From: greearb@candelatech.com
To: linux-wireless@vger.kernel.org
Cc: Ben Greear <greearb@candelatech.com>,
Carson Vandegriffe <carson.vandegriffe@candelatech.com>
Subject: [RFC] Revert "mt76: use IEEE80211_OFFLOAD_ENCAP_ENABLED instead of MT_DRV_AMSDU_OFFLOAD"
Date: Sat, 1 Oct 2022 07:19:20 -0700 [thread overview]
Message-ID: <20221001141920.17592-1-greearb@candelatech.com> (raw)
From: Ben Greear <greearb@candelatech.com>
This reverts commit f17f4864504d754bcbf31e4c89412cdf9946c409.
The reverted commit significantly decreases performance when running
a test where two MTK7915 radios have 16 station vdevs each, configured
for AC mode, and transmitting UDP traffic to AP.
Reported-by: Carson Vandegriffe <carson.vandegriffe@candelatech.com>
Signed-off-by: Ben Greear <greearb@candelatech.com>
---
NOTE: This patch fixes the performance regression we see, but maybe
there is a better way to fix it.
Test run before this patch:
http://www.candelatech.com/downloads/tr-398-2022-10-01-07-03-52-bad/
After this patch:
http://www.candelatech.com/downloads/tr-398-2022-10-01-07-04-50-good/
drivers/net/wireless/mediatek/mt76/mac80211.c | 8 ++++++--
drivers/net/wireless/mediatek/mt76/mt76.h | 1 +
drivers/net/wireless/mediatek/mt76/mt7915/mmio.c | 3 ++-
drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 3 ++-
4 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index 088f6de28500..bde99a478777 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -444,8 +444,12 @@ mt76_phy_init(struct mt76_phy *phy, struct ieee80211_hw *hw)
ieee80211_hw_set(hw, SUPPORTS_CLONED_SKBS);
ieee80211_hw_set(hw, SUPPORTS_AMSDU_IN_AMPDU);
ieee80211_hw_set(hw, SUPPORTS_REORDERING_BUFFER);
- ieee80211_hw_set(hw, TX_AMSDU);
- ieee80211_hw_set(hw, TX_FRAG_LIST);
+
+ if (!(dev->drv->drv_flags & MT_DRV_AMSDU_OFFLOAD)) {
+ ieee80211_hw_set(hw, TX_AMSDU);
+ ieee80211_hw_set(hw, TX_FRAG_LIST);
+ }
+
ieee80211_hw_set(hw, MFP_CAPABLE);
ieee80211_hw_set(hw, AP_LINK_PS);
ieee80211_hw_set(hw, REPORTS_TX_ACK_STATUS);
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 531f52fa282f..82ef26ca5db1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -395,6 +395,7 @@ struct mt76_hw_cap {
#define MT_DRV_SW_RX_AIRTIME BIT(2)
#define MT_DRV_RX_DMA_HDR BIT(3)
#define MT_DRV_HW_MGMT_TXQ BIT(4)
+#define MT_DRV_AMSDU_OFFLOAD BIT(5)
struct mt76_driver_ops {
u32 drv_flags;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
index c41b5b00e939..24c9fda005ca 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
@@ -658,7 +658,8 @@ struct mt7915_dev *mt7915_mmio_probe(struct device *pdev,
static const struct mt76_driver_ops drv_ops = {
/* txwi_size = txd size + txp size */
.txwi_size = MT_TXD_SIZE + sizeof(struct mt76_connac_fw_txp),
- .drv_flags = MT_DRV_TXWI_NO_FREE | MT_DRV_HW_MGMT_TXQ,
+ .drv_flags = MT_DRV_TXWI_NO_FREE | MT_DRV_HW_MGMT_TXQ |
+ MT_DRV_AMSDU_OFFLOAD,
.survey_flags = SURVEY_INFO_TIME_TX |
SURVEY_INFO_TIME_RX |
SURVEY_INFO_TIME_BSS_RX,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index ea3069d18c35..6a8f43879422 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -231,7 +231,8 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
static const struct mt76_driver_ops drv_ops = {
/* txwi_size = txd size + txp size */
.txwi_size = MT_TXD_SIZE + sizeof(struct mt76_connac_hw_txp),
- .drv_flags = MT_DRV_TXWI_NO_FREE | MT_DRV_HW_MGMT_TXQ,
+ .drv_flags = MT_DRV_TXWI_NO_FREE | MT_DRV_HW_MGMT_TXQ |
+ MT_DRV_AMSDU_OFFLOAD,
.survey_flags = SURVEY_INFO_TIME_TX |
SURVEY_INFO_TIME_RX |
SURVEY_INFO_TIME_BSS_RX,
--
2.20.1
reply other threads:[~2022-10-01 14:19 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20221001141920.17592-1-greearb@candelatech.com \
--to=greearb@candelatech.com \
--cc=carson.vandegriffe@candelatech.com \
--cc=linux-wireless@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).