[PATCH] minstrel_ht: fix freeze with legacy hardware

[PATCH] minstrel_ht: fix freeze with legacy hardware

[Christian Lamparter]

Ever since minstrel_ht was prompted to be the default RC,
I had an this odd freeze in ieee80211_tx_status, 
whenever I tried to connect my p54*(legacy 11abg) to
a 802.11n network.

Apparently, minstrel_ht was not interested if the hardware
supported 802.11n or not?

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
---
Note: There's another possible way to fix this.
We can simply AND sta's ht_supported bool with
the setting in sband's ht_supported.

The advantage is that we won't need this patch
and the procedure would "match" to what we did
with the ht_cap flags.

This is up for debate... But in the meantime,
here's a fix.

P.S.: is there a valid operation situation
where sta is be NULL, but priv_sta isn't?
---
diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c
index b5ace24..8ad3630 100644
--- a/net/mac80211/rc80211_minstrel_ht.c
+++ b/net/mac80211/rc80211_minstrel_ht.c
@@ -635,8 +635,12 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband,
 	int stbc;
 	int i;
 
-	/* fall back to the old minstrel for legacy stations */
-	if (sta && !sta->ht_cap.ht_supported) {
+	/*
+	 * fall back to the old minstrel for legacy stations,
+	 * or on legacy hardware.
+	 */
+	if ((sta && !sta->ht_cap.ht_supported) ||
+	    !sband->ht_cap.ht_supported) {
 		msp->is_ht = false;
 		memset(&msp->legacy, 0, sizeof(msp->legacy));
 		msp->legacy.r = msp->ratelist;

Re: [PATCH] minstrel_ht: fix freeze with legacy hardware

[Felix Fietkau]

On 2010-07-14 11:43 PM, Christian Lamparter wrote:
> Ever since minstrel_ht was prompted to be the default RC,
> I had an this odd freeze in ieee80211_tx_status, 
> whenever I tried to connect my p54*(legacy 11abg) to
> a 802.11n network.
> 
> Apparently, minstrel_ht was not interested if the hardware
> supported 802.11n or not?
> 
> Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
> ---
> Note: There's another possible way to fix this.
> We can simply AND sta's ht_supported bool with
> the setting in sband's ht_supported.
> 
> The advantage is that we won't need this patch
> and the procedure would "match" to what we did
> with the ht_cap flags.
> 
> This is up for debate... But in the meantime,
> here's a fix.
> 
> P.S.: is there a valid operation situation
> where sta is be NULL, but priv_sta isn't?
Either of those two options is OK with me. I would prefer masking it in
mac80211 though, since the sta ht setting are already partially masked
by hw capabilities.

- Felix

[PATCH] mac80211: skip HT parsing if HW does not support HT

[Christian Lamparter]

This patch will also fix the odd freeze which occurred
when minstrel_ht connects to an 802.11n network with
legacy hardware.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
---
John,

As you know Felix has favored the alternative.
Therefore, can you please ignore
[PATCH] minstrel_ht: fix freeze with legacy hardware
and use this one instead?

Regards,
	Chr
---
diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c
index be928ef..9d101fb 100644
--- a/net/mac80211/ht.c
+++ b/net/mac80211/ht.c
@@ -29,7 +29,7 @@ void ieee80211_ht_cap_ie_to_sta_ht_cap(struct ieee80211_supported_band *sband,
 
 	memset(ht_cap, 0, sizeof(*ht_cap));
 
-	if (!ht_cap_ie)
+	if (!ht_cap_ie || !sband->ht_cap.ht_supported)
 		return;
 
 	ht_cap->ht_supported = true;