From: "Chen, Kenneth W" Following up with the work on shared page table done by Dave McCracken. This set of patch target shared page table for hugetlb memory only. The shared page table is particular useful in the situation of large number of independent processes sharing large shared memory segments. In the normal page case, the amount of memory saved from process' page table is quite significant. For hugetlb, the saving on page table memory is not the primary objective (as hugetlb itself already cuts down page table overhead significantly), instead, the purpose of using shared page table on hugetlb is to allow faster TLB refill and smaller cache pollution upon TLB miss. With PT sharing, pte entries are shared among hundreds of processes, the cache consumption used by all the page table is smaller and in return, application gets much higher cache hit ratio. One other effect is that cache hit ratio with hardware page walker hitting on pte in cache will be higher and this helps to reduce tlb miss latency. These two effects contribute to higher application performance. Signed-off-by: Ken Chen Cc: Hugh Dickins Signed-off-by: Andrew Morton --- arch/i386/mm/hugetlbpage.c | 112 +++++++++++++++++++++++++++++++- arch/ia64/mm/hugetlbpage.c | 5 + arch/powerpc/mm/hugetlbpage.c | 5 + arch/sh/mm/hugetlbpage.c | 5 + arch/sh64/mm/hugetlbpage.c | 5 + arch/sparc64/mm/hugetlbpage.c | 5 + fs/hugetlbfs/inode.c | 2 include/linux/hugetlb.h | 2 mm/hugetlb.c | 21 +++++- 9 files changed, 158 insertions(+), 4 deletions(-) diff -puN arch/i386/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/i386/mm/hugetlbpage.c --- a/arch/i386/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/i386/mm/hugetlbpage.c @@ -17,6 +17,113 @@ #include #include +static unsigned long page_table_shareable(struct vm_area_struct *svma, + struct vm_area_struct *vma, + unsigned long addr, pgoff_t idx) +{ + unsigned long saddr = ((idx - svma->vm_pgoff) << PAGE_SHIFT) + + svma->vm_start; + unsigned long sbase = saddr & PUD_MASK; + unsigned long s_end = sbase + PUD_SIZE; + + /* + * match the virtual addresses, permission and the alignment of the + * page table page. + */ + if (pmd_index(addr) != pmd_index(saddr) || + vma->vm_flags != svma->vm_flags || + sbase < svma->vm_start || svma->vm_end < s_end) + return 0; + + return saddr; +} + +static int vma_shareable(struct vm_area_struct *vma, unsigned long addr) +{ + unsigned long base = addr & PUD_MASK; + unsigned long end = base + PUD_SIZE; + + /* + * check on proper vm_flags and page table alignment + */ + if (vma->vm_flags & VM_MAYSHARE && + vma->vm_start <= base && end <= vma->vm_end) + return 1; + return 0; +} + +/* + * search for a shareable pmd page for hugetlb. + */ +static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) +{ + struct vm_area_struct *vma = find_vma(mm, addr); + struct address_space *mapping = vma->vm_file->f_mapping; + pgoff_t idx = ((addr - vma->vm_start) >> PAGE_SHIFT) + + vma->vm_pgoff; + struct prio_tree_iter iter; + struct vm_area_struct *svma; + unsigned long saddr; + pte_t *spte = NULL; + + if (!vma_shareable(vma, addr)) + return; + + spin_lock(&mapping->i_mmap_lock); + vma_prio_tree_foreach(svma, &iter, &mapping->i_mmap, idx, idx) { + if (svma == vma) + continue; + + saddr = page_table_shareable(svma, vma, addr, idx); + if (saddr) { + spte = huge_pte_offset(svma->vm_mm, saddr); + if (spte) { + get_page(virt_to_page(spte)); + break; + } + } + } + + if (!spte) + goto out; + + spin_lock(&mm->page_table_lock); + if (pud_none(*pud)) + pud_populate(mm, pud, (unsigned long) spte & PAGE_MASK); + else + put_page(virt_to_page(spte)); + spin_unlock(&mm->page_table_lock); +out: + spin_unlock(&mapping->i_mmap_lock); +} + +/* + * unmap huge page backed by shared pte. + * + * Hugetlb pte page is ref counted at the time of mapping. If pte is shared + * indicated by page_count > 1, unmap is achieved by clearing pud and + * decrementing the ref count. If count == 1, the pte page is not shared. + * + * called with vma->vm_mm->page_table_lock held. + * + * returns: 1 successfully unmapped a shared pte page + * 0 the underlying pte page is not shared, or it is the last user + */ +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + pgd_t *pgd = pgd_offset(mm, *addr); + pud_t *pud = pud_offset(pgd, *addr); + + BUG_ON(page_count(virt_to_page(ptep)) == 0); + if (page_count(virt_to_page(ptep)) == 1) + return 0; + + pud_clear(pud); + put_page(virt_to_page(ptep)); + *addr = ALIGN(*addr, HPAGE_SIZE * PTRS_PER_PTE) - HPAGE_SIZE; + return 1; +} + pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; @@ -25,8 +132,11 @@ pte_t *huge_pte_alloc(struct mm_struct * pgd = pgd_offset(mm, addr); pud = pud_alloc(mm, pgd, addr); - if (pud) + if (pud) { + if (pud_none(*pud)) + huge_pmd_share(mm, addr, pud); pte = (pte_t *) pmd_alloc(mm, pud, addr); + } BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte)); return pte; diff -puN arch/ia64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/ia64/mm/hugetlbpage.c --- a/arch/ia64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/ia64/mm/hugetlbpage.c @@ -64,6 +64,11 @@ huge_pte_offset (struct mm_struct *mm, u return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* diff -puN arch/powerpc/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/powerpc/mm/hugetlbpage.c --- a/arch/powerpc/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/powerpc/mm/hugetlbpage.c @@ -146,6 +146,11 @@ pte_t *huge_pte_alloc(struct mm_struct * return hugepte_offset(hpdp, addr); } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp) { pte_t *hugepte = hugepd_page(*hpdp); diff -puN arch/sh/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/sh/mm/hugetlbpage.c --- a/arch/sh/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/sh/mm/hugetlbpage.c @@ -63,6 +63,11 @@ pte_t *huge_pte_offset(struct mm_struct return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff -puN arch/sh64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/sh64/mm/hugetlbpage.c --- a/arch/sh64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/sh64/mm/hugetlbpage.c @@ -53,6 +53,11 @@ pte_t *huge_pte_offset(struct mm_struct return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff -puN arch/sparc64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 arch/sparc64/mm/hugetlbpage.c --- a/arch/sparc64/mm/hugetlbpage.c~shared-page-table-for-hugetlb-page-v4 +++ a/arch/sparc64/mm/hugetlbpage.c @@ -235,6 +235,11 @@ pte_t *huge_pte_offset(struct mm_struct return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff -puN fs/hugetlbfs/inode.c~shared-page-table-for-hugetlb-page-v4 fs/hugetlbfs/inode.c --- a/fs/hugetlbfs/inode.c~shared-page-table-for-hugetlb-page-v4 +++ a/fs/hugetlbfs/inode.c @@ -293,7 +293,7 @@ hugetlb_vmtruncate_list(struct prio_tree if (h_vm_pgoff >= h_pgoff) v_offset = 0; - unmap_hugepage_range(vma, + __unmap_hugepage_range(vma, vma->vm_start + v_offset, vma->vm_end); } } diff -puN include/linux/hugetlb.h~shared-page-table-for-hugetlb-page-v4 include/linux/hugetlb.h --- a/include/linux/hugetlb.h~shared-page-table-for-hugetlb-page-v4 +++ a/include/linux/hugetlb.h @@ -17,6 +17,7 @@ int hugetlb_sysctl_handler(struct ctl_ta int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *); int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, struct page **, struct vm_area_struct **, unsigned long *, int *, int); void unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); +void __unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); int hugetlb_prefault(struct address_space *, struct vm_area_struct *); int hugetlb_report_meminfo(char *); int hugetlb_report_node_meminfo(int, char *); @@ -34,6 +35,7 @@ extern int sysctl_hugetlb_shm_group; pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr); pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr); +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep); struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write); struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, diff -puN mm/hugetlb.c~shared-page-table-for-hugetlb-page-v4 mm/hugetlb.c --- a/mm/hugetlb.c~shared-page-table-for-hugetlb-page-v4 +++ a/mm/hugetlb.c @@ -356,8 +356,8 @@ nomem: return -ENOMEM; } -void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end) +void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) { struct mm_struct *mm = vma->vm_mm; unsigned long address; @@ -381,6 +381,9 @@ void unmap_hugepage_range(struct vm_area if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; + pte = huge_ptep_get_and_clear(mm, address, ptep); if (pte_none(pte)) continue; @@ -398,6 +401,16 @@ void unmap_hugepage_range(struct vm_area } } +void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + if (vma->vm_file) { + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); + __unmap_hugepage_range(vma, start, end); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); + } +} + static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, pte_t pte) { @@ -632,11 +645,14 @@ void hugetlb_change_protection(struct vm BUG_ON(address >= end); flush_cache_range(vma, address, end); + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); spin_lock(&mm->page_table_lock); for (; address < end; address += HPAGE_SIZE) { ptep = huge_pte_offset(mm, address); if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; if (!pte_none(*ptep)) { pte = huge_ptep_get_and_clear(mm, address, ptep); pte = pte_mkhuge(pte_modify(pte, newprot)); @@ -645,6 +661,7 @@ void hugetlb_change_protection(struct vm } } spin_unlock(&mm->page_table_lock); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); flush_tlb_range(vma, start, end); } _