Skip to contentSkip to conversation

rt2800: fix RT5592 TX power settings regression

PATCH In review Ronald Wahl 3 participants 6 messages Nov 5, 2014 – Nov 25, 2014 Networking drivers (wireless)

lore
Share & tools
lore

Ronald Wahl

author
#0 lore raw

PATCH 1/1 rt2800: fix RT5592 TX power settings regression

Commit cee2c7315f60beeff6137ee59e99acc77d636eeb (rt2800: fix RT5390 & RT3290 TX power settings regression) needs to be extended for the RT5592 chipset as well. But at least for the RT5592 the existing regression fix is not right because the value returned from rt2800_get_gain_calibration_delta() is bogus as it is generated by an unappropriate algorithm. This can cause severe connection issues with sticks that have external ALC enabled like the Netis WF2150 because of too low TX power at least during the scan process.

So the fix for now is not to call rt2800_get_gain_calibration_delta() for the RT5592 chipset. I do not touch the existing regression fix for RT5390 & RT3290 but I think they may need a rework as well.

Signed-off-by Ronald Wahl <[email protected]>
---
 drivers/net/wireless/rt2x00/rt2800lib.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/wireless/rt2x00/rt2800lib.c +5 -1
index 9f57a2d..66f3546 100644
@@ -4118,8 +4118,12 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * to temperature or maybe other factors) is smaller or bigger than
 	 * expected. We adjust it, based on TSSI reference and boundaries values
 	 * provided in EEPROM.
+	 *
+	 * TODO: add different temperature compensation code for RT5592
 	 */
-	delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	if (!rt2x00_rt(rt2x00dev, RT5592)) {
+	    delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	}
 
 	/*
 	 * Decrease power according to user settings, on devices with unknown
Signature · 1 line
1.9.3
#1 lore raw
Quoted 11 lines · Ronald Wahl

Commit cee2c7315f60beeff6137ee59e99acc77d636eeb (rt2800: fix RT5390 & RT3290 TX power settings regression) needs to be extended for the RT5592 chipset as well. But at least for the RT5592 the existing regression fix is not right because the value returned from rt2800_get_gain_calibration_delta() is bogus as it is generated by an unappropriate algorithm. This can cause severe connection issues with sticks that have external ALC enabled like the Netis WF2150 because of too low TX power at least during the scan process.

So the fix for now is not to call rt2800_get_gain_calibration_delta() for the RT5592 chipset. I do not touch the existing regression fix for RT5390 & RT3290 but I think they may need a rework as well.

Thanks for the patch, but I prefer to call rt2800_get_gain_calibration_delta() on chips that we know it is needed for them, something like in attached patch

Mike, since you are cee2c7315f reporter, could you test the attached patch does not break driver functioning on your H/W.

Thanks
Stanislaw

drivers/net/wireless/rt2x00/rt2800lib.c +26 -19
index 9f57a2d..81ee481 100644
@@ -4119,7 +4119,20 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * expected. We adjust it, based on TSSI reference and boundaries values
 	 * provided in EEPROM.
 	 */
-	delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	switch (rt2x00dev->chip.rt) {
+	case RT2860:
+	case RT2872:
+	case RT2883:
+	case RT3070:
+	case RT3071:
+	case RT3090:
+	case RT3572:
+		delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+		break;
+	default:
+		/* TODO: temperature compensation code for other chips. */
+		break;
+	}
 
