[syslinux] ISOLINUX boot problem: request for help
H. Peter Anvin
hpa at zytor.com
Thu Jul 26 14:06:49 PDT 2007
Peter Bartke wrote:
>
> I'm also suspicious of the BIOS of the ASUS P5GD2 board.
> This is the output of `lshw':
> description: BIOS
> vendor: American Megatrends Inc.
> physical id: 0
> version: 1005.009 (02/03/2005)
> size: 64KB
> capacity: 448KB
> capabilities: isa pci pnp apm upgrade shadowing escd cdboot
> bootselect socketedrom edd int13floppy1200 int13floppy720
> int13floppy2880 int5printscreen int9keyboard int14serial int17printer
> int10video acpi usb agp ls120boot zipboot biosbootspecification
>
> The Kernel is 2.6.21.5. slax-6.0.0-rc5.iso with this kernel apparently
> boots for a lot of people.
The BIOS is almost certainly chief suspect #1. There are BIOSes which
have been known to be sensitive to the exact location to the kernel
image on the CD -- why, I don't know (my best guess would involve
inappropriate timeouts; CDs can take a long time seeking.) That is part
of why I recommend trying to force mkisofs to put the kernel and the
isolinux file near the start of the disc.
> The first lines of the successful boot form harddisk are:
>
> Jul 25 20:41:01 ravel kernel: Linux version 2.6.21.5-smp (root at midas)
> (gcc version 4.1.2) #2 SMP Tue Jun 19 14:58:11 CDT 2007
> Jul 25 20:41:01 ravel kernel: BIOS-provided physical RAM map:
> Jul 25 20:41:01 ravel kernel: 127MB HIGHMEM available.
> Jul 25 20:41:01 ravel kernel: 896MB LOWMEM available.
> Jul 25 20:41:01 ravel kernel: found SMP MP-table at 000ff780
>
> Could there lurk a memory fragmentation problem as in ancient ages?
No, that's a pretty typical readout for a 1 GB machine.
However, tweaking a mem= option might answer that for sure, e.g. mem=800M.
-hpa
More information about the Syslinux
mailing list