[PATCH v3] kernel-device: expect non-NULL 'vendor' and 'product' argument in get_device_ids()
Ben Chan
benchan at chromium.org
Tue Aug 15 16:10:44 UTC 2017
get_device_ids() in mm-kernel-device-udev.c accepts a NULL 'vendor' or
'product' argument, but the current implementation could result in a
potential NULL dereferences of the 'vendor' argument. Given that
get_device_ids() is a local helper and its only caller provides a
non-NULL 'vendor' and 'product' argument, this patch removes the NULL
checks (i.e. get_device_ids() expects non-NULL 'vendor' and 'product').
This patch also rearranges the code such that the 'vendor' argument is
updated only when the function returns TRUE, just like how the 'product'
argument is handled.
---
src/kerneldevice/mm-kernel-device-udev.c | 22 +++++++---------------
1 file changed, 7 insertions(+), 15 deletions(-)
diff --git a/src/kerneldevice/mm-kernel-device-udev.c b/src/kerneldevice/mm-kernel-device-udev.c
index e8763f7e..b9f057bd 100644
--- a/src/kerneldevice/mm-kernel-device-udev.c
+++ b/src/kerneldevice/mm-kernel-device-udev.c
@@ -119,29 +119,21 @@ get_device_ids (GUdevDevice *device,
if (strlen (vid) != 4)
goto out;
- if (vendor) {
- *vendor = (guint16) (mm_utils_hex2byte (vid + 2) & 0xFF);
- *vendor |= (guint16) ((mm_utils_hex2byte (vid) & 0xFF) << 8);
- }
-
if (!pid)
pid = g_udev_device_get_property (device, "ID_MODEL_ID");
- if (!pid) {
- *vendor = 0;
+ if (!pid)
goto out;
- }
if (strncmp (pid, "0x", 2) == 0)
pid += 2;
- if (strlen (pid) != 4) {
- *vendor = 0;
+ if (strlen (pid) != 4)
goto out;
- }
- if (product) {
- *product = (guint16) (mm_utils_hex2byte (pid + 2) & 0xFF);
- *product |= (guint16) ((mm_utils_hex2byte (pid) & 0xFF) << 8);
- }
+ *vendor = (guint16) (mm_utils_hex2byte (vid + 2) & 0xFF);
+ *vendor |= (guint16) ((mm_utils_hex2byte (vid) & 0xFF) << 8);
+
+ *product = (guint16) (mm_utils_hex2byte (pid + 2) & 0xFF);
+ *product |= (guint16) ((mm_utils_hex2byte (pid) & 0xFF) << 8);
success = TRUE;
--
2.14.0.434.g98096fd7a8-goog
More information about the ModemManager-devel
mailing list