In mt7996_set_coverage_class(), each phy is iterated over in calling
mt7996_mac_set_coverage_class(). Thus, the phy2 and phy3 configuration
logic in mt7996_mac_set_coverage_class() can be dropped.
Signed-off-by: Dylan Eskew <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
commit b478e162f227 ("PCI/ASPM: Consolidate link state defines") changed
PCIE_LINK_STATE_L0S (1) to (BIT(0) | BIT(1)). PCI_EXP_LNKCTL_ASPM_L0S (1)
and PCI_EXP_LNKCTL_ASPM_L1 (2) are no longer matched with
PCIE_LINK_STATE_L0S (3) and PCIE_LINK_STATE_L1 (4).
On the platform enabling ASPM L0s and L1, mt76_pci_disable_aspm is not able
to disable L1. Fix this by transforming aspm_conf to pcie link state.
Signed-off-by: Jiajia Liu <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Since mt7925_mac_sta_add publishes wcid, add publish check in mt76_sta_add
to avoid reinitializing the wcid->poll_list.
Found dev->sta_poll_list corruption when using mt7925 and 7.1-rc4.
According to the corruption information, prev->next was changed to itself.
wlan0: disconnect from AP 90:fb:5d:94:8b:e3 for new auth to 90:fb:5d:94:8b:e2
wlan0: authenticate with 90:fb:5d:94:8b:e2 (local address=84:9e:56:9c:7e:6b)
wlan0: send auth to 90:fb:5d:94:8b:e2 (try 1/3)
slab kmalloc-8k start ffff8c80958a6000 pointer offset 4160 size 8192
list_add corruption. prev->next should be next (ffff8c808a7488f8), but was ffff8c80958a7040. (prev=ffff8c80958a7040).
mt76_wcid_add_poll+0x95/0xd0 [mt76]
mt7925_mac_add_txs.part.0+0xa5/0xe0 [mt7925_common]
mt7925_rx_check+0xa7/0xc0 [mt7925_common]
mt76_dma_rx_poll+0x50d/0x790 [mt76]
mt792x_poll_rx+0x52/0xe0 [mt792x_lib]
Signed-off-by: Jiajia Liu <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Drop the unnecessary pdev->bus NULL check in mt7996_pci_probe() since
the pointer is already dereferenced earlier in mt76_pci_disable_aspm(),
making the check dead code. Silences the related Smatch warning.
Fixes: 377aa17d2aed ("wifi: mt76: mt7996: Add NPU offload support to MT7996 driver")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
In mt7925_mac_link_bss_add(), the MT7927 branch added an else-if that
checked mvif->phy->mt76->chandef.chan before assigning it to the local
chan pointer. If that check failed, chan stayed NULL and chan->band was
dereferenced unconditionally a line later.
phy->chandef.chan is set during phy registration in mt76_check_sband()
and is never cleared, so the check is a no-op. Drop it and use the
chandef channel as an unconditional fallback, matching the assumption
made elsewhere in the driver (e.g. the existing basic_rates_idx
selection right below this block and mt7925/mac.c).
Reported-by: kernel test robot <[email protected]>
Reported-by: Dan Carpenter <[email protected]>
Closes: https://lore.kernel.org/r/[email protected]/
Signed-off-by: Felix Fietkau <[email protected]>
mt76_tx_status_skb_add() zeroes the mt76_tx_cb struct stored at
info->status.status_driver_data via memset(). Since info->control and
info->status are members of the same union in ieee80211_tx_info,
this overwrites info->control.flags.
In mt7996_tx_prepare_skb(), mt76_tx_status_skb_add() is called before
mt7996_mac_write_txwi(), which re-reads info->control.flags to extract
IEEE80211_TX_CTRL_MLO_LINK. Because the field has been zeroed, the
link_id always resolves to 0 for frames using global_wcid, leading to
incorrect TXWI configuration.
