[PATCH v2 0/8] cfg80211/mac80211: Add support for 6GHZ STA for various modes : LPI, SP and VLP

[PATCH v2 0/8] cfg80211/mac80211: Add support for 6GHZ STA for various modes : LPI, SP and VLP

[Wen Gong]

v2: change per comments of johannes.
    including code style, code logic, patch merge, commit log... 

It introduced some new concept:
power type of AP(STANDARD_POWER_AP, INDOOR_AP, VERY_LOW_POWER_AP)
power type of STATION(DEFAULT_CLIENT, SUBORDINATE_CLIENT)
power spectral density(psd)

This patchset for cfg80211/mac80211 is to add the definition of new
concept of 6G and add basic parse of IE(transmit power envelope
element) in beacon and save power spectral density(psd) reported
by lower-driver for 6G channel, the info will be passed to lower
driver when connecting to 6G AP.

Wen Gong (8):
  cfg80211: add power type definition for 6 GHz
  mac80211: add definition of regulatory info in 6 GHz operation
    information
  mac80211: add parse regulatory info in 6 GHz operation information
  cfg80211: add definition for 6 GHz power spectral density(psd)
  cfg80211: save power spectral density(psd) of regulatory rule
  mac80211: add definition for transmit power envelope element
  mac80211: add parse transmit power envelope element
  mac80211: save transmit power envelope element and power constraint

 include/linux/ieee80211.h    | 43 +++++++++++++++++++++++++++++++++++-
 include/net/cfg80211.h       |  7 ++++++
 include/net/mac80211.h       |  6 +++++
 include/net/regulatory.h     |  1 +
 include/uapi/linux/nl80211.h | 36 ++++++++++++++++++++++++++++++
 net/mac80211/chan.c          |  9 ++++++++
 net/mac80211/ieee80211_i.h   |  3 +++
 net/mac80211/mlme.c          | 26 ++++++++++++++++++++++
 net/mac80211/util.c          | 19 ++++++++++++++++
 net/wireless/reg.c           | 14 ++++++++++++
 10 files changed, 163 insertions(+), 1 deletion(-)

-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Wen Gong]

6 GHz regulatory domains introduces different modes for 6 GHz AP
operations Low Power Indoor(LPI), Standard Power(SP) and Very Low
Power(VLP). 6 GHz STAs could be operated as either Regular or
Subordinate clients. This patch is define the flags for power type
of AP and STATION mode.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 include/net/cfg80211.h       |  2 ++
 include/uapi/linux/nl80211.h | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 36 insertions(+)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 58c2cd417e89..f17a4d1202fc 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -676,6 +676,7 @@ struct key_params {
  *	chan will define the primary channel and all other
  *	parameters are ignored.
  * @freq1_offset: offset from @center_freq1, in KHz
+ * @power_type: power type of BSS for 6 GHz
  */
 struct cfg80211_chan_def {
 	struct ieee80211_channel *chan;
@@ -684,6 +685,7 @@ struct cfg80211_chan_def {
 	u32 center_freq2;
 	struct ieee80211_edmg edmg;
 	u16 freq1_offset;
+	enum nl80211_ap_reg_power power_type;
 };
 
 /*
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index f962c06e9818..ab1d4aa85272 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -4088,6 +4088,40 @@ enum nl80211_dfs_regions {
 	NL80211_DFS_JP		= 3,
 };
 
+/**
+ * enum nl80211_ap_reg_power - regulatory power for a Access Point
+ *
+ * @NL80211_REG_UNSET_AP: Access Point has no regulatory power mode
+ * @NL80211_REG_LPI: Indoor Access Point
+ * @NL80211_REG_SP: Standard power Access Point
+ * @NL80211_REG_VLP: Very low power Access Point
+ */
+enum nl80211_ap_reg_power {
+	NL80211_REG_UNSET_AP,
+	NL80211_REG_LPI_AP,
+	NL80211_REG_SP_AP,
+	NL80211_REG_VLP_AP,
+	NL80211_REG_AP_POWER_AFTER_LAST,
+	NL80211_REG_AP_POWER_MAX =
+		NL80211_REG_AP_POWER_AFTER_LAST - 1,
+};
+
+/**
+ * enum nl80211_client_reg_power - regulatory power for a client
+ *
+ * @NL80211_REG_UNSET_CLIENT: Client has no regulatory power mode
+ * @NL80211_REG_DEFAULT_CLIENT: Default Client
+ * @NL80211_REG_SUBORDINATE_CLIENT: Subordinate Client
+ */
+enum nl80211_client_reg_power {
+	NL80211_REG_UNSET_CLIENT,
+	NL80211_REG_DEFAULT_CLIENT,
+	NL80211_REG_SUBORDINATE_CLIENT,
+	NL80211_REG_CLIENT_POWER_AFTER_LAST,
+	NL80211_REG_CLIENT_POWER_MAX =
+		NL80211_REG_CLIENT_POWER_AFTER_LAST - 1,
+};
+
 /**
  * enum nl80211_user_reg_hint_type - type of user regulatory hint
  *
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Johannes Berg]

>  struct cfg80211_chan_def {
>  	struct ieee80211_channel *chan;
> @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
>  	u32 center_freq2;
>  	struct ieee80211_edmg edmg;
>  	u16 freq1_offset;
> +	enum nl80211_ap_reg_power power_type;

I'm not sure why this should be in the chandef, there's no way that
anything in cfg80211 is ever using it there, at least in your patches.

> +/**
> + * enum nl80211_ap_reg_power - regulatory power for a Access Point
> + *
> + * @NL80211_REG_UNSET_AP: Access Point has no regulatory power mode
> + * @NL80211_REG_LPI: Indoor Access Point
> + * @NL80211_REG_SP: Standard power Access Point
> + * @NL80211_REG_VLP: Very low power Access Point
> + */
> +enum nl80211_ap_reg_power {
> +	NL80211_REG_UNSET_AP,
> +	NL80211_REG_LPI_AP,
> +	NL80211_REG_SP_AP,
> +	NL80211_REG_VLP_AP,
> +	NL80211_REG_AP_POWER_AFTER_LAST,
> +	NL80211_REG_AP_POWER_MAX =
> +		NL80211_REG_AP_POWER_AFTER_LAST - 1,
> +};

Similarly here (and the other enum), why is this in nl80211 if it's
never used in nl80211?

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Johannes Berg]

On Thu, 2021-08-26 at 10:20 +0200, Johannes Berg wrote:
> >  struct cfg80211_chan_def {
> >  	struct ieee80211_channel *chan;
> > @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
> >  	u32 center_freq2;
> >  	struct ieee80211_edmg edmg;
> >  	u16 freq1_offset;
> > +	enum nl80211_ap_reg_power power_type;
> 
> I'm not sure why this should be in the chandef, there's no way that
> anything in cfg80211 is ever using it there, at least in your patches.

Does it even *apply* to a channel? What if I'm connecting to two APs on
the same channel (two client interfaces)?

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Wen Gong]

On 2021-08-26 16:22, Johannes Berg wrote:
> On Thu, 2021-08-26 at 10:20 +0200, Johannes Berg wrote:
>> >  struct cfg80211_chan_def {
>> >  	struct ieee80211_channel *chan;
>> > @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
>> >  	u32 center_freq2;
>> >  	struct ieee80211_edmg edmg;
>> >  	u16 freq1_offset;
>> > +	enum nl80211_ap_reg_power power_type;
>> 
>> I'm not sure why this should be in the chandef, there's no way that
>> anything in cfg80211 is ever using it there, at least in your patches.
> 
> Does it even *apply* to a channel? What if I'm connecting to two APs on
> the same channel (two client interfaces)?
> 
this is one copy for each connection, each client has its own 
cfg80211_chan_def.
also it can be moved to struct ieee80211_bss_conf.
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Johannes Berg]

On Thu, 2021-08-26 at 19:02 +0800, Wen Gong wrote:
> On 2021-08-26 16:22, Johannes Berg wrote:
> > On Thu, 2021-08-26 at 10:20 +0200, Johannes Berg wrote:
> > > >  struct cfg80211_chan_def {
> > > >  	struct ieee80211_channel *chan;
> > > > @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
> > > >  	u32 center_freq2;
> > > >  	struct ieee80211_edmg edmg;
> > > >  	u16 freq1_offset;
> > > > +	enum nl80211_ap_reg_power power_type;
> > > 
> > > I'm not sure why this should be in the chandef, there's no way that
> > > anything in cfg80211 is ever using it there, at least in your patches.
> > 
> > Does it even *apply* to a channel? What if I'm connecting to two APs on
> > the same channel (two client interfaces)?
> > 
> this is one copy for each connection, each client has its own 
> cfg80211_chan_def.

