Enable enhanced bmiss detection if the firmware supports it. This
feature is only enabled on some firmwares since it comes with a power
cost.
Also add a few missing command ids to keep the enums straight.
Signed-off-by: Thomas Pedersen <[email protected]>
---
v2:
comments (Kalle)
indentation (Kalle)
drivers/net/wireless/ath/ath6kl/cfg80211.c | 29 ++++++++++++++++++++++++++++
drivers/net/wireless/ath/ath6kl/cfg80211.h | 2 +
drivers/net/wireless/ath/ath6kl/core.h | 3 ++
drivers/net/wireless/ath/ath6kl/init.c | 3 ++
drivers/net/wireless/ath/ath6kl/wmi.c | 19 ++++++++++++++++++
drivers/net/wireless/ath/ath6kl/wmi.h | 11 ++++++++++
6 files changed, 67 insertions(+), 0 deletions(-)
diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c
index b869a35..af0c48a 100644
--- a/drivers/net/wireless/ath/ath6kl/cfg80211.c
+++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c
@@ -575,6 +575,8 @@ static int ath6kl_cfg80211_connect(struct wiphy *wiphy, struct net_device *dev,
}
vif->nw_type = vif->next_mode;
+ /* enable enhanced bmiss detection if applicable */
+ ath6kl_cfg80211_sta_bmiss_enhance(vif, true);
if (vif->wdev.iftype == NL80211_IFTYPE_P2P_CLIENT)
nw_subtype = SUBTYPE_P2PCLIENT;
@@ -1512,6 +1514,9 @@ static int ath6kl_cfg80211_change_iface(struct wiphy *wiphy,
}
}
+ /* need to clean up enhanced bmiss detection fw state */
+ ath6kl_cfg80211_sta_bmiss_enhance(vif, false);
+
set_iface_type:
switch (type) {
case NL80211_IFTYPE_STATION:
@@ -2614,6 +2619,30 @@ static int ath6kl_set_channel(struct wiphy *wiphy, struct net_device *dev,
return 0;
}
+void ath6kl_cfg80211_sta_bmiss_enhance(struct ath6kl_vif *vif, bool enable)
+{
+ int err;
+
+ if (WARN_ON(!test_bit(WMI_READY, &vif->ar->flag)))
+ return;
+
+ if (vif->nw_type != INFRA_NETWORK)
+ return;
+
+ if (!test_bit(ATH6KL_FW_CAPABILITY_BMISS_ENHANCE,
+ vif->ar->fw_capabilities))
+ return;
+
+ ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "%s fw bmiss enhance\n",
+ enable ? "enable" : "disable");
+ err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
+ vif->fw_vif_idx, enable);
+ if (err)
+ ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
+ "failed to %s enhanced bmiss detection: %d\n",
+ enable ? "enable" : "disable", err);
+}
+
static int ath6kl_get_rsn_capab(struct cfg80211_beacon_data *beacon,
u8 *rsn_capab)
{
diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.h b/drivers/net/wireless/ath/ath6kl/cfg80211.h
index 5ea8cbb..b992046 100644
--- a/drivers/net/wireless/ath/ath6kl/cfg80211.h
+++ b/drivers/net/wireless/ath/ath6kl/cfg80211.h
@@ -62,5 +62,7 @@ void ath6kl_cfg80211_cleanup(struct ath6kl *ar);
struct ath6kl *ath6kl_cfg80211_create(void);
void ath6kl_cfg80211_destroy(struct ath6kl *ar);
+/* TODO: remove this once ath6kl_vif_cleanup() is moved to cfg80211.c */
+void ath6kl_cfg80211_sta_bmiss_enhance(struct ath6kl_vif *vif, bool enable);
#endif /* ATH6KL_CFG80211_H */
diff --git a/drivers/net/wireless/ath/ath6kl/core.h b/drivers/net/wireless/ath/ath6kl/core.h
