From: Robin Murphy <robin.murphy@xxxxxxx> commit 5b61343b50590fb04a3f6be2cdc4868091757262 upstream. For various reasons based on the allocator behaviour and typical use-cases at the time, when the max32_alloc_size optimisation was introduced it seemed reasonable to couple the reset of the tracked size to the update of cached32_node upon freeing a relevant IOVA. However, since subsequent optimisations focused on helping genuine 32-bit devices make best use of even more limited address spaces, it is now a lot more likely for cached32_node to be anywhere in a "full" 32-bit address space, and as such more likely for space to become available from IOVAs below that node being freed. At this point, the short-cut in __cached_rbnode_delete_update() really doesn't hold up any more, and we need to fix the logic to reliably provide the expected behaviour. We still want cached32_node to only move upwards, but we should reset the allocation size if *any* 32-bit space has become available. Reported-by: Yunfei Wang <yf.wang@xxxxxxxxxxxx> Signed-off-by: Robin Murphy <robin.murphy@xxxxxxx> Reviewed-by: Miles Chen <miles.chen@xxxxxxxxxxxx> Link: https://lore.kernel.org/r/033815732d83ca73b13c11485ac39336f15c3b40.1646318408.git.robin.murphy@xxxxxxx Signed-off-by: Joerg Roedel <jroedel@xxxxxxx> Cc: Miles Chen <miles.chen@xxxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/iommu/iova.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -95,10 +95,11 @@ __cached_rbnode_delete_update(struct iov cached_iova = to_iova(iovad->cached32_node); if (free == cached_iova || (free->pfn_hi < iovad->dma_32bit_pfn && - free->pfn_lo >= cached_iova->pfn_lo)) { + free->pfn_lo >= cached_iova->pfn_lo)) iovad->cached32_node = rb_next(&free->node); + + if (free->pfn_lo < iovad->dma_32bit_pfn) iovad->max32_alloc_size = iovad->dma_32bit_pfn; - } cached_iova = to_iova(iovad->cached_node); if (free->pfn_lo >= cached_iova->pfn_lo)
