HID: pantherlord: validate output report details
authorKees Cook <keescook@chromium.org>
Wed, 28 Aug 2013 20:30:49 +0000 (22:30 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 27 Sep 2013 00:21:45 +0000 (17:21 -0700)
commit 412f30105ec6735224535791eed5cdc02888ecb4 upstream.

A HID device could send a malicious output report that would cause the
pantherlord HID driver to write beyond the output report allocation
during initialization, causing a heap overflow:

[  310.939483] usb 1-1: New USB device found, idVendor=0e8f, idProduct=0003
...
[  315.980774] BUG kmalloc-192 (Tainted: G        W   ): Redzone overwritten

CVE-2013-2892

Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/hid/hid-pl.c

index d29112f..2dcd7d9 100644 (file)
@@ -132,8 +132,14 @@ static int plff_init(struct hid_device *hid)
                        strong = &report->field[0]->value[2];
                        weak = &report->field[0]->value[3];
                        debug("detected single-field device");
-               } else if (report->maxfield >= 4 && report->field[0]->maxusage == 1 &&
-                               report->field[0]->usage[0].hid == (HID_UP_LED | 0x43)) {
+               } else if (report->field[0]->maxusage == 1 &&
+                          report->field[0]->usage[0].hid ==
+                               (HID_UP_LED | 0x43) &&
+                          report->maxfield >= 4 &&
+                          report->field[0]->report_count >= 1 &&
+                          report->field[1]->report_count >= 1 &&
+                          report->field[2]->report_count >= 1 &&
+                          report->field[3]->report_count >= 1) {
                        report->field[0]->value[0] = 0x00;
                        report->field[1]->value[0] = 0x00;
                        strong = &report->field[2]->value[0];