pagemap: hide physical addresses from non-privileged users

This patch makes pagemap readable for normal users and hides physical
addresses from them.  For some use-cases PFN isn't required at all.

See http://lkml.kernel.org/r/1425935472-17949-1-git-send-email-kirill@shutemov.name

Fixes: ab676b7d6fbf ("pagemap: do not leak physical addresses to non-privileged userspace")
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reviewed-by: Mark Williamson <mwilliamson@undo-software.com>
Tested-by:  Mark Williamson <mwilliamson@undo-software.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 14 insertions, 11 deletions

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 98ba9ea96b19..bc651644b1b2 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -939,6 +939,7 @@ typedef struct {
 struct pagemapread {
         int pos, len;           /* units: PM_ENTRY_BYTES, not bytes */
         pagemap_entry_t *buffer;
+        bool show_pfn;
 };
 
 #define PAGEMAP_WALK_SIZE       (PMD_SIZE)
@@ -1015,7 +1016,8 @@ static pagemap_entry_t pte_to_pagemap_entry(struct pagemapread *pm,
         struct page *page = NULL;
 
         if (pte_present(pte)) {
-                frame = pte_pfn(pte);
+                if (pm->show_pfn)
+                        frame = pte_pfn(pte);
                 flags |= PM_PRESENT;
                 page = vm_normal_page(vma, addr, pte);
                 if (pte_soft_dirty(pte))
@@ -1065,8 +1067,9 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end,
                  */
                 if (pmd_present(pmd)) {
                         flags |= PM_PRESENT;
-                        frame = pmd_pfn(pmd) +
-                                ((addr & ~PMD_MASK) >> PAGE_SHIFT);
+                        if (pm->show_pfn)
+                                frame = pmd_pfn(pmd) +
+                                        ((addr & ~PMD_MASK) >> PAGE_SHIFT);
                 }
 
                 for (; addr != end; addr += PAGE_SIZE) {
@@ -1075,7 +1078,7 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end,
                         err = add_to_pagemap(addr, &pme, pm);
                         if (err)
                                 break;
-                        if (flags & PM_PRESENT)
+                        if (pm->show_pfn && (flags & PM_PRESENT))
                                 frame++;
                 }
                 spin_unlock(ptl);
@@ -1129,8 +1132,9 @@ static int pagemap_hugetlb_range(pte_t *ptep, unsigned long hmask,
                         flags |= PM_FILE;
 
                 flags |= PM_PRESENT;
-                frame = pte_pfn(pte) +
-                        ((addr & ~hmask) >> PAGE_SHIFT);
+                if (pm->show_pfn)
+                        frame = pte_pfn(pte) +
+                                ((addr & ~hmask) >> PAGE_SHIFT);
         }
 
         for (; addr != end; addr += PAGE_SIZE) {
@@ -1139,7 +1143,7 @@ static int pagemap_hugetlb_range(pte_t *ptep, unsigned long hmask,
                 err = add_to_pagemap(addr, &pme, pm);
                 if (err)
                         return err;
-                if (flags & PM_PRESENT)
+                if (pm->show_pfn && (flags & PM_PRESENT))
                         frame++;
         }
 
@@ -1198,6 +1202,9 @@ static ssize_t pagemap_read(struct file *file, char __user *buf,
         if (!count)
                 goto out_mm;
 
+        /* do not disclose physical addresses: attack vector */
+        pm.show_pfn = file_ns_capable(file, &init_user_ns, CAP_SYS_ADMIN);
+
         pm.len = (PAGEMAP_WALK_SIZE >> PAGE_SHIFT);
         pm.buffer = kmalloc(pm.len * PM_ENTRY_BYTES, GFP_TEMPORARY);
         ret = -ENOMEM;
@@ -1267,10 +1274,6 @@ static int pagemap_open(struct inode *inode, struct file *file)
 {
         struct mm_struct *mm;
 
-        /* do not disclose physical addresses: attack vector */
-        if (!capable(CAP_SYS_ADMIN))
-                return -EPERM;
-
         mm = proc_mem_open(inode, PTRACE_MODE_READ);
         if (IS_ERR(mm))
                 return PTR_ERR(mm);