调用 mmap 系统调用 , 先检查 " 偏移 " 是否是 " 内存页大小 " 的 " 整数倍 " , 如果偏移是内存页大小的整数倍 , 则调用 sys_mmap_pgoff 函数 , 继续向下执行 ;
在 sys_mmap_pgoff 系统调用函数 中 , 最后调用了 vm_mmap_pgoff 函数 , 继续向下执行 ;
在 vm_mmap_pgoff 函数 中 , 核心处理过程就是调用 do_mmap 函数 , 这是 " 内存映射 " 创建的主要函数逻辑 ;
在 do_mmap 函数中 , 调用了 mmap_region 函数 , 创建 " 虚拟内存区域 " ;
一、mmap_region 函数执行流程
1、检查内存申请是否合法
在 mmap_region 函数中 ,
首先 , 调用 may_expand_vm 函数 , 针对 进程 的 " 用户虚拟地址空间 " 地址进行检查 , 查看是否超出了地址区间限制 ;
/* Check against address space limit. */
if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) {
unsigned long nr_pages;
/*
* MAP_FIXED may remove pages of mappings that intersects with
* requested mapping. Account for the pages it would unmap.
*/
nr_pages = count_vma_pages_range(mm, addr, addr + len);
if (!may_expand_vm(mm, vm_flags,
(len >> PAGE_SHIFT) - nr_pages))
return -ENOMEM;
}
源码路径 : linux-4.12\mm\mmap.c#1612
然后 , 检查 申请的 内存空间 , 是否与旧的 虚拟内存空间 有 重叠 , 如果有重叠部分 , 则调用 do_munmap 函数 , 删除内存空间 ;
/* Clear old maps */
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link,
&rb_parent)) {
if (do_munmap(mm, addr, len, uf))
return -ENOMEM;
}
源码路径 : linux-4.12\mm\mmap.c#1627
再后 , 检查是否有 私有的写映射 , 即检查内存是否可用 ;
/*
* Private writable mapping: check memory availability
*/
if (accountable_mapping(file, vm_flags)) {
charged = len >> PAGE_SHIFT;
if (security_vm_enough_memory_mm(mm, charged))
return -ENOMEM;
vm_flags |= VM_ACCOUNT;
}
源码路径 : linux-4.12\mm\mmap.c#1637
最后 , 检查是否可以与已有的 " 虚拟内存区域 " 进行合并 , 避免内存空间浪费 ;
/*
* Can we just expand an old mapping?
*/
vma = vma_merge(mm, prev, addr, addr + len, vm_flags,
NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX);
if (vma)
goto out;
源码路径 : linux-4.12\mm\mmap.c#1647
检查完毕后 , 开始 创建 " 虚拟内存区域 " ;
2、创建 " 虚拟内存区域 "
首先 , 检查是否可以与已有的 " 虚拟内存区域 " 进行合并 , 避免内存空间浪费 ;
- 如果可以合并 , 则调用
vma_merge空间进行内存合并 ; - 如果不可以合并 , 则创建新的虚拟内存区域 , 即跳转到
out位置进行执行 ;- 如果是 文件映射 , 调用
call_mmap函数 , 创建内存区域 ; - 如果是 匿名映射 , 调用
shmem_zero_setup函数 , 创建内存区域 ;
- 如果是 文件映射 , 调用
/*
* Can we just expand an old mapping?
*/
vma = vma_merge(mm, prev, addr, addr + len, vm_flags,
NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX);
if (vma)
goto out;
源码路径 : linux-4.12\mm\mmap.c#1647
在创建 " 虚拟内存区域 " 时 ,
- 如果命中
file分支 , 则是 创建 " 文件映射 " 内存区域 , 核心是调用 调用call_mmap函数 ; - 如果命中
else分支 , 就是创建 " 匿名映射 " 内存区域 , 核心是 调用shmem_zero_setup函数 ;
if (file) {
if (vm_flags & VM_DENYWRITE) {
error = deny_write_access(file);
if (error)
goto free_vma;
}
if (vm_flags & VM_SHARED) {
error = mapping_map_writable(file->f_mapping);
if (error)
goto allow_write_and_free_vma;
}
/* ->mmap() can change vma->vm_file, but must guarantee that
* vma_link() below can deny write-access if VM_DENYWRITE is set
* and map writably if VM_SHARED is set. This usually means the
* new file must not have been exposed to user-space, yet.
*/
vma->vm_file = get_file(file);
error = call_mmap(file, vma);
if (error)
goto unmap_and_free_vma;
/* Can addr have changed??
