Hi Tim, On Wednesday 18 March 2015 10:19:15 Tim Nordell wrote: > Laurent - > > On 03/18/15 09:59, Laurent Pinchart wrote: > > Hi Tim, > > The names might be a bit misleading, vb2-dma-contig requires contiguous > > memory in the device memory space, not in physical memory. The IOMMU, > > managed through dma_map_sg_attrs, should have mapped the userptr buffer > > contiguously in the ISP DMA address space. If it hasn't, that's what need > > to be investigated. > > I see now that it's the sg_dma_address(...) call that it's using and I > was assuming that was the physical memory address for the memory backing > up the buffer. > > So the vb2_dc_get_contiguous_size(...) should be against the view of > memory that the IOMMU presents to the OMAP3 ISP? That is, the DMA > addresses that the OMAP3 ISP can see? Correct. sg_dma_address() should contain the DMA virtual address when an IOMMU is used. > I was assuming it was checking the physical memory layout to it without > looking too closely to the code. Armed with that knowledge, I'll dig a > little deeper to see if I can figure out what happened here. Thank you. -- Regards, Laurent Pinchart -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