That depends on where you check it - but you're basically saying "use
this only from vif->bss_conf.chandef (or something, didn't check now),
but chandef shows up in many other places and you don't maintain it
anywhere else.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Wen Gong]

On 2021-08-26 16:20, Johannes Berg wrote:
>>  struct cfg80211_chan_def {
>>  	struct ieee80211_channel *chan;
>> @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
>>  	u32 center_freq2;
>>  	struct ieee80211_edmg edmg;
>>  	u16 freq1_offset;
>> +	enum nl80211_ap_reg_power power_type;
> 
> I'm not sure why this should be in the chandef, there's no way that
> anything in cfg80211 is ever using it there, at least in your patches.
> 
It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse regulatory 
info in 6 GHz operation information.
should i move it to mac80211's .h file?
>> +/**
>> + * enum nl80211_ap_reg_power - regulatory power for a Access Point
>> + *
>> + * @NL80211_REG_UNSET_AP: Access Point has no regulatory power mode
>> + * @NL80211_REG_LPI: Indoor Access Point
>> + * @NL80211_REG_SP: Standard power Access Point
>> + * @NL80211_REG_VLP: Very low power Access Point
>> + */
>> +enum nl80211_ap_reg_power {
>> +	NL80211_REG_UNSET_AP,
>> +	NL80211_REG_LPI_AP,
>> +	NL80211_REG_SP_AP,
>> +	NL80211_REG_VLP_AP,
>> +	NL80211_REG_AP_POWER_AFTER_LAST,
>> +	NL80211_REG_AP_POWER_MAX =
>> +		NL80211_REG_AP_POWER_AFTER_LAST - 1,
>> +};
> 
> Similarly here (and the other enum), why is this in nl80211 if it's
> never used in nl80211?
> 
It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse regulatory 
info in 6 GHz operation information.
should i move it to mac80211's .h file?
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Johannes Berg]

On Thu, 2021-08-26 at 18:57 +0800, Wen Gong wrote:
> On 2021-08-26 16:20, Johannes Berg wrote:
> > >  struct cfg80211_chan_def {
> > >  	struct ieee80211_channel *chan;
> > > @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
> > >  	u32 center_freq2;
> > >  	struct ieee80211_edmg edmg;
> > >  	u16 freq1_offset;
> > > +	enum nl80211_ap_reg_power power_type;
> > 
> > I'm not sure why this should be in the chandef, there's no way that
> > anything in cfg80211 is ever using it there, at least in your patches.
> > 
> It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse regulatory 
> info in 6 GHz operation information.
> should i move it to mac80211's .h file?
> > > +/**
> > > + * enum nl80211_ap_reg_power - regulatory power for a Access Point
[...]
> > 
> It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse regulatory 
> info in 6 GHz operation information.
> should i move it to mac80211's .h file?

Yeah I saw both of them are used, but why are they defined as nl80211
API? Do you have any intention to set them through nl80211?

And like I said, I'm not really convinced this belongs into struct
cfg80211_chan_def either. Maybe it should be in bss_conf too?

johannes
> 


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 1/8] cfg80211: add power type definition for 6 GHz

[Wen Gong]

On 2021-08-26 18:59, Johannes Berg wrote:
> On Thu, 2021-08-26 at 18:57 +0800, Wen Gong wrote:
>> On 2021-08-26 16:20, Johannes Berg wrote:
>> > >  struct cfg80211_chan_def {
>> > >  	struct ieee80211_channel *chan;
>> > > @@ -684,6 +685,7 @@ struct cfg80211_chan_def {
>> > >  	u32 center_freq2;
>> > >  	struct ieee80211_edmg edmg;
>> > >  	u16 freq1_offset;
>> > > +	enum nl80211_ap_reg_power power_type;
>> >
>> > I'm not sure why this should be in the chandef, there's no way that
>> > anything in cfg80211 is ever using it there, at least in your patches.
>> >
>> It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse 
>> regulatory
>> info in 6 GHz operation information.
>> should i move it to mac80211's .h file?
>> > > +/**
>> > > + * enum nl80211_ap_reg_power - regulatory power for a Access Point
> [...]
>> >
>> It is used in mac80211 of [PATCH v2 3/8] mac80211: add parse 
>> regulatory
>> info in 6 GHz operation information.
>> should i move it to mac80211's .h file?
> 
> Yeah I saw both of them are used, but why are they defined as nl80211
> API? Do you have any intention to set them through nl80211?
> 
> And like I said, I'm not really convinced this belongs into struct
> cfg80211_chan_def either. Maybe it should be in bss_conf too?
yes, I also want to move it to struct ieee80211_bss_conf.
> 
> johannes
>> 

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 2/8] mac80211: add definition of regulatory info in 6 GHz operation information

[Wen Gong]

IEEE Std 802.11ax™-2021 added regulatory info subfield in HE operation
element, this patch is to add it in mac80211 definition.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 include/linux/ieee80211.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 68d8b9cdd3b8..deb2f536d104 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2264,6 +2264,9 @@ ieee80211_he_ppe_size(u8 ppe_thres_hdr, const u8 *phy_cap_info)
 #define IEEE80211_HE_OPERATION_PARTIAL_BSS_COLOR		0x40000000
 #define IEEE80211_HE_OPERATION_BSS_COLOR_DISABLED		0x80000000
 
+#define IEEE80211_6GHZ_CTRL_REG_LPI_AP	0
+#define IEEE80211_6GHZ_CTRL_REG_SP_AP	1
+
 /**
  * ieee80211_he_6ghz_oper - HE 6 GHz operation Information field
  * @primary: primary channel
@@ -2280,6 +2283,7 @@ struct ieee80211_he_6ghz_oper {
 #define		IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_80MHZ	2
 #define		IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_160MHZ	3
 #define IEEE80211_HE_6GHZ_OPER_CTRL_DUP_BEACON	0x4
+#define IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO	0x38
 	u8 control;
 	u8 ccfs0;
 	u8 ccfs1;
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 2/8] mac80211: add definition of regulatory info in 6 GHz operation information

[Johannes Berg]

On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
> IEEE Std 802.11ax™-2021 added regulatory info subfield in HE operation
> element, this patch is to add it in mac80211 definition.
> 
> Signed-off-by: Wen Gong <wgong@codeaurora.org>
> ---
>  include/linux/ieee80211.h | 4 ++++
>  1 file changed, 4 insertions(+)

Not really mac80211 (subject), but I'll fix.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 3/8] mac80211: add parse regulatory info in 6 GHz operation information

[Wen Gong]

This patch is to convert the regulatory info subfield in HE operation
element to power type and save in struct cfg80211_chan_def.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 net/mac80211/util.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 93d96a4f9c3e..cb1c35d8ef48 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3395,6 +3395,16 @@ bool ieee80211_chandef_he_6ghz_oper(struct ieee80211_sub_if_data *sdata,
 					      NL80211_BAND_6GHZ);
 	he_chandef.chan = ieee80211_get_channel(sdata->local->hw.wiphy, freq);
 