Fix this by passing link_id as an explicit parameter to
mt7996_mac_write_txwi(). In mt7996_tx_prepare_skb(), the link_id is
already extracted from info->control.flags before the destructive
mt76_tx_status_skb_add() call. For the beacon and inband discovery
callers in mcu.c, use link_conf->link_id directly.
Fixes: f0b0b239b8f36 ("wifi: mt76: mt7996: rework mt7996_mac_write_txwi() for MLO support")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Link: https://patch.msgid.link/20260531-mt76_tx_status_skb_add-overwrite-fix-v2-1-b73c4b4a9798@kernel.org
Signed-off-by: Felix Fietkau <[email protected]>
For injected frames (e.g. via radiotap), mac80211 can pass
info->control.vif = NULL, as explicitly noted in struct ieee80211_tx_info.
Check vif pointer before executing ieee80211_vif_is_mld() in
mt7996_mac_write_txwi_80211 routine in order to avoid a possible NULL
pointer dereference.
Fixes: f0b0b239b8f36 ("wifi: mt76: mt7996: rework mt7996_mac_write_txwi() for MLO support")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Link: https://patch.msgid.link/20260531-mt7996_mac_write_txwi_80211-null-ptr-deref-v1-1-6dd38e1d3422@kernel.org
Signed-off-by: Felix Fietkau <[email protected]>
In mt7915_mcu_get_txpower_sku(), the response skb from
mt76_mcu_send_and_get_msg() is used in memcpy without validating
its length:
For TX_POWER_INFO_RATE:
memcpy(res, skb->data + 4, sizeof(res));
where sizeof(res) is MT7915_SKU_RATE_NUM * 2 = 322 bytes.
For TX_POWER_INFO_PATH:
memcpy(txpower, skb->data + 4, len);
In both cases, if the firmware returns a response shorter than
the expected size, the memcpy reads beyond the skb data buffer.
The data surfaces to userspace via debugfs (txpower_sku and
txpower_path).
Add length checks for both code paths before the memcpy.
Signed-off-by: Aviel Zohar <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
In mt7925_tm_query(), the response skb from mt76_mcu_send_and_get_msg()
is used in a memcpy without validating its length:
memcpy(evt_resp, skb->data + 8, MT7925_EVT_RSP_LEN);
where MT7925_EVT_RSP_LEN is 512. If the firmware returns a response
shorter than 520 bytes (8 + 512), this reads beyond the skb data
buffer. The over-read data is then returned to userspace via nla_put()
in mt7925_testmode_dump().
Add a length check before the memcpy to ensure the skb contains
sufficient data.
Fixes: c948b5da6bbe ("wifi: mt76: mt7925: add Mediatek Wi-Fi7 driver for mt7925 chips")
Signed-off-by: Aviel Zohar <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
mt792x_tx() rewrites addr1/addr2/addr3 by treating skb->data as
an 802.11 header for MLD traffic.
That is only valid for native 802.11 frames. Direct 802.3 TX can also
reach this path with IEEE80211_TX_CTL_HW_80211_ENCAP set, where
skb->data is not an 802.11 header.
Skip the MLD header rewrite for HW-encap packets to avoid corrupting
802.3 frame contents.
Fixes: ebb1406813c6 ("wifi: mt76: mt7925: add link handling to txwi")
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
With MLO, traffic for one TID can be sent on any active link. Programming
BA state only on the default link leaves the other active links out of
sync.
Program BA state on all active links instead.
Fixes: 766ea2cf5a39 ("Revert "wifi: mt76: mt7925: Update mt7925_mcu_uni_[tx,rx]_ba for MLO"")
Tested-by: Yao Ting Hsieh <[email protected]>
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
mt7925_mcu_sta_ba() currently hides the BA target behind an implicit WCID
lookup.
Pass the WCID explicitly so the caller controls which BA context is being
programmed.
No functional change intended.
