Merge Linus' latest into master
[safe/jmp/linux-2.6] / mm / nommu.c
index f3bfd01..ed75bc9 100644 (file)
@@ -10,6 +10,7 @@
  *  Copyright (c) 2000-2003 David McCullough <davidm@snapgear.com>
  *  Copyright (c) 2000-2001 D Jeff Dionne <jeff@uClinux.org>
  *  Copyright (c) 2002      Greg Ungerer <gerg@snapgear.com>
+ *  Copyright (c) 2007      Paul Mundt <lethal@linux-sh.org>
  */
 
 #include <linux/module.h>
@@ -21,7 +22,7 @@
 #include <linux/pagemap.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
-#include <linux/ptrace.h>
+#include <linux/tracehook.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/mount.h>
@@ -38,7 +39,7 @@ struct page *mem_map;
 unsigned long max_mapnr;
 unsigned long num_physpages;
 unsigned long askedalloc, realalloc;
-atomic_t vm_committed_space = ATOMIC_INIT(0);
+atomic_long_t vm_committed_space = ATOMIC_LONG_INIT(0);
 int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
 int sysctl_overcommit_ratio = 50; /* default is 50% */
 int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
@@ -104,16 +105,27 @@ unsigned int kobjsize(const void *objp)
 {
        struct page *page;
 
-       if (!objp || !((page = virt_to_page(objp))))
+       /*
+        * If the object we have should not have ksize performed on it,
+        * return size of 0
+        */
+       if (!objp || !virt_addr_valid(objp))
                return 0;
 
+       page = virt_to_head_page(objp);
+
+       /*
+        * If the allocator sets PageSlab, we know the pointer came from
+        * kmalloc().
+        */
        if (PageSlab(page))
                return ksize(objp);
 
-       BUG_ON(page->index < 0);
-       BUG_ON(page->index >= MAX_ORDER);
-
-       return (PAGE_SIZE << page->index);
+       /*
+        * The ksize() function is only guaranteed to work for pointers
+        * returned by kmalloc(). So handle arbitrary pointers here.
+        */
+       return PAGE_SIZE << compound_order(page);
 }
 
 /*
@@ -183,6 +195,26 @@ void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 }
 EXPORT_SYMBOL(__vmalloc);
 
+void *vmalloc_user(unsigned long size)
+{
+       void *ret;
+
+       ret = __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
+                       PAGE_KERNEL);
+       if (ret) {
+               struct vm_area_struct *vma;
+
+               down_write(&current->mm->mmap_sem);
+               vma = find_vma(current->mm, (unsigned long)ret);
+               if (vma)
+                       vma->vm_flags |= VM_USERMAP;
+               up_write(&current->mm->mmap_sem);
+       }
+
+       return ret;
+}
+EXPORT_SYMBOL(vmalloc_user);
+
 struct page *vmalloc_to_page(const void *addr)
 {
        return virt_to_page(addr);
@@ -234,6 +266,27 @@ void *vmalloc_node(unsigned long size, int node)
 }
 EXPORT_SYMBOL(vmalloc_node);
 
+#ifndef PAGE_KERNEL_EXEC
+# define PAGE_KERNEL_EXEC PAGE_KERNEL
+#endif
+
+/**
+ *     vmalloc_exec  -  allocate virtually contiguous, executable memory
+ *     @size:          allocation size
+ *
+ *     Kernel-internal function to allocate enough pages to cover @size
+ *     the page level allocator and map them into contiguous and
+ *     executable kernel virtual space.
+ *
+ *     For tight control over page level allocator and protection flags
+ *     use __vmalloc() instead.
+ */
+
+void *vmalloc_exec(unsigned long size)
+{
+       return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC);
+}
+
 /**
  * vmalloc_32  -  allocate virtually contiguous memory (32bit addressable)
  *     @size:          allocation size
@@ -253,10 +306,17 @@ EXPORT_SYMBOL(vmalloc_32);
  *
  * The resulting memory area is 32bit addressable and zeroed so it can be
  * mapped to userspace without leaking data.
+ *
+ * VM_USERMAP is set on the corresponding VMA so that subsequent calls to
+ * remap_vmalloc_range() are permissible.
  */
 void *vmalloc_32_user(unsigned long size)
 {
-       return __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+       /*
+        * We'll have to sort out the ZONE_DMA bits for 64-bit,
+        * but for now this can simply use vmalloc_user() directly.
+        */
+       return vmalloc_user(size);
 }
 EXPORT_SYMBOL(vmalloc_32_user);
 
