diff options
Diffstat (limited to 'mm/huge_memory.c')
| -rw-r--r-- | mm/huge_memory.c | 73 |
1 files changed, 38 insertions, 35 deletions
diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 470dcda10ad..81532f297fd 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1139,7 +1139,7 @@ static int __split_huge_page_splitting(struct page *page, * We can't temporarily set the pmd to null in order * to split it, the pmd must remain marked huge at all * times or the VM won't take the pmd_trans_huge paths - * and it won't wait on the anon_vma->root->lock to + * and it won't wait on the anon_vma->root->mutex to * serialize against split_huge_page*. */ pmdp_splitting_flush_notify(vma, address, pmd); @@ -1333,7 +1333,7 @@ static int __split_huge_page_map(struct page *page, return ret; } -/* must be called with anon_vma->root->lock hold */ +/* must be called with anon_vma->root->mutex hold */ static void __split_huge_page(struct page *page, struct anon_vma *anon_vma) { @@ -1408,6 +1408,9 @@ out: return ret; } +#define VM_NO_THP (VM_SPECIAL|VM_INSERTPAGE|VM_MIXEDMAP|VM_SAO| \ + VM_HUGETLB|VM_SHARED|VM_MAYSHARE) + int hugepage_madvise(struct vm_area_struct *vma, unsigned long *vm_flags, int advice) { @@ -1416,11 +1419,7 @@ int hugepage_madvise(struct vm_area_struct *vma, /* * Be somewhat over-protective like KSM for now! */ - if (*vm_flags & (VM_HUGEPAGE | - VM_SHARED | VM_MAYSHARE | - VM_PFNMAP | VM_IO | VM_DONTEXPAND | - VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | - VM_MIXEDMAP | VM_SAO)) + if (*vm_flags & (VM_HUGEPAGE | VM_NO_THP)) return -EINVAL; *vm_flags &= ~VM_NOHUGEPAGE; *vm_flags |= VM_HUGEPAGE; @@ -1436,11 +1435,7 @@ int hugepage_madvise(struct vm_area_struct *vma, /* * Be somewhat over-protective like KSM for now! */ - if (*vm_flags & (VM_NOHUGEPAGE | - VM_SHARED | VM_MAYSHARE | - VM_PFNMAP | VM_IO | VM_DONTEXPAND | - VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | - VM_MIXEDMAP | VM_SAO)) + if (*vm_flags & (VM_NOHUGEPAGE | VM_NO_THP)) return -EINVAL; *vm_flags &= ~VM_HUGEPAGE; *vm_flags |= VM_NOHUGEPAGE; @@ -1574,10 +1569,14 @@ int khugepaged_enter_vma_merge(struct vm_area_struct *vma) * page fault if needed. */ return 0; - if (vma->vm_file || vma->vm_ops) + if (vma->vm_ops) /* khugepaged not yet working on file or special mappings */ return 0; - VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + /* + * If is_pfn_mapping() is true is_learn_pfn_mapping() must be + * true too, verify it here. + */ + VM_BUG_ON(is_linear_pfn_mapping(vma) || vma->vm_flags & VM_NO_THP); hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; hend = vma->vm_end & HPAGE_PMD_MASK; if (hstart < hend) @@ -1772,12 +1771,9 @@ static void collapse_huge_page(struct mm_struct *mm, VM_BUG_ON(address & ~HPAGE_PMD_MASK); #ifndef CONFIG_NUMA + up_read(&mm->mmap_sem); VM_BUG_ON(!*hpage); new_page = *hpage; - if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) { - up_read(&mm->mmap_sem); - return; - } #else VM_BUG_ON(*hpage); /* @@ -1792,22 +1788,26 @@ static void collapse_huge_page(struct mm_struct *mm, */ new_page = alloc_hugepage_vma(khugepaged_defrag(), vma, address, node, __GFP_OTHER_NODE); + + /* + * After allocating the hugepage, release the mmap_sem read lock in + * preparation for taking it in write mode. + */ + up_read(&mm->mmap_sem); if (unlikely(!new_page)) { - up_read(&mm->mmap_sem); count_vm_event(THP_COLLAPSE_ALLOC_FAILED); *hpage = ERR_PTR(-ENOMEM); return; } +#endif + count_vm_event(THP_COLLAPSE_ALLOC); if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) { - up_read(&mm->mmap_sem); +#ifdef CONFIG_NUMA put_page(new_page); +#endif return; } -#endif - - /* after allocating the hugepage upgrade to mmap_sem write mode */ - up_read(&mm->mmap_sem); /* * Prevent all access to pagetables with the exception of @@ -1828,12 +1828,15 @@ static void collapse_huge_page(struct mm_struct *mm, (vma->vm_flags & VM_NOHUGEPAGE)) goto out; - /* VM_PFNMAP vmas may have vm_ops null but vm_file set */ - if (!vma->anon_vma || vma->vm_ops || vma->vm_file) + if (!vma->anon_vma || vma->vm_ops) goto out; if (is_vma_temporary_stack(vma)) goto out; - VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + /* + * If is_pfn_mapping() is true is_learn_pfn_mapping() must be + * true too, verify it here. + */ + VM_BUG_ON(is_linear_pfn_mapping(vma) || vma->vm_flags & VM_NO_THP); pgd = pgd_offset(mm, address); if (!pgd_present(*pgd)) @@ -2066,13 +2069,16 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, progress++; continue; } - /* VM_PFNMAP vmas may have vm_ops null but vm_file set */ - if (!vma->anon_vma || vma->vm_ops || vma->vm_file) + if (!vma->anon_vma || vma->vm_ops) goto skip; if (is_vma_temporary_stack(vma)) goto skip; - - VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + /* + * If is_pfn_mapping() is true is_learn_pfn_mapping() + * must be true too, verify it here. + */ + VM_BUG_ON(is_linear_pfn_mapping(vma) || + vma->vm_flags & VM_NO_THP); hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; hend = vma->vm_end & HPAGE_PMD_MASK; @@ -2228,11 +2234,8 @@ static void khugepaged_loop(void) while (likely(khugepaged_enabled())) { #ifndef CONFIG_NUMA hpage = khugepaged_alloc_hugepage(); - if (unlikely(!hpage)) { - count_vm_event(THP_COLLAPSE_ALLOC_FAILED); + if (unlikely(!hpage)) break; - } - count_vm_event(THP_COLLAPSE_ALLOC); #else if (IS_ERR(hpage)) { khugepaged_alloc_sleep(); |