Re: [PATCH 3/3] ath6kl: Remove few deadcode in main.c

On Tue, Jan 03, 2012 at 09:12:50AM +0100, Johannes Berg wrote:

On Tue, 2012-01-03 at 11:10 +0530, Vasanthakumar Thiagarajan wrote:

quoted

quoted

quoted

@@ -62,8 +62,7 @@ static void ath6kl_add_new_sta(struct ath6kl *ar, u8 *mac, u16 aid, u8 *wpaie,
 
 	sta = &ar->sta_list[free_slot];
 	memcpy(sta->mac, mac, ETH_ALEN);
-	if (ielen <= ATH6KL_MAX_IE)
-		memcpy(sta->wpa_ie, wpaie, ielen);
+	memcpy(sta->wpa_ie, wpaie, ielen);

quoted

And then someone changes the u8 ielen to something else and accidentally
adds a bug. I would prefer to have an explicit check for the ielen to
make it obvious what's the maximum length.

I don't really like using u8 for ielen either. IMHO size_t or similar
would be better.

The length of any IE is 1 byte, u8 is the appropriate one.

No -- this can contain multiple IEs.

Right, it can, but as of now this is any one these ies, wpa, wapi, wpa/wpa2.
Ok, i don't see any disadvantage in converting this u8 to size_t,
i'll send the next version.

Vasanth