As I understand the code, KGPE-d16 doesn't use AGESA part (nor any?).
Any reason why this value would be defaulting to enabled for whole sb700 dependents?
--- a/src/vendorcode/amd/cimx/sb700/SBTYPE.h
+++ b/src/vendorcode/amd/cimx/sb700/SBTYPE.h
@@ -133,7 +133,7 @@ typedef struct _AMDSBCFG
UINT32 SataPortMultCap :1; //6, 0:OFF 1:ON
UINT32 SataReserved :2; //8:7, Reserved
UINT32 SataClkAutoOff :1; //9, AutoClockOff for IDE modes 0:Disabled, 1:Enabled
- UINT32 SataIdeCombinedMode :1; //10, SataIDECombinedMode 0:Disabled, 1:Enabled
+ UINT32 SataIdeCombinedMode :0; //10, SataIDECombinedMode 0:Disabled, 1:Enabled
UINT32 SataIdeCombMdPriSecOpt:1; //11, Combined Mode, SATA as primary or secondary 0:primary 1:secondary
UINT32 SataReserved1 :6; //17:12, Not used currently
UINT32 SataEspPort :6; //23:18 SATA port is external accessiable on a signal only connector (eSATA:)
Thierry Laurion wrote:
> ENABLE_IDE_COMBINED_MODE available for sp800 but not for sp700, for ewhich
> sp5100 is derived from:
..
> Suggested Workaround
> Disable combined mode by setting a platform BIOS callback option to CIMx
> called "SataIdeCombinedMode" to 0.
..
> Is there something i'm missing? Is it possible to disable combined
> sata mode for sb700 from coreboot?
SB700 mainboard support seems copypasted rather than engineered. The
sustainable solution is to move sb700_cfg.c from src/mainboard/*/
to src/southbridge/amd/cimx/sb700/ and hook the configuration in that
file into Kconfig.
Until someone does that, you could indeed change the assignment of
that option by looking for
SataIdeCombinedMode
in the source code, if a sb700_cimx_config() function is called for
this board - but that might not be the case if the board port chose
not to use that part of AGESA.
//Peter
--
coreboot mailing list: coreboot@coreboot.org
https://mail.coreboot.org/mailman/listinfo/coreboot