[coreboot] Coreboot fails to initialize on ASUS A8V-E SE

Knut Kujat knuku at gap.upv.es
Mon Dec 7 18:05:09 CET 2009


Myles Watson escribió:
>
>
>     worked fine now I'm getting a little further ;) :
>
> Good.
>  
>
>     coreboot-2.3 Fri Dec  4 20:15:37 CET 2009
>     starting...                                                                        
>
>     now booting...
>     real_main                                                                                                     
>
>     core0 started:
>     now booting... Core0 started
>     started ap apicid:
>     SBLink=00
>     NC node|link=00
>     K8T890 found at LDT 00 Agreed on width: 01 CPU programmed to HT
>     freq: 06 VIA HT caps: 0075
>     ht reset -
>     soft reset
>
>
>     coreboot-2. normal_image replaced with fallback_imageWelcome to
>     the real_main!
>
>     coreboot-2.3 Fri Dec  4 20:15:37 CET 2009 starting...
>     now booting... real_main
>     core0 started:
>     now booting... Core0 started
>     started ap apicid:
>     SBLink=00
>     NC node|link=00
>     K8T890 found at LDT 00 Agreed on width: 01 CPU programmed to HT
>     freq: 06 VIA HT caps: 0075
>     Current fid_cur: 0x10, fid_max: 0x10
>     Requested fid_new: 0x10
>     Debug: after init_fidvid_bsp
>     Debug: after allow_all_aps_stop
>     Debug: after fill_mem_ctrl
>     Debug: after enable_smbus
>     Debug: after memreset_setup
>     Ram1.00
>     Ram2.00
>     Device error
>     Device error
>     No memory for this cpu
>     Ram3
>     No memory
>
>     but now it seems that coreboot doesn't find any ram. I already
>     tried with patched Kconfig and without same result.
>
>
> I haven't played with RAM initialization much.  The "Device error"
> message is coming from the smbus, so I would try to see how that is
> configured and put more debugging statements in there.  I see that
> enable_smbus ran, so I don't know what's missing.  Hopefully someone
> who knows more will chime in.
>
> Is there a cmos option that could be causing trouble?  I'd check that
> too since the board was choosing to do a normal boot.
>
> Thanks,
> Myles
Hi,

thanks for your help so far. I cleared the CMOS before restart but nothing.

Ram1.00
setting up CPU00 northbridge registers
done.
Ram2.00
Debug: in smbus_read_byte
Debug: after smbus_reset
Debug: after SMBUS_DELAY (first)
Debug: after smbus_wait-until_ready (first)
Debug: after SMBUS_DELAY (second)
Device error
Debug: smbus_wait-until_ready (second)
Debug: out smbus_read_byte
Debug: in smbus_read_byte
Debug: after smbus_reset
Debug: after SMBUS_DELAY (first)
Debug: after smbus_wait-until_ready (first)
Debug: after SMBUS_DELAY (second)
Debug: smbus_wait-until_ready (second)
Debug: out smbus_read_byte
Debug: in smbus_read_byte
Debug: after smbus_reset
Debug: after SMBUS_DELAY (first)
Debug: after smbus_wait-until_ready (first)
Debug: after SMBUS_DELAY (second)
Device error
Debug: smbus_wait-until_ready (second)
Debug: out smbus_read_byte
Debug: in smbus_read_byte
Debug: after smbus_reset
Debug: after SMBUS_DELAY (first)
Debug: after smbus_wait-until_ready (first)
Debug: after SMBUS_DELAY (second)
Debug: smbus_wait-until_ready (second)
Debug: out smbus_read_byte
No memory for this cpu
Ram3
No memory

This board has 4 sockets but only 2 are populated with 1GB dimms each.
Could the other empty two sockets be the device error, which is
generated by the smbus_wait_until_ready function?  But why would it say
that there is no memory at all?

Yet another question, I can see PRINT_DEBUG in vt8237r_early_smbus.c but
i can't see them on the serial although I have the highest message
level. Why?

thx,
Knut Kujat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.coreboot.org/pipermail/coreboot/attachments/20091207/cd4d5216/attachment.html>


More information about the coreboot mailing list