git.openwrt.org Git - openwrt/staging/blogic.git/blob

   1 From 23dc11dfbd8e78a2450120e9afc83c1d32a8e683 Mon Sep 17 00:00:00 2001
   2 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <rafal@milecki.pl>
   3 Date: Mon, 12 Sep 2016 19:02:34 +0200
   4 Subject: [PATCH] Revert "nl80211: Remove duplicated check in
   5  nl80211_setup_ap()"
   6 MIME-Version: 1.0
   7 Content-Type: text/plain; charset=UTF-8
   8 Content-Transfer-Encoding: 8bit
   9
  10 This reverts commit 647862eb60c324015ea31293cc052558b5185ca4.
  11
  12 The second check of device_ap_sme looks like duplicated, but it isn't
  13 actually. The trick is nl80211_create_monitor_interface may change that
  14 variable value and the second evaluation may give a different result.
  15
  16 This definitely isn't a very clear code, but that change caused a
  17 regression for drivers that:
  18 1) Don't report NL80211_ATTR_DEVICE_AP_SME
  19 2) Don't support monitor mode
  20 3) Don't support subscribing for PROBE_REQ and/or ACTION frames
  21 like brcmfmac. With such drivers hostapd doesn't start anymore.
  22
  23 Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
  24 ---
  25  src/drivers/driver_nl80211.c | 3 ++-
  26  1 file changed, 2 insertions(+), 1 deletion(-)
  27
  28 --- a/src/drivers/driver_nl80211.c
  29 +++ b/src/drivers/driver_nl80211.c
  30 @@ -4434,7 +4434,8 @@ static int nl80211_setup_ap(struct i802_
  31                                    "nl80211: Failed to subscribe for mgmt frames from SME driver - trying to run without it");
  32
  33         if (!drv->device_ap_sme && drv->use_monitor &&
  34 -           nl80211_create_monitor_interface(drv))
  35 +           nl80211_create_monitor_interface(drv) &&
  36 +           !drv->device_ap_sme)
  37                 return -1;
  38
  39         if (drv->device_ap_sme &&