Hi all, Linus complained about the complex flow in dma_direct_alloc, so this tries to simplify it a bit, and while I was at it I also made sure that unencrypted pages never leak back into the page allocator.
Changes since v2: - don't call dma_set_decrypted on remapped memory - move the leak printk into dma_set_encrypted - add another local variable to clean up dma_direct_alloc - return NULL when the is no way to make the memory coherent Changes since v1: - fix a missing return - add a new patch to fix a pre-existing missing unmap - various additional cleanups Diffstat: direct.c | 234 +++++++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 138 insertions(+), 96 deletions(-) _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
