[PATCH] Staging: rtl8188eu: rtw_mlme: Fix uninitialized variable authmode

Larry Finger Larry.Finger at lwfinger.net
Fri Jul 24 17:56:22 UTC 2020


On 7/24/20 8:28 AM, Dinghao Liu wrote:
> The variable authmode will keep uninitialized if neither if
> statements used to initialize this variable are not triggered.

Besides Greg's comment, you need to re-parse this sentence. I realize that 
English is probably not your first language, but this one is not what you meant.

You likely meant "The variable authmode will remain uninitialized if all 
statements used to initialize this variable are not triggered."

A possible (line-wrapped) patch to quiet the tools would be:

diff --git a/drivers/staging/rtl8188eu/core/rtw_mlme.c 
b/drivers/staging/rtl8188eu/core/rtw_mlme.c
index 9de2d421f6b1..9e4d78bc9a2e 100644
--- a/drivers/staging/rtl8188eu/core/rtw_mlme.c
+++ b/drivers/staging/rtl8188eu/core/rtw_mlme.c
@@ -1729,9 +1729,11 @@ int rtw_restruct_sec_ie(struct adapter *adapter, u8 
*in_ie, u8 *out_ie, uint in_
         if ((ndisauthmode == Ndis802_11AuthModeWPA) ||
             (ndisauthmode == Ndis802_11AuthModeWPAPSK))
                 authmode = _WPA_IE_ID_;
-       if ((ndisauthmode == Ndis802_11AuthModeWPA2) ||
-           (ndisauthmode == Ndis802_11AuthModeWPA2PSK))
+       else if ((ndisauthmode == Ndis802_11AuthModeWPA2) ||
+                (ndisauthmode == Ndis802_11AuthModeWPA2PSK))
                 authmode = _WPA2_IE_ID_;
+       else
+               authmode = 0;

         if (check_fwstate(pmlmepriv, WIFI_UNDER_WPS)) {
                 memcpy(out_ie + ielength, psecuritypriv->wps_ie, 
psecuritypriv->wps_ie_len);


Yes, in this routine, it would be possible for authmode to not be set; however, 
later code only compares it to either _WPA_IE_ID_ or _WPA2_IE_ID_. It is never 
used in a way that an unset value could make the program flow be different by 
arbitrarily setting the value to zero. Thus your statement "Then authmode may 
contain a garbage value and influence the execution flow of this function." is 
false.

Larry


More information about the devel mailing list