[PATCH v4 0/6] Relocate U-Boot in the last bank

[Tom Rini]

On Fri, May 15, 2026 at 04:52:21PM +0300, Ilias Apalodimas wrote:

> Hi This is v4 of [1][2][3]!
> 
> There was a discussion recently on the mailing lists regarding our
> management of memory above ram_top [0]. The tl;dr is that we have two problems.

FWIW, you dropped the reference to [0] which is:
https://lore.kernel.org/u-boot/CAC_iWjKFAzpj3B_MEW7-dnOrcAV-rfkhXXo8Bv0KgLNP2VJxRA@mail.gmail.com/

> The first one is that U-Boot always relocates to the top of the first available
> bank unless there's special board code to sidestep that. The second is we don't
> successfully deal with devices that can only do 32-bit DMA.
> 
> This patch series deals with the first problem by adding a Kconfig option
> allowing platforms to relocate to the top of the last discovered bank.
> 
> It's worth noting that this is easily testable with QEMU
> 
> qemu-system-aarch64 -m 8192 -smp 2 -nographic -cpu cortex-a57 \
>     -machine virt,secure=off \
>     -bios u-boot.bin \
>     -device virtio-rng-pci \
>     -drive id=os,if=none,file="$image" \
>     -device virtio-blk-device,drive=os \
>     -object memory-backend-ram,id=ram0,size=4G \
>     -object memory-backend-ram,id=ram1,size=4G \
>     -numa node,memdev=ram0 \
>     -numa node,memdev=ram1

So, now all of the sandbox runs in CI are failing over at least one of
the existing tests, and the Pi 4 platforms over in the sage lab are also
now failing the tftp kernel test.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20260527/0e74530d/attachment.sig>