On Fri, 2024-10-18 at 14:54 +0300, Kirill A. Shutemov wrote:

On Fri, Oct 18, 2024 at 01:22:35PM +0200, Roberto Sassu wrote:

quoted

On Fri, 2024-10-18 at 14:05 +0300, Kirill A. Shutemov wrote:

quoted

On Fri, Oct 18, 2024 at 12:00:22PM +0100, Lorenzo Stoakes wrote:

quoted

+ Liam, Jann

On Fri, Oct 18, 2024 at 01:49:06PM +0300, Kirill A. Shutemov wrote:

quoted

On Fri, Oct 18, 2024 at 11:24:06AM +0200, Roberto Sassu wrote:

quoted

Probably it is hard, @Kirill would there be any way to safely move
security_mmap_file() out of the mmap_lock lock?

What about something like this (untested):

diff --git a/mm/mmap.c b/mm/mmap.c
index dd4b35a25aeb..03473e77d356 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c

@@ -1646,6 +1646,26 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
 	if (pgoff + (size >> PAGE_SHIFT) < pgoff)
 		return ret;

+	if (mmap_read_lock_killable(mm))
+		return -EINTR;
+
+	vma = vma_lookup(mm, start);
+
+	if (!vma || !(vma->vm_flags & VM_SHARED)) {
+		mmap_read_unlock(mm);
+		return -EINVAL;
+	}
+
+	file = get_file(vma->vm_file);
+
+	mmap_read_unlock(mm);
+
+	ret = security_mmap_file(vma->vm_file, prot, flags);

Accessing VMA fields without any kind of lock is... very much not advised.

I'm guessing you meant to say:

	ret = security_mmap_file(file, prot, flags);

Here? :)

Sure. My bad.

Patch with all fixups:

Thanks a lot! Let's wait a bit until the others have a chance to
comment. Meanwhile, I will test it.

Do you want me to do the final patch, or will you be proposing it?

You can post it if it works:

Signed-off-by: Kirill A. Shutemov <redacted>

Thanks! Ok.

Roberto