mm: bring in additional flag for fixup_user_fault to signal unlock

[linux-beck.git] / mm / gup.c

diff --git a/mm/gup.c b/mm/gup.c
index aa21c4b865a50d24e758ccc0f9d10d01e2114c95..b64a36175884e07604b0e216bc2d545a2892dcb7 100644 (file)
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -618,6 +618,8 @@ EXPORT_SYMBOL(__get_user_pages);
  * @mm:                mm_struct of target mm
  * @address:   user address
  * @fault_flags:flags to pass down to handle_mm_fault()
+ * @unlocked:  did we unlock the mmap_sem while retrying, maybe NULL if caller
+ *             does not allow retry
  *
  * This is meant to be called in the specific scenario where for locking reasons
  * we try to access user memory in atomic context (within a pagefault_disable()
@@ -629,22 +631,28 @@ EXPORT_SYMBOL(__get_user_pages);
  * The main difference with get_user_pages() is that this function will
  * unconditionally call handle_mm_fault() which will in turn perform all the
  * necessary SW fixup of the dirty and young bits in the PTE, while
- * handle_mm_fault() only guarantees to update these in the struct page.
+ * get_user_pages() only guarantees to update these in the struct page.
  *
  * This is important for some architectures where those bits also gate the
  * access permission to the page because they are maintained in software.  On
  * such architectures, gup() will not be enough to make a subsequent access
  * succeed.
  *
- * This has the same semantics wrt the @mm->mmap_sem as does filemap_fault().
+ * This function will not return with an unlocked mmap_sem. So it has not the
+ * same semantics wrt the @mm->mmap_sem as does filemap_fault().
  */
 int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
-                    unsigned long address, unsigned int fault_flags)
+                    unsigned long address, unsigned int fault_flags,
+                    bool *unlocked)
 {
        struct vm_area_struct *vma;
        vm_flags_t vm_flags;
-       int ret;
+       int ret, major = 0;
+
+       if (unlocked)
+               fault_flags |= FAULT_FLAG_ALLOW_RETRY;
 
+retry:
        vma = find_extend_vma(mm, address);
        if (!vma || address < vma->vm_start)
                return -EFAULT;
@@ -654,6 +662,7 @@ int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
                return -EFAULT;
 
        ret = handle_mm_fault(mm, vma, address, fault_flags);
+       major |= ret & VM_FAULT_MAJOR;
        if (ret & VM_FAULT_ERROR) {
                if (ret & VM_FAULT_OOM)
                        return -ENOMEM;
@@ -663,8 +672,19 @@ int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
                        return -EFAULT;
                BUG();
        }
+
+       if (ret & VM_FAULT_RETRY) {
+               down_read(&mm->mmap_sem);
+               if (!(fault_flags & FAULT_FLAG_TRIED)) {
+                       *unlocked = true;
+                       fault_flags &= ~FAULT_FLAG_ALLOW_RETRY;
+                       fault_flags |= FAULT_FLAG_TRIED;
+                       goto retry;
+               }
+       }
+
        if (tsk) {
-               if (ret & VM_FAULT_MAJOR)
+               if (major)
                        tsk->maj_flt++;
                else
                        tsk->min_flt++;