fork: lock VMAs of the parent process when forking

When forking a child process, the parent write-protects anonymous pages
and COW-shares them with the child being forked using copy_present_pte().

We must not take any concurrent page faults on the source vma's as they
are being processed, as we expect both the vma and the pte's behind it
to be stable.  For example, the anon_vma_fork() expects the parents
vma->anon_vma to not change during the vma copy.

A concurrent page fault on a page newly marked read-only by the page
copy might trigger wp_page_copy() and a anon_vma_prepare(vma) on the
source vma, defeating the anon_vma_clone() that wasn't done because the
parent vma originally didn't have an anon_vma, but we now might end up
copying a pte entry for a page that has one.

Before the per-vma lock based changes, the mmap_lock guaranteed
exclusion with concurrent page faults.  But now we need to do a
vma_start_write() to make sure no concurrent faults happen on this vma
while it is being processed.

This fix can potentially regress some fork-heavy workloads.  Kernel
build time did not show noticeable regression on a 56-core machine while
a stress test mapping 10000 VMAs and forking 5000 times in a tight loop
shows ~5% regression.  If such fork time regression is unacceptable,
disabling CONFIG_PER_VMA_LOCK should restore its performance.  Further
optimizations are possible if this regression proves to be problematic.

Suggested-by: David Hildenbrand <>
Reported-by: Jiri Slaby <>
Reported-by: Holger Hoffstätte <>
Reported-by: Jacob Young <>
Fixes: 0bff0aaea03e ("x86/mm: try VMA lock-based page fault handling first")
Signed-off-by: Suren Baghdasaryan <>
Signed-off-by: Linus Torvalds <>
diff --git a/kernel/fork.c b/kernel/fork.c
index b85814e..d2e12b6 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -686,6 +686,7 @@
 	for_each_vma(old_vmi, mpnt) {
 		struct file *file;
+		vma_start_write(mpnt);
 		if (mpnt->vm_flags & VM_DONTCOPY) {
 			vm_stat_account(mm, mpnt->vm_flags, -vma_pages(mpnt));