Skip to contentSkip to conversation

auxdisplay/cfag12864bfb.c: Convert to use vm_map_pages_zero()

PATCH In review Souptick Joarder 2 participants 2 messages May 26, 2019 – Jun 20, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 auxdisplay/cfag12864bfb.c: Convert to use vm_map_pages_zero()

While using mmap, the incorrect values of length and vm_pgoff are ignored and this driver goes ahead with mapping cfag12864b_buffer to user vma.

Convert vm_insert_pages() to use vm_map_pages_zero(). We could later "fix" these drivers to behave according to the normal vm_pgoff offsetting simply by removing the _zero suffix on the function name and if that causes regressions, it gives us an easy way to revert.

---
v2: Fixed minor typo.

 drivers/auxdisplay/cfag12864bfb.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/auxdisplay/cfag12864bfb.c +3 -2
index 40c8a55..4074886 100644
@@ -52,8 +52,9 @@
 
 static int cfag12864bfb_mmap(struct fb_info *info, struct vm_area_struct *vma)
 {
-	return vm_insert_page(vma, vma->vm_start,
-		virt_to_page(cfag12864b_buffer));
+	struct page *pages = virt_to_page(cfag12864b_buffer);
+
+	return vm_map_pages_zero(vma, &pages, 1);
 }
 
 static struct fb_ops cfag12864bfb_ops = {
Signature · 1 line
1.9.1
#1 lore raw
Quoted 5 lines · Souptick Joarder

While using mmap, the incorrect values of length and vm_pgoff are ignored and this driver goes ahead with mapping cfag12864b_buffer to user vma.

Quoted 8 lines

Convert vm_insert_pages() to use vm_map_pages_zero(). We could later "fix" these drivers to behave according to the normal vm_pgoff offsetting simply by removing the _zero suffix on the function name and if that causes regressions, it gives us an easy way to revert.

---
v2: Fixed minor typo.

Almost forgot about these two -- picking both v2s for a PR later. Not sure what happened with the mm/ related one I sent, is it in Andrew's tree?

Cheers,
Miguel