* [PATCH 0/2] Make ampdu tx work correctly
@ 2021-05-31 9:02 chris.chiu
2021-05-31 9:02 ` [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL chris.chiu
2021-05-31 9:02 ` [PATCH 2/2] rtl8xxxu: Fix ampdu_action to get block ack session work chris.chiu
0 siblings, 2 replies; 4+ messages in thread
From: chris.chiu @ 2021-05-31 9:02 UTC (permalink / raw)
To: Jes.Sorensen, kvalo, davem, kuba
Cc: code, linux-wireless, netdev, linux-kernel, Chris Chiu
From: Chris Chiu <chris.chiu@canonical.com>
The rtl8xxxu is the driver based on mac80211 framework, but the
ampdu tx is never working. Fix the ampdu_action and the hw capability
to enable the ampdu tx.
Chris Chiu (2):
rtl8xxxu: unset the hw capability HAS_RATE_CONTROL
rtl8xxxu: Fix ampdu_action to get block ack session work
.../net/wireless/realtek/rtl8xxxu/rtl8xxxu.h | 1 +
.../wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 28 ++++++++++++-------
2 files changed, 19 insertions(+), 10 deletions(-)
--
2.20.1
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL
2021-05-31 9:02 [PATCH 0/2] Make ampdu tx work correctly chris.chiu
@ 2021-05-31 9:02 ` chris.chiu
2021-06-03 9:51 ` Kalle Valo
2021-05-31 9:02 ` [PATCH 2/2] rtl8xxxu: Fix ampdu_action to get block ack session work chris.chiu
1 sibling, 1 reply; 4+ messages in thread
From: chris.chiu @ 2021-05-31 9:02 UTC (permalink / raw)
To: Jes.Sorensen, kvalo, davem, kuba
Cc: code, linux-wireless, netdev, linux-kernel, Chris Chiu
From: Chris Chiu <chris.chiu@canonical.com>
The HAS_RATE_CONTROL hw capability needs to be unset for the rate
control of mac80211 to work. Since the ieee80211_start_tx_ba_session
is started by the method .get_rate of rate_control_ops. We need to
unset it so the ampdu can be handled by mac80211.
Signed-off-by: Chris Chiu <chris.chiu@canonical.com>
---
drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 9ff09cf7eb62..4cf13d2f86b1 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -6678,7 +6678,6 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
/*
* The firmware handles rate control
*/
- ieee80211_hw_set(hw, HAS_RATE_CONTROL);
ieee80211_hw_set(hw, AMPDU_AGGREGATION);
wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST);
--
2.20.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH 2/2] rtl8xxxu: Fix ampdu_action to get block ack session work
2021-05-31 9:02 [PATCH 0/2] Make ampdu tx work correctly chris.chiu
2021-05-31 9:02 ` [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL chris.chiu
@ 2021-05-31 9:02 ` chris.chiu
1 sibling, 0 replies; 4+ messages in thread
From: chris.chiu @ 2021-05-31 9:02 UTC (permalink / raw)
To: Jes.Sorensen, kvalo, davem, kuba
Cc: code, linux-wireless, netdev, linux-kernel, Chris Chiu
From: Chris Chiu <chris.chiu@canonical.com>
The TID is not handled in the ampdu actions. Fix the ampdu_action
to handle the ampdu operations according to the TID. The ampdu
stop also needs to be handled by ieee80211_stop_tx_ba_cb_irqsafe
for the mac80211 to respond accordingly.
Signed-off-by: Chris Chiu <chris.chiu@canonical.com>
---
.../net/wireless/realtek/rtl8xxxu/rtl8xxxu.h | 1 +
.../wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 27 ++++++++++++-------
2 files changed, 19 insertions(+), 9 deletions(-)
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index d1a566cc0c9e..ebd69c161899 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1383,6 +1383,7 @@ struct rtl8xxxu_priv {
u8 no_pape:1;
u8 int_buf[USB_INTR_CONTENT_LENGTH];
u8 rssi_level;
+ u8 tid_bitmap;
/*
* Only one virtual interface permitted because only STA mode
* is supported and no iface_combinations are provided.
