| Message ID | 1559561865-27928-1-git-send-email-srirrama@codeaurora.org (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | b0d742f32bc3cf341a58720d04cee63fb307190c |
| Delegated to: | Kalle Valo |
| Headers | show |
| Series | ath11k: Fix vdev start failure when operating in CN, KR regdomain | expand |
Sriram R <srirrama@codeaurora.org> wrote: > Currently, firmware expects the correct regdomain to be set in > vdev start command. As firmware supports regulatory offload, it can track > the current regdomain. This info is received from firmware via > REG_CHAN_LIST_CC event during bringup or after setting country code. > Currently the dfs_region received is mapped to nl80211_dfs_regions > and stored in default_regd or new_regd and is used for vdev start. > But since nl80211_dfs_regions doesn't have a value enumerated for CN > KR, the firmware rejects VDEV start with an 'invalid regdomain' error. > Hence, send the appropriate regdomain received from firmware during > REG_CHAN_LIST_CC event in vdev start to avoid vdev start failure. > > Signed-off-by: Sriram R <srirrama@codeaurora.org> Patch applied to ath.git, thanks. b0d742f32bc3 ath11k: Fix vdev start failure when operating in CN, KR regdomain
Kalle Valo <kvalo@codeaurora.org> writes: > Sriram R <srirrama@codeaurora.org> wrote: > >> Currently, firmware expects the correct regdomain to be set in >> vdev start command. As firmware supports regulatory offload, it can track >> the current regdomain. This info is received from firmware via >> REG_CHAN_LIST_CC event during bringup or after setting country code. >> Currently the dfs_region received is mapped to nl80211_dfs_regions >> and stored in default_regd or new_regd and is used for vdev start. >> But since nl80211_dfs_regions doesn't have a value enumerated for CN >> KR, the firmware rejects VDEV start with an 'invalid regdomain' error. >> Hence, send the appropriate regdomain received from firmware during >> REG_CHAN_LIST_CC event in vdev start to avoid vdev start failure. >> >> Signed-off-by: Sriram R <srirrama@codeaurora.org> > > Patch applied to ath.git, thanks. > > b0d742f32bc3 ath11k: Fix vdev start failure when operating in CN, KR regdomain This added a new warning: drivers/net/wireless/ath/ath11k/wmi.c:4655:34: warning: mixing different enum types drivers/net/wireless/ath/ath11k/wmi.c:4655:34: unsigned int enum dfs_reg versus drivers/net/wireless/ath/ath11k/wmi.c:4655:34: unsigned int enum ath11k_dfs_region Please send a followup patch to fix that.
On 2019-06-06 22:18, Kalle Valo wrote: > Kalle Valo <kvalo@codeaurora.org> writes: > >> Sriram R <srirrama@codeaurora.org> wrote: >> >>> Currently, firmware expects the correct regdomain to be set in >>> vdev start command. As firmware supports regulatory offload, it can >>> track >>> the current regdomain. This info is received from firmware via >>> REG_CHAN_LIST_CC event during bringup or after setting country code. >>> Currently the dfs_region received is mapped to nl80211_dfs_regions >>> and stored in default_regd or new_regd and is used for vdev start. >>> But since nl80211_dfs_regions doesn't have a value enumerated for CN >>> KR, the firmware rejects VDEV start with an 'invalid regdomain' >>> error. >>> Hence, send the appropriate regdomain received from firmware during >>> REG_CHAN_LIST_CC event in vdev start to avoid vdev start failure. >>> >>> Signed-off-by: Sriram R <srirrama@codeaurora.org> >> >> Patch applied to ath.git, thanks. >> >> b0d742f32bc3 ath11k: Fix vdev start failure when operating in CN, KR >> regdomain > > This added a new warning: > > drivers/net/wireless/ath/ath11k/wmi.c:4655:34: warning: mixing > different enum types > drivers/net/wireless/ath/ath11k/wmi.c:4655:34: unsigned int enum > dfs_reg versus > drivers/net/wireless/ath/ath11k/wmi.c:4655:34: unsigned int enum > ath11k_dfs_region > > Please send a followup patch to fix that. Sure Kalle. These warnings got subsided during my compilation and they look valid. I'll send a followup patch. Regards, Sriram.R
diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h index 148c9a4..f7d2d96 100644 --- a/drivers/net/wireless/ath/ath11k/core.h +++ b/drivers/net/wireless/ath/ath11k/core.h @@ -302,6 +302,16 @@ enum ath11k_stats_type { ATH11K_STATS_TYPE_MAX, }; +enum ath11k_dfs_region { + ATH11K_DFS_REG_UNSET, + ATH11K_DFS_REG_FCC, + ATH11K_DFS_REG_ETSI, + ATH11K_DFS_REG_MKK, + ATH11K_DFS_REG_CN, + ATH11K_DFS_REG_KR, + ATH11K_DFS_REG_UNDEF, +}; + struct ath11k_htt_data_stats { u64 legacy[ATH11K_COUNTER_TYPE_MAX][ATH11K_LEGACY_NUM]; u64 ht[ATH11K_COUNTER_TYPE_MAX][ATH11K_HT_MCS_NUM]; @@ -622,6 +632,9 @@ struct ath11k_base { * This may or may not be used during the runtime */ struct ieee80211_regdomain *new_regd[MAX_RADIOS]; + + /* Current DFS Regulatory */ + enum ath11k_dfs_region dfs_region; #ifdef CONFIG_ATH11K_DEBUGFS struct dentry *debugfs_soc; #endif diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index cb0de8e..87a420e 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -3915,15 +3915,7 @@ ath11k_mac_vdev_start_restart(struct ath11k_vif *arvif, arg.channel.passive = arg.channel.chan_radar; spin_lock_bh(&ab->data_lock); - - /* Use the new reg info if available */ - if (ar->ab->new_regd[ar->pdev_idx]) - arg.regdomain = - ar->ab->new_regd[ar->pdev_idx]->dfs_region; - else - arg.regdomain = - ar->ab->default_regd[ar->pdev_idx]->dfs_region; - + arg.regdomain = ar->ab->dfs_region; spin_unlock_bh(&ab->data_lock); /* TODO: Notify if secondary 80Mhz also needs radar detection */ diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c index 9ca44c4..3ac1e1c 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.c +++ b/drivers/net/wireless/ath/ath11k/wmi.c @@ -4645,6 +4645,7 @@ static int ath11k_reg_chan_list_event(struct ath11k_base *ab, u8 *evt_buf, u32 l /* This regd would be applied during mac registration */ ab->default_regd[pdev_idx] = regd; } + ab->dfs_region = reg_info->dfs_region; spin_unlock(&ab->data_lock); goto mem_free; @@ -4748,6 +4749,8 @@ static inline const char *ath11k_wmi_vdev_resp_print(u32 vdev_resp_status) return "not supported"; case WMI_VDEV_START_RESPONSE_DFS_VIOLATION: return "dfs violation"; + case WMI_VDEV_START_RESPONSE_INVALID_REGDOMAIN: + return "invalid regdomain"; default: return "unknown"; } diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h index 366a136..0d36c8a 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.h +++ b/drivers/net/wireless/ath/ath11k/wmi.h @@ -4221,6 +4221,7 @@ enum wmi_vdev_start_resp_status_code { WMI_VDEV_START_RESPONSE_INVALID_VDEVID = 1, WMI_VDEV_START_RESPONSE_NOT_SUPPORTED = 2, WMI_VDEV_START_RESPONSE_DFS_VIOLATION = 3, + WMI_VDEV_START_RESPONSE_INVALID_REGDOMAIN = 4, }; ;
Currently, firmware expects the correct regdomain to be set in vdev start command. As firmware supports regulatory offload, it can track the current regdomain. This info is received from firmware via REG_CHAN_LIST_CC event during bringup or after setting country code. Currently the dfs_region received is mapped to nl80211_dfs_regions and stored in default_regd or new_regd and is used for vdev start. But since nl80211_dfs_regions doesn't have a value enumerated for CN KR, the firmware rejects VDEV start with an 'invalid regdomain' error. Hence, send the appropriate regdomain received from firmware during REG_CHAN_LIST_CC event in vdev start to avoid vdev start failure. Signed-off-by: Sriram R <srirrama@codeaurora.org> --- drivers/net/wireless/ath/ath11k/core.h | 13 +++++++++++++ drivers/net/wireless/ath/ath11k/mac.c | 10 +--------- drivers/net/wireless/ath/ath11k/wmi.c | 3 +++ drivers/net/wireless/ath/ath11k/wmi.h | 1 + 4 files changed, 18 insertions(+), 9 deletions(-)