Re: [PATCH] fix infoleak in wireless
From: Greg Kroah-Hartman <hidden>
Date: 2016-05-03 21:34:06
Also in:
linux-wireless, lkml
On Tue, May 03, 2016 at 05:11:07PM -0400, Kangjie Lu wrote:
quoted hunk
Opps, I did not notice the patch is not attached. From 34a82a734388d07eb10f91770f86938e38f7575a Mon Sep 17 00:00:00 2001 From: Kangjie Lu <redacted> Date: Tue, 3 May 2016 14:15:18 -0400 Subject: [PATCH] fix infoleak in wireless MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The 6-bytes array “mac_addr” is not initialized in the dump_station implementations of “drivers/staging/wilc1000/wilc_wfi_cfgoperations.c” and “drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c”, so all 6 bytes may be leaked. Signed-off-by: Kangjie Lu <redacted> --- net/wireless/nl80211.c | 1 + 1 file changed, 1 insertion(+)diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 056a730..2e92d14 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c@@ -3905,6 +3905,7 @@ static int nl80211_dump_station(struct sk_buff *skb,while (1) { memset(&sinfo, 0, sizeof(sinfo)); + eth_zero_addr(mac_addr); err = rdev_dump_station(rdev, wdev->netdev, sta_idx, mac_addr, &sinfo); if (err == -ENOENT)
Patch is corrupted :( Why not fix up the staging drivers, they are the real problem here, which is what I think the networking maintainers were telling you to do. thanks, greg k-h