On 20.06.2009 00:02, Myles Watson wrote: >>> Field Descriptions >>> Scratch Data (Data)-Bits 31-0. >>> >>> So you get one per socket, first one at PCI_DEV(0,0x18,2) 0x9C >>> >>> >> AFAICS this is not MMIO, so it's unusable for locking. >> > > ? I thought you were looking for config space. Why does it need to be > MMIO? >
I'm unaware of any method to perform atomic value swaps or atomic increments in config space. Unless I'm mistaken, such atomic operations are required for locking. >> I did look at the >> Family 0Fh BKDG. Should I have looked elsewhere? >> > > I'm not an expert. I'd just seen this scratch register recently. > Thanks for digging it up. It seems that some CPU revisions change the definition to be partially a scratch register. Regards, Carl-Daniel -- http://www.hailfinger.org/ -- coreboot mailing list: [email protected] http://www.coreboot.org/mailman/listinfo/coreboot