Tested-by: Yao Ting Hsieh <[email protected]>
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
For MLO, the same TID can run over different links. Keeping TX BA state in
a link WCID makes the state depend on which link starts aggregation first.
Store it in the primary WCID instead, so the BA state stays stable across
links.
Fixes: 44eb173bdd4f ("wifi: mt76: mt7925: add link handling in mt7925_txwi_free")
Tested-by: Yao Ting Hsieh <[email protected]>
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Add the MT7927 USB device ID and normalize the raw 0x6639 CHIPID to
MT7927 so the shared mt7925u code can use the proper chip-specific paths.
Restore the required band setup after USB MAC reset before restarting
the device.
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
MT7927 needs DBDC enabled and uses a fixed firmware band assignment for
2.4GHz and 5/6GHz BSS contexts.
Reprogram the STA dev context when the channel context is assigned so the
firmware sees the updated band_idx before the BSS is used. This avoids
stale band programming after band changes.
Co-developed-by: Javier Tia <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Add the MT7927-specific PCIe DMA queue layout and local DMA setup,
including the extra RX ring and WFDMA/prefetch programming.
MT7927 does not follow the mt7925 DMA configuration, so provide a
separate DMA init path while reusing the existing common queue setup.
This keeps the chip-specific changes local and prepares the driver
for later MT7927 enablement without affecting existing devices.
Co-developed-by: Javier Tia <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Convert mt7925 DMA init to use the common mt792x queue allocation
helper and isolate the mt7925-specific prefetch/WFDMA setup.
This removes duplicated DMA setup code and prepares the path for
follow-up chip-specific DMA support with smaller deltas.
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
The mt792x PCIe DMA setup uses the same standard queue allocation
sequence for data, MCU, firmware download and RX rings.
Factor this part out into a small common helper so later chip support
can reuse the existing flow without duplicating the queue setup logic.
This keeps the common DMA skeleton in one place and makes follow-up
chip-specific DMA changes smaller and easier to review.
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Add a dedicated MT7927 WFSYS reset path in mt792x_wfsys_reset().
Unlike the existing connac2/connac3 reset flow that toggles the WFSYS
software reset bit and waits for init-done, MT7927 reset is driven
through CBInfra and requires polling ROMCODE_INDEX until the MCU returns
to the idle value after reset.
Keep this dormant for now: no MT7927 PCI IDs are added by this patch, so
it only prepares the reset logic without making the driver bind to MT7927
hardware yet.
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Firmware may return wlan_failure when deactivating a BSS that is already
inactive. This is a valid teardown case and should not fail the remove
path.
Keep activation failures unchanged since they still indicate that firmware
failed to create or activate the BSS state.
Signed-off-by: Sean Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Disable PCIe ASPM unconditionally for MT7927. The CONNINFRA power
domain and WFDMA register access are unreliable with PCIe L1 active,
causing throughput to drop from 1+ Gbps to ~200 Mbps.
Disable runtime PM and deep sleep for MT7927. The combo chip shares
a CONNINFRA power domain between WiFi (PCIe) and BT (USB).
SET_OWN/CLR_OWN transitions on the LPCTL register crash the BT
firmware, requiring a full power cycle to recover. PM enablement will
be addressed in a follow-up once safe power state transitions are
determined.
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
The mac_reset and resume paths use the hardcoded MT_INT_RX_DONE_ALL
constant (bits 0-2) to re-enable RX interrupts. This is correct for
MT7925 (RX rings 0, 1, 2) but wrong for chips using different ring
indices.
Define a per-chip irq_map with the correct RX interrupt enable bits and
replace hardcoded MT_INT_RX_DONE_ALL with irq_map field reads in the
resume and mac_reset paths. Add the MT7927 irq_map with interrupt bits
matching its RX ring layout (rings 4, 6, 7), selected at probe time
based on PCI device ID.
This ensures the correct interrupt bits are enabled regardless of the
chip variant.
