Re: port-evbarm/48669: evbarm/imx23_olinuxino (ARM9E) doesn't boot

[Petri Laakso <petri.laakso%asd.fi@localhost>]

On Thu, 20 Mar 2014 16:40:00 +0000 (UTC)
petri.laakso%asd.fi@localhost wrote:

> >Number:         48669
> >Category:       port-evbarm
> >Synopsis:       evbarm/imx23_olinuxino (ARM9E) doesn't boot
> >Confidential:   no
> >Severity:       serious
> >Priority:       medium
> >Responsible:    port-evbarm-maintainer
> >State:          open
> >Class:          sw-bug
> >Submitter-Id:   net
> >Arrival-Date:   Thu Mar 20 16:40:00 +0000 2014
> >Originator:     Petri Laakso
> >Release:        current
> >Organization:
> >Environment:
> >Description:
> I was reported problem to me that evbarm/imx23_olinuxino doesn't work.
> 
> All user is going to see is endless loop of:
> 
> bootimx23: HW revision TA5, boot reason 5V, power source 5V
> Enabling 4P2 regulator...done
> Enabling 4P2 regulator output to DCDC...done
> Enabling VDDMEM...done
> Powering VDDD from DCDC...done
> Powering VDDA from DCDC...done
> Powering VDDIO from DCDC...done
> Enabling clocks...done
> Configuring pins...done
> Configuring EMI...done
> Press any key to drop into boot prompt...
> 
> NetBSD/evbarm (OLinuXino) booting ...
> panic: pmap_map_chunk: no L2 table for VA 0xc0400000
> 
> Undefined Instruction
> r14_un
> 
> >How-To-Repeat:
> Try to run evbarm/imx23_olinuxino
> >Fix:
> 

System started to boot again when I reverted from 1.24 to 1.22.
src/sys/arch/arm/arm32/arm32_kvminit.c so problem lies somewhere
around there.

Petri