On Sun, Oct 14, 2018 at 09:46:44AM +0200, Takashi Iwai wrote: > Takashi Iwai wrote: > > Christoph Hellwig wrote: > > > The DMA API does its own zone decisions based on the coherent_dma_mask. > > > Signed-off-by: Christoph Hellwig > > Reviewed-by: Takashi Iwai > Mark, let me know whether you'd take this via your tree or may I take > via mine (with a minor correction of the subject line). I don't have the patch... in fact, having found it in list archives I see there's also a SPI patch I'll have to go hunting for so I guess I'll apply this one when I find that :( Christoph, please CC maintainers. > FWIW, the driver sets the 31bit (!) DMA mask, and that's the likely > reason of GFP_DMA passed there in the original code. Likely, yes.