+	switch (u8_get_bits(he_6ghz_oper->control,
+			    IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
+	case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
+		he_chandef.power_type = NL80211_REG_LPI_AP;
+		break;
+	case IEEE80211_6GHZ_CTRL_REG_SP_AP:
+		he_chandef.power_type = NL80211_REG_SP_AP;
+		break;
+	}
+
 	switch (u8_get_bits(he_6ghz_oper->control,
 			    IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH)) {
 	case IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_20MHZ:
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 3/8] mac80211: add parse regulatory info in 6 GHz operation information

[Johannes Berg]

On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
> This patch is to convert the regulatory info subfield in HE operation
> element to power type and save in struct cfg80211_chan_def.
> 
> Signed-off-by: Wen Gong <wgong@codeaurora.org>
> ---
>  net/mac80211/util.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/net/mac80211/util.c b/net/mac80211/util.c
> index 93d96a4f9c3e..cb1c35d8ef48 100644
> --- a/net/mac80211/util.c
> +++ b/net/mac80211/util.c
> @@ -3395,6 +3395,16 @@ bool ieee80211_chandef_he_6ghz_oper(struct ieee80211_sub_if_data *sdata,
>  					      NL80211_BAND_6GHZ);
>  	he_chandef.chan = ieee80211_get_channel(sdata->local->hw.wiphy, freq);
>  
> 
> +	switch (u8_get_bits(he_6ghz_oper->control,
> +			    IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
> +	case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
> +		he_chandef.power_type = NL80211_REG_LPI_AP;
> +		break;
> +	case IEEE80211_6GHZ_CTRL_REG_SP_AP:
> +		he_chandef.power_type = NL80211_REG_SP_AP;
> +		break;
> +	}

Now here you put it in the chandef, but don't propagate it anywhere if
channels are merged, etc. I don't think this can work correctly in all
but the most trivial use cases.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 4/8] cfg80211: add definition for 6 GHz power spectral density(psd)

[Wen Gong]

6 GHz regulatory domains introduces power spectral density(psd). This
patch is define the flags for psd.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 include/net/cfg80211.h       | 5 +++++
 include/net/regulatory.h     | 1 +
 include/uapi/linux/nl80211.h | 2 ++
 3 files changed, 8 insertions(+)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f17a4d1202fc..27326e5d5ff3 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -108,6 +108,8 @@ struct wiphy;
  *	on this channel.
  * @IEEE80211_CHAN_16MHZ: 16 MHz bandwidth is permitted
  *	on this channel.
+ * @IEEE80211_CHAN_PSD: power spectral density (in dBm)
+ *	on this channel.
  *
  */
 enum ieee80211_channel_flags {
@@ -130,6 +132,7 @@ enum ieee80211_channel_flags {
 	IEEE80211_CHAN_4MHZ		= 1<<16,
 	IEEE80211_CHAN_8MHZ		= 1<<17,
 	IEEE80211_CHAN_16MHZ		= 1<<18,
+	IEEE80211_CHAN_PSD		= 1<<19,
 };
 
 #define IEEE80211_CHAN_NO_HT40 \
@@ -163,6 +166,7 @@ enum ieee80211_channel_flags {
  *	on this channel.
  * @dfs_state_entered: timestamp (jiffies) when the dfs state was entered.
  * @dfs_cac_ms: DFS CAC time in milliseconds, this is valid for DFS channels.
+ * @psd: power spectral density (in dBm)
  */
 struct ieee80211_channel {
 	enum nl80211_band band;
@@ -179,6 +183,7 @@ struct ieee80211_channel {
 	enum nl80211_dfs_state dfs_state;
 	unsigned long dfs_state_entered;
 	unsigned int dfs_cac_ms;
+	s8 psd;
 };
 
 /**
diff --git a/include/net/regulatory.h b/include/net/regulatory.h
index 47f06f6f5a67..ed20004fb6a9 100644
--- a/include/net/regulatory.h
+++ b/include/net/regulatory.h
@@ -221,6 +221,7 @@ struct ieee80211_reg_rule {
 	u32 flags;
 	u32 dfs_cac_ms;
 	bool has_wmm;
+	s8 psd;
 };
 
 struct ieee80211_regdomain {
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index ab1d4aa85272..9ac1dc551cbc 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -4043,6 +4043,7 @@ enum nl80211_sched_scan_match_attr {
  * @NL80211_RRF_NO_80MHZ: 80MHz operation not allowed
  * @NL80211_RRF_NO_160MHZ: 160MHz operation not allowed
  * @NL80211_RRF_NO_HE: HE operation not allowed
+ * @NL80211_RRF_PSD: channels has power spectral density value
  */
 enum nl80211_reg_rule_flags {
 	NL80211_RRF_NO_OFDM		= 1<<0,
@@ -4061,6 +4062,7 @@ enum nl80211_reg_rule_flags {
 	NL80211_RRF_NO_80MHZ		= 1<<15,
 	NL80211_RRF_NO_160MHZ		= 1<<16,
 	NL80211_RRF_NO_HE		= 1<<17,
+	NL80211_RRF_PSD		= 1<<18,
 };
 
 #define NL80211_RRF_PASSIVE_SCAN	NL80211_RRF_NO_IR
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 5/8] cfg80211: save power spectral density(psd) of regulatory rule

[Wen Gong]

The power spectral density(psd) of regulatory rule should be take
effect to the channels. This patch is to save the values to the
channel which has psd value.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 net/wireless/reg.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 0406ce7334fa..602d95e8bde6 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -1583,6 +1583,8 @@ static u32 map_regdom_flags(u32 rd_flags)
 		channel_flags |= IEEE80211_CHAN_NO_160MHZ;
 	if (rd_flags & NL80211_RRF_NO_HE)
 		channel_flags |= IEEE80211_CHAN_NO_HE;
+	if (rd_flags & NL80211_RRF_PSD)
+		channel_flags |= IEEE80211_CHAN_PSD;
 	return channel_flags;
 }
 
@@ -1787,6 +1789,9 @@ static void handle_channel_single_rule(struct wiphy *wiphy,
 				chan->dfs_cac_ms = reg_rule->dfs_cac_ms;
 		}
 
+		if (chan->flags & IEEE80211_CHAN_PSD)
+			chan->psd = reg_rule->psd;
+
 		return;
 	}
 
@@ -1807,6 +1812,9 @@ static void handle_channel_single_rule(struct wiphy *wiphy,
 			chan->dfs_cac_ms = IEEE80211_DFS_MIN_CAC_TIME_MS;
 	}
 
+	if (chan->flags & IEEE80211_CHAN_PSD)
+		chan->psd = reg_rule->psd;
+
 	if (chan->orig_mpwr) {
 		/*
 		 * Devices that use REGULATORY_COUNTRY_IE_FOLLOW_POWER
@@ -1876,6 +1884,9 @@ static void handle_channel_adjacent_rules(struct wiphy *wiphy,
 							 rrule2->dfs_cac_ms);
 		}
 
+		if (chan->flags & IEEE80211_CHAN_PSD)
+			chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
+
 		return;
 	}
 
@@ -2533,6 +2544,9 @@ static void handle_channel_custom(struct wiphy *wiphy,
 			chan->dfs_cac_ms = IEEE80211_DFS_MIN_CAC_TIME_MS;
 	}
 
+	if (chan->flags & IEEE80211_CHAN_PSD)
+		chan->psd = reg_rule->psd;
+
 	chan->max_power = chan->max_reg_power;
 }
 
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 5/8] cfg80211: save power spectral density(psd) of regulatory rule

[Johannes Berg]

I was going to apply this and patch 4 squashed, saying:


    cfg80211: regulatory: handle 6 GHz power spectral density (PSD)
    
    6 GHz regulatory domains introduce power spectral density (PSD).
    Allow wiphy-specific regulatory rules to specify these values.

but ...
> 
> +		if (chan->flags & IEEE80211_CHAN_PSD)
> +			chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
> +

This is obviously wrong?

