24.10-nss #41
Closed
24.10-nss #41
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
feeds: Switch to fork that requires nss qdisc Without it, NSS based SQM doesn't work.
Signed-off-by: bitthief <[email protected]> package: kernel: nat46: add kernel 6.1 support Signed-off-by: bitthief <[email protected]> Co-Developed-by: Sean Khan <[email protected]>
Instead of 'install', 'cp', use standard build functions Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: bitthief <[email protected]> Co-Developed-by: Sean Khan <[email protected]>
Signed-off-by: bitthief <[email protected]> qualcommax: dts: provide label for NSS reserved-memory Provide a label for the NSS reserved-memory node so it can be easily passed to the NSS DRV instead of having to global match by name which is fragile. Signed-off-by: bitthief <[email protected]>
Signed-off-by: bitthief <[email protected]>
Signed-off-by: bitthief <[email protected]>
Attempt at getting NSS WiFi Offload working on kernel 6.1 (backport 6.5) These patches are a mix of @Ansuel branch for 5.15 (ipq807x-5.15-ecm-wifi) Ansuel@3ecaee7 And QUIC's upstream NSS patches on [NHSS.QSDK.12.4.5.r3](https://git.codelinaro.org/clo/qsdk/oss/system/feeds/wlan-open.git) For the actual package setup (Makfiles,ath.mk) Qualcomm's branch is under 'wlan-open/NHSS.QSDK.12.4.5.r3' Looks like they've swtiched to using backports 6.5-rc3, and figured to use a mix of their patches on 'NHSS.QSDK.12.4.5.r3' (nss) 'wlan-open/NHSS.QSDK.12.4.5.r3' (syncing with openwrt main). Booting with "nss_offload=1 frame_mode=2" Logs show: [ 16.606282] WARNING: CPU: 2 PID: 3524 at ath11k_nss_tx+0x1d4/0x1e0 [ath11k] [ 16.611060] Modules linked in: ecm pppoe ppp_async nft_fib_inet nf_flow_table_inet ath11k_ahb ath11k pptp pppox ppp_generic nft_reject_ipv6 nft_reject_ipv4 nft_reject_inet nft_reject nft_redir nft_quota nft_objref nft_numgen nft_nat nft_masq nft_log nft_limit nft_hash nft_fullcone nft_flow_offload nft_fib_ipv6 nft_fib_ipv4 nft_fib nft_ct nft_compat nft_chain_nat nf_tables nf_nat nf_flow_table nf_conntrack_netlink nf_conntrack mac80211 iptable_mangle iptable_filter ipt_REJECT ipt_ECN ip_tables cfg80211 xt_time xt_tcpudp xt_tcpmss xt_statistic xt_multiport xt_mark xt_mac xt_limit xt_length xt_hl xt_ecn xt_dscp xt_comment xt_TCPMSS xt_LOG xt_HL xt_DSCP xt_CLASSIFY x_tables wireguard slhc sch_cake qrtr_smd qrtr qmi_helpers nfnetlink nf_reject_ipv6 nf_reject_ipv4 nf_log_syslog nf_defrag_ipv4 mhi libcrc32c libchacha20poly1305 compat sch_tbf sch_ingress sch_htb sch_hfsc em_u32 cls_u32 cls_route cls_matchall cls_fw cls_flow cls_basic act_skbedit act_mirred act_gact qca_nss_cfi_cryptoapi [ 16.611215] qca_nss_crypto qca_nss_qdisc qca_nss_wifi_meshmgr qca_nss_gre ledtrig_usbport qca_mcs bonding ip6_gre ip_gre gre ifb nat46 nf_defrag_ipv6 sit qca_nss_drv ip6_tunnel tunnel6 tunnel4 nls_utf8 nls_iso8859_1 nls_cp437 vxlan seqiv michael_mic uas usb_storage leds_gpio xhci_plat_hcd xhci_pci xhci_hcd dwc3 dwc3_qcom qca_nss_dp qca_ssdk ramoops reed_solomon pstore gpio_button_hotplug ext4 mbcache jbd2 aquantia hwmon crc_ccitt crc32c_generic [ 16.721723] CPU: 2 PID: 3524 Comm: hostapd Not tainted 6.1.68 #0 [ 16.743958] Hardware name: Dynalink DL-WRX36 (DT) [ 16.750032] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 16.754635] pc : ath11k_nss_tx+0x1d4/0x1e0 [ath11k] [ 16.761403] lr : ath11k_nss_tx+0x1bc/0x1e0 [ath11k] [ 16.766264] sp : ffffffc00b1db790 [ 16.771122] x29: ffffffc00b1db790 x28: 0000000000000038 