From: "Liam R. Howlett" Subject: kernel/fork: use maple tree for dup_mmap() during forking The maple tree was already tracking VMAs in this function by an earlier commit, but the rbtree iterator was being used to iterate the list. Change the iterator to use a maple tree native iterator and switch to the maple tree advanced API to avoid multiple walks of the tree during insert operations. Unexport the now-unused vma_store() function. For performance reasons we bulk allocate the maple tree nodes. The node calculations are done internally to the tree and use the VMA count and assume the worst-case node requirements. The VM_DONT_COPY flag does not allow for the most efficient copy method of the tree and so a bulk loading algorithm is used. Link: https://lkml.kernel.org/r/20220426150616.3937571-16-Liam.Howlett@oracle.com Signed-off-by: Liam R. Howlett Signed-off-by: Matthew Wilcox (Oracle) Acked-by: Vlastimil Babka Cc: Catalin Marinas Cc: David Howells Cc: Will Deacon Cc: Yu Zhao Signed-off-by: Andrew Morton --- include/linux/mm.h | 2 -- kernel/fork.c | 24 ++++++++++++++++++------ mm/mmap.c | 2 +- 3 files changed, 19 insertions(+), 9 deletions(-) --- a/include/linux/mm.h~kernel-fork-use-maple-tree-for-dup_mmap-during-forking +++ a/include/linux/mm.h @@ -2607,8 +2607,6 @@ extern bool arch_has_descending_max_zone /* nommu.c */ extern atomic_long_t mmap_pages_allocated; extern int nommu_shrink_inode_mappings(struct inode *, size_t, size_t); -/* mmap.c */ -void vma_store(struct mm_struct *mm, struct vm_area_struct *vma); /* interval_tree.c */ void vma_interval_tree_insert(struct vm_area_struct *node, --- a/kernel/fork.c~kernel-fork-use-maple-tree-for-dup_mmap-during-forking +++ a/kernel/fork.c @@ -583,7 +583,9 @@ static __latent_entropy int dup_mmap(str struct vm_area_struct *mpnt, *tmp, *prev, **pprev; struct rb_node **rb_link, *rb_parent; int retval; - unsigned long charge; + unsigned long charge = 0; + MA_STATE(old_mas, &oldmm->mm_mt, 0, 0); + MA_STATE(mas, &mm->mm_mt, 0, 0); LIST_HEAD(uf); uprobe_start_dup_mmap(); @@ -615,7 +617,12 @@ static __latent_entropy int dup_mmap(str khugepaged_fork(mm, oldmm); prev = NULL; - for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) { + + retval = mas_expected_entries(&mas, oldmm->map_count); + if (retval) + goto out; + + mas_for_each(&old_mas, mpnt, ULONG_MAX) { struct file *file; if (mpnt->vm_flags & VM_DONTCOPY) { @@ -629,7 +636,7 @@ static __latent_entropy int dup_mmap(str */ if (fatal_signal_pending(current)) { retval = -EINTR; - goto out; + goto loop_out; } if (mpnt->vm_flags & VM_ACCOUNT) { unsigned long len = vma_pages(mpnt); @@ -695,7 +702,9 @@ static __latent_entropy int dup_mmap(str rb_parent = &tmp->vm_rb; /* Link the vma into the MT */ - vma_store(mm, tmp); + mas.index = tmp->vm_start; + mas.last = tmp->vm_end - 1; + mas_store(&mas, tmp); mm->map_count++; if (!(tmp->vm_flags & VM_WIPEONFORK)) @@ -705,10 +714,13 @@ static __latent_entropy int dup_mmap(str tmp->vm_ops->open(tmp); if (retval) - goto out; + goto loop_out; + } /* a new mm has just been created */ retval = arch_dup_mmap(oldmm, mm); +loop_out: + mas_destroy(&mas); out: mmap_write_unlock(mm); flush_tlb_mm(oldmm); @@ -724,7 +736,7 @@ fail_nomem_policy: fail_nomem: retval = -ENOMEM; vm_unacct_memory(charge); - goto out; + goto loop_out; } static inline int mm_alloc_pgd(struct mm_struct *mm) --- a/mm/mmap.c~kernel-fork-use-maple-tree-for-dup_mmap-during-forking +++ a/mm/mmap.c @@ -756,7 +756,7 @@ static inline void vma_mt_szero(struct m * @mm: The struct_mm * @vma: The vm_area_struct to store in the maple tree. */ -void vma_store(struct mm_struct *mm, struct vm_area_struct *vma) +static void vma_store(struct mm_struct *mm, struct vm_area_struct *vma) { MA_STATE(mas, &mm->mm_mt, 0, 0); _