On Thu, 19 Nov 2020 18:53:52 +0100, Nicolas Saenz Julienne wrote:
> Using two distinct DMA zones turned out to be problematic. Here's an
> attempt go back to a saner default.
>
> I tested this on both a RPi4 and QEMU.
Applied to arm64 (for-next/zone-dma-default-32-bit), thanks!
[1/7] arm64: mm: Move reserve_crashkernel() into mem_init()
https://git.kernel.org/arm64/c/0a30c53573b0
[2/7] arm64: mm: Move zone_dma_bits initialization into zone_sizes_init()
https://git.kernel.org/arm64/c/9804f8c69b04
[3/7] of/address: Introduce of_dma_get_max_cpu_address()
https://git.kernel.org/arm64/c/964db79d6c18
[4/7] of: unittest: Add test for of_dma_get_max_cpu_address()
https://git.kernel.org/arm64/c/07d13a1d6120
[5/7] arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges
https://git.kernel.org/arm64/c/8424ecdde7df
[6/7] arm64: mm: Set ZONE_DMA size based on early IORT scan
https://git.kernel.org/arm64/c/2b8652936f0c
[7/7] mm: Remove examples from enum zone_type comment
https://git.kernel.org/arm64/c/04435217f968
--
Catalin
