[coreboot] v3 memtest Unexpected Interrupt - Halting

Carl-Daniel Hailfinger c-d.hailfinger.devel.2006 at gmx.net
Fri May 9 12:21:30 CEST 2008

On 09.05.2008 07:01, Peter Stuge wrote:
> A long v3 memtest was requested. It doesn't take long before Memtest
> stops, sometimes it is immediate after startup. The serial console
> looks like this:

I tried to make sense of the output, but I'm not completely sure I got
it right.

>       Memtest-86 v3.4       | Pass 10% ###

Which gcc? Newer gcc miscompile memtest86 and memtest86+ due to
undefined statements in the memtest code. This not only causes memtest
to throw errors which don't exist, in theory it also could lead to
overwriting of its own stack and code.

> [...]
> Unexpected Interrupt - Halting
>    Lowest Error Address: 00000fffffc
>  Type: Gen_Prot     eax: 3430d8ec
>    PC: 00003bb9     ebx: 00000000
>    CS: 00000010     ecx: ffffffff
> Eflag: 00010012s    edx: 037dfefc
>  Code: 00000000e    edi: ffffffff
>                     esi: 0001bab8
>                     ebp: 00000001
>                     esp: 0001b874
> CS:EIP: 89 07 39 fa 75 f1 83 ef 04 eb

It seems we definitely need memtest to use libpayload and display
graphical results. That would at least rule out problems caused by
serial output. Then again, having memtest send its results as POST codes
should work well if you can capture POST codes.


More information about the coreboot mailing list