From: Wey-Yi Guy <wey-yi.w.guy@intel.com>
To: linville@tuxdriver.com
Cc: linux-wireless@vger.kernel.org,
ipw3945-devel@lists.sourceforge.net,
Shanyu Zhao <shanyu.zhao@intel.com>,
Wey-Yi Guy <wey-yi.w.guy@intel.com>
Subject: [PATCH 06/16] iwlwifi: avoid race condition in channel change
Date: Wed, 18 Aug 2010 09:22:41 -0700 [thread overview]
Message-ID: <1282148571-9258-7-git-send-email-wey-yi.w.guy@intel.com> (raw)
In-Reply-To: <1282148571-9258-1-git-send-email-wey-yi.w.guy@intel.com>
From: Shanyu Zhao <shanyu.zhao@intel.com>
When iwl_mac_config() is called by mac80211, the channel pointer
hw->conf->channel can potentially change, resulting in mismatch
band and channel number when configuring RXON command. To avoid
this situation, save the channel pointer in local variables
and validate the channel before using it. Note that priv->mutex
is locked during the whole function so the local variables are safe.
Same change is applied to iwl_mac_channel_switch() since basically
it copies code from iwl_mac_config().
Also removed an outdated comment in the flow.
Signed-off-by: Shanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
---
drivers/net/wireless/iwlwifi/iwl-agn.c | 14 ++++++--------
drivers/net/wireless/iwlwifi/iwl-core.c | 14 ++++++--------
2 files changed, 12 insertions(+), 16 deletions(-)
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c
index 00cbb83..8011105 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn.c
@@ -3632,6 +3632,7 @@ static void iwl_mac_channel_switch(struct ieee80211_hw *hw,
struct iwl_priv *priv = hw->priv;
const struct iwl_channel_info *ch_info;
struct ieee80211_conf *conf = &hw->conf;
+ struct ieee80211_channel *channel = ch_switch->channel;
struct iwl_ht_config *ht_conf = &priv->current_ht_config;
u16 ch;
unsigned long flags = 0;
@@ -3655,10 +3656,10 @@ static void iwl_mac_channel_switch(struct ieee80211_hw *hw,
mutex_lock(&priv->mutex);
if (priv->cfg->ops->lib->set_channel_switch) {
- ch = ch_switch->channel->hw_value;
+ ch = channel->hw_value;
if (le16_to_cpu(priv->active_rxon.channel) != ch) {
ch_info = iwl_get_channel_info(priv,
- conf->channel->band,
+ channel->band,
ch);
if (!is_channel_valid(ch_info)) {
IWL_DEBUG_MAC80211(priv, "invalid channel\n");
@@ -3687,15 +3688,12 @@ static void iwl_mac_channel_switch(struct ieee80211_hw *hw,
} else
ht_conf->is_40mhz = false;
- /* if we are switching from ht to 2.4 clear flags
- * from any ht related info since 2.4 does not
- * support ht */
- if ((le16_to_cpu(priv->staging_rxon.channel) != ch))
+ if (le16_to_cpu(priv->staging_rxon.channel) != ch)
priv->staging_rxon.flags = 0;
- iwl_set_rxon_channel(priv, conf->channel);
+ iwl_set_rxon_channel(priv, channel);
iwl_set_rxon_ht(priv, ht_conf);
- iwl_set_flags_for_band(priv, conf->channel->band,
+ iwl_set_flags_for_band(priv, channel->band,
priv->vif);
spin_unlock_irqrestore(&priv->lock, flags);
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 221a211..fc78690 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -1947,6 +1947,7 @@ int iwl_mac_config(struct ieee80211_hw *hw, u32 changed)
struct iwl_priv *priv = hw->priv;
const struct iwl_channel_info *ch_info;
struct ieee80211_conf *conf = &hw->conf;
+ struct ieee80211_channel *channel = conf->channel;
struct iwl_ht_config *ht_conf = &priv->current_ht_config;
unsigned long flags = 0;
int ret = 0;
@@ -1956,7 +1957,7 @@ int iwl_mac_config(struct ieee80211_hw *hw, u32 changed)
mutex_lock(&priv->mutex);
IWL_DEBUG_MAC80211(priv, "enter to channel %d changed 0x%X\n",
- conf->channel->hw_value, changed);
+ channel->hw_value, changed);
if (unlikely(!priv->cfg->mod_params->disable_hw_scan &&
test_bit(STATUS_SCANNING, &priv->status))) {
@@ -1987,8 +1988,8 @@ int iwl_mac_config(struct ieee80211_hw *hw, u32 changed)
if (scan_active)
goto set_ch_out;
- ch = conf->channel->hw_value;
- ch_info = iwl_get_channel_info(priv, conf->channel->band, ch);
+ ch = channel->hw_value;
+ ch_info = iwl_get_channel_info(priv, channel->band, ch);
if (!is_channel_valid(ch_info)) {
IWL_DEBUG_MAC80211(priv, "leave - invalid channel\n");
ret = -EINVAL;
@@ -2019,16 +2020,13 @@ int iwl_mac_config(struct ieee80211_hw *hw, u32 changed)
* from BSS config in iwl_ht_conf */
ht_conf->ht_protection = IEEE80211_HT_OP_MODE_PROTECTION_NONE;
- /* if we are switching from ht to 2.4 clear flags
- * from any ht related info since 2.4 does not
- * support ht */
if ((le16_to_cpu(priv->staging_rxon.channel) != ch))
priv->staging_rxon.flags = 0;
- iwl_set_rxon_channel(priv, conf->channel);
+ iwl_set_rxon_channel(priv, channel);
iwl_set_rxon_ht(priv, ht_conf);
- iwl_set_flags_for_band(priv, conf->channel->band, priv->vif);
+ iwl_set_flags_for_band(priv, channel->band, priv->vif);
spin_unlock_irqrestore(&priv->lock, flags);
if (priv->cfg->ops->lib->update_bcast_station)
--
1.7.0.4
next prev parent reply other threads:[~2010-08-18 16:18 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-18 16:22 [PATCH 00/16] iwlwifi update for 2.6.37 Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 01/16] iwlwifi: remove unused define Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 02/16] iwlwifi: additional comments in iwl_cfg Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 03/16] iwlwifi: max/min aggregation time limit Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 04/16] iwlwifi: make aggregation time limit configurable Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 05/16] iwlwifi: do not call ieee80211_frequency_to_channel Wey-Yi Guy
2010-08-18 16:22 ` Wey-Yi Guy [this message]
2010-08-18 16:22 ` [PATCH 07/16] iwlwifi: cleanup iwl_set_rxon_channel() Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 08/16] iwlwifi: refactor iwl_setup_rxon_timing Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 09/16] iwlagn: adjust starting action for rate scale Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 10/16] iwlagn: continue perform rate scale when error detected Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 11/16] iwlwifi: additional parameter in REPLY_RX_PHY_CMD Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 12/16] iwlwifi: more generic name for rssi calc in iwlagn Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 13/16] iwlwifi: configurable ampdu factor and density Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 14/16] iwlwifi: remove unused iwl_send_card_state function Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 15/16] iwlwifi: enable experimental ucode support Wey-Yi Guy
2010-08-18 16:22 ` [PATCH 16/16] iwlwifi: move debug options into submenu Wey-Yi Guy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1282148571-9258-7-git-send-email-wey-yi.w.guy@intel.com \
--to=wey-yi.w.guy@intel.com \
--cc=ipw3945-devel@lists.sourceforge.net \
--cc=linux-wireless@vger.kernel.org \
--cc=linville@tuxdriver.com \
--cc=shanyu.zhao@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox