* [PATCH 00/11] rtw88: some driver fixes
@ 2019-12-20 9:21 yhchuang
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
` (10 more replies)
0 siblings, 11 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
This set includes some driver fixes
Chien-Hsun Liao (1):
rtw88: 8822c: modify rf protection setting
Chin-Yen Lee (1):
rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop()
Ping-Ke Shih (3):
rtw88: fix rate mask for 1SS chip
rtw88: fix TX secondary channel offset of 40M if current bw is 20M or
40M
rtw88: Use secondary channel offset enumeration
Tzu-En Huang (2):
rtw88: 8822c: update power sequence to v15
rtw88: remove unused spinlock
Yan-Hsuan Chuang (3):
rtw88: remove unused variable 'in_lps'
rtw88: remove unused vif pointer in struct rtw_vif
rtw88: assign NULL to skb after being kfree()'ed
Zong-Zhe Yang (1):
rtw88: change max_num_of_tx_queue() definition to inline in pci.h
drivers/net/wireless/realtek/rtw88/fw.c | 5 +++-
drivers/net/wireless/realtek/rtw88/mac.c | 10 ++++---
drivers/net/wireless/realtek/rtw88/mac80211.c | 2 --
drivers/net/wireless/realtek/rtw88/main.c | 29 +++++++++----------
drivers/net/wireless/realtek/rtw88/main.h | 15 ++++++----
drivers/net/wireless/realtek/rtw88/pci.h | 2 +-
drivers/net/wireless/realtek/rtw88/phy.c | 10 -------
drivers/net/wireless/realtek/rtw88/rtw8822b.c | 2 +-
drivers/net/wireless/realtek/rtw88/rtw8822c.c | 23 ++++++++++++---
drivers/net/wireless/realtek/rtw88/rtw8822c.h | 5 ++++
10 files changed, 59 insertions(+), 44 deletions(-)
--
2.17.1
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 01/11] rtw88: fix rate mask for 1SS chip
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-24 7:10 ` Chris Chiu
2020-01-26 15:38 ` Kalle Valo
2019-12-20 9:21 ` [PATCH 02/11] rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M yhchuang
` (9 subsequent siblings)
10 siblings, 2 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Ping-Ke Shih <pkshih@realtek.com>
The rate mask is used to tell firmware the supported rate depends on
negotiation. We loop 2 times for all VHT/HT 2SS rate mask first, and then
only keep the part according to chip's NSS.
This commit fixes the logic error of '&' operations for VHT/HT rate, and
we should run this logic before adding legacy rate.
To access HT MCS map, index 0/1 represent MCS 0-7/8-15 respectively. Use
NL80211_BAND_xxx is incorrect, so fix it as well.
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/main.c | 12 +++++-------
1 file changed, 5 insertions(+), 7 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index ae61415e1665..f369ddca953a 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -706,8 +706,8 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
if (sta->vht_cap.cap & IEEE80211_VHT_CAP_SHORT_GI_80)
is_support_sgi = true;
} else if (sta->ht_cap.ht_supported) {
- ra_mask |= (sta->ht_cap.mcs.rx_mask[NL80211_BAND_5GHZ] << 20) |
- (sta->ht_cap.mcs.rx_mask[NL80211_BAND_2GHZ] << 12);
+ ra_mask |= (sta->ht_cap.mcs.rx_mask[1] << 20) |
+ (sta->ht_cap.mcs.rx_mask[0] << 12);
if (sta->ht_cap.cap & IEEE80211_HT_CAP_RX_STBC)
stbc_en = HT_STBC_EN;
if (sta->ht_cap.cap & IEEE80211_HT_CAP_LDPC_CODING)
@@ -717,6 +717,9 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
is_support_sgi = true;
}
+ if (efuse->hw_cap.nss == 1)
+ ra_mask &= RA_MASK_VHT_RATES_1SS | RA_MASK_HT_RATES_1SS;
+
if (hal->current_band_type == RTW_BAND_5G) {
ra_mask |= (u64)sta->supp_rates[NL80211_BAND_5GHZ] << 4;
if (sta->vht_cap.vht_supported) {
@@ -750,11 +753,6 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
wireless_set = 0;
}
- if (efuse->hw_cap.nss == 1) {
- ra_mask &= RA_MASK_VHT_RATES_1SS;
- ra_mask &= RA_MASK_HT_RATES_1SS;
- }
-
switch (sta->bandwidth) {
case IEEE80211_STA_RX_BW_80:
bw_mode = RTW_CHANNEL_WIDTH_80;
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 02/11] rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 03/11] rtw88: Use secondary channel offset enumeration yhchuang
` (8 subsequent siblings)
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Ping-Ke Shih <pkshih@realtek.com>
TX secondary channel offset is valid only if current bandwidth is 80M,
otherwise leave this value as zero. The wrong value of txsc40 causes
MAC unpredictable behavior.
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/mac.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 507970387b2a..ed9bb427ba60 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -16,10 +16,12 @@ void rtw_set_channel_mac(struct rtw_dev *rtwdev, u8 channel, u8 bw,
u8 value8;
txsc20 = primary_ch_idx;
- if (txsc20 == 1 || txsc20 == 3)
- txsc40 = 9;
- else
- txsc40 = 10;
+ if (bw == RTW_CHANNEL_WIDTH_80) {
+ if (txsc20 == 1 || txsc20 == 3)
+ txsc40 = 9;
+ else
+ txsc40 = 10;
+ }
rtw_write8(rtwdev, REG_DATA_SC,
BIT_TXSC_20M(txsc20) | BIT_TXSC_40M(txsc40));
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 03/11] rtw88: Use secondary channel offset enumeration
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
2019-12-20 9:21 ` [PATCH 02/11] rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-24 7:18 ` Chris Chiu
2019-12-20 9:21 ` [PATCH 04/11] rtw88: 8822c: update power sequence to v15 yhchuang
` (7 subsequent siblings)
10 siblings, 1 reply; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Ping-Ke Shih <pkshih@realtek.com>
The hardware value of secondary channel offset isn't very intuitive. This
commit adds enumeration, so we can easier to check the logic with the
suffix of enumeration name, likes _UPPER or _LOWER.
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/mac.c | 6 +++---
drivers/net/wireless/realtek/rtw88/main.c | 14 +++++++-------
drivers/net/wireless/realtek/rtw88/main.h | 10 ++++++++++
drivers/net/wireless/realtek/rtw88/rtw8822b.c | 2 +-
drivers/net/wireless/realtek/rtw88/rtw8822c.c | 2 +-
5 files changed, 22 insertions(+), 12 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index ed9bb427ba60..94d1b179e2e1 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -17,10 +17,10 @@ void rtw_set_channel_mac(struct rtw_dev *rtwdev, u8 channel, u8 bw,
txsc20 = primary_ch_idx;
if (bw == RTW_CHANNEL_WIDTH_80) {
- if (txsc20 == 1 || txsc20 == 3)
- txsc40 = 9;
+ if (txsc20 == RTW_SC_20_UPPER || txsc20 == RTW_SC_20_UPPERST)
+ txsc40 = RTW_SC_40_UPPER;
else
- txsc40 = 10;
+ txsc40 = RTW_SC_40_LOWER;
}
rtw_write8(rtwdev, REG_DATA_SC,
BIT_TXSC_20M(txsc20) | BIT_TXSC_40M(txsc40));
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index f369ddca953a..cc61c390226c 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -317,15 +317,15 @@ void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
case NL80211_CHAN_WIDTH_20_NOHT:
case NL80211_CHAN_WIDTH_20:
bandwidth = RTW_CHANNEL_WIDTH_20;
- primary_chan_idx = 0;
+ primary_chan_idx = RTW_SC_DONT_CARE;
break;
case NL80211_CHAN_WIDTH_40:
bandwidth = RTW_CHANNEL_WIDTH_40;
if (primary_freq > center_freq) {
- primary_chan_idx = 1;
+ primary_chan_idx = RTW_SC_20_UPPER;
center_chan -= 2;
} else {
- primary_chan_idx = 2;
+ primary_chan_idx = RTW_SC_20_LOWER;
center_chan += 2;
}
break;
@@ -333,10 +333,10 @@ void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
bandwidth = RTW_CHANNEL_WIDTH_80;
if (primary_freq > center_freq) {
if (primary_freq - center_freq == 10) {
- primary_chan_idx = 1;
+ primary_chan_idx = RTW_SC_20_UPPER;
center_chan -= 2;
} else {
- primary_chan_idx = 3;
+ primary_chan_idx = RTW_SC_20_UPPERST;
center_chan -= 6;
}
/* assign the center channel used
@@ -345,10 +345,10 @@ void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
cch_by_bw[RTW_CHANNEL_WIDTH_40] = center_chan + 4;
} else {
if (center_freq - primary_freq == 10) {
- primary_chan_idx = 2;
+ primary_chan_idx = RTW_SC_20_LOWER;
center_chan += 2;
} else {
- primary_chan_idx = 4;
+ primary_chan_idx = RTW_SC_20_LOWEST;
center_chan += 6;
}
/* assign the center channel used
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index d012eefcd0da..144fbab38fd5 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -95,6 +95,16 @@ enum rtw_bandwidth {
RTW_CHANNEL_WIDTH_10 = 6,
};
+enum rtw_sc_offset {
+ RTW_SC_DONT_CARE = 0,
+ RTW_SC_20_UPPER = 1,
+ RTW_SC_20_LOWER = 2,
+ RTW_SC_20_UPPERST = 3,
+ RTW_SC_20_LOWEST = 4,
+ RTW_SC_40_UPPER = 9,
+ RTW_SC_40_LOWER = 10,
+};
+
enum rtw_net_type {
RTW_NET_NO_LINK = 0,
RTW_NET_AD_HOC = 1,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822b.c b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
index 4bc14b1a6340..2eed777ee692 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822b.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
@@ -645,7 +645,7 @@ static void rtw8822b_set_channel_bb(struct rtw_dev *rtwdev, u8 channel, u8 bw,
rtw_write32_mask(rtwdev, REG_ADC160, BIT(30), 0x1);
break;
case RTW_CHANNEL_WIDTH_40:
- if (primary_ch_idx == 1)
+ if (primary_ch_idx == RTW_SC_20_UPPER)
rtw_write32_set(rtwdev, REG_RXSB, BIT(4));
else
rtw_write32_clr(rtwdev, REG_RXSB, BIT(4));
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 174029836833..57faef21ea52 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -1482,7 +1482,7 @@ static void rtw8822c_set_channel_bb(struct rtw_dev *rtwdev, u8 channel, u8 bw,
break;
case RTW_CHANNEL_WIDTH_40:
rtw_write32_mask(rtwdev, REG_CCKSB, BIT(4),
- (primary_ch_idx == 1 ? 1 : 0));
+ (primary_ch_idx == RTW_SC_20_UPPER ? 1 : 0));
rtw_write32_mask(rtwdev, REG_TXBWCTL, 0xf, 0x5);
rtw_write32_mask(rtwdev, REG_TXBWCTL, 0xc0, 0x0);
rtw_write32_mask(rtwdev, REG_TXBWCTL, 0xff00,
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 04/11] rtw88: 8822c: update power sequence to v15
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (2 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 03/11] rtw88: Use secondary channel offset enumeration yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 05/11] rtw88: 8822c: modify rf protection setting yhchuang
` (6 subsequent siblings)
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Tzu-En Huang <tehuang@realtek.com>
Update card enable power sequence flow, to fix CMD11 fail after
reboot and wrong PLL clock.
Signed-off-by: Tzu-En Huang <tehuang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/rtw8822c.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 57faef21ea52..7c8db951a5bc 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -3487,12 +3487,12 @@ static struct rtw_pwr_seq_cmd trans_cardemu_to_act_8822c[] = {
RTW_PWR_CUT_ALL_MSK,
RTW_PWR_INTF_ALL_MSK,
RTW_PWR_ADDR_MAC,
- RTW_PWR_CMD_WRITE, BIT(7), 0},
- {0x0005,
+ RTW_PWR_CMD_WRITE, (BIT(4) | BIT(3)), 0},
+ {0x1018,
RTW_PWR_CUT_ALL_MSK,
RTW_PWR_INTF_ALL_MSK,
RTW_PWR_ADDR_MAC,
- RTW_PWR_CMD_WRITE, (BIT(4) | BIT(3)), 0},
+ RTW_PWR_CMD_WRITE, BIT(2), BIT(2)},
{0x0005,
RTW_PWR_CUT_ALL_MSK,
RTW_PWR_INTF_ALL_MSK,
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 05/11] rtw88: 8822c: modify rf protection setting
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (3 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 04/11] rtw88: 8822c: update power sequence to v15 yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-24 7:39 ` Chris Chiu
2019-12-20 9:21 ` [PATCH 06/11] rtw88: remove unused spinlock yhchuang
` (5 subsequent siblings)
10 siblings, 1 reply; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Chien-Hsun Liao <ben.liao@realtek.com>
According to some experiments, the original rf protection
setting can not perfectly make sure that there is no hardware
pi write during the direct write. So, modify the setting so
that the hardware block of pi would be turned off during the
direct write.
Signed-off-by: Chien-Hsun Liao <ben.liao@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/phy.c | 10 ----------
drivers/net/wireless/realtek/rtw88/rtw8822c.c | 15 +++++++++++++++
drivers/net/wireless/realtek/rtw88/rtw8822c.h | 5 +++++
3 files changed, 20 insertions(+), 10 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/phy.c b/drivers/net/wireless/realtek/rtw88/phy.c
index a3e1e9578b65..4b2f11be60cf 100644
--- a/drivers/net/wireless/realtek/rtw88/phy.c
+++ b/drivers/net/wireless/realtek/rtw88/phy.c
@@ -749,20 +749,10 @@ bool rtw_phy_write_rf_reg(struct rtw_dev *rtwdev, enum rtw_rf_path rf_path,
direct_addr = base_addr[rf_path] + (addr << 2);
mask &= RFREG_MASK;
- if (addr == RF_CFGCH) {
- rtw_write32_mask(rtwdev, REG_RSV_CTRL, BITS_RFC_DIRECT, DISABLE_PI);
- rtw_write32_mask(rtwdev, REG_WLRF1, BITS_RFC_DIRECT, DISABLE_PI);
- }
-
rtw_write32_mask(rtwdev, direct_addr, mask, data);
udelay(1);
- if (addr == RF_CFGCH) {
- rtw_write32_mask(rtwdev, REG_RSV_CTRL, BITS_RFC_DIRECT, ENABLE_PI);
- rtw_write32_mask(rtwdev, REG_WLRF1, BITS_RFC_DIRECT, ENABLE_PI);
- }
-
return true;
}
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 7c8db951a5bc..4231f94d515e 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -1289,6 +1289,17 @@ static int rtw8822c_mac_init(struct rtw_dev *rtwdev)
return 0;
}
+static void rtw8822c_rstb_3wire(struct rtw_dev *rtwdev, bool enable)
+{
+ if (enable) {
+ rtw_write32_mask(rtwdev, REG_RSTB, BIT_RSTB_3WIRE, 0x1);
+ rtw_write32_mask(rtwdev, REG_ANAPAR_A, BIT_ANAPAR_UPDATE, 0x1);
+ rtw_write32_mask(rtwdev, REG_ANAPAR_B, BIT_ANAPAR_UPDATE, 0x1);
+ } else {
+ rtw_write32_mask(rtwdev, REG_RSTB, BIT_RSTB_3WIRE, 0x0);
+ }
+}
+
static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
{
#define RF18_BAND_MASK (BIT(16) | BIT(9) | BIT(8))
@@ -1337,6 +1348,8 @@ static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
break;
}
+ rtw8822c_rstb_3wire(rtwdev, false);
+
rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWE2, 0x04, 0x01);
rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWA, 0x1f, 0x12);
rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWD0, 0xfffff, rf_rxbb);
@@ -1349,6 +1362,8 @@ static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
rtw_write_rf(rtwdev, RF_PATH_A, RF_CFGCH, RFREG_MASK, rf_reg18);
rtw_write_rf(rtwdev, RF_PATH_B, RF_CFGCH, RFREG_MASK, rf_reg18);
+
+ rtw8822c_rstb_3wire(rtwdev, true);
}
static void rtw8822c_toggle_igi(struct rtw_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.h b/drivers/net/wireless/realtek/rtw88/rtw8822c.h
index abd9f300bedd..dfd8662a0c0e 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.h
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.h
@@ -190,6 +190,8 @@ const struct rtw_table name ## _tbl = { \
#define BIT_3WIRE_TX_EN BIT(0)
#define BIT_3WIRE_RX_EN BIT(1)
#define BIT_3WIRE_PI_ON BIT(28)
+#define REG_ANAPAR_A 0x1830
+#define BIT_ANAPAR_UPDATE BIT(29)
#define REG_RXAGCCTL0 0x18ac
#define BITS_RXAGC_CCK GENMASK(15, 12)
#define BITS_RXAGC_OFDM GENMASK(8, 4)
@@ -223,6 +225,8 @@ const struct rtw_table name ## _tbl = { \
#define BIT_CCK_BLK_EN BIT(1)
#define BIT_CCK_OFDM_BLK_EN (BIT(0) | BIT(1))
#define REG_CCAMSK 0x1c80
+#define REG_RSTB 0x1c90
+#define BIT_RSTB_3WIRE BIT(8)
#define REG_RX_BREAK 0x1d2c
#define BIT_COM_RX_GCK_EN BIT(31)
#define REG_RXFNCTL 0x1d30
@@ -243,6 +247,7 @@ const struct rtw_table name ## _tbl = { \
#define REG_OFDM_TXCNT 0x2de0
#define REG_ORITXCODE2 0x4100
#define REG_3WIRE2 0x410c
+#define REG_ANAPAR_B 0x4130
#define REG_RXAGCCTL 0x41ac
#define REG_DCKB_I_0 0x41bc
#define REG_DCKB_I_1 0x41c0
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 06/11] rtw88: remove unused spinlock
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (4 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 05/11] rtw88: 8822c: modify rf protection setting yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 07/11] rtw88: remove unused variable 'in_lps' yhchuang
` (4 subsequent siblings)
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Tzu-En Huang <tehuang@realtek.com>
dm_lock is never used. Thus, remove this redundant spinlock.
Signed-off-by: Tzu-En Huang <tehuang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/main.c | 1 -
drivers/net/wireless/realtek/rtw88/main.h | 3 ---
2 files changed, 4 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index cc61c390226c..d6c9cc1806f7 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -1339,7 +1339,6 @@ int rtw_core_init(struct rtw_dev *rtwdev)
skb_queue_head_init(&rtwdev->coex.queue);
skb_queue_head_init(&rtwdev->tx_report.queue);
- spin_lock_init(&rtwdev->dm_lock);
spin_lock_init(&rtwdev->rf_lock);
spin_lock_init(&rtwdev->h2c.lock);
spin_lock_init(&rtwdev->txq_lock);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 144fbab38fd5..b2cc7479d49b 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1544,9 +1544,6 @@ struct rtw_dev {
/* ensures exclusive access from mac80211 callbacks */
struct mutex mutex;
- /* lock for dm to use */
- spinlock_t dm_lock;
-
/* read/write rf register */
spinlock_t rf_lock;
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 07/11] rtw88: remove unused variable 'in_lps'
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (5 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 06/11] rtw88: remove unused spinlock yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 08/11] rtw88: remove unused vif pointer in struct rtw_vif yhchuang
` (3 subsequent siblings)
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Unused, will not be used neither, because the hardware/firmware
can only support one vif for LPS currnetly. If there's more than
one vif, than driver will never enter LPS. So remove it.