*
* Answer: Yes, several device drivers can do it in their
* f_op->mmap method. -DaveM
* Bug: If addr is changed, prev, rb_link, rb_parent should
* be updated for vma_link()
*/
WARN_ON_ONCE(addr != vma->vm_start);
addr = vma->vm_start;
vm_flags = vma->vm_flags;
} else if (vm_flags & VM_SHARED) {
error = shmem_zero_setup(vma);
if (error)
goto free_vma;
}
源码路径 : linux-4.12\mm\mmap.c#1671
在后 , 调用 vma_link 函数 和 vma_set_page_prot 函数 , 完成 " 虚拟内存区域 " 创建 ;
vma_link(mm, vma, prev, rb_link, rb_parent);
源码路径 : linux-4.12\mm\mmap.c#1710
vma_set_page_prot(vma);
源码路径 : linux-4.12\mm\mmap.c#1743
二、mmap_region 函数源码
mmap_region 函数源码 定义在 Linux 内核源码的 linux-4.12\mm\mmap.c#1602 位置 ;
mmap_region 函数源码如下 :
unsigned long mmap_region(struct file *file, unsigned long addr,
unsigned long len, vm_flags_t vm_flags, unsigned long pgoff,
struct list_head *uf)
{
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma, *prev;
int error;
struct rb_node **rb_link, *rb_parent;
unsigned long charged = 0;
/* Check against address space limit. */
if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) {
unsigned long nr_pages;
/*
* MAP_FIXED may remove pages of mappings that intersects with
* requested mapping. Account for the pages it would unmap.
*/
nr_pages = count_vma_pages_range(mm, addr, addr + len);
if (!may_expand_vm(mm, vm_flags,
(len >> PAGE_SHIFT) - nr_pages))
return -ENOMEM;
}
/* Clear old maps */
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link,
&rb_parent)) {
if (do_munmap(mm, addr, len, uf))
return -ENOMEM;
}
/*
* Private writable mapping: check memory availability
*/
if (accountable_mapping(file, vm_flags)) {
charged = len >> PAGE_SHIFT;
if (security_vm_enough_memory_mm(mm, charged))
return -ENOMEM;
vm_flags |= VM_ACCOUNT;
}
/*
* Can we just expand an old mapping?
*/
vma = vma_merge(mm, prev, addr, addr + len, vm_flags,
NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX);
if (vma)
goto out;
/*
* Determine the object being mapped and call the appropriate
* specific mapper. the address has already been validated, but
* not unmapped, but the maps are removed from the list.
*/
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
if (!vma) {
error = -ENOMEM;
goto unacct_error;
}
vma->vm_mm = mm;
vma->vm_start = addr;
vma->vm_end = addr + len;
vma->vm_flags = vm_flags;
vma->vm_page_prot = vm_get_page_prot(vm_flags);
vma->vm_pgoff = pgoff;
INIT_LIST_HEAD(&vma->anon_vma_chain);
if (file) {
if (vm_flags & VM_DENYWRITE) {
error = deny_write_access(file);
if (error)
goto free_vma;
}
if (vm_flags & VM_SHARED) {
error = mapping_map_writable(file->f_mapping);
if (error)
goto allow_write_and_free_vma;
}
/* ->mmap() can change vma->vm_file, but must guarantee that
* vma_link() below can deny write-access if VM_DENYWRITE is set
* and map writably if VM_SHARED is set. This usually means the
* new file must not have been exposed to user-space, yet.
*/
vma->vm_file = get_file(file);
error = call_mmap(file, vma);
if (error)
goto unmap_and_free_vma;
/* Can addr have changed??
*
* Answer: Yes, several device drivers can do it in their
* f_op->mmap method. -DaveM
* Bug: If addr is changed, prev, rb_link, rb_parent should
* be updated for vma_link()
*/
WARN_ON_ONCE(addr != vma->vm_start);
addr = vma->vm_start;
vm_flags = vma->vm_flags;
} else if (vm_flags & VM_SHARED) {
error = shmem_zero_setup(vma);
if (error)
goto free_vma;
}
vma_link(mm, vma, prev, rb_link, rb_parent);
/* Once vma denies write, undo our temporary denial count */
if (file) {
if (vm_flags & VM_SHARED)
mapping_unmap_writable(file->f_mapping);
if (vm_flags & VM_DENYWRITE)
allow_write_access(file);
}
file = vma->vm_file;
out:
perf_event_mmap(vma);
vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT);
if (vm_flags & VM_LOCKED) {
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) ||
vma == get_gate_vma(current->mm)))
mm->locked_vm += (len >> PAGE_SHIFT);
else
vma->vm_flags &= VM_LOCKED_CLEAR_MASK;
}
if (file)
uprobe_mmap(vma);
/*
* New (or expanded) vma always get soft dirty status.
* Otherwise user-space soft-dirty page tracker won't
* be able to distinguish situation when vma area unmapped,
* then new mapped in-place (which must be aimed as
* a completely new data area).
*/
vma->vm_flags |= VM_SOFTDIRTY;
vma_set_page_prot(vma);
return addr;
unmap_and_free_vma:
vma->vm_file = NULL;
fput(file);
/* Undo any partial mapping done by a device driver. */
unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end);
charged = 0;
if (vm_flags & VM_SHARED)
mapping_unmap_writable(file->f_mapping);
allow_write_and_free_vma:
if (vm_flags & VM_DENYWRITE)
allow_write_access(file);
free_vma:
kmem_cache_free(vm_area_cachep, vma);
unacct_error:
if (charged)
vm_unacct_memory(charged);
return error;
}
源码路径 : linux-4.12\mm\mmap.c#1602
