Re: [coreboot] Seabios stop at Harddisk boot.

I found (hint by ruik=Rudolf ?) it is  due to the memory I reserved
"004000 - 0000000000029000 = 2".

After I remove it from coreboot-v2 coreboot_table.c.  seabios+windows
work fine.

Seabios is very cool.

I  will try to install windows driver and try s3 then.

________________________________

From: Jason Zhao 
Sent: Thursday, November 13, 2008 3:30 PM
To: 'Kevin O'Connor'
Cc: [e-mail]
Subject: Seabios stop at Harddisk boot.

The same Coreboot and the same board had boot Linux ok.

I try seabios then.  Vga has normal output on screen.

After I choose to boot from Harddisk, it stop at "fail
handle_15XX:314(00000086):"

Any idea about this?

------------------------------------------------------------------------
----------------------------------------------------------------

Copying ACPI RSDP from 1bef0000 to 000ff330

Found 1 cpu(s)

SMBIOS table addr=0x000ff350

ram_size=0x1bef0000

Scan for VGA option rom

Running option rom at 0000c000:00000003

Turning on vga console

Starting SeaBIOS

keyboard initialized

e820 map has 7 items:

0: 0000000000000000 - 0000000000004000 = 1

1: 0000000000004000 - 0000000000029000 = 2 (((((((((I reserved this,
can remove it if needed.))))))))))

2: 0000000000029000 - 000000000009fc00 = 1

3: 000000000009fc00 - 00000000000a0000 = 2

4: 00000000000f0000 - 0000000000100000 = 2

5: 0000000000100000 - 000000001bef0000 = 1

6: 000000001bef0000 - 000000001bf00000 = 2

final bios_table_addr: 0x000ff45a (used 48%)

ATA controller 0 at 000001f0/000003f0

ATA controller 1 at 00000170/00000370

ata1-0: PCHS=16383/16/63 translation=lba LCHS=1024/255/63

ata1 master: ST3120213A ATA-7 Hard-Disk (111 GiBytes)

ata1  slave: Unknown device

Scan for option roms

Press F12 for boot menu.

Select boot device:

1. Floppy

2. Hard Disk

3. CD-Rom

(((((((((I chose Hard Disk))))))))))

Changing serial settings was 00000003/00000002 now 00000003/00000000

enter handle_19:

NULL

Booting from Hard Disk...

Booting from 00000000:00007c00

enter handle_12:

a=00000000 b=00000000 c=00000000 d=00000080 si=00002118 di=0000402c

ds=00000000 es=00000000 ip=000024c1 cs=00000000 f=00000297 r=00001fd2

enter handle_12:

a=00000000 b=00000000 c=00000000 d=00022780 si=00002118 di=00054480

ds=00000000 es=00000000 ip=00008a01 cs=00000000 f=00000297 r=00001fd2

fail get_device:610(00000001):

a=00004100 b=000055aa c=fffabad4 d=00054581 si=00067fc8 di=00000081

ds=00000000 es=00000000 ip=000088b5 cs=00000000 f=00000246 r=00001fd2

fail get_device:610(00000001):

a=00000800 b=00000000 c=fffabad4 d=00054581 si=00067fc8 di=00000081

ds=00000000 es=00000000 ip=000088e9 cs=00000000 f=00000246 r=00001fd2

fail handle_15XX:314(00000086):

a=000000c0 b=00000000 c=fffaba03 d=00000001 si=00054538 di=00067fc8

ds=00000000 es=00000000 ip=00008ab6 cs=00000000 f=00000246 r=00001fd2

------------------------------------------------------------------------
------------------------------

Seabios stop at Harddisk boot.

[Rudolf Marek]

[e-mail] wrote:
> I found (hint by ruik=Rudolf ?) it is  due to the memory I reserved 
> “004000 - 0000000000029000 = 2”.

Yes thats me ;) I have nick Ruik but using Rudolf as for emails. Please note 
that windows dislike to be reserved memory "a000 and above" leave the hole 
there. In other words leave it as it is ;)

Also if you will move the coreboot to some higher location keep in mind that 
windows/ACPI needs first 8MB without any reserved memory. As we discussed on IRC 
best would be to move the coreboot to some TOPMEM-VGAMEM-something And then use 
the backup approach for first 1MB.

I recall someone had the "poison memory patch" to observe what memory was 
touched by coreboot? Uwe? Carldani?

Thanks,
Rudolf

Seabios stop at Harddisk boot.

[Carl-Daniel Hailfinger]

On 13.11.2008 11:13, Rudolf Marek wrote:
> Also if you will move the coreboot to some higher location keep in
> mind that windows/ACPI needs first 8MB without any reserved memory. As
> we discussed on IRC best would be to move the coreboot to some
> TOPMEM-VGAMEM-something And then use the backup approach for first 1MB.
>
> I recall someone had the "poison memory patch" to observe what memory
> was touched by coreboot? Uwe? Carldani?

Yes, that was me. I will resend the patch.

Regards,
Carl-Daniel