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/mac80211.c | 1 -
drivers/net/wireless/realtek/rtw88/main.h | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 34a1c3b53cd4..ddde3b855893 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -157,7 +157,6 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
rtwvif->stats.rx_unicast = 0;
rtwvif->stats.tx_cnt = 0;
rtwvif->stats.rx_cnt = 0;
- rtwvif->in_lps = false;
memset(&rtwvif->bfee, 0, sizeof(struct rtw_bfee));
rtwvif->conf = &rtw_vif_port[port];
rtw_txq_init(rtwdev, vif->txq);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index b2cc7479d49b..a80fa38fb404 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -737,7 +737,6 @@ struct rtw_vif {
const struct rtw_vif_port *conf;
struct rtw_traffic_stats stats;
- bool in_lps;
struct rtw_bfee bfee;
};
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 08/11] rtw88: remove unused vif pointer in struct rtw_vif
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (6 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 07/11] rtw88: remove unused variable 'in_lps' yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 09/11] rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop() yhchuang
` (2 subsequent siblings)
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
As driver can easily get vif with container_of(), we can
just remove it.
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/mac80211.c | 1 -
drivers/net/wireless/realtek/rtw88/main.h | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index ddde3b855893..76f85640ff0d 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -152,7 +152,6 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
u8 bcn_ctrl = 0;
rtwvif->port = port;
- rtwvif->vif = vif;
rtwvif->stats.tx_unicast = 0;
rtwvif->stats.rx_unicast = 0;
rtwvif->stats.tx_cnt = 0;
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index a80fa38fb404..760192442198 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -726,7 +726,6 @@ struct rtw_bf_info {
};
struct rtw_vif {
- struct ieee80211_vif *vif;
enum rtw_net_type net_type;
u16 aid;
u8 mac_addr[ETH_ALEN];
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 09/11] rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop()
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (7 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 08/11] rtw88: remove unused vif pointer in struct rtw_vif yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 10/11] rtw88: assign NULL to skb after being kfree()'ed yhchuang
2019-12-20 9:21 ` [PATCH 11/11] rtw88: change max_num_of_tx_queue() definition to inline in pci.h yhchuang
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Chin-Yen Lee <timlee@realtek.com>
Fix typo, should use rtw_hci_stop()
Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index d6c9cc1806f7..bbdd6aaaf89e 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -879,7 +879,7 @@ int rtw_core_start(struct rtw_dev *rtwdev)
static void rtw_power_off(struct rtw_dev *rtwdev)
{
- rtwdev->hci.ops->stop(rtwdev);
+ rtw_hci_stop(rtwdev);
rtw_mac_power_off(rtwdev);
}
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 10/11] rtw88: assign NULL to skb after being kfree()'ed
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (8 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 09/11] rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop() yhchuang
@ 2019-12-20 9:21 ` yhchuang
2019-12-20 9:21 ` [PATCH 11/11] rtw88: change max_num_of_tx_queue() definition to inline in pci.h yhchuang
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Should assign NULL to skb after kfree(), in case of driver
trying to free the same skb again.
This could happen if driver failed to allocate an skb when
building reserved page.
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/fw.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index b8c581161f61..92cefb79e09b 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -857,13 +857,16 @@ static u8 *rtw_build_rsvd_page(struct rtw_dev *rtwdev,
page += rtw_len_to_page(rsvd_pkt->skb->len, page_size);
kfree_skb(rsvd_pkt->skb);
+ rsvd_pkt->skb = NULL;
}
return buf;
release_skb:
- list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list)
+ list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
kfree_skb(rsvd_pkt->skb);
+ rsvd_pkt->skb = NULL;
+ }
return NULL;
}
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* [PATCH 11/11] rtw88: change max_num_of_tx_queue() definition to inline in pci.h
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
` (9 preceding siblings ...)
2019-12-20 9:21 ` [PATCH 10/11] rtw88: assign NULL to skb after being kfree()'ed yhchuang
@ 2019-12-20 9:21 ` yhchuang
10 siblings, 0 replies; 20+ messages in thread
From: yhchuang @ 2019-12-20 9:21 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, briannorris
From: Zong-Zhe Yang <kevin_yang@realtek.com>
It's more reasonable to define max_num_of_tx_queue() as an inline function.
Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
---
drivers/net/wireless/realtek/rtw88/pci.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/realtek/rtw88/pci.h b/drivers/net/wireless/realtek/rtw88/pci.h
index 49bf29a92152..1580cfc57361 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.h
+++ b/drivers/net/wireless/realtek/rtw88/pci.h
@@ -210,7 +210,7 @@ struct rtw_pci {
void __iomem *mmap;
};
-static u32 max_num_of_tx_queue(u8 queue)
+static inline u32 max_num_of_tx_queue(u8 queue)
{
u32 max_num;
--
2.17.1
^ permalink raw reply related [flat|nested] 20+ messages in thread
* Re: [PATCH 01/11] rtw88: fix rate mask for 1SS chip
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
@ 2019-12-24 7:10 ` Chris Chiu
2020-01-26 15:38 ` Kalle Valo
1 sibling, 0 replies; 20+ messages in thread
From: Chris Chiu @ 2019-12-24 7:10 UTC (permalink / raw)
To: Tony Chuang; +Cc: Kalle Valo, linux-wireless, Brian Norris
On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
>
> From: Ping-Ke Shih <pkshih@realtek.com>
>
> The rate mask is used to tell firmware the supported rate depends on
> negotiation. We loop 2 times for all VHT/HT 2SS rate mask first, and then
> only keep the part according to chip's NSS.
>
> This commit fixes the logic error of '&' operations for VHT/HT rate, and
> we should run this logic before adding legacy rate.
>
> To access HT MCS map, index 0/1 represent MCS 0-7/8-15 respectively. Use
> NL80211_BAND_xxx is incorrect, so fix it as well.
