Change in coreboot[master]: mb/supermicro/x11-lga1151-series: switch from dev.init to mb_ops.init

[Arthur Heymans (Code Review)]

Arthur Heymans has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/48083 )

Change subject: mb/supermicro/x11-lga1151-series: switch from dev.init to mb_ops.init
......................................................................

Patch Set 6:

(1 comment)

https://review.coreboot.org/c/coreboot/+/48083/5//COMMIT_MSG 
Commit Message:

https://review.coreboot.org/c/coreboot/+/48083/5//COMMIT_MSG@9 
PS5, Line 9: Set mainboard_ops.init directly instead of using the indirection via a
           : mainboard_enable function.
> huh? this does not change any behaviour. mainboard_ops. […]

So there is a difference between struct device_operations and struct chip_operations. Both have an init function which is probably confusing you. The ramstage are as follows:

BS_PRE_DEVICE,
    BS_DEV_INIT_CHIPS,
    BS_DEV_ENUMERATE,
    BS_DEV_RESOURCES,
    BS_DEV_ENABLE,
    BS_DEV_INIT,
    BS_POST_DEVICE,
    BS_OS_RESUME_CHECK,
    BS_OS_RESUME,
    BS_WRITE_TABLES,
    BS_PAYLOAD_LOAD,
    BS_PAYLOAD_BOOT,

So if you place this in the struct chip_operations it will run before any resource allocation. This is why the mainboard code often places init code in the struct device_operations like the code does here (before this change).
Now granted, it looks like this mainboard init code only thouches prc, which is likely ok to do, as a general rule you want that to happen after resource allocation.