On Wed, 23 Aug 2000, Michael Peddemors wrote:
> On Wed, 23 Aug 2000, Mike Galbraith wrote:
>
> > Try mem=31M for grins (hunch-o-meter says memory detection changes)
> GREAT HUNCH!
(good.. hunch-o-meter still works)
> Well, that got rid of the initrd extended beyond message...
>
> And later it at least FINDS a compressed image....(Just used a known good
> filesystem and it boots)
>
> Is this a bug report that it doesn't see the memsize correctly compared to
> previous kernels? Can't be putting mem= in my distro..
I would say it's a bug if I didn't know that so many BIOS suck.
> rts 4096, the zone(1) reports 3840, and zone(2) is 0)
>
> > LOADER_TYPE 49 is the identifier for syslinux.
> > from ldlinux.asm:
> > syslinux_id equ 031h ; SYSLINUX (3) version 1.x (1)
> >
> > At a glance, it looks like INITRD_SIZE is set wrong.
> >
> > Please verify INITRD_START, and add INITRD_SIZE to your printk.
> > INITRD_START=3276800 means INITRD_SIZE=30249773.. and that doesn't
> > jibe with floppy ;-) Here, syslinux always loads (16k aligned)
> > the initrd at top of memory and gets INITRD_SIZE right.
> >
> > > Strange why this (2.4 vs 2.2) kernel works different tho.. Something MUST
> > > have changed.
>
> When I added the INITRD_SIZE paramater to the printk statement, I get `692836`
> That is the EXACT size of my initrd.gz file.
(Which still doesn't jibe with INITRD_START=3276800)
> So now we know what is going on, but what is the BUG resolution...
No ideas yet. Best bet would be to ask David Parsons. He knows a
bunch about booting. :>> hpa knows quite a lot about the subject too..
maybe he'll have a comment when he get's a chance to read mail again.
In the meantime... scirtch scritch scritch.
-Mike
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Aug 23 2000 - 21:00:10 EST