>
> Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
> ---
Reviewed-by: Chris Chiu <chiu@endlessm.com>
> drivers/net/wireless/realtek/rtw88/main.c | 12 +++++-------
> 1 file changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
> index ae61415e1665..f369ddca953a 100644
> --- a/drivers/net/wireless/realtek/rtw88/main.c
> +++ b/drivers/net/wireless/realtek/rtw88/main.c
> @@ -706,8 +706,8 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
> if (sta->vht_cap.cap & IEEE80211_VHT_CAP_SHORT_GI_80)
> is_support_sgi = true;
> } else if (sta->ht_cap.ht_supported) {
> - ra_mask |= (sta->ht_cap.mcs.rx_mask[NL80211_BAND_5GHZ] << 20) |
> - (sta->ht_cap.mcs.rx_mask[NL80211_BAND_2GHZ] << 12);
> + ra_mask |= (sta->ht_cap.mcs.rx_mask[1] << 20) |
> + (sta->ht_cap.mcs.rx_mask[0] << 12);
> if (sta->ht_cap.cap & IEEE80211_HT_CAP_RX_STBC)
> stbc_en = HT_STBC_EN;
> if (sta->ht_cap.cap & IEEE80211_HT_CAP_LDPC_CODING)
> @@ -717,6 +717,9 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
> is_support_sgi = true;
> }
>
> + if (efuse->hw_cap.nss == 1)
> + ra_mask &= RA_MASK_VHT_RATES_1SS | RA_MASK_HT_RATES_1SS;
> +
> if (hal->current_band_type == RTW_BAND_5G) {
> ra_mask |= (u64)sta->supp_rates[NL80211_BAND_5GHZ] << 4;
> if (sta->vht_cap.vht_supported) {
> @@ -750,11 +753,6 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
> wireless_set = 0;
> }
>
> - if (efuse->hw_cap.nss == 1) {
> - ra_mask &= RA_MASK_VHT_RATES_1SS;
> - ra_mask &= RA_MASK_HT_RATES_1SS;
> - }
> -
> switch (sta->bandwidth) {
> case IEEE80211_STA_RX_BW_80:
> bw_mode = RTW_CHANNEL_WIDTH_80;
> --
> 2.17.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 03/11] rtw88: Use secondary channel offset enumeration
2019-12-20 9:21 ` [PATCH 03/11] rtw88: Use secondary channel offset enumeration yhchuang
@ 2019-12-24 7:18 ` Chris Chiu
2019-12-24 7:44 ` Tony Chuang
0 siblings, 1 reply; 20+ messages in thread
From: Chris Chiu @ 2019-12-24 7:18 UTC (permalink / raw)
To: Tony Chuang; +Cc: Kalle Valo, linux-wireless, Brian Norris
On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
>
> From: Ping-Ke Shih <pkshih@realtek.com>
>
> The hardware value of secondary channel offset isn't very intuitive. This
> commit adds enumeration, so we can easier to check the logic with the
> suffix of enumeration name, likes _UPPER or _LOWER.
>
> Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
> ---
> drivers/net/wireless/realtek/rtw88/mac.c | 6 +++---
> drivers/net/wireless/realtek/rtw88/main.c | 14 +++++++-------
> drivers/net/wireless/realtek/rtw88/main.h | 10 ++++++++++
> drivers/net/wireless/realtek/rtw88/rtw8822b.c | 2 +-
> drivers/net/wireless/realtek/rtw88/rtw8822c.c | 2 +-
> 5 files changed, 22 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
> index ed9bb427ba60..94d1b179e2e1 100644
> --- a/drivers/net/wireless/realtek/rtw88/mac.c
> +++ b/drivers/net/wireless/realtek/rtw88/mac.c
> @@ -17,10 +17,10 @@ void rtw_set_channel_mac(struct rtw_dev *rtwdev, u8 channel, u8 bw,
>
> txsc20 = primary_ch_idx;
> if (bw == RTW_CHANNEL_WIDTH_80) {
> - if (txsc20 == 1 || txsc20 == 3)
> - txsc40 = 9;
> + if (txsc20 == RTW_SC_20_UPPER || txsc20 == RTW_SC_20_UPPERST)
> + txsc40 = RTW_SC_40_UPPER;
> else
> - txsc40 = 10;
> + txsc40 = RTW_SC_40_LOWER;
> }
> rtw_write8(rtwdev, REG_DATA_SC,
> BIT_TXSC_20M(txsc20) | BIT_TXSC_40M(txsc40));
> diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
> index f369ddca953a..cc61c390226c 100644
> --- a/drivers/net/wireless/realtek/rtw88/main.c
> +++ b/drivers/net/wireless/realtek/rtw88/main.c
> @@ -333,10 +333,10 @@ void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
> bandwidth = RTW_CHANNEL_WIDTH_80;
> if (primary_freq > center_freq) {
> if (primary_freq - center_freq == 10) {
> - primary_chan_idx = 1;
> + primary_chan_idx = RTW_SC_20_UPPER;
> center_chan -= 2;
> } else {
> - primary_chan_idx = 3;
> + primary_chan_idx = RTW_SC_20_UPPERST;
> center_chan -= 6;
> }
> /* assign the center channel used
> @@ -345,10 +345,10 @@ void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
> cch_by_bw[RTW_CHANNEL_WIDTH_40] = center_chan + 4;
> } else {
> if (center_freq - primary_freq == 10) {
> - primary_chan_idx = 2;
> + primary_chan_idx = RTW_SC_20_LOWER;
> center_chan += 2;
> } else {
> - primary_chan_idx = 4;
> + primary_chan_idx = RTW_SC_20_LOWEST;
> center_chan += 6;
> }
> /* assign the center channel used
> diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
> index d012eefcd0da..144fbab38fd5 100644
> --- a/drivers/net/wireless/realtek/rtw88/main.h
> +++ b/drivers/net/wireless/realtek/rtw88/main.h
> @@ -95,6 +95,16 @@ enum rtw_bandwidth {
> RTW_CHANNEL_WIDTH_10 = 6,
> };
>
> +enum rtw_sc_offset {
> + RTW_SC_DONT_CARE = 0,
> + RTW_SC_20_UPPER = 1,
> + RTW_SC_20_LOWER = 2,
> + RTW_SC_20_UPPERST = 3,
I'll suppgest RTW_SC_20_UPMOST instead or simply RTW_SC_20_UPPEREST.
