Changes since v3: - Dropped patch 02/10: succeeded by patch 10/15 in this series - Added patches 11-15 to correctly handle syncing/mapping dmabuf buffers for CPU access. This was never done correctly before. Many thanks to Pawel Osciak for helping me with this during the media mini-summit last week. The patch series adds an allocation context to dma-sg and uses that to move dma_(un)map_sg into the vb2 framework, which is where it belongs. Some drivers needs to fixup the buffers before giving it back to userspace (or before handing it over to the kernel). Document that this can be done in buf_prepare and buf_finish. The last 5 patches make this more strict by requiring all cpu access to be bracketed by calls to vb2_plane_begin/end_cpu_access() which replaces the old vb2_plane_vaddr() call. Note: two drivers still use the vb2_plane_addr() call: coda and exynos4-is/fimc-capture.c. For both drivers I will need some help since I am not sure where to put the begin/end calls. Patch 14 removes the vb2_plane_vaddr call, so obviously those two drivers won't compile after that. DMABUF export support is added to dma-sg and vmalloc, so now all memory models support DMABUF importing and exporting. I am inclined to make a pull request for patches 1-10 if there are no new comments. The issues that patches 11-15 address are separate from the patches 1-10 and this is only an issue when using dmabuf with drivers that need cpu access. Reviews are very welcome. I plan on doing more testing next week, in particular with the patch series posted for vivid that add dma-contig. It's now easy to extend with dma-sg as well, so that should make testing a lot simpler. Regards, Hans -- 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