index 4d9c6f1..cb79378 100644
--- a/drivers/net/wireless/ath/ath6kl/core.h
+++ b/drivers/net/wireless/ath/ath6kl/core.h
@@ -100,6 +100,9 @@ enum ath6kl_fw_capability {
/* Firmware has support to override rsn cap of rsn ie */
ATH6KL_FW_CAPABILITY_RSN_CAP_OVERRIDE,
+ /* Firmware supports enhanced bmiss detection */
+ ATH6KL_FW_CAPABILITY_BMISS_ENHANCE,
+
/* this needs to be last */
ATH6KL_FW_CAPABILITY_MAX,
};
diff --git a/drivers/net/wireless/ath/ath6kl/init.c b/drivers/net/wireless/ath/ath6kl/init.c
index 7eb0515..10de132 100644
--- a/drivers/net/wireless/ath/ath6kl/init.c
+++ b/drivers/net/wireless/ath/ath6kl/init.c
@@ -1659,6 +1659,9 @@ void ath6kl_cleanup_vif(struct ath6kl_vif *vif, bool wmi_ready)
cfg80211_scan_done(vif->scan_req, true);
vif->scan_req = NULL;
}
+
+ /* need to clean up enhanced bmiss detection fw state */
+ ath6kl_cfg80211_sta_bmiss_enhance(vif, false);
}
void ath6kl_stop_txrx(struct ath6kl *ar)
diff --git a/drivers/net/wireless/ath/ath6kl/wmi.c b/drivers/net/wireless/ath/ath6kl/wmi.c
index ee8ec23..09a9e0c 100644
--- a/drivers/net/wireless/ath/ath6kl/wmi.c
+++ b/drivers/net/wireless/ath/ath6kl/wmi.c
@@ -2997,6 +2997,25 @@ int ath6kl_wmi_add_del_mcast_filter_cmd(struct wmi *wmi, u8 if_idx,
return ret;
}
+int ath6kl_wmi_sta_bmiss_enhance_cmd(struct wmi *wmi, u8 if_idx, bool enhance)
+{
+ struct sk_buff *skb;
+ struct wmi_sta_bmiss_enhance_cmd *cmd;
+ int ret;
+
+ skb = ath6kl_wmi_get_new_buf(sizeof(*cmd));
+ if (!skb)
+ return -ENOMEM;
+
+ cmd = (struct wmi_sta_bmiss_enhance_cmd *) skb->data;
+ cmd->enable = enhance ? 1 : 0;
+
+ ret = ath6kl_wmi_cmd_send(wmi, if_idx, skb,
+ WMI_STA_BMISS_ENHANCE_CMDID,
+ NO_SYNC_WMIFLAG);
+ return ret;
+}
+
s32 ath6kl_wmi_get_rate(s8 rate_index)
{
if (rate_index == RATE_AUTO)
diff --git a/drivers/net/wireless/ath/ath6kl/wmi.h b/drivers/net/wireless/ath/ath6kl/wmi.h
index 9076bec..014f3dd 100644
--- a/drivers/net/wireless/ath/ath6kl/wmi.h
+++ b/drivers/net/wireless/ath/ath6kl/wmi.h
@@ -624,6 +624,10 @@ enum wmi_cmd_id {
WMI_SEND_MGMT_CMDID,
WMI_BEGIN_SCAN_CMDID,
+ WMI_SET_BLACK_LIST,
+ WMI_SET_MCASTRATE,
+
+ WMI_STA_BMISS_ENHANCE_CMDID,
};
enum wmi_mgmt_frame_type {
@@ -1017,6 +1021,11 @@ struct wmi_bmiss_time_cmd {
__le16 num_beacons;
};
+/* WMI_STA_ENHANCE_BMISS_CMDID */
+struct wmi_sta_bmiss_enhance_cmd {
+ u8 enable;
+} __packed;
+
/* WMI_SET_POWER_MODE_CMDID */
enum wmi_power_mode {
REC_POWER = 0x01,
@@ -2547,6 +2556,8 @@ int ath6kl_wmi_set_roam_mode_cmd(struct wmi *wmi, enum wmi_roam_mode mode);
int ath6kl_wmi_mcast_filter_cmd(struct wmi *wmi, u8 if_idx, bool mc_all_on);
int ath6kl_wmi_add_del_mcast_filter_cmd(struct wmi *wmi, u8 if_idx,
u8 *filter, bool add_filter);
+int ath6kl_wmi_sta_bmiss_enhance_cmd(struct wmi *wmi, u8 if_idx, bool enable);
+
/* AP mode uAPSD */
int ath6kl_wmi_ap_set_apsd(struct wmi *wmi, u8 if_idx, u8 enable);
--
1.7.4.1
On 05/14/2012 09:03 PM, Pedersen, Thomas wrote:
> On Mon, May 14, 2012 at 10:56:59AM -0700, Joe Perches wrote:
>
>> Why 2 messages when 1 message might do?