> + RTW_SC_20_LOWEST = 4,
> + RTW_SC_40_UPPER = 9,
> + RTW_SC_40_LOWER = 10,
> +};
> +
> enum rtw_net_type {
> RTW_NET_NO_LINK = 0,
> RTW_NET_AD_HOC = 1,
> --
> 2.17.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 05/11] rtw88: 8822c: modify rf protection setting
2019-12-20 9:21 ` [PATCH 05/11] rtw88: 8822c: modify rf protection setting yhchuang
@ 2019-12-24 7:39 ` Chris Chiu
2019-12-24 7:50 ` Tony Chuang
0 siblings, 1 reply; 20+ messages in thread
From: Chris Chiu @ 2019-12-24 7:39 UTC (permalink / raw)
To: Tony Chuang; +Cc: Kalle Valo, linux-wireless, Brian Norris
On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
>
> From: Chien-Hsun Liao <ben.liao@realtek.com>
>
> According to some experiments, the original rf protection
> setting can not perfectly make sure that there is no hardware
> pi write during the direct write. So, modify the setting so
> that the hardware block of pi would be turned off during the
> direct write.
>
Sorry, I don't really understand this part. Does it mean rtw8822c_rstb_3wire()
is to disable/enable the hardware block of PI? In this patch, I can only
see the code block of ENABLE_PI/DISABLE_PI been removed and some
rtw_write_rf()s been protected by new rtw8822c_rstb_3wire(). If the new
function is to replace the ENABLE_PI/DISABLE_PI, maybe they should be
removed in the reg.h. And It seems rtw8822c_rstb_3wire() is only for 8822c,
means there's no such problem for 8822b?
Chris
> Signed-off-by: Chien-Hsun Liao <ben.liao@realtek.com>
> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
> ---
> drivers/net/wireless/realtek/rtw88/phy.c | 10 ----------
> drivers/net/wireless/realtek/rtw88/rtw8822c.c | 15 +++++++++++++++
> drivers/net/wireless/realtek/rtw88/rtw8822c.h | 5 +++++
> 3 files changed, 20 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/net/wireless/realtek/rtw88/phy.c b/drivers/net/wireless/realtek/rtw88/phy.c
> index a3e1e9578b65..4b2f11be60cf 100644
> --- a/drivers/net/wireless/realtek/rtw88/phy.c
> +++ b/drivers/net/wireless/realtek/rtw88/phy.c
> @@ -749,20 +749,10 @@ bool rtw_phy_write_rf_reg(struct rtw_dev *rtwdev, enum rtw_rf_path rf_path,
> direct_addr = base_addr[rf_path] + (addr << 2);
> mask &= RFREG_MASK;
>
> - if (addr == RF_CFGCH) {
> - rtw_write32_mask(rtwdev, REG_RSV_CTRL, BITS_RFC_DIRECT, DISABLE_PI);
> - rtw_write32_mask(rtwdev, REG_WLRF1, BITS_RFC_DIRECT, DISABLE_PI);
> - }
> -
> rtw_write32_mask(rtwdev, direct_addr, mask, data);
>
> udelay(1);
>
> - if (addr == RF_CFGCH) {
> - rtw_write32_mask(rtwdev, REG_RSV_CTRL, BITS_RFC_DIRECT, ENABLE_PI);
> - rtw_write32_mask(rtwdev, REG_WLRF1, BITS_RFC_DIRECT, ENABLE_PI);
> - }
> -
> return true;
> }
>
> diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
> index 7c8db951a5bc..4231f94d515e 100644
> --- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
> +++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
> @@ -1289,6 +1289,17 @@ static int rtw8822c_mac_init(struct rtw_dev *rtwdev)
> return 0;
> }
>
> +static void rtw8822c_rstb_3wire(struct rtw_dev *rtwdev, bool enable)
> +{
> + if (enable) {
> + rtw_write32_mask(rtwdev, REG_RSTB, BIT_RSTB_3WIRE, 0x1);
> + rtw_write32_mask(rtwdev, REG_ANAPAR_A, BIT_ANAPAR_UPDATE, 0x1);
> + rtw_write32_mask(rtwdev, REG_ANAPAR_B, BIT_ANAPAR_UPDATE, 0x1);
> + } else {
> + rtw_write32_mask(rtwdev, REG_RSTB, BIT_RSTB_3WIRE, 0x0);
> + }
> +}
> +
> static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
> {
> #define RF18_BAND_MASK (BIT(16) | BIT(9) | BIT(8))
> @@ -1337,6 +1348,8 @@ static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
> break;
> }
>
> + rtw8822c_rstb_3wire(rtwdev, false);
> +
> rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWE2, 0x04, 0x01);
> rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWA, 0x1f, 0x12);
> rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTWD0, 0xfffff, rf_rxbb);
> @@ -1349,6 +1362,8 @@ static void rtw8822c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
>
> rtw_write_rf(rtwdev, RF_PATH_A, RF_CFGCH, RFREG_MASK, rf_reg18);
> rtw_write_rf(rtwdev, RF_PATH_B, RF_CFGCH, RFREG_MASK, rf_reg18);
> +
> + rtw8822c_rstb_3wire(rtwdev, true);
> }
>
> static void rtw8822c_toggle_igi(struct rtw_dev *rtwdev)
> --
> 2.17.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* RE: [PATCH 03/11] rtw88: Use secondary channel offset enumeration
2019-12-24 7:18 ` Chris Chiu
@ 2019-12-24 7:44 ` Tony Chuang
0 siblings, 0 replies; 20+ messages in thread
From: Tony Chuang @ 2019-12-24 7:44 UTC (permalink / raw)
To: Chris Chiu; +Cc: Kalle Valo, linux-wireless, Brian Norris
> On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
> >
> > From: Ping-Ke Shih <pkshih@realtek.com>
> >
> > The hardware value of secondary channel offset isn't very intuitive. This
> > commit adds enumeration, so we can easier to check the logic with the
> > suffix of enumeration name, likes _UPPER or _LOWER.
