[PATCH v5] arm64: kdump: simplify the reservation behaviour of crashkernel=,high

[Catalin Marinas]

On Thu, Apr 13, 2023 at 03:45:50PM +0800, Baoquan He wrote:
> I am OK with this version, or the version with min(SZ_4G,
> arm64_dma_phys_limit), or v4. Please help point out if I got your idea
> correctly. Thanks a lot.

I think we should stick to this patch. The disabling of the ZONE_DMA(32)
is fairly specialised and you are right that we should not introduce an
artificial 4GB crashkernel boundary on such systems. The slight
confusion may be that ,high triggers a search above 4GB where there's
not such boundary but this would match the documentation anyway.

-- 
Catalin