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 4cf13d2f86b1..790be4ecc3d0 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -4805,6 +4805,8 @@ rtl8xxxu_fill_txdesc_v1(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr,
struct ieee80211_rate *tx_rate = ieee80211_get_tx_rate(hw, tx_info);
struct rtl8xxxu_priv *priv = hw->priv;
struct device *dev = &priv->udev->dev;
+ u8 *qc = ieee80211_get_qos_ctl(hdr);
+ u8 tid = qc[0] & IEEE80211_QOS_CTL_TID_MASK;
u32 rate;
u16 rate_flags = tx_info->control.rates[0].flags;
u16 seq_number;
@@ -4828,7 +4830,8 @@ rtl8xxxu_fill_txdesc_v1(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr,
tx_desc->txdw3 = cpu_to_le32((u32)seq_number << TXDESC32_SEQ_SHIFT);
- if (ampdu_enable)
+ if (ampdu_enable && (priv->tid_bitmap & BIT(tid)) &&
+ (tx_info->flags & IEEE80211_TX_CTL_AMPDU))
tx_desc->txdw1 |= cpu_to_le32(TXDESC32_AGG_ENABLE);
else
tx_desc->txdw1 |= cpu_to_le32(TXDESC32_AGG_BREAK);
@@ -4876,6 +4879,8 @@ rtl8xxxu_fill_txdesc_v2(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr,
struct rtl8xxxu_priv *priv = hw->priv;
struct device *dev = &priv->udev->dev;
struct rtl8xxxu_txdesc40 *tx_desc40;
+ u8 *qc = ieee80211_get_qos_ctl(hdr);
+ u8 tid = qc[0] & IEEE80211_QOS_CTL_TID_MASK;
u32 rate;
u16 rate_flags = tx_info->control.rates[0].flags;
u16 seq_number;
@@ -4902,7 +4907,8 @@ rtl8xxxu_fill_txdesc_v2(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr,
tx_desc40->txdw9 = cpu_to_le32((u32)seq_number << TXDESC40_SEQ_SHIFT);
- if (ampdu_enable)
+ if (ampdu_enable && (priv->tid_bitmap & BIT(tid)) &&
+ (tx_info->flags & IEEE80211_TX_CTL_AMPDU))
tx_desc40->txdw2 |= cpu_to_le32(TXDESC40_AGG_ENABLE);
else
tx_desc40->txdw2 |= cpu_to_le32(TXDESC40_AGG_BREAK);
@@ -6089,6 +6095,7 @@ rtl8xxxu_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
struct device *dev = &priv->udev->dev;
u8 ampdu_factor, ampdu_density;
struct ieee80211_sta *sta = params->sta;
+ u16 tid = params->tid;
enum ieee80211_ampdu_mlme_action action = params->action;
switch (action) {
@@ -6101,17 +6108,19 @@ rtl8xxxu_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
dev_dbg(dev,
"Changed HT: ampdu_factor %02x, ampdu_density %02x\n",
ampdu_factor, ampdu_density);
- break;
+ return IEEE80211_AMPDU_TX_START_IMMEDIATE;
+ case IEEE80211_AMPDU_TX_STOP_CONT:
case IEEE80211_AMPDU_TX_STOP_FLUSH:
- dev_dbg(dev, "%s: IEEE80211_AMPDU_TX_STOP_FLUSH\n", __func__);
- rtl8xxxu_set_ampdu_factor(priv, 0);
- rtl8xxxu_set_ampdu_min_space(priv, 0);
- break;
case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
- dev_dbg(dev, "%s: IEEE80211_AMPDU_TX_STOP_FLUSH_CONT\n",
- __func__);
+ dev_dbg(dev, "%s: IEEE80211_AMPDU_TX_STOP\n", __func__);
rtl8xxxu_set_ampdu_factor(priv, 0);
rtl8xxxu_set_ampdu_min_space(priv, 0);
+ priv->tid_bitmap &= ~BIT(tid);
+ ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
+ break;
+ case IEEE80211_AMPDU_TX_OPERATIONAL:
+ dev_dbg(dev, "%s: IEEE80211_AMPDU_TX_OPERATIONAL\n", __func__);
+ priv->tid_bitmap |= BIT(tid);
break;
case IEEE80211_AMPDU_RX_START:
dev_dbg(dev, "%s: IEEE80211_AMPDU_RX_START\n", __func__);
--
2.20.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL
2021-05-31 9:02 ` [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL chris.chiu
@ 2021-06-03 9:51 ` Kalle Valo
0 siblings, 0 replies; 4+ messages in thread
From: Kalle Valo @ 2021-06-03 9:51 UTC (permalink / raw)
To: chris.chiu
Cc: Jes.Sorensen, davem, kuba, code, linux-wireless, netdev, linux-kernel
chris.chiu@canonical.com writes:
> From: Chris Chiu <chris.chiu@canonical.com>
>
> The HAS_RATE_CONTROL hw capability needs to be unset for the rate
> control of mac80211 to work. Since the ieee80211_start_tx_ba_session
> is started by the method .get_rate of rate_control_ops. We need to
> unset it so the ampdu can be handled by mac80211.
The commit log is not really describing in detail _why_ you are doing
this. Switching the rate control from hardware/firmware to mac80211 is a
major change and I want to see a good explanation why this is the right
thing to do and does not cause any regressions.
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-06-03 9:51 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-05-31 9:02 [PATCH 0/2] Make ampdu tx work correctly chris.chiu
2021-05-31 9:02 ` [PATCH 1/2] rtl8xxxu: unset the hw capability HAS_RATE_CONTROL chris.chiu
2021-06-03 9:51 ` Kalle Valo
2021-05-31 9:02 ` [PATCH 2/2] rtl8xxxu: Fix ampdu_action to get block ack session work chris.chiu
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).