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 5/8] cfg80211: save power spectral density(psd) of regulatory rule

[Wen Gong]

On 2021-08-26 16:25, Johannes Berg wrote:
> I was going to apply this and patch 4 squashed, saying:
> 
> 
>     cfg80211: regulatory: handle 6 GHz power spectral density (PSD)
> 
>     6 GHz regulatory domains introduce power spectral density (PSD).
>     Allow wiphy-specific regulatory rules to specify these values.
> 
> but ...
>> 
>> +		if (chan->flags & IEEE80211_CHAN_PSD)
>> +			chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
>> +
> 
> This is obviously wrong?
Yes it should change like this:

	if ((rrule1->flags & NL80211_RRF_PSD) && (rrule1->flags & 
NL80211_RRF_PSD))
		chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
	else
		chan->flags &= ~NL80211_RRF_PSD;

> 
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 5/8] cfg80211: save power spectral density(psd) of regulatory rule

[Johannes Berg]

On Thu, 2021-08-26 at 18:43 +0800, Wen Gong wrote:
> On 2021-08-26 16:25, Johannes Berg wrote:
> > I was going to apply this and patch 4 squashed, saying:
> > 
> > 
> >     cfg80211: regulatory: handle 6 GHz power spectral density (PSD)
> > 
> >     6 GHz regulatory domains introduce power spectral density (PSD).
> >     Allow wiphy-specific regulatory rules to specify these values.
> > 
> > but ...
> > > 
> > > +		if (chan->flags & IEEE80211_CHAN_PSD)
> > > +			chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
> > > +
> > 
> > This is obviously wrong?
> Yes it should change like this:
> 
> 	if ((rrule1->flags & NL80211_RRF_PSD) && (rrule1->flags & 
> NL80211_RRF_PSD))
> 		chan->psd = min_t(s8, rrule1->psd, rrule1->psd);

One of those still should be rrule2 :)

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 5/8] cfg80211: save power spectral density(psd) of regulatory rule

[Wen Gong]

On 2021-08-26 18:56, Johannes Berg wrote:
> On Thu, 2021-08-26 at 18:43 +0800, Wen Gong wrote:
>> On 2021-08-26 16:25, Johannes Berg wrote:
>> > I was going to apply this and patch 4 squashed, saying:
>> >
>> >
>> >     cfg80211: regulatory: handle 6 GHz power spectral density (PSD)
>> >
>> >     6 GHz regulatory domains introduce power spectral density (PSD).
>> >     Allow wiphy-specific regulatory rules to specify these values.
>> >
>> > but ...
>> > >
>> > > +		if (chan->flags & IEEE80211_CHAN_PSD)
>> > > +			chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
>> > > +
>> >
>> > This is obviously wrong?
>> Yes it should change like this:
>> 
>> 	if ((rrule1->flags & NL80211_RRF_PSD) && (rrule1->flags &
>> NL80211_RRF_PSD))
>> 		chan->psd = min_t(s8, rrule1->psd, rrule1->psd);
> 
> One of those still should be rrule2 :)
> 
yes
     if ((rrule1->flags & NL80211_RRF_PSD) && (rrule2->flags & 
NL80211_RRF_PSD))
         chan->psd = min_t(s8, rrule1->psd, rrule2->psd);
     else
         chan->flags &= ~NL80211_RRF_PSD;
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 6/8] mac80211: add definition for transmit power envelope element

[Wen Gong]

IEEE Std 802.11ax™-2021 have some change for transmit power envelope
element. This patch to add the definition of it.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 include/linux/ieee80211.h | 39 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 38 insertions(+), 1 deletion(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index deb2f536d104..65f5100052df 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2290,6 +2290,43 @@ struct ieee80211_he_6ghz_oper {
 	u8 minrate;
 } __packed;
 
+/**
+ * In "9.4.2.161 Transmit Power Envelope element" of "IEEE Std 802.11ax-2021",
+ * it show 4 types in "Table 9-275a-Maximum Transmit Power Interpretation subfield encoding".
+ * And it has 2 category for each type in "Table E-12-Regulatory Info subfield encoding in the United States".
+ * So it it totally max 8 Transmit Power Envelope element.
+ */
+#define IEEE80211_TPE_MAX_IE_COUNT	8
+/**
+ * In "Table 9-277—Meaning of Maximum Transmit Power Count subfield"
+ * of "IEEE Std 802.11ax™‐2021", the max power level is 8.
+ */
+#define IEEE80211_MAX_NUM_PWR_LEVEL	8
+
+#define IEEE80211_TPE_MAX_POWER_COUNT	8
+
+/* transmit power interpretation type of transmit power envelope element*/
+enum ieee80211_tx_power_intrpt_type {
+	IEEE80211_TPE_LOCAL_EIRP,
+	IEEE80211_TPE_LOCAL_EIRP_PSD,
+	IEEE80211_TPE_REG_CLIENT_EIRP,
+	IEEE80211_TPE_REG_CLIENT_EIRP_PSD,
+};
+
+/**
+ * struct ieee80211_tx_pwr_env
+ *
+ * This structure represents the "Transmit Power Envelope element"
+ */
+struct ieee80211_tx_pwr_env {
+	u8 tx_power_info;
+	s8 tx_power[IEEE80211_TPE_MAX_POWER_COUNT];
+} __packed;
+
+#define IEEE80211_TX_PWR_ENV_INFO_COUNT 0x7
+#define IEEE80211_TX_PWR_ENV_INFO_INTERPRET 0x38
+#define IEEE80211_TX_PWR_ENV_INFO_CATEGORY 0xC0
+
 /*
  * ieee80211_he_oper_size - calculate 802.11ax HE Operations IE size
  * @he_oper_ie: byte data of the He Operations IE, stating from the byte
@@ -2871,7 +2908,7 @@ enum ieee80211_eid {
 	WLAN_EID_VHT_OPERATION = 192,
 	WLAN_EID_EXTENDED_BSS_LOAD = 193,
 	WLAN_EID_WIDE_BW_CHANNEL_SWITCH = 194,
-	WLAN_EID_VHT_TX_POWER_ENVELOPE = 195,
+	WLAN_EID_TX_POWER_ENVELOPE = 195,
 	WLAN_EID_CHANNEL_SWITCH_WRAPPER = 196,
 	WLAN_EID_AID = 197,
 	WLAN_EID_QUIET_CHANNEL = 198,
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 6/8] mac80211: add definition for transmit power envelope element

[Johannes Berg]

On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
> IEEE Std 802.11ax™-2021 have some change for transmit power envelope
> element. This patch to add the definition of it.
> 

Generally: I'm sure Kalle has some page on this, but your commit
messages could use some work - you don't need to say "this patch" for
example, that's entirely obvious :)


> 
> +/**

not actually kernel-doc, but I can fix 

> + * In "9.4.2.161 Transmit Power Envelope element" of "IEEE Std 802.11ax-2021",
> + * it show 4 types in "Table 9-275a-Maximum Transmit Power Interpretation subfield encoding".
> + * And it has 2 category for each type in "Table E-12-Regulatory Info subfield encoding in the United States".
> + * So it it totally max 8 Transmit Power Envelope element.

very long lines, will fix that too

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 7/8] mac80211: add parse transmit power envelope element

[Wen Gong]

The transmit power envelope elements in beacon is used to calculate
the power limit by lower driver, and sometimes it has more than
one elements in a beacon frame.

