Skip to contentSkip to conversation

iwlagn: use cts-to-self protection on 5000 adapters series

PATCH Superseded Stanislaw Gruszka 2 participants 4 messages May 25, 2011 – May 26, 2011 Networking drivers (wireless)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iwlagn: use cts-to-self protection on 5000 adapters series

This patch fixes 802.11n stability and performance regression we have since 2.6.35. It boost performance on my 5GHz N-only network from about 5MB/s to 8MB/s. Similar percentage boost can be observed on 2.4 GHz.

These are test results of 5x downloading of approximately 700MB iso image:

vanilla: 5.27 5.22 4.94 4.47 5.31 ; avr 5.0420 std 0.35110
patched: 8.07 7.95 8.06 7.99 7.96 ; avr 8.0060 std 0.055946

This was achieved with NetworkManager configured to do not perform periodical scans, by configuring constant BSSID. With periodical scans, after some time, performance downgrade to unpached driver level, like in example below:

patched: 7.40 7.61 4.28 4.37 4.80 avr 5.6920 std 1.6683

However patch still make better here, since similar test on unpached driver make link disconnects with below messages after some time:

wlan1: authenticate with 00:23:69:35:d1:3f (try 1)
wlan1: authenticate with 00:23:69:35:d1:3f (try 2)
wlan1: authenticate with 00:23:69:35:d1:3f (try 3)
wlan1: authentication with 00:23:69:35:d1:3f timed out

On 2.6.35 kernel patch helps against connection hangs with messages:

iwlagn 0000:20:00.0: queue 10 stuck 3 time. Fw reload. iwlagn 0000:20:00.0: On demand firmware reload
iwlagn 0000:20:00.0: Stopping AGG while state not ON or starting

---
 drivers/net/wireless/iwlwifi/iwl-5000.c     |    1 -
 drivers/net/wireless/iwlwifi/iwl-agn-hcmd.c |   12 ++----------
 drivers/net/wireless/iwlwifi/iwl-agn-rxon.c |    8 ++++++++
 3 files changed, 10 insertions(+), 11 deletions(-)
drivers/net/wireless/iwlwifi/iwl-5000.c +0 -1
index a70b8cf..f96eb5b 100644
@@ -425,7 +425,6 @@ static struct iwl_base_params iwl5000_base_params = {
 };
 static struct iwl_ht_params iwl5000_ht_params = {
 	.ht_greenfield_support = true,
-	.use_rts_for_aggregation = true, /* use rts/cts protection */
 };
 
 #define IWL_DEVICE_5000						\
drivers/net/wireless/iwlwifi/iwl-agn-hcmd.c +2 -10
index b12c72d..23fa93d 100644
@@ -163,17 +163,9 @@ static void iwlagn_tx_cmd_protection(struct iwl_priv *priv,
 				     __le16 fc, __le32 *tx_flags)
 {
 	if (info->control.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS ||
-	    info->control.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT) {
+	    info->control.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT ||
+	    info->flags & IEEE80211_TX_CTL_AMPDU)
 		*tx_flags |= TX_CMD_FLG_PROT_REQUIRE_MSK;
-		return;
-	}
-
-	if (priv->cfg->ht_params &&
-	    priv->cfg->ht_params->use_rts_for_aggregation &&
-	    info->flags & IEEE80211_TX_CTL_AMPDU) {
-		*tx_flags |= TX_CMD_FLG_PROT_REQUIRE_MSK;
-		return;
-	}
 }
 
 /* Calc max signal level (dBm) among 3 possible receivers */
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +8 -0
index a95ad84..0a1941d 100644
@@ -325,6 +325,14 @@ int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx)
 			return 0;
 	}
 
+	/*
+	 * force CTS-to-self frames protection if RTS-CTS is not preferred
+	 * one aggregation protection method
+	 */
+	if (!(priv->cfg->ht_params &&
+	      priv->cfg->ht_params->use_rts_for_aggregation))
+		ctx->staging.flags |= RXON_FLG_SELF_CTS_EN;
+
 	if ((ctx->vif && ctx->vif->bss_conf.use_short_slot) ||
 	    !(ctx->staging.flags & RXON_FLG_BAND_24G_MSK))
 		ctx->staging.flags |= RXON_FLG_SHORT_SLOT_MSK;
Signature · 1 line
1.7.1
#1 lore raw
Quoted 7 lines · Stanislaw Gruszka

This patch fixes 802.11n stability and performance regression we have since 2.6.35. It boost performance on my 5GHz N-only network from about 5MB/s to 8MB/s. Similar percentage boost can be observed on 2.4 GHz.

