Re: [PATCH] fbdev: udlfb: validate vendor descriptor items

On 7/18/26 20:26, Helge Deller wrote:

On 7/6/26 11:30, Pengpeng Hou wrote:

quoted

dlfb_parse_vendor_descriptor() walks key-length-value items inside the
DisplayLink vendor descriptor.

Require each item to contain its key, length and declared value bytes
before reading item-specific fields such as max_area.

Signed-off-by: Pengpeng Hou <redacted>
---
  drivers/video/fbdev/udlfb.c | 22 ++++++++++++++++------
  1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/video/fbdev/udlfb.c b/drivers/video/fbdev/udlfb.c
index fdbb8671a810..e78d6f95c9c5 100644
--- a/drivers/video/fbdev/udlfb.c
+++ b/drivers/video/fbdev/udlfb.c

@@ -1586,19 +1586,29 @@ static int dlfb_parse_vendor_descriptor(struct dlfb_data *dlfb,

          desc += 5; /* the fixed header we've already parsed */
          while (desc < desc_end) {
+            char *value;
              u8 length;
              u16 key;
-            key = *desc++;
-            key |= (u16)*desc++ << 8;
+            if (desc_end - desc < sizeof(key) + sizeof(length))
+                goto unrecognized;
+
+            key = get_unaligned_le16(desc);

Is there a reason why you switch to unconditional little-endian reads?
Is this "vendor descriptor" always little-endian?
If yes, then your patch is probably correct.
If not, I think your patch will most likely break big-endian machines.

Please ignore my comments above.
I should have looked more closely.
Your patch is of course correct!

The patch is now added to fbdev git tree.

Thanks!
Helge