This is to add parse the transmit power envelope elements, then it
will be saved and transfer to lower driver to calculate the power
limit.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 net/mac80211/ieee80211_i.h | 3 +++
 net/mac80211/util.c        | 9 +++++++++
 2 files changed, 12 insertions(+)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 648696b49f89..bb62de5e3758 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1494,6 +1494,7 @@ struct ieee802_11_elems {
 	const struct ieee80211_he_spr *he_spr;
 	const struct ieee80211_mu_edca_param_set *mu_edca_param_set;
 	const struct ieee80211_he_6ghz_capa *he_6ghz_capa;
+	const struct ieee80211_tx_pwr_env *tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT];
 	const u8 *uora_element;
 	const u8 *mesh_id;
 	const u8 *peering;
@@ -1544,6 +1545,8 @@ struct ieee802_11_elems {
 	u8 perr_len;
 	u8 country_elem_len;
 	u8 bssid_index_len;
+	u8 tx_pwr_env_len[IEEE80211_TPE_MAX_IE_COUNT];
+	u8 tx_pwr_env_num;
 
 	/* whether a parse error occurred while retrieving these elements */
 	bool parse_error;
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index cb1c35d8ef48..12c70cc30461 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1336,6 +1336,15 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 			elems->rsnx = pos;
 			elems->rsnx_len = elen;
 			break;
+		case WLAN_EID_TX_POWER_ENVELOPE:
+			if (elems->tx_pwr_env_num >= ARRAY_SIZE(elems->tx_pwr_env) ||
+			    elen < 1)
+				break;
+
+			elems->tx_pwr_env[elems->tx_pwr_env_num] = (void *)pos;
+			elems->tx_pwr_env_len[elems->tx_pwr_env_num] = elen;
+			elems->tx_pwr_env_num++;
+			break;
 		case WLAN_EID_EXTENSION:
 			ieee80211_parse_extension_element(calc_crc ?
 								&crc : NULL,
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 7/8] mac80211: add parse transmit power envelope element

[Johannes Berg]

On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
> 
> +		case WLAN_EID_TX_POWER_ENVELOPE:
> +			if (elems->tx_pwr_env_num >= ARRAY_SIZE(elems->tx_pwr_env) ||
> +			    elen < 1)
> +				break;

IMHO this should also check the max length, but I'll fix that.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

[PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

This is to save the transmit power envelope element and power
constraint in struct ieee80211_bss_conf for 6 GHz. Lower driver
will use this info to calculate the power limit.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
---
 include/net/mac80211.h |  6 ++++++
 net/mac80211/chan.c    |  9 +++++++++
 net/mac80211/mlme.c    | 26 ++++++++++++++++++++++++++
 3 files changed, 41 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index e89530d0d9c6..6e11e122e364 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -631,6 +631,9 @@ struct ieee80211_fils_discovery {
  * @s1g: BSS is S1G BSS (affects Association Request format).
  * @beacon_tx_rate: The configured beacon transmit rate that needs to be passed
  *	to driver when rate control is offloaded to firmware.
+ * @tx_pwr_env: transmit power envelope array of BSS.
+ * @tx_pwr_env_num: number of @tx_pwr_env.
+ * @pwr_reduction: power constraint of BSS.
  */
 struct ieee80211_bss_conf {
 	const u8 *bssid;
@@ -700,6 +703,9 @@ struct ieee80211_bss_conf {
 	u32 unsol_bcast_probe_resp_interval;
 	bool s1g;
 	struct cfg80211_bitrate_mask beacon_tx_rate;
+	struct ieee80211_tx_pwr_env tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT];
+	u8 tx_pwr_env_num;
+	u8 pwr_reduction;
 };
 
 /**
diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 907bb1f748a1..149d4ac798f6 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -721,6 +721,15 @@ static int ieee80211_assign_vif_chanctx(struct ieee80211_sub_if_data *sdata,
 					 lockdep_is_held(&local->chanctx_mtx));
 
 	if (conf) {
+		if (conf->def.chan->band == NL80211_BAND_6GHZ) {
+			struct ieee80211_bss_conf *bss_conf;
+
+			bss_conf = &sdata->vif.bss_conf;
+			bss_conf->pwr_reduction = 0;
+			bss_conf->tx_pwr_env_num = 0;
+			memset(bss_conf->tx_pwr_env, 0, sizeof(bss_conf->tx_pwr_env));
+		}
+
 		curr_ctx = container_of(conf, struct ieee80211_chanctx, conf);
 
 		drv_unassign_vif_chanctx(local, sdata, curr_ctx);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 2480bd0577bb..a6d66b4ad7ee 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -5070,6 +5070,7 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 
 	if (!(ifmgd->flags & IEEE80211_STA_DISABLE_HE)) {
 		const struct cfg80211_bss_ies *ies;
+		struct ieee80211_bss_conf *bss_conf;
 		const u8 *he_oper_ie;
 
 		ies = rcu_dereference(cbss->ies);
@@ -5081,6 +5082,31 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 		else
 			he_oper = NULL;
 
+		if (is_6ghz) {
+			struct ieee802_11_elems elems;
+			u8 i, j = 0;
+
+			ieee802_11_parse_elems(ies->data, ies->len, false, &elems,
+					       NULL, NULL);
+
+			if (elems.pwr_constr_elem)
+				bss_conf->pwr_reduction = *elems.pwr_constr_elem;
+
+			BUILD_BUG_ON(ARRAY_SIZE(bss_conf->tx_pwr_env) !=
+				     ARRAY_SIZE(elems.tx_pwr_env));
+
+			for (i = 0; i < elems.tx_pwr_env_num; i++) {
+				if (elems.tx_pwr_env_len[i] >
+				    sizeof(bss_conf->tx_pwr_env[j]))
+					continue;
+
+				bss_conf->tx_pwr_env_num++;
+				memcpy(&bss_conf->tx_pwr_env[j], elems.tx_pwr_env[i],
+				       elems.tx_pwr_env_len[i]);
+				j++;
+			}
+		}
+
 		if (!ieee80211_verify_sta_he_mcs_support(sband, he_oper))
 			ifmgd->flags |= IEEE80211_STA_DISABLE_HE;
 	}
-- 
2.31.1


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
> 
>  	if (!(ifmgd->flags & IEEE80211_STA_DISABLE_HE)) {
>  		const struct cfg80211_bss_ies *ies;
> +		struct ieee80211_bss_conf *bss_conf;

am I missing where you set this?

> +		if (is_6ghz) {
> +			struct ieee802_11_elems elems;

This is pretty big, not sure we want it on the stack (causes warnings
for me in build). Also, if we're doing this anyway, then we can change
the code above (perhaps as a separate patch) to not do
cfg80211_find_ext_ie() but rather take it out of the parsed.

> +			u8 i, j = 0;
> +
> +			ieee802_11_parse_elems(ies->data, ies->len, false, &elems,

(line too long)

> +					       NULL, NULL);
> +
> +			if (elems.pwr_constr_elem)
> +				bss_conf->pwr_reduction = *elems.pwr_constr_elem;

before using it?

> +
> +			BUILD_BUG_ON(ARRAY_SIZE(bss_conf->tx_pwr_env) !=
> +				     ARRAY_SIZE(elems.tx_pwr_env));
> +
> +			for (i = 0; i < elems.tx_pwr_env_num; i++) {
> +				if (elems.tx_pwr_env_len[i] >
> +				    sizeof(bss_conf->tx_pwr_env[j]))
> +					continue;

I did that in the parsing itself.

> +
> +				bss_conf->tx_pwr_env_num++;
> +				memcpy(&bss_conf->tx_pwr_env[j], elems.tx_pwr_env[i],
> +				       elems.tx_pwr_env_len[i]);

You're never resetting any of this for the next connection (if it's not
6 GHz for example, or doesn't have any data) - should probably memset
all the new members to 0 before the "if (is_6ghz)" or so?

