Hi Christoph Thanks for the series! I have a couple of questions: - Is there any platform where dma_alloc_noncontiguos can fail? This is, !ops->alloc_noncontiguous and !dev->coherent_dma_mask If yes then we need to add a function to let the driver know in advance that it has to use the coherent allocator (usb_alloc_coherent for uvc) - In dma_alloc_noncontiguos, on the dma_alloc_pages fallback. If we have a device where the dma happens in only one direction, could not get more performance with DMA_FROM/TO_DEVICE instead of DMA_BIDIRECTIONAL ? Then I have tried to use the API, and I have encountered a problem: on uvcvideo the device passed to the memory allocator is different for DMA_PAGES and NON_CONTIGUOUS: https://github.com/ribalda/linux/blob/042cd497739f71c8d4a83a67ee970369e2baca4a/drivers/media/usb/uvc/uvc_video.c#L1236 I need to dig a bit tomorrow to figure out why this is, I have hardware to test both paths, so it should not be too difficult. Thanks again On Tue, Jan 26, 2021 at 6:07 PM . Christoph Hellwig <hch@xxxxxx> wrote: > > Please take a quick look at this branch: > > http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/dma_alloc_noncontiguous > > Warning: hot off the press, and only with the v4l conversion as that > seemed at little easier than uvcvideo. -- Ricardo Ribalda
