On Fri, 26 Jul 2013 09:18:03 +0200 Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net wrote:
I still have to search/replace Hudson with Yangtze, but this patch will help a lot finding out more about the various FIFO related registers.
A45/A50M/A55E (aka Hudson-1) has the same id as sb7xx-sb9xx: 0x439d (doc id 47777, page 301).
SB6xx is using 0x438d, which Carl-Daniel uses in his patch to test for hudson... how come? I guess because the block diagram in figure 2 in 47777 shows this, but IMHO that's wrong.
So what about Kabini aka Hudson-3? The BKDG (48751) lists 0x780e for the LPC bridge, which is true on my ASRock A180-H too: 00:14.3 ISA bridge [0601]: Advanced Micro Devices [AMD] FCH LPC Bridge [1022:780e] (rev 11) Subsystem: ASRock Incorporation Device [1849:780e]
The problem is Hudson-2... no public data sheets. :/ Also interesting: http://pci-ids.ucw.cz/read/PC/1022/780e
I have not looked at the coreboot source yet... maybe there is stuff.
I assume that all Hudson-2 and newer are 0x780e and use the new interface. I would therefore propose to use "FCH" as a name for anything related to this class of spi controllers.