The patch titled hugetlb: prepare_hugepage_range check offset too has been added to the -mm tree. Its filename is hugetlb-prepare_hugepage_range-check-offset-too.patch See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find out what to do about this ------------------------------------------------------ Subject: hugetlb: prepare_hugepage_range check offset too From: Hugh Dickins <hugh@xxxxxxxxxxx> (David:) If hugetlbfs_file_mmap() returns a failure to do_mmap_pgoff() - for example, because the given file offset is not hugepage aligned - then do_mmap_pgoff will go to the unmap_and_free_vma backout path. But at this stage the vma hasn't been marked as hugepage, and the backout path will call unmap_region() on it. That will eventually call down to the non-hugepage version of unmap_page_range(). On ppc64, at least, that will cause serious problems if there are any existing hugepage pagetable entries in the vicinity - for example if there are any other hugepage mappings under the same PUD. unmap_page_range() will trigger a bad_pud() on the hugepage pud entries. I suspect this will also cause bad problems on ia64, though I don't have a machine to test it on. (Hugh:) prepare_hugepage_range() should check file offset alignment when it checks virtual address and length, to stop MAP_FIXED with a bad huge offset from unmapping before it fails further down. PowerPC should apply the same prepare_hugepage_range alignment checks as ia64 and all the others do. Then none of the alignment checks in hugetlbfs_file_mmap are required (nor is the check for too small a mapping); but even so, move up setting of VM_HUGETLB and add a comment to warn of what David Gibson discovered - if hugetlbfs_file_mmap fails before setting it, do_mmap_pgoff's unmap_region when unwinding from error will go the non-huge way, which may cause bad behaviour on architectures (powerpc and ia64) which segregate their huge mappings into a separate region of the address space. Signed-off-by: Hugh Dickins <hugh@xxxxxxxxxxx> Cc: "Luck, Tony" <tony.luck@xxxxxxxxx> Cc: "David S. Miller" <davem@xxxxxxxxxxxxx> Acked-by: Adam Litke <agl@xxxxxxxxxx> Acked-by: David Gibson <david@xxxxxxxxxxxxxxxxxxxxx> Cc: Paul Mackerras <paulus@xxxxxxxxx> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxx> --- arch/ia64/mm/hugetlbpage.c | 4 +++- arch/powerpc/mm/hugetlbpage.c | 8 ++++++-- fs/hugetlbfs/inode.c | 21 ++++++++------------- include/linux/hugetlb.h | 10 +++++++--- mm/mmap.c | 2 +- 5 files changed, 25 insertions(+), 20 deletions(-) diff -puN arch/ia64/mm/hugetlbpage.c~hugetlb-prepare_hugepage_range-check-offset-too arch/ia64/mm/hugetlbpage.c --- a/arch/ia64/mm/hugetlbpage.c~hugetlb-prepare_hugepage_range-check-offset-too +++ a/arch/ia64/mm/hugetlbpage.c @@ -70,8 +70,10 @@ huge_pte_offset (struct mm_struct *mm, u * Don't actually need to do any preparation, but need to make sure * the address is in the right region. */ -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) diff -puN arch/powerpc/mm/hugetlbpage.c~hugetlb-prepare_hugepage_range-check-offset-too arch/powerpc/mm/hugetlbpage.c --- a/arch/powerpc/mm/hugetlbpage.c~hugetlb-prepare_hugepage_range-check-offset-too +++ a/arch/powerpc/mm/hugetlbpage.c @@ -491,11 +491,15 @@ static int open_high_hpage_areas(struct return 0; } -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { int err = 0; - if ( (addr+len) < addr ) + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) return -EINVAL; if (addr < 0x100000000UL) diff -puN fs/hugetlbfs/inode.c~hugetlb-prepare_hugepage_range-check-offset-too fs/hugetlbfs/inode.c --- a/fs/hugetlbfs/inode.c~hugetlb-prepare_hugepage_range-check-offset-too +++ a/fs/hugetlbfs/inode.c @@ -62,24 +62,19 @@ static int hugetlbfs_file_mmap(struct fi loff_t len, vma_len; int ret; - if (vma->vm_pgoff & (HPAGE_SIZE / PAGE_SIZE - 1)) - return -EINVAL; - - if (vma->vm_start & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end - vma->vm_start < HPAGE_SIZE) - return -EINVAL; + /* + * vma alignment has already been checked by prepare_hugepage_range. + * If you add any error returns here, do so after setting VM_HUGETLB, + * so is_vm_hugetlb_page tests below unmap_region go the right way + * when do_mmap_pgoff unwinds (may be important on powerpc and ia64). + */ + vma->vm_flags |= VM_HUGETLB | VM_RESERVED; + vma->vm_ops = &hugetlb_vm_ops; vma_len = (loff_t)(vma->vm_end - vma->vm_start); mutex_lock(&inode->i_mutex); file_accessed(file); - vma->vm_flags |= VM_HUGETLB | VM_RESERVED; - vma->vm_ops = &hugetlb_vm_ops; ret = -ENOMEM; len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); diff -puN include/linux/hugetlb.h~hugetlb-prepare_hugepage_range-check-offset-too include/linux/hugetlb.h --- a/include/linux/hugetlb.h~hugetlb-prepare_hugepage_range-check-offset-too +++ a/include/linux/hugetlb.h @@ -60,8 +60,11 @@ void hugetlb_free_pgd_range(struct mmu_g * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) @@ -69,7 +72,8 @@ static inline int prepare_hugepage_range return 0; } #else -int prepare_hugepage_range(unsigned long addr, unsigned long len); +int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff); #endif #ifndef ARCH_HAS_SETCLEAR_HUGE_PTE @@ -107,7 +111,7 @@ static inline unsigned long hugetlb_tota #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL -#define prepare_hugepage_range(addr, len) (-EINVAL) +#define prepare_hugepage_range(addr,len,pgoff) (-EINVAL) #define pmd_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; }) diff -puN mm/mmap.c~hugetlb-prepare_hugepage_range-check-offset-too mm/mmap.c --- a/mm/mmap.c~hugetlb-prepare_hugepage_range-check-offset-too +++ a/mm/mmap.c @@ -1379,7 +1379,7 @@ get_unmapped_area(struct file *file, uns * Check if the given range is hugepage aligned, and * can be made suitable for hugepages. */ - ret = prepare_hugepage_range(addr, len); + ret = prepare_hugepage_range(addr, len, pgoff); } else { /* * Ensure that a normal request is not falling in a _ Patches currently in -mm which might be from hugh@xxxxxxxxxxx are hugetlb-prepare_hugepage_range-check-offset-too.patch __unmap_hugepage_range-add-comment.patch shared-page-table-for-hugetlb-page-v4.patch htlb-forget-rss-with-pt-sharing.patch mlock-cleanup.patch honour-mnt_noexec-for-access.patch generic-bug-implementation.patch generic-bug-for-i386.patch generic-bug-for-x86-64.patch bug-test-1.patch tty-switch-to-ktermios-powerpc-fix.patch - To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html