 	/*
 	 * Decrease power according to user settings, on devices with unknown

@@ -4136,25 +4149,19 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * TODO: we do not use +6 dBm option to do not increase power beyond
 	 * regulatory limit, however this could be utilized for devices with
 	 * CAPABILITY_POWER_LIMIT.
-	 *
-	 * TODO: add different temperature compensation code for RT3290 & RT5390
-	 * to allow to use BBP_R1 for those chips.
-	 */
-	if (!rt2x00_rt(rt2x00dev, RT3290) &&
-	    !rt2x00_rt(rt2x00dev, RT5390)) {
-		rt2800_bbp_read(rt2x00dev, 1, &r1);
-		if (delta <= -12) {
-			power_ctrl = 2;
-			delta += 12;
-		} else if (delta <= -6) {
-			power_ctrl = 1;
-			delta += 6;
-		} else {
-			power_ctrl = 0;
-		}
-		rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
-		rt2800_bbp_write(rt2x00dev, 1, r1);
+	 */
+	if (delta <= -12) {
+		power_ctrl = 2;
+		delta += 12;
+	} else if (delta <= -6) {
+		power_ctrl = 1;
+		delta += 6;
+	} else {
+		power_ctrl = 0;
 	}
+	rt2800_bbp_read(rt2x00dev, 1, &r1);
+	rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
+	rt2800_bbp_write(rt2x00dev, 1, r1);
 
 	offset = TX_PWR_CFG_0;
#2 lore raw

Any comments on this patch? Ron, does it work for you?

Quoted 94 lines · Stanislaw Gruszka
Quoted 11 lines · Ronald Wahl

Commit cee2c7315f60beeff6137ee59e99acc77d636eeb (rt2800: fix RT5390 & RT3290 TX power settings regression) needs to be extended for the RT5592 chipset as well. But at least for the RT5592 the existing regression fix is not right because the value returned from rt2800_get_gain_calibration_delta() is bogus as it is generated by an unappropriate algorithm. This can cause severe connection issues with sticks that have external ALC enabled like the Netis WF2150 because of too low TX power at least during the scan process.

So the fix for now is not to call rt2800_get_gain_calibration_delta() for the RT5592 chipset. I do not touch the existing regression fix for RT5390 & RT3290 but I think they may need a rework as well.

Thanks for the patch, but I prefer to call rt2800_get_gain_calibration_delta() on chips that we know it is needed for them, something like in attached patch

Mike, since you are cee2c7315f reporter, could you test the attached patch does not break driver functioning on your H/W.

Thanks
Stanislaw

drivers/net/wireless/rt2x00/rt2800lib.c +26 -19
index 9f57a2d..81ee481 100644
@@ -4119,7 +4119,20 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * expected. We adjust it, based on TSSI reference and boundaries values
 	 * provided in EEPROM.
 	 */
-	delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	switch (rt2x00dev->chip.rt) {
+	case RT2860:
+	case RT2872:
+	case RT2883:
+	case RT3070:
+	case RT3071:
+	case RT3090:
+	case RT3572:
+		delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+		break;
+	default:
+		/* TODO: temperature compensation code for other chips. */
+		break;
+	}
 
 	/*
 	 * Decrease power according to user settings, on devices with unknown

@@ -4136,25 +4149,19 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * TODO: we do not use +6 dBm option to do not increase power beyond
 	 * regulatory limit, however this could be utilized for devices with
 	 * CAPABILITY_POWER_LIMIT.
-	 *
-	 * TODO: add different temperature compensation code for RT3290 & RT5390
-	 * to allow to use BBP_R1 for those chips.
-	 */
-	if (!rt2x00_rt(rt2x00dev, RT3290) &&
-	    !rt2x00_rt(rt2x00dev, RT5390)) {
-		rt2800_bbp_read(rt2x00dev, 1, &r1);
-		if (delta <= -12) {
-			power_ctrl = 2;
-			delta += 12;
-		} else if (delta <= -6) {
-			power_ctrl = 1;
-			delta += 6;
-		} else {
-			power_ctrl = 0;
-		}
-		rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
-		rt2800_bbp_write(rt2x00dev, 1, r1);
+	 */
+	if (delta <= -12) {
+		power_ctrl = 2;
+		delta += 12;
+	} else if (delta <= -6) {
+		power_ctrl = 1;
+		delta += 6;
+	} else {
+		power_ctrl = 0;
 	}
+	rt2800_bbp_read(rt2x00dev, 1, &r1);
+	rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
+	rt2800_bbp_write(rt2x00dev, 1, r1);
 
 	offset = TX_PWR_CFG_0;

_______________________________________________
users mailing list
[email protected]
http://rt2x00.serialmonkey.com/mailman/listinfo/users_rt2x00.serialmonkey.com

Signature · 2 lines
John W. Linville		Someday the world will need a hero, and you
[email protected]			might be all we have.  Be ready.
author
#3 lore raw

For me the patch will work but there is a change for RT3290 and RT5390. On both chipsets BBP 1 was not written before but with this patch it is going to be written now and the value depends on the return of rt2800_get_txpower_bw_comp(). It should be checked that the function does the correct things for these chips.

