> In article <linux.kernel.Pine.LNX.4.10.9909211108450.2094-100000@badlands.lexington.ibm.com>,
> Richard A Nelson <cowboy@vnet.ibm.com> wrote:
> >Do the ac series have Parson's memory detection?
>
> I hope so!
>
> That's *fascinating*; your output looks like your machine is punting
> the memory region code altogether and falling back to E801.
>
> Can you hand-apply a cookie? In arch/i386/kernel/setup.c, can you
> add a
>
> printk("e820: E820_MAP_NR = %d\n", E820_MAP_NR);
>
> before the
>
> if (E820_MAP_NR > 1)
>
> line in setup_memory_region (I'm sorry; I'd give you a patch except
> I put a K7 into my build machine and it promptly killed the disk :-()
> and tell me what it says?
No problem, Done...
> >This box has 96m, but only 64m was seen (like 2.2.x)
> >
> >memory region: 639k @ 00000000
> >memory region: 65535k @ 00100000
> >Memory: 63584k/66556k available (1100k kernel code, 412k reserved, 1356k data, 1
>
> Are you getting any e820: messages?
Not before your patch, but now I get:
Linux version 2.3.18ac8 (root@badlands) (gcc version 2.7.2.3) #7 Wed Sep 22 12:04:11 EDT 1999
e820: E820_MAP_NR = 1
Initializing CPU#0
Detected 199433617 Hz processor.
Console: colour VGA+ 80x34
Calibrating delay loop... 398.13 BogoMIPS
memory region: 639k @ 00000000
memory region: 65535k @ 00100000
Memory: 63560k/66556k available (1120k kernel code, 416k reserved, 1356k data, 100k init, 16636k bigmem)
This is an older IBM PC 300GL with a 430VX mobo
sigh... /me wonders what benefits are gained via customized boards/BIOSes that
keeps a company doing things like this
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/