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-26 16:29, Johannes Berg wrote:
> On Fri, 2021-08-20 at 08:20 -0400, Wen Gong wrote:
>> 
>>  	if (!(ifmgd->flags & IEEE80211_STA_DISABLE_HE)) {
>>  		const struct cfg80211_bss_ies *ies;
>> +		struct ieee80211_bss_conf *bss_conf;
> 
> am I missing where you set this?
sorry, i lost it in v2 patch.
bss_conf = &sdata->vif.bss_conf;
> 
>> +		if (is_6ghz) {
>> +			struct ieee802_11_elems elems;
> 
> This is pretty big, not sure we want it on the stack (causes warnings
> for me in build). Also, if we're doing this anyway, then we can change
> the code above (perhaps as a separate patch) to not do
> cfg80211_find_ext_ie() but rather take it out of the parsed.
will change it.
> 
>> +			u8 i, j = 0;
>> +
>> +			ieee802_11_parse_elems(ies->data, ies->len, false, &elems,
> 
> (line too long)
> 
>> +					       NULL, NULL);
>> +
>> +			if (elems.pwr_constr_elem)
>> +				bss_conf->pwr_reduction = *elems.pwr_constr_elem;
> 
> before using it?
> 
>> +
>> +			BUILD_BUG_ON(ARRAY_SIZE(bss_conf->tx_pwr_env) !=
>> +				     ARRAY_SIZE(elems.tx_pwr_env));
>> +
>> +			for (i = 0; i < elems.tx_pwr_env_num; i++) {
>> +				if (elems.tx_pwr_env_len[i] >
>> +				    sizeof(bss_conf->tx_pwr_env[j]))
>> +					continue;
> 
> I did that in the parsing itself.
> 
>> +
>> +				bss_conf->tx_pwr_env_num++;
>> +				memcpy(&bss_conf->tx_pwr_env[j], elems.tx_pwr_env[i],
>> +				       elems.tx_pwr_env_len[i]);
> 
> You're never resetting any of this for the next connection (if it's not
> 6 GHz for example, or doesn't have any data) - should probably memset
> all the new members to 0 before the "if (is_6ghz)" or so?
> 
it is memset here i this patch:
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -721,6 +721,15 @@ static int ieee80211_assign_vif_chanctx(struct 
ieee80211_sub_if_data *sdata,
                                          
lockdep_is_held(&local->chanctx_mtx));

         if (conf) {
+               if (conf->def.chan->band == NL80211_BAND_6GHZ) {
+                       struct ieee80211_bss_conf *bss_conf;
+
+                       bss_conf = &sdata->vif.bss_conf;
+                       bss_conf->pwr_reduction = 0;
+                       bss_conf->tx_pwr_env_num = 0;
+                       memset(bss_conf->tx_pwr_env, 0, 
sizeof(bss_conf->tx_pwr_env));
+               }
+
                 curr_ctx = container_of(conf, struct ieee80211_chanctx, 
conf);

                 drv_unassign_vif_chanctx(local, sdata, curr_ctx);

> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Thu, 2021-08-26 at 18:50 +0800, Wen Gong wrote:
> > 
> it is memset here i this patch:

Oops, missed that.

But is that really a good place for it? Doesn't really seem to belong to
assigning a channel context - maybe put it into set_disassoc()?

johannes




-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-26 18:57, Johannes Berg wrote:
> On Thu, 2021-08-26 at 18:50 +0800, Wen Gong wrote:
>> >
>> it is memset here i this patch:
> 
> Oops, missed that.
> 
> But is that really a good place for it? Doesn't really seem to belong 
> to
> assigning a channel context - maybe put it into set_disassoc()?
> 
it is correct.
you can see it is place together with "drv_unassign_vif_chanctx(local, 
sdata, curr_ctx)"
in ieee80211_assign_vif_chanctx(), it is for disconnect.
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Thu, 2021-08-26 at 19:00 +0800, Wen Gong wrote:
> On 2021-08-26 18:57, Johannes Berg wrote:
> > On Thu, 2021-08-26 at 18:50 +0800, Wen Gong wrote:
> > > > 
> > > it is memset here i this patch:
> > 
> > Oops, missed that.
> > 
> > But is that really a good place for it? Doesn't really seem to belong 
> > to
> > assigning a channel context - maybe put it into set_disassoc()?
> > 
> it is correct.
> you can see it is place together with "drv_unassign_vif_chanctx(local, 
> sdata, curr_ctx)"
> in ieee80211_assign_vif_chanctx(), it is for disconnect.

Yes, I know it's *correct*, but that doesn't mean it's *good*?

Look at that code - it does nothing with bss_conf. Nobody is ever going
to look there.

johannes
> 


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-26 19:10, Johannes Berg wrote:
> On Thu, 2021-08-26 at 19:00 +0800, Wen Gong wrote:
>> On 2021-08-26 18:57, Johannes Berg wrote:
>> > On Thu, 2021-08-26 at 18:50 +0800, Wen Gong wrote:
>> > > >
>> > > it is memset here i this patch:
>> >
>> > Oops, missed that.
>> >
>> > But is that really a good place for it? Doesn't really seem to belong
>> > to
>> > assigning a channel context - maybe put it into set_disassoc()?
>> >
>> it is correct.
>> you can see it is place together with "drv_unassign_vif_chanctx(local,
>> sdata, curr_ctx)"
>> in ieee80211_assign_vif_chanctx(), it is for disconnect.
> 
> Yes, I know it's *correct*, but that doesn't mean it's *good*?
> 
> Look at that code - it does nothing with bss_conf. Nobody is ever going
> to look there.
> 
Yes,
seems it is better do memset() in ieee80211_set_disassoc().
I will change it in next verion.
> johannes
>> 

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

Hi Johannes,
On 8/26/2021 6:57 PM, Johannes Berg wrote:
> On Thu, 2021-08-26 at 18:50 +0800, Wen Gong wrote:
>> it is memset here i this patch:
> Oops, missed that.
>
> But is that really a good place for it? Doesn't really seem to belong to
> assigning a channel context - maybe put it into set_disassoc()?
>
> johannes
We hit buffer overflow issue while connecting to 6 GHz AP fail and fail.
Will you fix it? Or do you have any suggestion to fix it?

[  227.539928] wlp90s0: authenticate with 02:03:7f:12:66:66
[  227.601846] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  227.633902] wlp90s0: authenticate with 02:03:7f:12:66:66
[  227.633906] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  227.657203] wlp90s0: 02:03:7f:12:66:66 denied authentication (status 1)
...
[  263.014661] wlp90s0: authenticate with 02:03:7f:12:66:66
[  263.075667] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  263.112427] wlp90s0: authenticate with 02:03:7f:12:66:66
[  263.112433] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  263.132507] wlp90s0: 02:03:7f:12:66:66 denied authentication (status 1)
[  279.668551] wlp90s0: authenticate with 02:03:7f:12:66:66
[  279.728848] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  279.763685] wlp90s0: authenticate with 02:03:7f:12:66:66
[  279.763696] wlp90s0: send auth to 02:03:7f:12:66:66 (try 1/3)
[  279.790867] wlp90s0: 02:03:7f:12:66:66 denied authentication (status 1)

for above fail, ieee80211_set_disassoc() is not called, so the
bss_conf->tx_pwr_env_num never reset and it is increased each
time in ieee80211_prep_channel().

Finally the bss_conf->tx_pwr_env_num arrived to 20+(it should be 1 for 
correct
value), it has exceeded the max value IEEE80211_TPE_MAX_IE_COUNT(8), and
lead access bss_conf->tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT] overflow.