> >
> > Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
> > Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
> > ---
> > drivers/net/wireless/realtek/rtw88/mac.c | 6 +++---
> > drivers/net/wireless/realtek/rtw88/main.c | 14 +++++++-------
> > drivers/net/wireless/realtek/rtw88/main.h | 10 ++++++++++
> > drivers/net/wireless/realtek/rtw88/rtw8822b.c | 2 +-
> > drivers/net/wireless/realtek/rtw88/rtw8822c.c | 2 +-
> > 5 files changed, 22 insertions(+), 12 deletions(-)
> >
> > diff --git a/drivers/net/wireless/realtek/rtw88/mac.c
> b/drivers/net/wireless/realtek/rtw88/mac.c
> > index ed9bb427ba60..94d1b179e2e1 100644
> > --- a/drivers/net/wireless/realtek/rtw88/mac.c
> > +++ b/drivers/net/wireless/realtek/rtw88/mac.c
> > @@ -17,10 +17,10 @@ void rtw_set_channel_mac(struct rtw_dev *rtwdev,
> u8 channel, u8 bw,
> >
> > txsc20 = primary_ch_idx;
> > if (bw == RTW_CHANNEL_WIDTH_80) {
> > - if (txsc20 == 1 || txsc20 == 3)
> > - txsc40 = 9;
> > + if (txsc20 == RTW_SC_20_UPPER || txsc20 ==
> RTW_SC_20_UPPERST)
> > + txsc40 = RTW_SC_40_UPPER;
> > else
> > - txsc40 = 10;
> > + txsc40 = RTW_SC_40_LOWER;
> > }
> > rtw_write8(rtwdev, REG_DATA_SC,
> > BIT_TXSC_20M(txsc20) | BIT_TXSC_40M(txsc40));
> > diff --git a/drivers/net/wireless/realtek/rtw88/main.c
> b/drivers/net/wireless/realtek/rtw88/main.c
> > index f369ddca953a..cc61c390226c 100644
> > --- a/drivers/net/wireless/realtek/rtw88/main.c
> > +++ b/drivers/net/wireless/realtek/rtw88/main.c
> > @@ -333,10 +333,10 @@ void rtw_get_channel_params(struct
> cfg80211_chan_def *chandef,
> > bandwidth = RTW_CHANNEL_WIDTH_80;
> > if (primary_freq > center_freq) {
> > if (primary_freq - center_freq == 10) {
> > - primary_chan_idx = 1;
> > + primary_chan_idx =
> RTW_SC_20_UPPER;
> > center_chan -= 2;
> > } else {
> > - primary_chan_idx = 3;
> > + primary_chan_idx =
> RTW_SC_20_UPPERST;
> > center_chan -= 6;
> > }
> > /* assign the center channel used
> > @@ -345,10 +345,10 @@ void rtw_get_channel_params(struct
> cfg80211_chan_def *chandef,
> > cch_by_bw[RTW_CHANNEL_WIDTH_40] =
> center_chan + 4;
> > } else {
> > if (center_freq - primary_freq == 10) {
> > - primary_chan_idx = 2;
> > + primary_chan_idx =
> RTW_SC_20_LOWER;
> > center_chan += 2;
> > } else {
> > - primary_chan_idx = 4;
> > + primary_chan_idx =
> RTW_SC_20_LOWEST;
> > center_chan += 6;
> > }
> > /* assign the center channel used
> > diff --git a/drivers/net/wireless/realtek/rtw88/main.h
> b/drivers/net/wireless/realtek/rtw88/main.h
> > index d012eefcd0da..144fbab38fd5 100644
> > --- a/drivers/net/wireless/realtek/rtw88/main.h
> > +++ b/drivers/net/wireless/realtek/rtw88/main.h
> > @@ -95,6 +95,16 @@ enum rtw_bandwidth {
> > RTW_CHANNEL_WIDTH_10 = 6,
> > };
> >
> > +enum rtw_sc_offset {
> > + RTW_SC_DONT_CARE = 0,
> > + RTW_SC_20_UPPER = 1,
> > + RTW_SC_20_LOWER = 2,
> > + RTW_SC_20_UPPERST = 3,
>
> I'll suppgest RTW_SC_20_UPMOST instead or simply RTW_SC_20_UPPEREST.
>
OK, that makes sense :)
Yan-Hsuan
^ permalink raw reply [flat|nested] 20+ messages in thread
* RE: [PATCH 05/11] rtw88: 8822c: modify rf protection setting
2019-12-24 7:39 ` Chris Chiu
@ 2019-12-24 7:50 ` Tony Chuang
2019-12-24 8:17 ` Chris Chiu
0 siblings, 1 reply; 20+ messages in thread
From: Tony Chuang @ 2019-12-24 7:50 UTC (permalink / raw)
To: Chris Chiu; +Cc: Kalle Valo, linux-wireless, Brian Norris
> On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
> >
> > From: Chien-Hsun Liao <ben.liao@realtek.com>
> >
> > According to some experiments, the original rf protection
> > setting can not perfectly make sure that there is no hardware
> > pi write during the direct write. So, modify the setting so
> > that the hardware block of pi would be turned off during the
> > direct write.
> >
>
> Sorry, I don't really understand this part. Does it mean rtw8822c_rstb_3wire()
> is to disable/enable the hardware block of PI? In this patch, I can only
> see the code block of ENABLE_PI/DISABLE_PI been removed and some
> rtw_write_rf()s been protected by new rtw8822c_rstb_3wire(). If the new
> function is to replace the ENABLE_PI/DISABLE_PI, maybe they should be
> removed in the reg.h. And It seems rtw8822c_rstb_3wire() is only for 8822c,
> means there's no such problem for 8822b?
>
Yes, rtw8822c_rstb_3wire() is to disable/enable PI. Only 8822c uses mix mode.
That means, 8822c uses direct write for RF registers except for register 0x0.
And 8822b uses sipi write (indirect). So 8822b doesn't have such problem.
Yan-Hsuan
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 05/11] rtw88: 8822c: modify rf protection setting
2019-12-24 7:50 ` Tony Chuang
@ 2019-12-24 8:17 ` Chris Chiu
0 siblings, 0 replies; 20+ messages in thread
From: Chris Chiu @ 2019-12-24 8:17 UTC (permalink / raw)
To: Tony Chuang; +Cc: Kalle Valo, linux-wireless, Brian Norris
On Tue, Dec 24, 2019 at 3:50 PM Tony Chuang <yhchuang@realtek.com> wrote:
>
> > On Fri, Dec 20, 2019 at 5:22 PM <yhchuang@realtek.com> wrote:
> > >
> > > From: Chien-Hsun Liao <ben.liao@realtek.com>
> > >
> > > According to some experiments, the original rf protection
> > > setting can not perfectly make sure that there is no hardware
> > > pi write during the direct write. So, modify the setting so
> > > that the hardware block of pi would be turned off during the
> > > direct write.
