| Message ID | 55813ec6-b99f-dd25-a1e2-7af9e3be3117@gmail.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [1/2] wifi: rtl8xxxu: Report the RSSI to the firmware | expand |
> -----Original Message----- > From: Bitterblue Smith <rtl8821cerfe2@gmail.com> > Sent: Saturday, December 31, 2022 12:54 AM > To: linux-wireless@vger.kernel.org > Cc: Jes Sorensen <Jes.Sorensen@gmail.com>; Ping-Ke Shih <pkshih@realtek.com> > Subject: [PATCH 2/2] wifi: rtl8xxxu: Use a longer retry limit of 48 > > The Realtek rate control algorithm goes back and forth a lot between > the highest and the lowest rate it's allowed to use. This is due to > a lot of frames being dropped because the retry limits set by > IEEE80211_CONF_CHANGE_RETRY_LIMITS are too low. (Experimentally, they > are 4 for long frames and 7 for short frames.) > > The vendor drivers hardcode the value 48 for both retry limits (for > station mode), which makes dropped frames very rare and thus the rate > control is more stable. I have similar feeling with Realtek 802.11ac chips as well, but I didn't dig further years ago. Maybe, it could have TX EVM problem. > > Because most Realtek chips handle the rate control in the firmware, > which can't be modified, ignore the limits set by > IEEE80211_CONF_CHANGE_RETRY_LIMITS and use the value 48 (set during > chip initialisation), same as the vendor drivers. > > Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)") No sure if "Fixes" is suitable to this patch, because original looks well but bad performance in real. > Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com> Reviewed-by: Ping-Ke Shih <pkshih@realtek.com> > --- > drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 9 --------- > 1 file changed, 9 deletions(-) > > diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > index f5fd3c448587..9bca5e837583 100644 > --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > @@ -6184,7 +6184,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) > { > struct rtl8xxxu_priv *priv = hw->priv; > struct device *dev = &priv->udev->dev; > - u16 val16; > int ret = 0, channel; > bool ht40; > > @@ -6194,14 +6193,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) > __func__, hw->conf.chandef.chan->hw_value, > changed, hw->conf.chandef.width); > > - if (changed & IEEE80211_CONF_CHANGE_RETRY_LIMITS) { > - val16 = ((hw->conf.long_frame_max_tx_count << > - RETRY_LIMIT_LONG_SHIFT) & RETRY_LIMIT_LONG_MASK) | > - ((hw->conf.short_frame_max_tx_count << > - RETRY_LIMIT_SHORT_SHIFT) & RETRY_LIMIT_SHORT_MASK); > - rtl8xxxu_write16(priv, REG_RETRY_LIMIT, val16); > - } > - > if (changed & IEEE80211_CONF_CHANGE_CHANNEL) { > switch (hw->conf.chandef.width) { > case NL80211_CHAN_WIDTH_20_NOHT: > -- > 2.38.0 > > ------Please consider the environment before printing this e-mail.
