Message ID | 20170929075234.dgyuiwbowu3jq6r4@mwanda (mailing list archive) |
---|---|
State | Accepted |
Commit | 64e79426c2041b931c053a11743277a4f7f279c4 |
Delegated to: | Kalle Valo |
Headers | show |
Dan Carpenter <dan.carpenter@oracle.com> wrote: > My static checker complains that we have an upper bound but no lower > bound. I suspect neither are really required but it doesn't hurt to add > a check for negatives. > > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> > > diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c > index c53cbf3d52bd..294a6b43d1bc 100644 > --- a/drivers/net/wireless/realtek/rtlwifi/core.c > +++ b/drivers/net/wireless/realtek/rtlwifi/core.c > @@ -453,7 +453,8 @@ static void _rtl_add_wowlan_patterns(struct ieee80211_hw *hw, > for (i = 0; i < wow->n_patterns; i++) { > memset(&rtl_pattern, 0, sizeof(struct rtl_wow_pattern)); > memset(mask, 0, MAX_WOL_BIT_MASK_SIZE); > - if (patterns[i].pattern_len > MAX_WOL_PATTERN_SIZE) { > + if (patterns[i].pattern_len < 0 || > + patterns[i].pattern_len > MAX_WOL_PATTERN_SIZE) { > RT_TRACE(rtlpriv, COMP_POWER, DBG_WARNING, > "Pattern[%d] is too long\n", i); > continue; Patch applied to wireless-drivers-next.git, thanks. 64e79426c204 rtlwifi: silence underflow warning
diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c index c53cbf3d52bd..294a6b43d1bc 100644 --- a/drivers/net/wireless/realtek/rtlwifi/core.c +++ b/drivers/net/wireless/realtek/rtlwifi/core.c @@ -453,7 +453,8 @@ static void _rtl_add_wowlan_patterns(struct ieee80211_hw *hw, for (i = 0; i < wow->n_patterns; i++) { memset(&rtl_pattern, 0, sizeof(struct rtl_wow_pattern)); memset(mask, 0, MAX_WOL_BIT_MASK_SIZE); - if (patterns[i].pattern_len > MAX_WOL_PATTERN_SIZE) { + if (patterns[i].pattern_len < 0 || + patterns[i].pattern_len > MAX_WOL_PATTERN_SIZE) { RT_TRACE(rtlpriv, COMP_POWER, DBG_WARNING, "Pattern[%d] is too long\n", i); continue;
My static checker complains that we have an upper bound but no lower bound. I suspect neither are really required but it doesn't hurt to add a check for negatives. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>