On Mon, Nov 12, 2018 at 04:15:46PM +0800, Peter Xu wrote: > On Fri, Nov 09, 2018 at 07:49:45PM +0800, Yu Zhang wrote: > > Currently, vIOMMU is using the value of IOVA address width, instead of > > the host address width(HAW) to calculate the number of reserved bits in > > data structures such as root entries, context entries, and entries of > > DMA paging structures etc. > > > > However values of IOVA address width and of the HAW may not equal. For > > example, a 48-bit IOVA can only be mapped to host addresses no wider than > > 46 bits. Using 48, instead of 46 to calculate the reserved bit may result > > in an invalid IOVA being accepted. > > > > To fix this, a new field - haw_bits is introduced in struct IntelIOMMUState, > > whose value is initialized based on the maximum physical address set to > > guest CPU. Also, definitions such as VTD_HOST_AW_39/48BIT etc. are renamed > > to clarify. > > IIRC I raised this question some time ago somewhere but no one > remembered to follow that up. Thanks for fixing it. > > It looks mostly good to me, only one tiny comment below... > > [...] > > > @@ -887,6 +887,7 @@ static int vtd_page_walk_level(dma_addr_t addr, > > uint64_t start, > > uint64_t iova = start; > > uint64_t iova_next; > > int ret = 0; > > + uint8_t haw = info->as->iommu_state->haw_bits; > > For now vtd_page_walk_info->aw_bits caches the GAW information and we > use a single vtd_page_walk_info during one page walk, maybe we can > also do the same for HAW instead of fetching it every time here from > info->as->iommu_state->haw_bits?
Thank you, Peter. And yes, using haw_bits in vtd_page_walk_info is better. :) > > Regards, > > -- > Peter Xu > B.R. Yu