@@ -706,7 +766,7 @@ static unsigned long determine_vm_flags(struct file *file,
         * it's being traced - otherwise breakpoints set in it may interfere
         * with another untraced process
         */
-       if ((flags & MAP_PRIVATE) && (current->ptrace & PT_PTRACED))
+       if ((flags & MAP_PRIVATE) && tracehook_expect_breakpoints(current))
                vm_flags &= ~VM_MAYSHARE;
 
        return vm_flags;
@@ -934,8 +994,13 @@ unsigned long do_mmap_pgoff(struct file *file,
 
        INIT_LIST_HEAD(&vma->anon_vma_node);
        atomic_set(&vma->vm_usage, 1);
-       if (file)
+       if (file) {
                get_file(file);
+               if (vm_flags & VM_EXECUTABLE) {
+                       added_exe_file_vma(current->mm);
+                       vma->vm_mm = current->mm;
+               }
+       }
        vma->vm_file    = file;
        vma->vm_flags   = vm_flags;
        vma->vm_start   = addr;
@@ -990,8 +1055,11 @@ unsigned long do_mmap_pgoff(struct file *file,
        up_write(&nommu_vma_sem);
        kfree(vml);
        if (vma) {
-               if (vma->vm_file)
+               if (vma->vm_file) {
                        fput(vma->vm_file);
+                       if (vma->vm_flags & VM_EXECUTABLE)
+                               removed_exe_file_vma(vma->vm_mm);
+               }
                kfree(vma);
        }
        return ret;
@@ -1021,7 +1089,7 @@ EXPORT_SYMBOL(do_mmap_pgoff);
 /*
  * handle mapping disposal for uClinux
  */
-static void put_vma(struct vm_area_struct *vma)
+static void put_vma(struct mm_struct *mm, struct vm_area_struct *vma)
 {
        if (vma) {
                down_write(&nommu_vma_sem);
@@ -1043,8 +1111,11 @@ static void put_vma(struct vm_area_struct *vma)
                        realalloc -= kobjsize(vma);
                        askedalloc -= sizeof(*vma);
 
-                       if (vma->vm_file)
+                       if (vma->vm_file) {
                                fput(vma->vm_file);
+                               if (vma->vm_flags & VM_EXECUTABLE)
+                                       removed_exe_file_vma(mm);
+                       }
                        kfree(vma);
                }
 
@@ -1081,7 +1152,7 @@ int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len)
  found:
        vml = *parent;
 
-       put_vma(vml->vma);
+       put_vma(mm, vml->vma);
 
        *parent = vml->next;
        realalloc -= kobjsize(vml);
@@ -1126,7 +1197,7 @@ void exit_mmap(struct mm_struct * mm)
 
                while ((tmp = mm->context.vmlist)) {
                        mm->context.vmlist = tmp->next;
-                       put_vma(tmp->vma);
+                       put_vma(mm, tmp->vma);
 
                        realalloc -= kobjsize(tmp);
                        askedalloc -= sizeof(*tmp);
@@ -1216,6 +1287,21 @@ int remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
 }
 EXPORT_SYMBOL(remap_pfn_range);
 
+int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
+                       unsigned long pgoff)
+{
+       unsigned int size = vma->vm_end - vma->vm_start;
+
+       if (!(vma->vm_flags & VM_USERMAP))
+               return -EINVAL;
+
+       vma->vm_start = (unsigned long)(addr + (pgoff << PAGE_SHIFT));
+       vma->vm_end = vma->vm_start + size;
+
+       return 0;
+}
+EXPORT_SYMBOL(remap_vmalloc_range);
+
 void swap_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
 {
 }
@@ -1352,7 +1438,7 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
         * cast `allowed' as a signed long because vm_committed_space
         * sometimes has a negative value
         */
-       if (atomic_read(&vm_committed_space) < (long)allowed)
+       if (atomic_long_read(&vm_committed_space) < (long)allowed)
                return 0;
 error:
        vm_unacct_memory(pages);