[  327.391621] wlp90s0: authenticate with 02:03:7f:12:66:66
[  327.434036] BUG: kernel NULL pointer dereference, address: 
0000000000000018
[  327.434039] #PF: supervisor read access in kernel mode
[  327.434040] #PF: error_code(0x0000) - not-present page
[  327.434042] PGD 0 P4D 0
[  327.434044] Oops: 0000 [#1] PREEMPT SMP NOPTI
[  327.434047] CPU: 2 PID: 804 Comm: wpa_supplicant Kdump: loaded 
Tainted: G        W          6.2.0-rc8-wt-ath+ #13
[  327.434050] Hardware name: Intel(R) Client Systems 
NUC11PHi7/NUC11PHBi7, BIOS PHTGL579.0063.2021.0707.1057 07/07/2021
[  327.434052] RIP: 0010:ath12k_mac_fill_reg_tpc_info+0x292/0x3b0 [ath12k]
[  327.434080] RSP: 0018:ffffb7330160f448 EFLAGS: 00010246
[  327.434081] RAX: 0000000000000000 RBX: 0000000000000006 RCX: 
00000000005a8f98
[  327.434082] RDX: ffff9c7de1cb7f00 RSI: 00000000006cdf18 RDI: 
ffff9c7de2000508
[  327.434084] RBP: ffffb7330160f4b8 R08: 0000000000000000 R09: 
ffff9c7de1cb7f00
[  327.434084] R10: ffff9c7de2000508 R11: ffffb7330160f0e0 R12: 
ffff9c7dda376090
[  327.434085] R13: 0000000000000001 R14: 0000000000000010 R15: 
ffff9c7de2002080
[  327.434086] FS:  00007f8258ee9c00(0000) GS:ffff9c8170480000(0000) 
knlGS:0000000000000000
[  327.434087] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  327.434088] CR2: 0000000000000018 CR3: 0000000108fec006 CR4: 
0000000000770ee0
[  327.434090] PKRU: 55555554
[  327.434090] Call Trace:
[  327.434091]  <TASK>
[  327.434094]  ath12k_mac_vdev_start_restart+0x6c8/0x7f0 [ath12k]
[  327.434104]  ? crypto_alloc_tfm_node+0x60/0x130
[  327.434108]  ath12k_mac_station_add+0x163/0x440 [ath12k]
[  327.434117]  ath12k_mac_handle_link_sta_state.isra.79+0x76/0x420 [ath12k]
[  327.434126]  ath12k_mac_op_sta_state+0x19f/0x350 [ath12k]
[  327.434136]  drv_sta_state+0x89/0x5d0 [mac80211]
[  327.434160]  sta_info_insert_rcu+0x222/0x580 [mac80211]
[  327.434176]  sta_info_insert+0xf/0x20 [mac80211]
[  327.434190]  ieee80211_prep_connection+0x200/0x490 [mac80211]
[  327.434214]  ieee80211_mgd_auth+0x2aa/0x4f0 [mac80211]
[  327.434237]  ? __local_bh_enable_ip+0x3b/0x80
[  327.434239]  ? _raw_spin_unlock_bh+0x1d/0x30
[  327.434243]  ieee80211_auth+0x18/0x20 [mac80211]
[  327.434263]  cfg80211_mlme_auth+0x94/0x180 [cfg80211]
[  327.434297]  nl80211_authenticate+0x392/0x3f0 [cfg80211]
[  327.434315]  genl_family_rcv_msg_doit.isra.19+0xf4/0x120
[  327.434318]  genl_rcv_msg+0x1a5/0x2a0
[  327.434320]  ? __cfg80211_rdev_from_attrs+0x1f0/0x1f0 [cfg80211]
[  327.434336]  ? cfg80211_prepare_cqm.isra.79+0x170/0x170 [cfg80211]
[  327.434352]  ? nl80211_put_signal.part.56+0xd0/0xd0 [cfg80211]
[  327.434368]  ? genl_get_cmd_both+0x60/0x60
[  327.434370]  netlink_rcv_skb+0x5a/0x110
[  327.434372]  genl_rcv+0x28/0x40
[  327.434374]  netlink_unicast+0x1be/0x290
[  327.434375]  netlink_sendmsg+0x377/0x4e0
[  327.434377]  sock_sendmsg+0x9a/0xa0
[  327.434380]  ____sys_sendmsg+0x22b/0x2f0
[  327.434382]  ___sys_sendmsg+0x88/0xd0
[  327.434384]  ? dput+0x5f/0x2e0
[  327.434386]  ? __fsnotify_parent+0x109/0x350
[  327.434389]  __sys_sendmsg+0x6c/0xc0
[  327.434391]  ? __sys_sendmsg+0x6c/0xc0
[  327.434393]  __x64_sys_sendmsg+0x1f/0x30
[  327.434395]  do_syscall_64+0x37/0x90
[  327.434398]  entry_SYSCALL_64_after_hwframe+0x63/0xcd
[  327.434400] RIP: 0033:0x7f8258927b17
>
>
>
>

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-26 16:29, Johannes Berg wrote:
...
> 
>> +		if (is_6ghz) {
>> +			struct ieee802_11_elems elems;
> 
> This is pretty big, not sure we want it on the stack (causes warnings
> for me in build). Also, if we're doing this anyway, then we can change
> the code above (perhaps as a separate patch) to not do
> cfg80211_find_ext_ie() but rather take it out of the parsed.
do you mean NOT use cfg80211_find_ext_ie()/cfg80211_find_ie() and still 
use "struct ieee802_11_elems elems" here and
move this code to a separate function/patch?
it has more than one tx_pwr_env in one beacon, if we use 
cfg80211_find_ext_ie()/cfg80211_find_ie(),
it need add more logic.
> 
...
> 
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-27 at 10:11 +0800, Wen Gong wrote:
> > Also, if we're doing this anyway, then we can change
> > the code above (perhaps as a separate patch) to not do
> > cfg80211_find_ext_ie() but rather take it out of the parsed.
> do you mean NOT use cfg80211_find_ext_ie()/cfg80211_find_ie() and still 
> use "struct ieee802_11_elems elems" here and
> move this code to a separate function/patch?

Well, there's an existing place in this function that uses
cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
it feels like if we're going to do the full parsing, we should switch
all the existing "look up an element" to also use the parsed data
instead.

Not the other way around :)

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-27 14:46, Johannes Berg wrote:
> On Fri, 2021-08-27 at 10:11 +0800, Wen Gong wrote:
>> > Also, if we're doing this anyway, then we can change
>> > the code above (perhaps as a separate patch) to not do
>> > cfg80211_find_ext_ie() but rather take it out of the parsed.
>> do you mean NOT use cfg80211_find_ext_ie()/cfg80211_find_ie() and 
>> still
>> use "struct ieee802_11_elems elems" here and
>> move this code to a separate function/patch?
> 
> Well, there's an existing place in this function that uses
> cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
> it feels like if we're going to do the full parsing, we should switch
> all the existing "look up an element" to also use the parsed data
> instead.
> 
> Not the other way around :)
ok.
so it ha 2 way to change, right?
1.
change ieee802_11_parse_elems() to ieee80211_bss_get_ie()

2.
still use ieee802_11_parse_elems(), and change others 
ieee80211_bss_get_ie()/cfg80211_find_ext_ie()
to use the result of ieee802_11_parse_elems()
> 
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-27 at 14:53 +0800, Wen Gong wrote:
> > 
> > Well, there's an existing place in this function that uses
> > cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
> > it feels like if we're going to do the full parsing, we should switch
> > all the existing "look up an element" to also use the parsed data
> > instead.

> ok.
> so it ha 2 way to change, right?
> 1.
> change ieee802_11_parse_elems() to ieee80211_bss_get_ie()

No why?

I think we should make a first patch (that doesn't add TPE yet) that
changes the function to ieee80211_parse_elems() and removes all the
ieee80211_bss_get_ie() / cfg80211_find_ext_ie() calls in favour of just
parsing once, and then looking at the elements there.

Then your TPE patch becomes trivial since the elems are already there?

> 
> 2.
> still use ieee802_11_parse_elems(), and change others 
> ieee80211_bss_get_ie()/cfg80211_find_ext_ie()
> to use the result of ieee802_11_parse_elems()
> 

Right!

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-27 14:55, Johannes Berg wrote:
> On Fri, 2021-08-27 at 14:53 +0800, Wen Gong wrote:
>> >
>> > Well, there's an existing place in this function that uses
>> > cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
>> > it feels like if we're going to do the full parsing, we should switch
>> > all the existing "look up an element" to also use the parsed data
>> > instead.
> 
>> ok.
>> so it ha 2 way to change, right?
>> 1.
>> change ieee802_11_parse_elems() to ieee80211_bss_get_ie()
> 
> No why?
> 
> I think we should make a first patch (that doesn't add TPE yet) that
> changes the function to ieee80211_parse_elems() and removes all the
> ieee80211_bss_get_ie() / cfg80211_find_ext_ie() calls in favour of just
> parsing once, and then looking at the elements there.
> 
> Then your TPE patch becomes trivial since the elems are already there?
this patch still needed, because the lower driver need the info.
and this patch is save the info to "struct ieee80211_bss_conf *bss_conf" 
and
pass it to lower driver.