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Tested-by: 张旭涵 <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
mt7925_init_eht_caps() only populates EHT MCS/NSS maps for BW <= 80
and BW = 160, but never sets BW = 320. This means iw phy shows no
320MHz MCS map entries even though the hardware supports 320MHz
operation in the 6GHz band.
Add the missing 320MHz capability bits for 6GHz:
- PHY_CAP0: IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ
- PHY_CAP1: beamformee SS for 320MHz
- PHY_CAP2: sounding dimensions for 320MHz
- PHY_CAP6: MCS15 support for 320MHz width
- MCS/NSS: populate bw._320 maps for 6GHz band
Introduce is_320mhz_supported() to gate 320MHz on MT7927 only, since
MT7925 does not support 320MHz operation.
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Tested-by: 张旭涵 <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
The sta_rec_eht structure has a mcs_map_bw320 field, and the channel
width mapping includes NL80211_CHAN_WIDTH_320, but the 320MHz MCS/NSS
map was never copied from the station's EHT capabilities to the MCU TLV.
This prevents negotiation of 320MHz channel width even when both the
hardware and firmware advertise support for it.
Add the missing memcpy for the 320MHz MCS map, matching the existing
pattern for BW20, BW80, and BW160.
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Tested-by: 张旭涵 <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
The RX vector (RXV) and TX status (TXS) parsing in mac.c lack handling
for 320MHz channel width. When the hardware reports 320MHz in the
bandwidth field, mt7925_mac_fill_rx_rate() returns -EINVAL and
mt7925_mac_add_txs_skb() records no bandwidth stats.
Add IEEE80211_STA_RX_BW_320 cases to both functions. The RXV parser
also handles BW_320+1 since the hardware can report 320MHz in two
adjacent encoding positions.
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Tested-by: 张旭涵 <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
bss_rlm_tlv() in mt7925_mcu_bss_rlm_tlv() has no case for
NL80211_CHAN_WIDTH_320. When associated to a 320MHz BSS, the switch
falls through to default and sends bw=0 (CMD_CBW_20MHZ) to firmware
via BSS_RLM TLV. Firmware then configures the RX radio for 20MHz
and cannot decode the AP's 320MHz frames, resulting in complete data
path failure at 320MHz.
Add the missing NL80211_CHAN_WIDTH_320 case with CMD_CBW_320MHZ and
center_chan2.
Tested on ASUS RT-BE92U: 320MHz throughput goes from 0 Mbps to
841 Mbps (iperf3 -t30 -P8), PHY 4803 Mbps EHT-MCS11.
Reported-by: 张旭涵 <[email protected]>
Closes: https://github.com/openwrt/mt76/issues/927
Tested-by: 张旭涵 <[email protected]>
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
In the error path of mt7925_change_vif_links(), the free: label iterates
over link_ids to clean up, but compares against `mconf` and `mlink`
which hold stale values from the last loop iteration rather than the
current link_id being freed.
Use array-indexed access (mconfs[link_id] / mlinks[link_id]) to compare
against the correct per-link pointers.
Fixes: 69acd6d910b0 ("wifi: mt76: mt7925: add mt7925_change_vif_links")
Tested-by: Marcin FM <[email protected]>
Tested-by: Cristian-Florin Radoi <[email protected]>
Tested-by: George Salukvadze <[email protected]>
Tested-by: Evgeny Kapusta <[email protected]>
Tested-by: Samu Toljamo <[email protected]>
Tested-by: Ariel Rosenfeld <[email protected]>
Tested-by: Chapuis Dario <[email protected]>
Tested-by: Thibaut François <[email protected]>
Tested-by: 张旭涵 <[email protected]>
Reviewed-by: Sean Wang <[email protected]>
Signed-off-by: Javier Tia <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
When MT76_NPU and CONFIG_NET_MEDIATEK_SOC_WED are enabled and
mt76 detects properly the Airoha NPU SoC, mt7996_init_tx_queues() will
dereference a NULL WED pointer.
