From 425333bf3a7743715c17e503049d0837d6c4a603 Mon Sep 17 00:00:00 2001 From: Alexey Kardashevskiy Date: Mon, 10 Sep 2018 18:29:07 +1000 Subject: KVM: PPC: Avoid marking DMA-mapped pages dirty in real mode At the moment the real mode handler of H_PUT_TCE calls iommu_tce_xchg_rm() which in turn reads the old TCE and if it was a valid entry, marks the physical page dirty if it was mapped for writing. Since it is in real mode, realmode_pfn_to_page() is used instead of pfn_to_page() to get the page struct. However SetPageDirty() itself reads the compound page head and returns a virtual address for the head page struct and setting dirty bit for that kills the system. This adds additional dirty bit tracking into the MM/IOMMU API for use in the real mode. Note that this does not change how VFIO and KVM (in virtual mode) set this bit. The KVM (real mode) changes include: - use the lowest bit of the cached host phys address to carry the dirty bit; - mark pages dirty when they are unpinned which happens when the preregistered memory is released which always happens in virtual mode; - add mm_iommu_ua_mark_dirty_rm() helper to set delayed dirty bit; - change iommu_tce_xchg_rm() to take the kvm struct for the mm to use in the new mm_iommu_ua_mark_dirty_rm() helper; - move iommu_tce_xchg_rm() to book3s_64_vio_hv.c (which is the only caller anyway) to reduce the real mode KVM and IOMMU knowledge across different subsystems. This removes realmode_pfn_to_page() as it is not used anymore. While we at it, remove some EXPORT_SYMBOL_GPL() as that code is for the real mode only and modules cannot call it anyway. Signed-off-by: Alexey Kardashevskiy Reviewed-by: David Gibson Signed-off-by: Paul Mackerras --- arch/powerpc/mm/mmu_context_iommu.c | 34 ++++++++++++++++++++++++++++++---- 1 file changed, 30 insertions(+), 4 deletions(-) (limited to 'arch/powerpc/mm/mmu_context_iommu.c') diff --git a/arch/powerpc/mm/mmu_context_iommu.c b/arch/powerpc/mm/mmu_context_iommu.c index c9ee9e23845f..56c2234cc6ae 100644 --- a/arch/powerpc/mm/mmu_context_iommu.c +++ b/arch/powerpc/mm/mmu_context_iommu.c @@ -18,11 +18,15 @@ #include #include #include +#include #include #include static DEFINE_MUTEX(mem_list_mutex); +#define MM_IOMMU_TABLE_GROUP_PAGE_DIRTY 0x1 +#define MM_IOMMU_TABLE_GROUP_PAGE_MASK ~(SZ_4K - 1) + struct mm_iommu_table_group_mem_t { struct list_head next; struct rcu_head rcu; @@ -263,6 +267,9 @@ static void mm_iommu_unpin(struct mm_iommu_table_group_mem_t *mem) if (!page) continue; + if (mem->hpas[i] & MM_IOMMU_TABLE_GROUP_PAGE_DIRTY) + SetPageDirty(page); + put_page(page); mem->hpas[i] = 0; } @@ -360,7 +367,6 @@ struct mm_iommu_table_group_mem_t *mm_iommu_lookup_rm(struct mm_struct *mm, return ret; } -EXPORT_SYMBOL_GPL(mm_iommu_lookup_rm); struct mm_iommu_table_group_mem_t *mm_iommu_find(struct mm_struct *mm, unsigned long ua, unsigned long entries) @@ -390,7 +396,7 @@ long mm_iommu_ua_to_hpa(struct mm_iommu_table_group_mem_t *mem, if (pageshift > mem->pageshift) return -EFAULT; - *hpa = *va | (ua & ~PAGE_MASK); + *hpa = (*va & MM_IOMMU_TABLE_GROUP_PAGE_MASK) | (ua & ~PAGE_MASK); return 0; } @@ -413,11 +419,31 @@ long mm_iommu_ua_to_hpa_rm(struct mm_iommu_table_group_mem_t *mem, if (!pa) return -EFAULT; - *hpa = *pa | (ua & ~PAGE_MASK); + *hpa = (*pa & MM_IOMMU_TABLE_GROUP_PAGE_MASK) | (ua & ~PAGE_MASK); return 0; } -EXPORT_SYMBOL_GPL(mm_iommu_ua_to_hpa_rm); + +extern void mm_iommu_ua_mark_dirty_rm(struct mm_struct *mm, unsigned long ua) +{ + struct mm_iommu_table_group_mem_t *mem; + long entry; + void *va; + unsigned long *pa; + + mem = mm_iommu_lookup_rm(mm, ua, PAGE_SIZE); + if (!mem) + return; + + entry = (ua - mem->ua) >> PAGE_SHIFT; + va = &mem->hpas[entry]; + + pa = (void *) vmalloc_to_phys(va); + if (!pa) + return; + + *pa |= MM_IOMMU_TABLE_GROUP_PAGE_DIRTY; +} long mm_iommu_mapped_inc(struct mm_iommu_table_group_mem_t *mem) { -- cgit v1.2.3