2023年2月22日 17:24，Robin Murphy [off-list ref] 写道：

[...]

AFAICS, all you should actually need is a single self-contained addition here, something like:

+ /*
+  * DT-based MIPS doesn't use OF_DMA_DEFAULT_COHERENT, but
+  * might override the system-wide default at runtime.
+  */
+#if defined(CONFIG_ARCH_HAS_SYNC_DMA_FOR_DEVICE) || \
+ defined(CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU) || \
+ defined(CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU_ALL)
+ is_coherent = dma_default_coherent;
+#endif

That makes more sense, thanks.
I’ll append CONFIG_MIPS as a condition here as well because it may break RISC-V whose dma_default_coherent
is not set to true from very beginning.

quoted

    node = of_node_get(np);
 

Then *after* that's fixed, we can do a more comprehensive refactoring to merge the two mechanisms properly. FWIW I think I'd prefer an approach closer to the first one, where config options control the initial value of dma_default_coherent rather than architectures having to override it unconditionally (and TBH I'd also like to have a generic config symbol for whether an arch supports per-device coherency or not).

Ok I’ll try to revert to the initial way.
Is there any reason that an arch can’t support per-device coherency?

Thanks
- Jiaxun

Thanks,
Robin.