These are test results of 5x downloading of approximately 700MB iso image:

Quoted 6 lines

vanilla: 5.27 5.22 4.94 4.47 5.31 ; avr 5.0420 std 0.35110
patched: 8.07 7.95 8.06 7.99 7.96 ; avr 8.0060 std 0.055946

This was achieved with NetworkManager configured to do not perform periodical scans, by configuring constant BSSID. With periodical scans, after some time, performance downgrade to unpached driver level, like

unpatched?

Quoted 5 lines

in example below:

patched: 7.40 7.61 4.28 4.37 4.80 avr 5.6920 std 1.6683

However patch still make better here, since similar test on unpached

unpatched?

Quoted 62 lines

driver make link disconnects with below messages after some time:

wlan1: authenticate with 00:23:69:35:d1:3f (try 1)
wlan1: authenticate with 00:23:69:35:d1:3f (try 2)
wlan1: authenticate with 00:23:69:35:d1:3f (try 3)
wlan1: authentication with 00:23:69:35:d1:3f timed out

On 2.6.35 kernel patch helps against connection hangs with messages:

iwlagn 0000:20:00.0: queue 10 stuck 3 time. Fw reload. iwlagn 0000:20:00.0: On demand firmware reload
iwlagn 0000:20:00.0: Stopping AGG while state not ON or starting

---
 drivers/net/wireless/iwlwifi/iwl-5000.c     |    1 -
 drivers/net/wireless/iwlwifi/iwl-agn-hcmd.c |   12 ++----------
 drivers/net/wireless/iwlwifi/iwl-agn-rxon.c |    8 ++++++++
 3 files changed, 10 insertions(+), 11 deletions(-)
drivers/net/wireless/iwlwifi/iwl-5000.c +0 -1
index a70b8cf..f96eb5b 100644
@@ -425,7 +425,6 @@ static struct iwl_base_params iwl5000_base_params = {
 };
 static struct iwl_ht_params iwl5000_ht_params = {
 	.ht_greenfield_support = true,
-	.use_rts_for_aggregation = true, /* use rts/cts protection */
 };
 
 #define IWL_DEVICE_5000						\
drivers/net/wireless/iwlwifi/iwl-agn-hcmd.c +2 -10
index b12c72d..23fa93d 100644
@@ -163,17 +163,9 @@ static void iwlagn_tx_cmd_protection(struct iwl_priv *priv,
 				     __le16 fc, __le32 *tx_flags)
 {
 	if (info->control.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS ||
-	    info->control.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT) {
+	    info->control.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT ||
+	    info->flags & IEEE80211_TX_CTL_AMPDU)
 		*tx_flags |= TX_CMD_FLG_PROT_REQUIRE_MSK;
-		return;
-	}
-
-	if (priv->cfg->ht_params &&
-	    priv->cfg->ht_params->use_rts_for_aggregation &&
-	    info->flags & IEEE80211_TX_CTL_AMPDU) {
-		*tx_flags |= TX_CMD_FLG_PROT_REQUIRE_MSK;
-		return;
-	}
 }
 
 /* Calc max signal level (dBm) among 3 possible receivers */
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +0 -0
index a95ad84..0a1941d 100644

drivers/net/wireless/iwlwifi/iwl-agn-rxon.c int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx) · line 330

Quoted 6 lines
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +3 -0
@@ -325,6 +325,14 @@ int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx)
 			return 0;
 	}
 
+	/*
+	 * force CTS-to-self frames protection if RTS-CTS is not preferred
+	 * one aggregation protection method
             on?