x27: 0000000000000000 [ 16.774602] x26: ffffff8006dd7430 x25: ffffff8007ea9dc8 x24: 0000000060000012 [ 16.781719] x23: ffffff8005c6a060 x22: ffffff8005c6a060 x21: ffffff80072f12d8 [ 16.788837] x20: ffffff8007ea9dc8 x19: ffffff8006082800 x18: 0000000000000005 [ 16.795957] x17: 6976312065707974 x16: 207061636e652062 x15: 6b73207874207373 [ 16.803075] x14: ffffffc00a0dd518 x13: 000000000000018b x12: 000000000000018b [ 16.810192] x11: 00000000ffffffea x10: ffffffc00a135518 x9 : 0000000000000001 [ 16.817310] x8 : 0000000000000001 x7 : 0000000000017fe8 x6 : c0000000ffffefff [ 16.824429] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000000 [ 16.831547] x2 : ffffff80072f1370 x1 : 0000000000000000 x0 : 0000000000000001 [ 16.838665] Call trace: [ 16.845774] ath11k_nss_tx+0x1d4/0x1e0 [ath11k] [ 16.848035] ath11k_mac_tx_mgmt_pending_free+0x3034/0x9600 [ath11k] [ 16.852551] ieee80211_handle_wake_tx_queue+0x68/0xb10 [mac80211] [ 16.858801] ieee80211_probereq_get+0xca4/0x11f0 [mac80211] [ 16.865049] ieee80211_tx_prepare_skb+0x1dc/0x240 [mac80211] [ 16.870432] ieee80211_xmit+0xcc/0x120 [mac80211] [ 16.876333] __ieee80211_subif_start_xmit+0x2b8/0x380 [mac80211] [ 16.880937] ieee80211_subif_start_xmit+0x40/0x3b0 [mac80211] [ 16.887013] ieee80211_subif_start_xmit_8023+0xb8/0x430 [mac80211] [ 16.892656] dev_hard_start_xmit+0x8c/0x110 [ 16.898728] __dev_queue_xmit+0x1f0/0xbc0 [ 16.902808] dev_queue_xmit+0x14/0x20 [ 16.906975] packet_sendmsg+0x768/0x1260 [ 16.910620] __sys_sendto+0xdc/0x140 [ 16.914613] __arm64_sys_sendto+0x28/0x40 [ 16.918174] invoke_syscall.constprop.0+0x5c/0x110 [ 16.922081] do_el0_svc+0x58/0x170 [ 16.926765] el0_svc+0x18/0x60 [ 16.930149] el0t_64_sync_handler+0x114/0x120 [ 16.933190] el0t_64_sync+0x174/0x178 [ 16.933197] ---[ end trace 0000000000000000 ]--- [ 16.933230] ath11k c000000.wifi: failed to transmit frame -22 [ 17.041495] ath11k c000000.wifi: encap mismatch in nss tx skb encap type 1vif encap type 2 [ 17.041578] ath11k c000000.wifi: failed to transmit frame -22 Which is confusing since the parameter is definitley set to '2'. ➤ cat /sys/module/ath11k/parameters/frame_mode 2 Booting with either "nss_offload=1 frame_mode=3" (RAW), or "nss_offload=1 frame_mode=1" (Native Wifi) Results in: [ 15.644742] ath11k c000000.wifi: peer not found for nss peer delete [ 15.744742] ath11k c000000.wifi: peer not found for nss peer delete [ 15.745742] ath11k c000000.wifi: peer not found for nss peer delete [ 15.746682] ath11k c000000.wifi: peer not found for nss peer delete Clients connect and then are immediately kicked off, stuck in a loop. ath11k-nss: Wifi offloading working (unstable) Wifi offloading seems to be working, however it is not stable. Logs will initially show a lot of flooding. ``` [Sun Dec 17 02:43:25 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:28 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:30 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:32 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:34 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:40 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:41 2023] ath11k c000000.wifi: peer not found for nss peer delete [Sun Dec 17 02:43:46 2023] ath11k c000000.wifi: peer not found for nss peer delete ``` But eventually ends up subsiding. Needs heavy testing... ath11k_nss: Add mac address to debug `nss peer delete` warnings also delete ath10k patch, and refresh. ath11k_nss: FIX `nss peer delete` + encap errors