Re: [RFC PATCH v2 15/27] mm/mprotect: Prevent mprotect from changing shadow stack

On Wed, 2018-07-11 at 11:12 +0200, Peter Zijlstra wrote:

On Tue, Jul 10, 2018 at 04:10:08PM -0700, Dave Hansen wrote:

quoted

On 07/10/2018 03:26 PM, Yu-cheng Yu wrote:

quoted

Signed-off-by: Yu-cheng Yu <redacted>

This still needs a changelog, even if you think it's simple.

quoted

--- a/mm/mprotect.c
+++ b/mm/mprotect.c

@@ -446,6 +446,15 @@ static int do_mprotect_pkey(unsigned long

start, size_t len,
 	error = -ENOMEM;
 	if (!vma)
 		goto out;
+
+	/*
+	 * Do not allow changing shadow stack memory.
+	 */
+	if (vma->vm_flags & VM_SHSTK) {
+		error = -EINVAL;
+		goto out;
+	}
+

I think this is a _bit_ draconian.  Why shouldn't we be able to use
protection keys with a shadow stack?  Or, set it to PROT_NONE?

Right, and then there's also madvise() and some of the other
accessors.

Why do we need to disallow this? AFAICT the worst that can happen is
that a process wrecks itself, so what?

Agree.  I will remove the patch.