> 
>> 
>> 2.
>> still use ieee802_11_parse_elems(), and change others
>> ieee80211_bss_get_ie()/cfg80211_find_ext_ie()
>> to use the result of ieee802_11_parse_elems()
>> 
> 
> Right!
> 
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-27 at 15:12 +0800, Wen Gong wrote:
> On 2021-08-27 14:55, Johannes Berg wrote:
> > On Fri, 2021-08-27 at 14:53 +0800, Wen Gong wrote:
> > > > 
> > > > Well, there's an existing place in this function that uses
> > > > cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
> > > > it feels like if we're going to do the full parsing, we should switch
> > > > all the existing "look up an element" to also use the parsed data
> > > > instead.
> > 
> > > ok.
> > > so it ha 2 way to change, right?
> > > 1.
> > > change ieee802_11_parse_elems() to ieee80211_bss_get_ie()
> > 
> > No why?
> > 
> > I think we should make a first patch (that doesn't add TPE yet) that
> > changes the function to ieee80211_parse_elems() and removes all the
> > ieee80211_bss_get_ie() / cfg80211_find_ext_ie() calls in favour of just
> > parsing once, and then looking at the elements there.
> > 
> > Then your TPE patch becomes trivial since the elems are already there?
> this patch still needed, because the lower driver need the info.
> and this patch is save the info to "struct ieee80211_bss_conf *bss_conf" 
> and
> pass it to lower driver.

Of course, but you don't have to deal with parsing etc. in that patch
then.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-27 15:38, Johannes Berg wrote:
> On Fri, 2021-08-27 at 15:12 +0800, Wen Gong wrote:
>> On 2021-08-27 14:55, Johannes Berg wrote:
>> > On Fri, 2021-08-27 at 14:53 +0800, Wen Gong wrote:
>> > > >
>> > > > Well, there's an existing place in this function that uses
>> > > > cfg80211_find_ext_ie(), and various uses of ieee80211_bss_get_ie(), so
>> > > > it feels like if we're going to do the full parsing, we should switch
>> > > > all the existing "look up an element" to also use the parsed data
>> > > > instead.
>> >
>> > > ok.
>> > > so it ha 2 way to change, right?
>> > > 1.
>> > > change ieee802_11_parse_elems() to ieee80211_bss_get_ie()
>> >
>> > No why?
>> >
>> > I think we should make a first patch (that doesn't add TPE yet) that
>> > changes the function to ieee80211_parse_elems() and removes all the
>> > ieee80211_bss_get_ie() / cfg80211_find_ext_ie() calls in favour of just
>> > parsing once, and then looking at the elements there.
>> >
>> > Then your TPE patch becomes trivial since the elems are already there?
>> this patch still needed, because the lower driver need the info.
>> and this patch is save the info to "struct ieee80211_bss_conf 
>> *bss_conf"
>> and
>> pass it to lower driver.
> 
> Of course, but you don't have to deal with parsing etc. in that patch
> then.
> 
yes.
then should I use "struct ieee802_11_elems elems = {0}" or "struct 
ieee802_11_elems *elems = kzalloc(sizeof(*elems))"
in the parsing patch?

> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-27 at 16:18 +0800, Wen Gong wrote:
> 
> then should I use "struct ieee802_11_elems elems = {0}" or "struct 
> ieee802_11_elems *elems = kzalloc(sizeof(*elems))"
> in the parsing patch?

Yeah, it's a good question ...

We keep adding stuff here, so it'll be safer to alloc it.

I get a fair number of stack size warnings on the build now (possibly
due to the addition of the TPE fields?), and while they're probably fine
for now (we get there from nl80211, so no deep stack), it's only going
to increase - we have EHT patches already now, for example.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-27 16:20, Johannes Berg wrote:
> On Fri, 2021-08-27 at 16:18 +0800, Wen Gong wrote:
>> 
>> then should I use "struct ieee802_11_elems elems = {0}" or "struct
>> ieee802_11_elems *elems = kzalloc(sizeof(*elems))"
>> in the parsing patch?
> 
> Yeah, it's a good question ...
> 
> We keep adding stuff here, so it'll be safer to alloc it.
> 
> I get a fair number of stack size warnings on the build now (possibly
> due to the addition of the TPE fields?), and while they're probably 
> fine
> for now (we get there from nl80211, so no deep stack), it's only going
> to increase - we have EHT patches already now, for example.
> 

the TPE is only 8 pointer in the struct ieee802_11_elems.
so stack size warnings should not caused by TPE.

#define IEEE80211_TPE_MAX_IE_COUNT	8
const struct ieee80211_tx_pwr_env 
*tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT];

> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Johannes Berg]

On Fri, 2021-08-27 at 16:28 +0800, Wen Gong wrote:
> 
> the TPE is only 8 pointer in the struct ieee802_11_elems.


I know, but it's showing _just_ above the threshold for the warning now,
so if it was _just_ below the threshold before, adding 64 bytes could
still do that. 64 bytes is a lot, after all, the threshold is only 512 I
think.

johannes


-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 8/8] mac80211: save transmit power envelope element and power constraint

[Wen Gong]

On 2021-08-27 16:30, Johannes Berg wrote:
> On Fri, 2021-08-27 at 16:28 +0800, Wen Gong wrote:
>> 
>> the TPE is only 8 pointer in the struct ieee802_11_elems.
> 
> 
> I know, but it's showing _just_ above the threshold for the warning 
> now,
> so if it was _just_ below the threshold before, adding 64 bytes could
> still do that. 64 bytes is a lot, after all, the threshold is only 512 
> I
> think.
> 
Yes, I will change to below in next version.
"struct ieee802_11_elems *elems = kzalloc(sizeof(*elems))"
> johannes

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

Re: [PATCH v2 0/8] cfg80211/mac80211: Add support for 6GHZ STA for various modes : LPI, SP and VLP

[Wen Gong]

Hi johannes,

Could I get your comments on this v2 patches?

On 2021-08-20 20:20, Wen Gong wrote:
> v2: change per comments of johannes.
>     including code style, code logic, patch merge, commit log...
> 
> It introduced some new concept:
> power type of AP(STANDARD_POWER_AP, INDOOR_AP, VERY_LOW_POWER_AP)
> power type of STATION(DEFAULT_CLIENT, SUBORDINATE_CLIENT)
> power spectral density(psd)
> 
> This patchset for cfg80211/mac80211 is to add the definition of new
> concept of 6G and add basic parse of IE(transmit power envelope
> element) in beacon and save power spectral density(psd) reported
> by lower-driver for 6G channel, the info will be passed to lower
> driver when connecting to 6G AP.
> 
> Wen Gong (8):
>   cfg80211: add power type definition for 6 GHz
>   mac80211: add definition of regulatory info in 6 GHz operation
>     information
>   mac80211: add parse regulatory info in 6 GHz operation information
>   cfg80211: add definition for 6 GHz power spectral density(psd)
>   cfg80211: save power spectral density(psd) of regulatory rule
>   mac80211: add definition for transmit power envelope element
>   mac80211: add parse transmit power envelope element
>   mac80211: save transmit power envelope element and power constraint
> 
>  include/linux/ieee80211.h    | 43 +++++++++++++++++++++++++++++++++++-
>  include/net/cfg80211.h       |  7 ++++++
>  include/net/mac80211.h       |  6 +++++
>  include/net/regulatory.h     |  1 +
>  include/uapi/linux/nl80211.h | 36 ++++++++++++++++++++++++++++++
>  net/mac80211/chan.c          |  9 ++++++++
>  net/mac80211/ieee80211_i.h   |  3 +++
>  net/mac80211/mlme.c          | 26 ++++++++++++++++++++++
>  net/mac80211/util.c          | 19 ++++++++++++++++
>  net/wireless/reg.c           | 14 ++++++++++++
>  10 files changed, 163 insertions(+), 1 deletion(-)

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k