ath5k: Linear PCDAC code fixes

PATCH In review Nick Kossifidis 4 participants 5 messages Jul 31, 2009 – Aug 4, 2009 Atheros ATH5K wireless driver

Share & tools

Nick Kossifidis

author

2009-07-31 18:34 UTC #0 lore raw

PATCH 2/4 ath5k: Linear PCDAC code fixes

* Set correct xpd curve indices for high/low gain curves during rfbuffer setup on RF5112B with both calibration curves available.

* Don't return zero min power when we have the same pcdac value twice because it breaks interpolation. Instead return the right x barrier as we do when we have equal power levels for 2 different pcdac values.

---
 drivers/net/wireless/ath/ath5k/phy.c |   24 +++++++++++++++++-------
 1 files changed, 17 insertions(+), 7 deletions(-)

drivers/net/wireless/ath/ath5k/phy.c +17 -7

index 2075ba9..6afba98 100644
@@ -740,13 +740,22 @@ int ath5k_hw_rfregs_init(struct ath5k_hw *ah, struct ieee80211_channel *channel,
 						AR5K_RF_XPD_GAIN, true);
 
 		} else {
-			/* TODO: Set high and low gain bits */
-			ath5k_hw_rfb_op(ah, rf_regs,
-						ee->ee_x_gain[ee_mode],
+			u8 *pdg_curve_to_idx = ee->ee_pdc_to_idx[ee_mode];
+			if (ee->ee_pd_gains[ee_mode] > 1) {
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
 						AR5K_RF_PD_GAIN_LO, true);
-			ath5k_hw_rfb_op(ah, rf_regs,
-						ee->ee_x_gain[ee_mode],
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[1],
 						AR5K_RF_PD_GAIN_HI, true);
+			} else {
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
+						AR5K_RF_PD_GAIN_LO, true);
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
+						AR5K_RF_PD_GAIN_HI, true);
+			}
 
 			/* Lower synth voltage on Rev 2 */
 			ath5k_hw_rfb_op(ah, rf_regs, 2,

@@ -1896,8 +1905,9 @@ ath5k_get_linear_pcdac_min(const u8 *stepL, const u8 *stepR,
 	s16 min_pwrL, min_pwrR;
 	s16 pwr_i;
 
-	if (WARN_ON(stepL[0] == stepL[1] || stepR[0] == stepR[1]))
-		return 0;
+	/* Some vendors write the same pcdac value twice !!! */
+	if (stepL[0] == stepL[1] || stepR[0] == stepR[1])
+		return max(pwrL[0], pwrR[0]);
 
 	if (pwrL[0] == pwrL[1])
 		min_pwrL = pwrL[0];

Luis R. Rodriguez

2009-07-31 18:07 UTC #1 lore raw

Quoted 10 lines · Nick Kossifidis

* Set correct xpd curve indices for high/low gain curves during rfbuffer setup on RF5112B with both calibration curves available.
* Don't return zero min power when we have the same pcdac value twice because it breaks interpolation. Instead return the right x barrier as we do when we have equal power levels for 2 different pcdac values.
---
 drivers/net/wireless/ath/ath5k/phy.c |   24 +++++++++++++++++-------
 1 files changed, 17 insertions(+), 7 deletions(-)

Missing SOB

Luis

Nick Kossifidis

maintainer

2009-07-31 18:12 UTC #2 lore raw

2009/7/31 Luis R. Rodriguez <[email protected]>:

Quoted 13 lines

Quoted 10 lines · Nick Kossifidis
* Set correct xpd curve indices for high/low gain curves during rfbuffer setup on RF5112B with both calibration curves available.
* Don't return zero min power when we have the same pcdac value twice because it breaks interpolation. Instead return the right x barrier as we do when we have equal power levels for 2 different pcdac values.
---
drivers/net/wireless/ath/ath5k/phy.c | 24 +++++++++++++++++-------
 1 files changed, 17 insertions(+), 7 deletions(-)

Quoted 3 lines

Missing SOB
Luis

Grrr, sorry ppl, resending...

Signature · 3 lines

GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick

Nick Kossifidis

author

2009-07-31 18:31 UTC #3 lore raw

In reply to Nick Kossifidis

* Set correct xpd curve indices for high/low gain curves during rfbuffer setup on RF5112B with both calibration curves available.

Signed-off-by Nick Kossifidis <[email protected]>

---
 drivers/net/wireless/ath/ath5k/phy.c |   24 +++++++++++++++++-------
 1 files changed, 17 insertions(+), 7 deletions(-)

drivers/net/wireless/ath/ath5k/phy.c +17 -7

index 2075ba9..6afba98 100644
@@ -740,13 +740,22 @@ int ath5k_hw_rfregs_init(struct ath5k_hw *ah, struct ieee80211_channel *channel,
 						AR5K_RF_XPD_GAIN, true);
 
 		} else {
-			/* TODO: Set high and low gain bits */
-			ath5k_hw_rfb_op(ah, rf_regs,
-						ee->ee_x_gain[ee_mode],
+			u8 *pdg_curve_to_idx = ee->ee_pdc_to_idx[ee_mode];
+			if (ee->ee_pd_gains[ee_mode] > 1) {
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
 						AR5K_RF_PD_GAIN_LO, true);
-			ath5k_hw_rfb_op(ah, rf_regs,
-						ee->ee_x_gain[ee_mode],
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[1],
 						AR5K_RF_PD_GAIN_HI, true);
+			} else {
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
+						AR5K_RF_PD_GAIN_LO, true);
+				ath5k_hw_rfb_op(ah, rf_regs,
+						pdg_curve_to_idx[0],
+						AR5K_RF_PD_GAIN_HI, true);
+			}
 
 			/* Lower synth voltage on Rev 2 */
 			ath5k_hw_rfb_op(ah, rf_regs, 2,

@@ -1896,8 +1905,9 @@ ath5k_get_linear_pcdac_min(const u8 *stepL, const u8 *stepR,
 	s16 min_pwrL, min_pwrR;
 	s16 pwr_i;
 
-	if (WARN_ON(stepL[0] == stepL[1] || stepR[0] == stepR[1]))
-		return 0;
+	/* Some vendors write the same pcdac value twice !!! */
+	if (stepL[0] == stepL[1] || stepR[0] == stepR[1])
+		return max(pwrL[0], pwrR[0]);
 
 	if (pwrL[0] == pwrL[1])
 		min_pwrL = pwrL[0];

Bob Copeland

Acked-by

2009-08-04 15:25 UTC #4 lore raw

Quoted 9 lines · Nick Kossifidis

* Set correct xpd curve indices for high/low gain curves during rfbuffer setup on RF5112B with both calibration curves available.
* Don't return zero min power when we have the same pcdac value twice because it breaks interpolation. Instead return the right x barrier as we do when we have equal power levels for 2 different pcdac values.
Signed-off-by Nick Kossifidis <[email protected]>

Acked-by Bob Copeland <[email protected]>

Signature · 1 line

Bob Copeland %% www.bobcopeland.com