head -r365677 and later do not have the xhci related DMA problem fixed

[Mark Millard]

On 2020-Sep-24, at 14:59, Robert Crowston <crowston at protonmail.com> wrote:

> Yes: aren’t you dependent on the unrelated D25219 to fix uefi?

I have had that patch in place right along. It does not fix the
xhci "> 3072 MiByte" DMA problem either.

> The DMA patch I submitted is at the PCI-e level; I haven’t touched xhci. Sorry for confusion!

Ahh. No central spot for dealing with such limitations related
to the xhci. Well, that makes uefi/ACPI support less likely
long term.

So, should I figure out the currently proper way to have a
u-boot based boot environment and try the huge-file duplicate
and diff/cmp test in such an environment?




Side note: While there is no CM4 compute module yet, they have
publicly stated that they currently intend on exposing the
PCIe on it once they have such as a product.

In fact, there is:

https://github.com/Hexxeh/rpi-firmware/blob/master/bcm2711-rpi-cm4.dtb

and:

https://github.com/raspberrypi/firmware/blob/master/boot/bcm2711-rpi-cm4.dtb

and even even official tagged builds with such:

https://github.com/raspberrypi/firmware/blob/1.20200902/boot/bcm2711-rpi-cm4.dtb


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