Re: SuSE 10.0 boot problems with CPU changed

On Sunday 27 August 2006 13:17, composlinuxmisc stood up and addressed the
masses in /comp.os.linux.misc/ as follows...:

Hi guys. One of my SuSE Linux 10.0 i586 boxes recently had its AMD XP
2500+ CPU damaged, perhaps by my recent overclocking adventures.

The computer seemed to initiate a thermal shutdown soon after bootup --
sometimes even in/before the BIOS menu at stock speeds -- and the
motherboard (Abit NF7) gave a two-tone siren.

Whilst it is being RMA'ed, I took an old AMD XP 1500+ CPU out to at
least get that box running, if more slowly. The BIOS detects the CPU
alright and when I shelved the 1500+ it was known to work.

Upon booting SuSE, I get a black screen and nothing else.

Does SuSE require a reinstall if the CPU is changed, or is it more
likely that other hardware on the comp. is somehow damaged?

Most likely the latter, I'd say. I don't see why you would need to
reinstall an entire distribution only because a CPU has been swapped for
another one, especially if the new CPU is of the same brand and generation
as the original CPU.

In your case, it wouldn't even make a difference if you were running Gentoo
- which is entirely compiled for your specific hardware, rather than to go
with a generic "one size fits all" binary code for everything in the
distribution.

As a sidenote, I've known entire installed distributions to work when the
hard disk was simply moved to another computer, which included moving the
disk from an Athlon XP based machine to a Pentium III. This variety of
target computers is exactly why distribution vendors typically compile a
"one size fits all" distro - this is of course less so for /x86-64,/ since
there are only two major types of CPU's in this class: the Intel EM64T and
the AMD64 platform.

Most stock distribution kernels for /IA32/ have been compiled to work on a
generic /i586/ or perhaps a /i686,/ and they contain enough generic /x86/
code to keep the kernel happy whether it is running on an Intel Pentium 4
or on an AMD K-6. The Athlon XP is an /i686/ and should therefore be able
to handle any generic /i586/ or /i686/ stock kernel and distribution. I
believe SuSE is compiled as /i586/ code.

There is a good reason as to why overclocking is generally discouraged, and
you are now experiencing that reason... Most likely you've burned more
than just your CPU...

With a bit of luck, it'll only be the memory modules, or perhaps you've
accidentally disconnected something or knocked something out of the socket
while you were replacing the CPU, but I suspect a lot more damage has been
done... :-/

While it is true that Intel CPU's can take less heat than a comparable AMD
CPU, an Intel CPU has an overheating protection, and most recent Intels all
have thermal throttling. As far as I know, AMD's don't have that and will
allow you to toast them.

Intel nags and freezes, but AMD goes all the way until it burns up. And
that's apparently what happened to your original CPU. However, imagine
what such a thermal torture does to the rest of the chips...

I myself have only had one machine going "black screen" on me before - even
the POST information didn't appear on the screen - and that was because the
machine had suffered a lightning strike through the cable modem. Needless
to say that most of the hardware was toast, all except for the floppy
drive, the Adaptec 29160 SCSI adapter, the IBM SCSI hard disk and the
Plextor SCSI CD-RW device. A true testimonial of quality for the SCSI
stuff - I'm actually using all of it in this very machine now, after it was
all tested electronically - and it's already been running without any
problems in this machine here for about two years now.

On the other hand, the Intel Pentium III and the i810 chipset-based
Chaintech motherboard, the Chaintech videocard, the Hitachi monitor - which
was in standby mode at the time of the strike - the Creative Soundblaster
and the Iomega 2 GB Jaz - which was also SCSI, but Jaz drives were never
really that good - were all waste.

Computers are very sophisticated pieces of hardware, and they're not exactly
given away for free in a pack of cookies either, so don't mess with the
clock speeds anymore... ;-)

Sorry for the lecture, but I guess it was necessary... ;-)

--
With kind regards,

*Aragorn*
(registered GNU/Linux user #223157)
.

Relevant Pages

  • mptscsih: ioc1: attempting task abort! (sc=d6e8a980)
    ... CPU 2: Machine Check Exception: 0000000000000004 ... OEM ID: INTEL Product ID: Bridge CRB APIC at: 0xFEE00000 ... CPU: Trace cache: 12K uops, ... SCSI device sda: 287132440 512-byte hdwr sectors ...
    (Linux-Kernel)
  • Sudden Reboot - Dell Dimension XPS
    ... 12MB Radeon 9800 Graphics Card ... The first event is my Intel Network card - E100B. ... a microcode update that is not applied by the computer's basic input/output ... Write down the following CPU information for each processor:o CPU Family ...
    (microsoft.public.windowsxp.hardware)
  • Re: Segmentation in real mode
    ... Other points to note is that the "segment registers" were introduced by ... Intel _SOLELY_ for the purpose of being the "upper part" of an address (not ... change the CPU, stick it a soundcard, etc....that you could slowly ...
    (alt.lang.asm)
  • DRM, Intel, Sony, virtualization and backdoors
    ... to my naive eye this is strikingly similar to Intel's DRM ... «While Intel steered clear of mentioning the new DRM ... technology at its Australian launch of the new products, ... that's is not merely a CPU id. ...
    (comp.os.linux.powerpc)
  • [SLE] 9.0 pro install hangs on ide detection
    ... 00:1f.1 IDE interface: Intel Corp. 82801EB Ultra ATA Storage Controller ... CPU: Trace cache: 12K uops, ... PCI: PCI BIOS revision 2.10 entry at 0xf0031, ...
    (SuSE)