drivers/net/wireless/iwlwifi/iwl-agn-rxon.c int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx) · line 338

Quoted 8 lines
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +5 -0
+	 */
+	if (!(priv->cfg->ht_params &&
+	      priv->cfg->ht_params->use_rts_for_aggregation))
+		ctx->staging.flags |= RXON_FLG_SELF_CTS_EN;
+
 	if ((ctx->vif && ctx->vif->bss_conf.use_short_slot) ||
 	    !(ctx->staging.flags & RXON_FLG_BAND_24G_MSK))
 		ctx->staging.flags |= RXON_FLG_SHORT_SLOT_MSK;

if I remember correctly, the RTS/CTS was added to fix 6000 performance issue 2 years ago. Did you see the similar performance downgrade issue on other devices beside 5000?

Thanks
Wey

#2 lore raw
Quoted 6 lines · wwguy
Quoted 1 line · Stanislaw Gruszka

after some time, performance downgrade to unpached driver level, like

unpatched?

Quoted 1 line

However patch still make better here, since similar test on unpached

unpatched?

I'll fix that.

Quoted 4 lines

drivers/net/wireless/iwlwifi/iwl-agn-rxon.c int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx) · line 330

Quoted 3 lines
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +3 -0
+	/*
+	 * force CTS-to-self frames protection if RTS-CTS is not preferred
+	 * one aggregation protection method
             on?

The sentence is not the best English, but replacing "one" to "on" will not help, I think.

Quoted 3 lines

if I remember correctly, the RTS/CTS was added to fix 6000 performance issue 2 years ago. Did you see the similar performance downgrade issue on other devices beside 5000?

No, I think only 5000 ware afected (also with random connection hangs). Patching and using cts-to-self on 6000, give slightly better performance here with 6300 adapter:

vanilla 6300 5Ghz: 8.13 8.03 8.13 8.03 8.19 avr 8.1020 std 0.070143 patched 6300 5Ghz: 8.60 8.59 8.60 8.41 8.57 avr 8.5540 std 0.081425

vanilla 6300 2GHz: 4.06 4.39 4.60 4.70 4.80 avr 4.5100 std 0.293770 patched 6300 2GHz: 4.94 4.90 4.94 5.05 4.88 avr 4.9420 std 0.065727

I plan to do more testing, but not yet (except 5000 adapters I have quite good performance).

Stanislaw

#3 lore raw
Quoted 26 lines · Stanislaw Gruszka
Quoted 6 lines · wwguy
Quoted 1 line · Stanislaw Gruszka

after some time, performance downgrade to unpached driver level, like

unpatched?

Quoted 1 line

However patch still make better here, since similar test on unpached

unpatched?

I'll fix that.

Quoted 4 lines

drivers/net/wireless/iwlwifi/iwl-agn-rxon.c int iwlagn_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx) · line 330

Quoted 3 lines
drivers/net/wireless/iwlwifi/iwl-agn-rxon.c +3 -0
+	/*
+	 * force CTS-to-self frames protection if RTS-CTS is not preferred
+	 * one aggregation protection method
             on?

The sentence is not the best English, but replacing "one" to "on" will not help, I think.

Quoted 3 lines

if I remember correctly, the RTS/CTS was added to fix 6000 performance issue 2 years ago. Did you see the similar performance downgrade issue on other devices beside 5000?

No, I think only 5000 ware afected (also with random connection hangs). Patching and using cts-to-self on 6000, give slightly better performance here with 6300 adapter:

Quoted 9 lines

vanilla 6300 5Ghz: 8.13 8.03 8.13 8.03 8.19 avr 8.1020 std 0.070143 patched 6300 5Ghz: 8.60 8.59 8.60 8.41 8.57 avr 8.5540 std 0.081425

vanilla 6300 2GHz: 4.06 4.39 4.60 4.70 4.80 avr 4.5100 std 0.293770 patched 6300 2GHz: 4.94 4.90 4.94 5.05 4.88 avr 4.9420 std 0.065727

I plan to do more testing, but not yet (except 5000 adapters I have quite good performance).

Thanks for testing. the patch looks good.

Wey