Fix the issue by always passing the WED pointer from mt7996_dma_init().
Fixes: cd7951f242a7 ("wifi: mt76: mt7996: Integrate MT7990 dma configuration for NPU")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Link: https://patch.msgid.link/[email protected]
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
mt76_put_vif_phy_link() frees the offchannel mlink with plain kfree()
after rcu_assign_pointer(NULL). However, rcu_assign_pointer only prevents
future RCU readers from obtaining the pointer -- it does not wait for
existing readers that already hold it via rcu_dereference.
The TX datapath (e.g. mt7996_mac_write_txwi) dereferences mlink->wcid
and mlink->idx under rcu_read_lock. If a TX softirq obtained the pointer
via rcu_dereference just before the NULL assignment, it will dereference
freed memory after the kfree.
struct mt76_vif_link already contains an rcu_head field that is unused at
this free site -- a developer oversight, since the adjacent
kfree_rcu_mightsleep call for rx_sc in the same function shows the
pattern was understood.
Replace kfree(mlink) with kfree_rcu(mlink, rcu_head).
Fixes: a8f424c1287c ("wifi: mt76: add multi-radio remain_on_channel functions")
Signed-off-by: Rajat Gupta <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
On a STATION vif, removing a TDLS peer takes the mt7925_mac_sta_remove
-> mt7925_mac_sta_remove_links path. The first loop in that function
calls mt7925_mcu_add_bss_info(..., enable=false) for every link of the
station being removed. For a non-MLO STATION vif there is exactly one
link, link 0, whose bss_conf is the AP's. TDLS peers do not have their
own bss_conf - they share the AP's BSS.
The result is that every TDLS peer teardown sends a BSS_INFO_UPDATE
with enable=0 for the AP's BSS to the firmware, which wipes the AP-side
rate-control context. The connection stays associated and TX from the
host still works at the negotiated rate, but the AP's downlink to us
collapses to the lowest mandatory OFDM rate (HE-MCS 0 / 6 Mbit/s OFDM)
and only slowly recovers as rate adaptation re-learns under sustained
traffic. With brief or bursty traffic the link can stay at 6-72 Mbit/s
indefinitely, requiring a manual reconnect.
mt7925_mac_link_sta_remove() already guards its own
mt7925_mcu_add_bss_info(..., false) call with
"vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls".
Add the equivalent guard at the top of the cleanup loop in
mt7925_mac_sta_remove_links(), above the link_sta / link_conf /
mlink / mconf lookups, so TDLS peer teardown skips the loop body
entirely without doing the per-link work that would just be thrown
away.
Verified on mt7925e by triggering Samsung-S938B auto-TDLS via iperf3
and watching iw rx bitrate after teardown:
Before: rx bitrate collapses to 6.0-72.0 Mbit/s, oscillates 17/72/
137/288/432 Mbit/s for 30+ seconds, no full recovery without
a manual reassoc.
After: rx bitrate stays at 1200.9 Mbit/s HE-MCS 11 NSS 2 80 MHz
across the entire TDLS lifecycle.
bpftrace confirms a single mt7925_mcu_add_bss_info(enable=0) call per
teardown before the fix; zero such calls after.
Fixes: 3878b4333602 ("wifi: mt76: mt7925: update mt7925_mac_link_sta_[add, assoc, remove] for MLO")
Cc: [email protected]
Signed-off-by: ElXreno <[email protected]>
Assisted-by: Claude:claude-opus-4-7 bpftrace
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
With HW TX encap offload enabled, the mt76 firmware builds the 802.11
header for the 802.3 frame using the per-WCID context. For a STATION
vif the HDR_TRANS TLV currently sets ToDS=1, which makes the firmware
default to the BSSID as A1 and emit STA->AP-formatted frames
regardless of which peer the WCID points to.
