Hi,

On Monday 28 December 2009 17:21:04 Jon Beniston wrote:
> Did you build the toolchain that they supply, or was you trying to use a
> latter version of GCC?

Their toolchain build script (make vendor_toolchains) does not work, so I used 
the binary toolchain that they supply instead (http://www.theobroma-
systems.com/assets/downloads/mico32/lm32linux-20080206toolchains_linux.tar.gz). 
I installed it according to the instructions in their PDF 
(http://www.latticesemi.com/dynamic/view_document.cfm?document_id=28115 p.18). 
So there is no good reason for the generated LM32 executables to be different 
from those in their sample initrd image; but in the facts, they are. The 
Theobroma Linux port is extremely confusing. At last resort, I could try to 
use binary diffing and analysis utilities to spot the problem.

> > Is there documentation anywhere about the LM32 FDPIC, still?
> 
> Not really. It is similar to
> http://gcc.gnu.org/ml/gcc/2008-02/msg00619.html, although only a GOT is
> used, not a PLT. Perhaps your problem is related to the .rofixup section.

Ok, will check that out. Thanks for the tip.

Sébastien
_______________________________________________
http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org
IRC: #milkym...@freenode
Webchat: www.milkymist.org/irc.html
Wiki: www.milkymist.org/wiki

Reply via email to