mm: madvise avoid exclusive mmap_sem

Avoid down_write of the mmap_sem in madvise when we can help it.

Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 29 insertions, 4 deletions

diff --git a/mm/madvise.c b/mm/madvise.c
index 603c525..e75096b 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -12,6 +12,24 @@
 #include <linux/hugetlb.h>
 
 /*
+ * Any behaviour which results in changes to the vma->vm_flags needs to
+ * take mmap_sem for writing. Others, which simply traverse vmas, need
+ * to only take it for reading.
+ */
+static int madvise_need_mmap_write(int behavior)
+{
+	switch (behavior) {
+	case MADV_REMOVE:
+	case MADV_WILLNEED:
+	case MADV_DONTNEED:
+		return 0;
+	default:
+		/* be safe, default to 1. list exceptions explicitly */
+		return 1;
+	}
+}
+
+/*
  * We can potentially split a vm area into separate
  * areas, each area with its own behavior.
  */
@@ -183,9 +201,9 @@ static long madvise_remove(struct vm_area_struct *vma,
 			+ ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
 
 	/* vmtruncate_range needs to take i_mutex and i_alloc_sem */
-	up_write(&current->mm->mmap_sem);
+	up_read(&current->mm->mmap_sem);
 	error = vmtruncate_range(mapping->host, offset, endoff);
-	down_write(&current->mm->mmap_sem);
+	down_read(&current->mm->mmap_sem);
 	return error;
 }
 
@@ -270,7 +288,10 @@ asmlinkage long sys_madvise(unsigned long start, size_t len_in, int behavior)
 	int error = -EINVAL;
 	size_t len;
 
-	down_write(&current->mm->mmap_sem);
+	if (madvise_need_mmap_write(behavior))
+		down_write(&current->mm->mmap_sem);
+	else
+		down_read(&current->mm->mmap_sem);
 
 	if (start & ~PAGE_MASK)
 		goto out;
@@ -332,6 +353,10 @@ asmlinkage long sys_madvise(unsigned long start, size_t len_in, int behavior)
 			vma = find_vma(current->mm, start);
 	}
 out:
-	up_write(&current->mm->mmap_sem);
+	if (madvise_need_mmap_write(behavior))
+		up_write(&current->mm->mmap_sem);
+	else
+		up_read(&current->mm->mmap_sem);
+
 	return error;
 }