>>
>> err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
>> vif->fw_vif_idx, enable);
>> ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
>> "%s enhanced fw bmiss detection: %s\n",
>> enable ? "enable" : "disable",
>> err ? "OK" : "failed");
>
> OK that seems nicer. Should we still print the error code, or maybe it
> doesn't really matter?
I missed this in the original review, but it's actually better to not
use ath6kl_dbg() for error messages. They are more difficult to notice
that way.
Or did you have a specific reason for using ath6kl_dbg()?
Kalle
On 05/14/2012 09:31 PM, Pedersen, Thomas wrote:
> On Mon, May 14, 2012 at 09:09:17PM +0300, Kalle Valo wrote:
>> On 05/14/2012 09:03 PM, Pedersen, Thomas wrote:
>>> On Mon, May 14, 2012 at 10:56:59AM -0700, Joe Perches wrote:
>>>
>>>> Why 2 messages when 1 message might do?
>>>>
>>>> err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
>>>> vif->fw_vif_idx, enable);
>>>> ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
>>>> "%s enhanced fw bmiss detection: %s\n",
>>>> enable ? "enable" : "disable",
>>>> err ? "OK" : "failed");
>>>
>>> OK that seems nicer. Should we still print the error code, or maybe it
>>> doesn't really matter?
>>
>> I missed this in the original review, but it's actually better to not
>> use ath6kl_dbg() for error messages. They are more difficult to notice
>> that way.
>>
>> Or did you have a specific reason for using ath6kl_dbg()?
>
> No, you're right. However, if we consolidate these messages they will
> both be under ath6kl_dbg() or ath6kl_err(), and neither one would be
> correct.
>
> How about just keeping these separate and printing the error through
> ath6kl_err()?
Sounds good to me. And if you can, try to print the error value in
ath6kl_err().
Kalle
On Mon, 2012-05-14 at 10:47 -0700, Thomas Pedersen wrote:
> Enable enhanced bmiss detection if the firmware supports it. This
> feature is only enabled on some firmwares since it comes with a power
> cost.
[]
> diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c
[]
> @@ -2614,6 +2619,30 @@ static int ath6kl_set_channel(struct wiphy *wiphy, struct net_device *dev,
> return 0;
> }
>
> +void ath6kl_cfg80211_sta_bmiss_enhance(struct ath6kl_vif *vif, bool enable)
> +{
> + int err;
> +
> + if (WARN_ON(!test_bit(WMI_READY, &vif->ar->flag)))
> + return;
> +
> + if (vif->nw_type != INFRA_NETWORK)
> + return;
> +
> + if (!test_bit(ATH6KL_FW_CAPABILITY_BMISS_ENHANCE,
> + vif->ar->fw_capabilities))
> + return;
> +
> + ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "%s fw bmiss enhance\n",
> + enable ? "enable" : "disable");
> + err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
> + vif->fw_vif_idx, enable);
> + if (err)
> + ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
> + "failed to %s enhanced bmiss detection: %d\n",
> + enable ? "enable" : "disable", err);
Why 2 messages when 1 message might do?
err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
vif->fw_vif_idx, enable);
ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
"%s enhanced fw bmiss detection: %s\n",
enable ? "enable" : "disable",
err ? "OK" : "failed");
On Mon, 2012-05-14 at 11:03 -0700, Pedersen, Thomas wrote:
> On Mon, May 14, 2012 at 10:56:59AM -0700, Joe Perches wrote:
> > On Mon, 2012-05-14 at 10:47 -0700, Thomas Pedersen wrote:
> > > Enable enhanced bmiss detection if the firmware supports it. This
> > > feature is only enabled on some firmwares since it comes with a power
> > > cost.