> > >
> >
> > Sorry, I don't really understand this part. Does it mean rtw8822c_rstb_3wire()
> > is to disable/enable the hardware block of PI? In this patch, I can only
> > see the code block of ENABLE_PI/DISABLE_PI been removed and some
> > rtw_write_rf()s been protected by new rtw8822c_rstb_3wire(). If the new
> > function is to replace the ENABLE_PI/DISABLE_PI, maybe they should be
> > removed in the reg.h. And It seems rtw8822c_rstb_3wire() is only for 8822c,
> > means there's no such problem for 8822b?
> >
>
> Yes, rtw8822c_rstb_3wire() is to disable/enable PI. Only 8822c uses mix mode.
> That means, 8822c uses direct write for RF registers except for register 0x0.
> And 8822b uses sipi write (indirect). So 8822b doesn't have such problem.
>
> Yan-Hsuan
Got it. Could you also explain this in the commit message so that we
won't misread
the code? Thanks.
Chris
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 01/11] rtw88: fix rate mask for 1SS chip
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
2019-12-24 7:10 ` Chris Chiu
@ 2020-01-26 15:38 ` Kalle Valo
2020-01-26 15:39 ` Kalle Valo
1 sibling, 1 reply; 20+ messages in thread
From: Kalle Valo @ 2020-01-26 15:38 UTC (permalink / raw)
To: yhchuang; +Cc: linux-wireless, briannorris
<yhchuang@realtek.com> wrote:
> From: Ping-Ke Shih <pkshih@realtek.com>
>
> The rate mask is used to tell firmware the supported rate depends on
> negotiation. We loop 2 times for all VHT/HT 2SS rate mask first, and then
> only keep the part according to chip's NSS.
>
> This commit fixes the logic error of '&' operations for VHT/HT rate, and
> we should run this logic before adding legacy rate.
>
> To access HT MCS map, index 0/1 represent MCS 0-7/8-15 respectively. Use
> NL80211_BAND_xxx is incorrect, so fix it as well.
>
> Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
> Reviewed-by: Chris Chiu <chiu@endlessm.com>
9 patches applied to wireless-drivers-next.git, thanks.
35a68fa5f96a rtw88: fix rate mask for 1SS chip
73a2d0b83424 rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M
e339b6493524 rtw88: 8822c: update power sequence to v15
3f43f10bd619 rtw88: remove unused spinlock
962562cde154 rtw88: remove unused variable 'in_lps'
65ae64d37575 rtw88: remove unused vif pointer in struct rtw_vif
fc83c616d4d9 rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop()
f48abf064ade rtw88: assign NULL to skb after being kfree()'ed
bbdd1d854e0a rtw88: change max_num_of_tx_queue() definition to inline in pci.h
--
https://patchwork.kernel.org/patch/11305077/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 01/11] rtw88: fix rate mask for 1SS chip
2020-01-26 15:38 ` Kalle Valo
@ 2020-01-26 15:39 ` Kalle Valo
0 siblings, 0 replies; 20+ messages in thread
From: Kalle Valo @ 2020-01-26 15:39 UTC (permalink / raw)
To: Kalle Valo; +Cc: yhchuang, linux-wireless, briannorris
Kalle Valo <kvalo@codeaurora.org> writes:
> <yhchuang@realtek.com> wrote:
>
>> From: Ping-Ke Shih <pkshih@realtek.com>
>>
>> The rate mask is used to tell firmware the supported rate depends on
>> negotiation. We loop 2 times for all VHT/HT 2SS rate mask first, and then
>> only keep the part according to chip's NSS.
>>
>> This commit fixes the logic error of '&' operations for VHT/HT rate, and
>> we should run this logic before adding legacy rate.
>>
>> To access HT MCS map, index 0/1 represent MCS 0-7/8-15 respectively. Use
>> NL80211_BAND_xxx is incorrect, so fix it as well.
>>
>> Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
>> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
>> Reviewed-by: Chris Chiu <chiu@endlessm.com>
>
> 9 patches applied to wireless-drivers-next.git, thanks.
>
> 35a68fa5f96a rtw88: fix rate mask for 1SS chip
> 73a2d0b83424 rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M
> e339b6493524 rtw88: 8822c: update power sequence to v15
> 3f43f10bd619 rtw88: remove unused spinlock
> 962562cde154 rtw88: remove unused variable 'in_lps'
> 65ae64d37575 rtw88: remove unused vif pointer in struct rtw_vif
> fc83c616d4d9 rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop()
> f48abf064ade rtw88: assign NULL to skb after being kfree()'ed
> bbdd1d854e0a rtw88: change max_num_of_tx_queue() definition to inline in pci.h
I dropped patches 3 and 5 as they had comments.
--
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply [flat|nested] 20+ messages in thread
end of thread, other threads:[~2020-01-26 15:39 UTC | newest]
Thread overview: 20+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-12-20 9:21 [PATCH 00/11] rtw88: some driver fixes yhchuang
2019-12-20 9:21 ` [PATCH 01/11] rtw88: fix rate mask for 1SS chip yhchuang
2019-12-24 7:10 ` Chris Chiu
2020-01-26 15:38 ` Kalle Valo
2020-01-26 15:39 ` Kalle Valo
2019-12-20 9:21 ` [PATCH 02/11] rtw88: fix TX secondary channel offset of 40M if current bw is 20M or 40M yhchuang
2019-12-20 9:21 ` [PATCH 03/11] rtw88: Use secondary channel offset enumeration yhchuang
2019-12-24 7:18 ` Chris Chiu
2019-12-24 7:44 ` Tony Chuang
2019-12-20 9:21 ` [PATCH 04/11] rtw88: 8822c: update power sequence to v15 yhchuang
2019-12-20 9:21 ` [PATCH 05/11] rtw88: 8822c: modify rf protection setting yhchuang
2019-12-24 7:39 ` Chris Chiu
2019-12-24 7:50 ` Tony Chuang
2019-12-24 8:17 ` Chris Chiu
2019-12-20 9:21 ` [PATCH 06/11] rtw88: remove unused spinlock yhchuang
2019-12-20 9:21 ` [PATCH 07/11] rtw88: remove unused variable 'in_lps' yhchuang
2019-12-20 9:21 ` [PATCH 08/11] rtw88: remove unused vif pointer in struct rtw_vif yhchuang
2019-12-20 9:21 ` [PATCH 09/11] rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop() yhchuang
2019-12-20 9:21 ` [PATCH 10/11] rtw88: assign NULL to skb after being kfree()'ed yhchuang
2019-12-20 9:21 ` [PATCH 11/11] rtw88: change max_num_of_tx_queue() definition to inline in pci.h yhchuang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).