For TDLS-paired peers this is wrong. Data frames go on air addressed
to the AP, the AP MAC-ACKs and silently drops them per IEEE 802.11z
(an AP must not forward to a TDLS-paired peer). Management and
control frames bypass the HW encap path and still reach the peer;
only user data fails.
Add MT_WCID_FLAG_TDLS_PEER, set it in mt7915, mt7921, mt7925 and
mt7996 sta-add paths when sta->tdls is true, and override the
HDR_TRANS TLV in mt76_connac_mcu_wtbl_hdr_trans_tlv() (Connac2 -
mt7915 / mt7921 / mt7922), mt7925_mcu_sta_hdr_trans_tlv() (mt7925)
and mt7996_mcu_sta_hdr_trans_tlv() (mt7996) to set ToDS=0, FromDS=0
when the flag is set. The 3-addr non-DS form matches what 802.11z
uses for direct links; the firmware then constructs the frame with
A1=peer rather than A1=BSSID. HW encap offload remains enabled for
AP and any non-TDLS traffic.
Verified on mt7925e + Samsung S938B over a 5 GHz HE 80 MHz channel
with iperf3 -t 30 to the TDLS peer:
before fix: over the TDLS direct link, 7 TDLS Setup action
frames and 3 RTS frames reach the peer; 0 QoS
Data frames make it through (mgmt/control paths
bypass HW encap, the data path does not). iperf3
stalls.
after fix: 2.90 GBytes transferred at 830 Mbit/s sustained,
0 TCP retransmits.
mt7915, mt7921, mt7922 and mt7996 are not regression-tested in this
change for lack of hardware. Their HDR_TRANS handling mirrors the
verified mt7925 change; the firmware behavior is shared across these
chips.
Signed-off-by: ElXreno <[email protected]>
Assisted-by: Claude:claude-opus-4-7 bpftrace tcpdump
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
This patch is based on a BUG as reported by Bongani Hlope at
https://lore.kernel.org/all/[email protected]/
When a channel-switch announcement (CSA) beacon is received,
cfg80211 queues a wiphy work item that eventually calls
mt7921_channel_switch_rx_beacon(). If the station disconnects
(or the channel context is otherwise torn down) between the
time the work is queued and the time it runs, the driver's
dev->new_ctx pointer can already have been cleared to NULL.
mt7921_channel_switch_rx_beacon() then dereferences new_ctx
unconditionally, triggering a NULL pointer dereference at
address 0x0:
BUG: kernel NULL pointer dereference, address: 0000000000000000
RIP: 0010:mt7921_channel_switch_rx_beacon+0x1f/0x100 [mt7921_common]
The same missing guard exists in mt7925_channel_switch_rx_beacon(),
which shares the same code pattern introduced by the same commit.
Add an early-return NULL check for dev->new_ctx in both
mt7921_channel_switch_rx_beacon() and
mt7925_channel_switch_rx_beacon(). When new_ctx is NULL there is
no pending channel switch to process, so returning immediately is
the correct and safe action.
Fixes: 8aa2f59260eb ("wifi: mt76: mt7921: introduce CSA support")
Reported-by: Bongani Hlope <[email protected]>
Oops-Analysis: http://oops.fenrus.org/reports/lkml/[email protected]/report.html
Link: https://lore.kernel.org/all/[email protected]/
Signed-off-by: Arjan van de Ven <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Felix Fietkau <[email protected]>
Cc: Lorenzo Bianconi <[email protected]>
Cc: Ryder Lee <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
When pcim_iomap_region() or devm_kmemdup() fail, the code returns
directly without cleaning up previously allocated resources:
- mt76_device allocated by mt76_alloc_device()
- pci irq vectors allocated by pci_alloc_irq_vectors()
Fix this by jumping to the existing error cleanup path instead of
returning directly.
Fixes: ee5bb35d2b83 ("wifi: mt76: mt7921: switch to pcim_iomap_region")
Signed-off-by: Hongling Zeng <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Felix Fietkau <[email protected]>
