Hi Andrew, This is a miscellany of fixes and tweaks to the common DMA mapping code which I've been collecting. They don't strictly depend on each other, but I figure I may as well send them all together for the sake of explaining them in one place: #1 is a straightforward and hopefully obvious bugfix. #2 has been posted before but I subsequently forgot to follow up on it. It's still possible to hit this at least on arm64 systems, with the SWIOTLB/pl330 combination blowing up just running dmatest. #3 follows on from a recent discussion about dma_sync_sg[0]; there's already a related patch in -next from Sakari clarifying the docs. #4 seemed worth posting now that the recent rework means it no longer has to be a sprawling touch-all-the-architectures patch for something so small. Thanks, Robin. [0]:http://thread.gmane.org/gmane.linux.kernel/2043117 Robin Murphy (4): dmapool: Fix overflow condition in pool_find_page dma-mapping: Tidy up dma_parms default handling dma-debug: Check nents in dma_sync_sg* dma-debug: Allow poisoning nonzero allocations include/asm-generic/dma-mapping-common.h | 2 +- include/linux/dma-debug.h | 6 ++++-- include/linux/dma-mapping.h | 17 ++++++++++------- include/linux/poison.h | 3 +++ lib/Kconfig.debug | 10 ++++++++++ lib/dma-debug.c | 14 +++++++++++++- mm/dmapool.c | 2 +- 7 files changed, 42 insertions(+), 12 deletions(-) -- 1.9.1 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href