Re: [PATCH 3/3] dma-pool: limit DMA and DMA32 zone size pools

[Michal Hocko]

On Mon 22-08-22 13:14:47, Christoph Hellwig wrote:
> On Mon, Aug 22, 2022 at 10:41:01AM +0200, Michal Hocko wrote:
> > ZONE_DMA existance is not runtime controlable (at least on x86).
> 
> This is more about ZONE_DMA32 that can disappear if all memory is
> inside ZONE_DMA.  That is highly unusual (but still possible) with a
> tiny x86-style ZONE_DMA, but can happen with a larger one other
> architectures.

OK, so you are worried about configurations that lack ZONE_DMA32 because
that memory range is covered by ZONE_DMA? I thought though simply do not
have CONFIG_ZONE_DMA32. Or am I wrong here?

-- 
Michal Hocko
SUSE Labs