- ron

Quoted 99 lines · John W. Linville

Any comments on this patch? Ron, does it work for you?

Quoted 94 lines · Stanislaw Gruszka
Quoted 11 lines · Ronald Wahl

Commit cee2c7315f60beeff6137ee59e99acc77d636eeb (rt2800: fix RT5390 & RT3290 TX power settings regression) needs to be extended for the RT5592 chipset as well. But at least for the RT5592 the existing regression fix is not right because the value returned from rt2800_get_gain_calibration_delta() is bogus as it is generated by an unappropriate algorithm. This can cause severe connection issues with sticks that have external ALC enabled like the Netis WF2150 because of too low TX power at least during the scan process.

So the fix for now is not to call rt2800_get_gain_calibration_delta() for the RT5592 chipset. I do not touch the existing regression fix for RT5390 & RT3290 but I think they may need a rework as well.

Thanks for the patch, but I prefer to call rt2800_get_gain_calibration_delta() on chips that we know it is needed for them, something like in attached patch

Mike, since you are cee2c7315f reporter, could you test the attached patch does not break driver functioning on your H/W.

Thanks
Stanislaw

drivers/net/wireless/rt2x00/rt2800lib.c +26 -19
index 9f57a2d..81ee481 100644
@@ -4119,7 +4119,20 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
  	 * expected. We adjust it, based on TSSI reference and boundaries values
  	 * provided in EEPROM.
  	 */
-	delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	switch (rt2x00dev->chip.rt) {
+	case RT2860:
+	case RT2872:
+	case RT2883:
+	case RT3070:
+	case RT3071:
+	case RT3090:
+	case RT3572:
+		delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+		break;
+	default:
+		/* TODO: temperature compensation code for other chips. */
+		break;
+	}
 
  	/*
  	 * Decrease power according to user settings, on devices with unknown

@@ -4136,25 +4149,19 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
  	 * TODO: we do not use +6 dBm option to do not increase power beyond
  	 * regulatory limit, however this could be utilized for devices with
  	 * CAPABILITY_POWER_LIMIT.
-	 *
-	 * TODO: add different temperature compensation code for RT3290 & RT5390
-	 * to allow to use BBP_R1 for those chips.
-	 */
-	if (!rt2x00_rt(rt2x00dev, RT3290) &&
-	    !rt2x00_rt(rt2x00dev, RT5390)) {
-		rt2800_bbp_read(rt2x00dev, 1, &r1);
-		if (delta <= -12) {
-			power_ctrl = 2;
-			delta += 12;
-		} else if (delta <= -6) {
-			power_ctrl = 1;
-			delta += 6;
-		} else {
-			power_ctrl = 0;
-		}
-		rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
-		rt2800_bbp_write(rt2x00dev, 1, r1);
+	 */
+	if (delta <= -12) {
+		power_ctrl = 2;
+		delta += 12;
+	} else if (delta <= -6) {
+		power_ctrl = 1;
+		delta += 6;
+	} else {
+		power_ctrl = 0;
  	}
+	rt2800_bbp_read(rt2x00dev, 1, &r1);
+	rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
+	rt2800_bbp_write(rt2x00dev, 1, r1);
 
  	offset = TX_PWR_CFG_0;

_______________________________________________
users mailing list
[email protected]
http://rt2x00.serialmonkey.com/mailman/listinfo/users_rt2x00.serialmonkey.com

Signature · 5 lines
Ronald Wahl - [email protected] - Phone +49 375271349-0 Fax -99
Raritan Deutschland GmbH, Kornmarkt 7, 08056 Zwickau, Germany
USt-IdNr. DE813094160, Steuer-Nr. 227/117/01749
Amtsgericht Chemnitz HRB 23605
Geschäftsführung: Stuart Hopper, Ralf Ploenes
#4 lore raw
Quoted 5 lines · Ronald Wahl

For me the patch will work but there is a change for RT3290 and RT5390. On both chipsets BBP 1 was not written before but with this patch it is going to be written now and the value depends on the return of rt2800_get_txpower_bw_comp(). It should be checked that the function does the correct things for these chips.