This should hopefullly fix the following errors: ``` [Sun Dec 24 22:03:21 2023] ath11k c000000.wifi: encap mismatch in nss tx skb encap type 1vif encap type 2 [Sun Dec 24 22:03:21 2023] ath11k c000000.wifi: failed to transmit frame -22 [Sun Dec 24 22:08:25 2023] ath11k c000000.wifi: encap mismatch in nss tx skb encap type 1vif encap type 2 [Sun Dec 24 22:08:25 2023] ath11k c000000.wifi: failed to transmit frame -22 ``` Clients should also be able to switch between 2Ghz and 5Ghz AP, as well as join using password (no longer required to boot with NSS off first, connect the client to acquire PSK, and reboot back with nss wifi offload enabled). Uptime has been 5 hours and so far no issues. ath11k_nss: Remove unecessary clang-tidy formatting ath11k_nss: add 'debug_mode' flag quiet warnings Will properly handle: ``` [Mon Dec 25 16:51:34 2023] ath11k c000000.wifi: encap mismatch in nss tx skb encap type 1 vif encap type 2 ``` should take the path for native wifi encap ath11k_nss: Rework depends, make `nss_redirect` optional 1.) Added a reworked `qca-nss-pbuf` init script from QSDK that will set sysctl NSS `n2hcfg` wifi options based on board type and available RAM. 2.) ath11k is fully capable of offloading wifi, it does not need mac80211 to create any NSS vifs. This lowers overhead and has actually shown considerable speed improvements for CPU programs on the router side (i.e. SMB). I believe ax3600's IoT (ath10k) would require it for offloading. I've added 2 new makemenu options, `MAC80211_NSS_SUPPORT` which enables NSS related features, and `MAC80211_NSS_REDIRECT` which provides the option to autoload the module on boot with `nss_redirect=1`. 3.) Reverted the option that deafulted `ATH11K_MEM_PROFILE_512M` to true. This was not required for platforms with 1GB or more memory (which is most of them). The default is 1GB. Select for platforms like Xiaomi AX3600. ath11k_nss: remove requirement for '/sys/kernel/debug/ath11k' ath11k_nss: set default values lower to avoid eating up memory ath11k_nss: Add NSS VLAN support + improvements 1) tweaked qca-nss-pbuf script to better handle different board configs 2) added some QSDK NSS patches for dynamic vlan, WDS. 3) added QSDK fixes for beacon, and bt-coex For devices that keep getting disconnected (mostly on 2G). It is recommended to use the following settings for wifi. Change for your country. config wifi-device 'radio0' option type 'mac80211' option path 'platform/soc/c000000.wifi' option band '5g' option txpower '24' option country 'PA' option channel '100' option htmode 'HE160' option cell_density '0' config wifi-iface 'default_radio0' option device 'radio0' option network 'lan' option mode 'ap' option encryption 'psk2' option key 'SOME_PASSWORD' option ssid 'EXAMPLE_5G' option dtim_period '3' option max_inactivity '86400' option disassoc_low_ack '0' config wifi-device 'radio1' option type 'mac80211' option path 'platform/soc/c000000.wifi+1' option band '2g' option htmode 'HT20' option channel '1' option txpower '36' option country 'PA' option cell_density '0' config wifi-iface 'default_radio1' option device 'radio1' option network 'lan' option mode 'ap' option ssid 'EXAMPLE_2G' option encryption 'psk2' option key 'SOME_PASSWORD' option max_inactivity '86400' option dtim_period '3' option disassoc_low_ack '0' WPA3 should be disabled, and left as WPA2. Experimenting with max_inactivity, dtim_period, and disassoc_low_ack to hopefully resolve the deauth disconnects.
