Re: [PATCH 1/2] dma-mapping: fix page attributes for dma_mmap_*

[Shawn Anastasio <shawn@anastas.io>]

On 8/7/19 8:04 AM, Christoph Hellwig wrote:
> Actually it is typical modern Linux style to just provide a prototype
> and then use "if (IS_ENABLED(CONFIG_FOO))" to guard the call(s) to it.

I see.

>> Also, like Will mentioned earlier, the function name isn't entirely
>> accurate anymore. I second the suggestion of using something like
>> arch_dma_noncoherent_pgprot().
> 
> As mentioned I plan to remove arch_dma_mmap_pgprot for 5.4, so I'd
> rather avoid churn for the short period of time.

Yeah, fair enough.

>> As for your idea of defining
>> pgprot_dmacoherent for all architectures as
>>
>> #ifndef pgprot_dmacoherent
>> #define pgprot_dmacoherent pgprot_noncached
>> #endif
>>
>> I think that the name here is kind of misleading too, since this
>> definition will only be used when there is no support for proper
>> DMA coherency.
> 
> Do you have a suggestion for a better name?  I'm pretty bad at naming,
> so just reusing the arm name seemed like a good way to avoid having
> to make naming decisions myself.

Good question. Perhaps something like `pgprot_dmacoherent_fallback`
would better convey that this is only used for devices that don't
support DMA coherency? Or maybe `pgprot_dma_noncoherent`?