On 03/01/2023 10:13, Ping-Ke Shih wrote: > >> -----Original Message----- >> From: Bitterblue Smith <rtl8821cerfe2@gmail.com> >> Sent: Saturday, December 31, 2022 12:54 AM >> To: linux-wireless@vger.kernel.org >> Cc: Jes Sorensen <Jes.Sorensen@gmail.com>; Ping-Ke Shih <pkshih@realtek.com> >> Subject: [PATCH 2/2] wifi: rtl8xxxu: Use a longer retry limit of 48 >> >> The Realtek rate control algorithm goes back and forth a lot between >> the highest and the lowest rate it's allowed to use. This is due to >> a lot of frames being dropped because the retry limits set by >> IEEE80211_CONF_CHANGE_RETRY_LIMITS are too low. (Experimentally, they >> are 4 for long frames and 7 for short frames.) >> >> The vendor drivers hardcode the value 48 for both retry limits (for >> station mode), which makes dropped frames very rare and thus the rate >> control is more stable. > > I have similar feeling with Realtek 802.11ac chips as well, but I didn't > dig further years ago. Maybe, it could have TX EVM problem. > >> >> Because most Realtek chips handle the rate control in the firmware, >> which can't be modified, ignore the limits set by >> IEEE80211_CONF_CHANGE_RETRY_LIMITS and use the value 48 (set during >> chip initialisation), same as the vendor drivers. >> >> Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)") > > No sure if "Fixes" is suitable to this patch, because original looks well but > bad performance in real. > Mostly I care about the Fixes tag because it gets the patch into the stable kernels, but I can use the Cc tag for that instead. >> Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com> > > Reviewed-by: Ping-Ke Shih <pkshih@realtek.com> > >> --- >> drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 9 --------- >> 1 file changed, 9 deletions(-) >> >> diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> index f5fd3c448587..9bca5e837583 100644 >> --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> @@ -6184,7 +6184,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) >> { >> struct rtl8xxxu_priv *priv = hw->priv; >> struct device *dev = &priv->udev->dev; >> - u16 val16; >> int ret = 0, channel; >> bool ht40; >> >> @@ -6194,14 +6193,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) >> __func__, hw->conf.chandef.chan->hw_value, >> changed, hw->conf.chandef.width); >> >> - if (changed & IEEE80211_CONF_CHANGE_RETRY_LIMITS) { >> - val16 = ((hw->conf.long_frame_max_tx_count << >> - RETRY_LIMIT_LONG_SHIFT) & RETRY_LIMIT_LONG_MASK) | >> - ((hw->conf.short_frame_max_tx_count << >> - RETRY_LIMIT_SHORT_SHIFT) & RETRY_LIMIT_SHORT_MASK); >> - rtl8xxxu_write16(priv, REG_RETRY_LIMIT, val16); >> - } >> - >> if (changed & IEEE80211_CONF_CHANGE_CHANNEL) { >> switch (hw->conf.chandef.width) { >> case NL80211_CHAN_WIDTH_20_NOHT: >> -- >> 2.38.0 >> >> ------Please consider the environment before printing this e-mail.
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index f5fd3c448587..9bca5e837583 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -6184,7 +6184,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) { struct rtl8xxxu_priv *priv = hw->priv; struct device *dev = &priv->udev->dev; - u16 val16; int ret = 0, channel; bool ht40; @@ -6194,14 +6193,6 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed) __func__, hw->conf.chandef.chan->hw_value, changed, hw->conf.chandef.width); - if (changed & IEEE80211_CONF_CHANGE_RETRY_LIMITS) { - val16 = ((hw->conf.long_frame_max_tx_count << - RETRY_LIMIT_LONG_SHIFT) & RETRY_LIMIT_LONG_MASK) | - ((hw->conf.short_frame_max_tx_count << - RETRY_LIMIT_SHORT_SHIFT) & RETRY_LIMIT_SHORT_MASK); - rtl8xxxu_write16(priv, REG_RETRY_LIMIT, val16); - } - if (changed & IEEE80211_CONF_CHANGE_CHANNEL) { switch (hw->conf.chandef.width) { case NL80211_CHAN_WIDTH_20_NOHT:
The Realtek rate control algorithm goes back and forth a lot between the highest and the lowest rate it's allowed to use. This is due to a lot of frames being dropped because the retry limits set by IEEE80211_CONF_CHANGE_RETRY_LIMITS are too low. (Experimentally, they are 4 for long frames and 7 for short frames.) The vendor drivers hardcode the value 48 for both retry limits (for station mode), which makes dropped frames very rare and thus the rate control is more stable. Because most Realtek chips handle the rate control in the firmware, which can't be modified, ignore the limits set by IEEE80211_CONF_CHANGE_RETRY_LIMITS and use the value 48 (set during chip initialisation), same as the vendor drivers. Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)") Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com> --- drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 9 --------- 1 file changed, 9 deletions(-)