> > []
> > > diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c
> > []
> > > @@ -2614,6 +2619,30 @@ static int ath6kl_set_channel(struct wiphy *wiphy, struct net_device *dev,
> > > return 0;
> > > }
> > >
> > > +void ath6kl_cfg80211_sta_bmiss_enhance(struct ath6kl_vif *vif, bool enable)
[]
> > Why 2 messages when 1 message might do?
> >
> > err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
> > vif->fw_vif_idx, enable);
> > ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
> > "%s enhanced fw bmiss detection: %s\n",
> > enable ? "enable" : "disable",
> > err ? "OK" : "failed");
>
> OK that seems nicer. Should we still print the error code, or maybe it
> doesn't really matter?
Hi Thomas.
That's up to you. I don't know the code at all.
btw: the err ?: output is badly reversed in my example.
On Mon, May 14, 2012 at 09:09:17PM +0300, Kalle Valo wrote:
> On 05/14/2012 09:03 PM, Pedersen, Thomas wrote:
> > On Mon, May 14, 2012 at 10:56:59AM -0700, Joe Perches wrote:
> >
> >> Why 2 messages when 1 message might do?
> >>
> >> err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
> >> vif->fw_vif_idx, enable);
> >> ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
> >> "%s enhanced fw bmiss detection: %s\n",
> >> enable ? "enable" : "disable",
> >> err ? "OK" : "failed");
> >
> > OK that seems nicer. Should we still print the error code, or maybe it
> > doesn't really matter?
>
> I missed this in the original review, but it's actually better to not
> use ath6kl_dbg() for error messages. They are more difficult to notice
> that way.
>
> Or did you have a specific reason for using ath6kl_dbg()?
No, you're right. However, if we consolidate these messages they will
both be under ath6kl_dbg() or ath6kl_err(), and neither one would be
correct.
How about just keeping these separate and printing the error through
ath6kl_err()?
Thomas
Hi Joe,
On Mon, May 14, 2012 at 10:56:59AM -0700, Joe Perches wrote:
> On Mon, 2012-05-14 at 10:47 -0700, Thomas Pedersen wrote:
> > Enable enhanced bmiss detection if the firmware supports it. This
> > feature is only enabled on some firmwares since it comes with a power
> > cost.
> []
> > diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c
> []
> > @@ -2614,6 +2619,30 @@ static int ath6kl_set_channel(struct wiphy *wiphy, struct net_device *dev,
> > return 0;
> > }
> >
> > +void ath6kl_cfg80211_sta_bmiss_enhance(struct ath6kl_vif *vif, bool enable)
> > +{
> > + int err;
> > +
> > + if (WARN_ON(!test_bit(WMI_READY, &vif->ar->flag)))
> > + return;
> > +
> > + if (vif->nw_type != INFRA_NETWORK)
> > + return;
> > +
> > + if (!test_bit(ATH6KL_FW_CAPABILITY_BMISS_ENHANCE,
> > + vif->ar->fw_capabilities))
> > + return;
> > +
> > + ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "%s fw bmiss enhance\n",
> > + enable ? "enable" : "disable");
> > + err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
> > + vif->fw_vif_idx, enable);
> > + if (err)
> > + ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
> > + "failed to %s enhanced bmiss detection: %d\n",
> > + enable ? "enable" : "disable", err);
>
> Why 2 messages when 1 message might do?
>
> err = ath6kl_wmi_sta_bmiss_enhance_cmd(vif->ar->wmi,
> vif->fw_vif_idx, enable);
> ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
> "%s enhanced fw bmiss detection: %s\n",
> enable ? "enable" : "disable",
> err ? "OK" : "failed");
OK that seems nicer. Should we still print the error code, or maybe it
doesn't really matter?
Thomas