FYI: Testing for RPi4B (and other BCM2711 device) SoC revisions

[Mark Millard via freebsd-arm]

Looks like EDK2 is getting work on dealing with the updated
BCM2711 vintages that do not require as many workarounds or
imposed limitations.

https://github.com/jlinton/edk2-platforms/commit/769d36d6d400b1dac2ec5e8a7b3eebd0a4751b8b

has AML code based on the following sort of logic

#define ID_CHIPREV 0xfc404000
. . .

OperationRegion (CHPR, SystemMemory, ID_CHIPREV, 0x4)
Field (CHPR, DWordAcc, NoLock, Preserve) {
SOCI, 32
}

if ((SOCI & 0xFF) >= 0x20)
. . .
else
. . .

(Not bothering with which file contains which text here.)
This (and the not shown code) is described with:

QUOTE
The newer BCM2711 SoC's don't have a DMA constraint on the emmc2
controller. So we don't need to do the 1M translation. Lets
allow the AML to detect the SoC revision and return a different
_DMA resource.
END QUOTE

My guess is that the above presumes that the context
has already been established to be a BCM2711 one.

Note: I'm not AML literate. So I'm guessing
interpretations for the code shown.

===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)