`dev.nss.n2hcfg.n2h_wifi_pool_buf` must be set BEFORE setting `dev.nss.n2hcfg.n2h_high_water_core0`, otherwise it resets the value.
ath11k_nss: bugfix change start order value 'dev.nss.n2hcfg.n2h_wifi_pool_buf' was not getting set, wait till after qca-nss-drv is loaded. ath11k_nss: finally fix n2hcfg values not being set
Memory allocations in the driver & mac80211 are logged and populate those values to the user space via debugfs. This stats will give the snapshot of the memory being used by the driver at the time of dumping these memory stats. Command: cat /sys/kernel/debug/ath11k/ipq8074\ hw2.0/memory_stats Sample output of the stats MEMORY STATS IN BYTES: malloc size : 6287583 ce_ring_alloc size: 109308 dma_alloc size:: 10831860 htc_skb_alloc size: 3840 wmi alloc size: 0 per peer object: 4644 rx_post_buf size: 5091840 Total size: 22329075 User can disable/enable the memory stats accounting with the below command. echo N > /sys/kernel/debug/ath11k/ipq8074\ hw2.0/enable_memory_stats where N = 0 to disable logging, 1 to enable the logging. Note: This should be enabled/disabled only after wifi is down. User shouldn't enable/disable when the wifi is up to avoid accounting the negative values which cause incorrect values in the memory stats. Command: cat /sys/kernel/debug/ieee80211/phyX/memory_stats memory stats: malloc_size: 108
In case you want to specify the memory profile to use rather than letting it autodetect. The following new uci option is introduced. pbuf. config general opt option memory_profile '' choices are 1gb, 512mb, 256mb. Delete option to let it autodetect. NOTE: You must reboot after changing these settings, for it to autoapply. ath11k_nss: fix compilation and rename some patches
also support offload for the following:
mac80211:
* vlan
* mesh (partial)
ath11k:
* vlan
bugfixes:
* Rreverted patch `105-ath11k-fix-monitor-crash-if-tx-offload-is-enabled.patch`.
Initial test show lower memory use. `ieee80211_tx_status_8023` was
also removed upstream for a reason as it wasn't being used, and the
logic behind it was flawed.
See https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]
* Mmemory profile for '256M' was missing logic that was in 512/1G.
Should build correctly now
* added more qsdk related patches to nss/offload
ath11k_nss: fix up patches
ath11k_nss: revert m3, fix coredump when rebooting
M3 SSR dump logic makes the router coredump. Just putting dummy case to
skip warning.
[12394.072384] Hardware name: Dynalink DL-WRX36 (DT)
[12394.079758] Call trace:
[12394.084354] dump_backtrace.part.0+0xbc/0xd0
[12394.086614] show_stack+0x18/0x30
[12394.091127] dump_stack_lvl+0x6c/0x88
[12394.094339] dump_stack+0x18/0x34
[12394.097985] bad_page+0xe0/0x110
[12394.101282] __free_pages_ok+0x33c/0x360
[12394.104582] __free_pages+0xbc/0xe0
[12394.108487] dma_direct_free+0xd0/0x140
[12394.111701] dma_free_attrs+0x90/0xb0
[12394.115519] ath11k_qmi_fwreset_from_cold_boot+0x9e0/0xa10 [ath11k]
[12394.119343] ath11k_qmi_deinit_service+0x64/0x21d0 [ath11k]
[12394.125419] ath11k_core_deinit+0xa4/0xc0 [ath11k]
[12394.130974] 0xffffffc000fb3600
[12394.135831] platform_shutdown+0x24/0x40
[12394.138871] device_shutdown+0x14c/0x240
[12394.143037] kernel_restart+0x40/0xb0
[12394.146944] __do_sys_reboot+0xcc/0x200
[12394.150502] __arm64_sys_reboot+0x24/0x30
[12394.154148] invoke_syscall.constprop.0+0x5c/0x110
[12394.158317] do_el0_svc+0x58/0x170
[12394.163001] el0_svc+0x18/0x60
[12394.166386] el0t_64_sync_handler+0x114/0x120
[12394.169426] el0t_64_sync+0x174/0x178
uci option pbuf.opt.memory_profile must be explicity set to auto, 1gb, 512m, 256m to run.
