 Doesn't boot LINUX SLS v1.02
Date: Mon, 02 Aug 1993 23:31:00

Ok,
I mentioned my problem in an earlier message but everybody
asked me to elaborate the issue. So here goes:

SYSTEM:
    486 System board using: AMI BIOS dated 06/06/92
                            486DX Processor at 33MHz
                            16MB (4x4MB SIMMs 70ns)
                            ISA BUS
                            SYMPHONY chipset
    ET4065 (ET4000; 64k colors) Video card without IRQ2
    Multi I/O with: 2x serial (I/O and IRQ standard)
                    1x parallel (I/O and IRQ for LPT1)
                    Floppy drive with 3.5" and 5.25" drives
                    IDE controller
        ST1144A (type 47; Cyl: 1001; Heads: 15; Sector: 15)
    Soundblaster2 (I/O: 220; IRQ: 2)
    NE2000 (I/O:300; IRQ: 5)

SoftWare which failes:
    SLS v1.02 kit (v0.99 pl8)
    LINUXpro (v0.99 pl11) (Fred van Kempen's toy)
    MCC Interim release

Software which works:
    SLS v1.01 kit (v0.99 pl6)
    HLU bootdisk (v0.99 pl7)

I tried at first with all cards installed: FAILED
    All I got was LILO telling it was booting. Then LINUX
    loading the RAMDISK. Then LINUX loading and
    uncompressing. Right after the message
        Booting LINUX
    my system freezes.

Ok so somebody sugested to play with RAM shadowing ROM
and so on. And so I did. Well I got just as far as
a line telling me: console: colour

The chips involved my systems are:
CPU: A80486DX-33
            L2523562
            SX729
            INTEL (M)(C) 1989
1: SYMPHONY
            SL82C461
            A29987.1
            1161 C_9246
2: SYMPHONY
            SL82C362
            2222220 JAPAN
3: SYMPHONY
            SL82C465
            4L02F2237
            JAPAN 9243EAI
keyboard: NEC
            (c) 1988-1991
            DISA 4236 297
            9224M7028

Main Memory: 4 SIMMs of 4MB with 9 chips at 70ns
    (The manual says 80ns would be fast enough)
Cache Memory: 9 chips of 32kx8 at 20ns
    (for 33MHz 20ns would do according the manual)
And there is a jumper which tells me:
    ZERO WAIT STATE CACHE WRITE HIT
this could also be place as:
    ONE WAIT STATE CACHE WRITE HIT

Problems occur even without the NE2000,
without the SoundBlaster,
with another Video card,
with an original 8042AH.

The system has been stripped to the bone just to see what is actualy causing the problem. Any item I couldn't do without has been exchanged by stuff from my 386SX (yep, even the keyboard and controller) EXCEPT for the systemboard itself, the RAM and the CPU.

I have no way of exchanging any of these yet.

If any one has cured a simular problem please specify faulty equipment and what part it has been exchanged by to fix it. Please specify as much as possible.

I hope to start a Systems query on the NET next weeks. I will make a sort of an approved hardware list so any future decisions can be made according this information. And we might even be able to put pressure on manufactures.

Thanx for your time,
Hugo.

 * Origin: Rust-Hoff; Home of the Brave. (2:512/32.23)