[coreboot] [GSoC] Re: malloc.c problems of option rom

Stefan Reinauer stepan at coresystems.de
Mon Aug 3 02:19:36 CEST 2009


Peter Stuge wrote:
> ron minnich wrote:
>   
>>> alloc 82  length:0x238,ptr=0x3077e0
>>> alloc 83 memory begin:0x77e0 end:0x577e0, hstart:0x300000
>>> set up error,*ptr=0xa2c0e2cc
>>> memory allocator panic!!! the size=0xc0e2cc. header=0xa2c0e2cc
>>>       
>> You're trying to allocate 1.25 Mbytes of memory?
>>     
>
> I think that's some kind of corruption.
>
> 0x238 is the original allocation length. We looked into this a little
> on IRC and the two values ptr and hstart (lines 82/83 above) should
> be the same. There's an assignment ptr=hstart right before the prints
> in the source Jason sent. It's possible that some source changes may
> have been left out in the running build - I don't know exactly what's
> going on.
>
> I suggested reverting back to a version of the memory allocator which
> has a known behavior, even if it is known to fail I think it will be
> easier to debug and fix than this problem.
>   

The latest upstream version is working fine with FILO here. 

It might be worth trying to start with FILO and reduce stack and heap
until it's possible to reproduce the problems...

Problems with parameter passing sounds a lot like stack corruption.

Stefan



-- 
coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br.
      Tel.: +49 761 7668825 • Fax: +49 761 7664613
Email: info at coresystems.dehttp://www.coresystems.de/
Registergericht: Amtsgericht Freiburg • HRB 7656
Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866





More information about the coreboot mailing list