I tested the patch on RT5390, it does not make troubles. Going to post it ...

Stanislaw

#5 lore raw

PATCH 1/1 rt2800: calculate tx power temperature compensation on selected chips

Currently implemented temperature compensation is only valid on some of supported chips. Other chips do not need temperature compensation or need different way to do this (not yet implemented in the rt2800 driver). Trying to do run rt2800_get_gain_calibration_delta() when this is not appropriate on particular chip gives bogus result of TX power and can make connection unstable.

This is follow up to commit 8c8d2017ba25c510ddf093419048460db1109bc4 "rt2800: fix RT5390 & RT3290 TX power settings regression". On that commit we avoid setting BBP_R1 register, but the real problem is wrong temperature compensation calculation.

Reported-and-tested-by Ronald Wahl <[email protected]>
Debugged-by Ronald Wahl <[email protected]>
Cc Mike Romberg <[email protected]>
Signed-off-by Stanislaw Gruszka <[email protected]>
---
 drivers/net/wireless/rt2x00/rt2800lib.c | 45 +++++++++++++++++++--------------
 1 file changed, 26 insertions(+), 19 deletions(-)
drivers/net/wireless/rt2x00/rt2800lib.c +26 -19
index 9f57a2d..81ee481 100644
@@ -4119,7 +4119,20 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * expected. We adjust it, based on TSSI reference and boundaries values
 	 * provided in EEPROM.
 	 */
-	delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+	switch (rt2x00dev->chip.rt) {
+	case RT2860:
+	case RT2872:
+	case RT2883:
+	case RT3070:
+	case RT3071:
+	case RT3090:
+	case RT3572:
+		delta += rt2800_get_gain_calibration_delta(rt2x00dev);
+		break;
+	default:
+		/* TODO: temperature compensation code for other chips. */
+		break;
+	}
 
 	/*
 	 * Decrease power according to user settings, on devices with unknown

@@ -4136,25 +4149,19 @@ static void rt2800_config_txpower_rt28xx(struct rt2x00_dev *rt2x00dev,
 	 * TODO: we do not use +6 dBm option to do not increase power beyond
 	 * regulatory limit, however this could be utilized for devices with
 	 * CAPABILITY_POWER_LIMIT.
-	 *
-	 * TODO: add different temperature compensation code for RT3290 & RT5390
-	 * to allow to use BBP_R1 for those chips.
-	 */
-	if (!rt2x00_rt(rt2x00dev, RT3290) &&
-	    !rt2x00_rt(rt2x00dev, RT5390)) {
-		rt2800_bbp_read(rt2x00dev, 1, &r1);
-		if (delta <= -12) {
-			power_ctrl = 2;
-			delta += 12;
-		} else if (delta <= -6) {
-			power_ctrl = 1;
-			delta += 6;
-		} else {
-			power_ctrl = 0;
-		}
-		rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
-		rt2800_bbp_write(rt2x00dev, 1, r1);
+	 */
+	if (delta <= -12) {
+		power_ctrl = 2;
+		delta += 12;
+	} else if (delta <= -6) {
+		power_ctrl = 1;
+		delta += 6;
+	} else {
+		power_ctrl = 0;
 	}
+	rt2800_bbp_read(rt2x00dev, 1, &r1);
+	rt2x00_set_field8(&r1, BBP1_TX_POWER_CTRL, power_ctrl);
+	rt2800_bbp_write(rt2x00dev, 1, r1);
 
 	offset = TX_PWR_CFG_0;
Signature · 1 line
1.8.3.1