On Tue, Jul 12, 2011 at 10:18 AM, xdrudis xdrudis@tinet.cat wrote:
On Sat, Jul 09, 2011 at 05:04:43PM +0200, Florentin Demetrescu wrote:
- my objective was to install coreboot on my new board MA785GMT-UDH2. I had bring with me a Phenom II 1055T CPU with 6 cores. Unfortunately I met big problems because:
[...]
coreboot and give it a run, but I will do that ASAP! Also I will investigate the problem of the 6 core Phenom II on this board..
Isn't it a fam 10 revision E CPU ?
Coreboot did not have any code specific for Fam 10 rev E last time I checked (mid feb 2011) (no errata workarounds, no specific initialization, just a small untested part in fidvid.c). There wasn't even a constant defined for rev E.
Back in August 2010 I asked how to extend the revision bitfield that's used as a trigger for rule based initializations and workarounds, and in a small thread it was suggested to get rid of it and use a struct, but I never did. http://www.coreboot.org/pipermail/coreboot/2010-August/059701.html
The C32 support might handle rev E. I don't know if there is much different from rev D (bigger differences between revC and revD). If more support is required, there may be some updates from AMD in the future.
Marc