ath11k_nss: fix build for 256/1G mem, and ath10k ath11k_nss: bump release version '8' mac80211: refactor NSS patches Since NSS requires patches to subsys, and ath*k directories. Move patches into a subset of nss for better tracking against QSDK, and modularization. ath11k_nss: rename patches ath11k_nss: clean up optional patches To reduce bug tracking headaches, I've remove the following patches, as they are not required for NSS offload and have been around the last 2-3 years without ever being upstreamed. nss/ath11k/235-001-ath11k-Add-support-for-beacon-tx-mode.patch nss/ath11k/237-002-ath11k-Add-provision-to-configure-rx-hashmap.patch nss/ath11k/902-020-ath11k-add-btcoex-config.patch nss/ath11k/902-022-ath11k-add-ap-ps-support.patch nss/ath11k/907-068-ath11k-add-rx-histogram-stats.patch nss/ath11k/907-108-ath11k-enable-ul-ofdma-ru-allocation-in-peer-stats.patch nss/ath11k/911-373-ath11k-Add-retry-mechanism-for-update_rx_qu.patch nss/ath11k/913-353-ath11k-ignore-frags-from-uninitialized-peer-in-dp.patch nss/ath11k/913-356-ath11k-invalid-desc-sanity-check.patch nss/ath11k/913-374-ath11k-Check-skb_headroom-before-using-skb_push.patch nss/ath11k/913-830-ath11k-Avoiding-memset-of-ppdu-info-for-next-skb.patch nss/subsys/235-002-mac80211-Add-support-for-beacon-tx-mode.patch nss/subsys/913-726-mac80211-fix-crash-when-accessing-null-pointer.patch ath11k_nss: Remove superfluous patches Remove patches unrelated to NSS offloading to minimize bloat and better track NSS related issues. ath11k_nss: Refactor patches to use upstream names Reworked patches to use upstream QSDK names. Allows for better tracking ath11k_nss: align wifi offload with qca-nss-drv The option in qca-nss-drv is actually 'NSS_DRV_WIFIOFFLOAD_ENABLE' use the same syntax. ath10k-ct: fix compile with NSS wifi ath11k_nss: Merge every NSS related feature + more * Added macro to disable NSS mesh offload * Added menuconfig option "ATH11K_NSS_MESH_SUPPORT" to selectivley build mesh support, as well it's depenacndy on nss-drv-wifimeshmgr. * Added option to disable HTT Stats, and STA stats (stations). * Reducing footprint by ~210KB. Debugfs minimal is still enabled. * Reworked a TON of patches, some my own, hopefully there should be a far less amount of WOA2/WPA3 connection issues. * Updated the /etc/init.d/pbuf script to be more robust. (handles tweaking ath11k and NSS settings better) NOTES: Although mesh package builds (nss-drv-wifimeshmgr), ath11k doesn't seem to support it yet. Not sure if NSS requires 3 radios to work, I'm at a dead end currently with that route. ATTENTION: the ptch `37-006-ath11k-Allow-fast-rx-by-bypassing-stats-update.patch` works well on nss with frame_mode=2. And on `ath11k frame_mode=1 nss_offload=0` And on `ath11k frame_mode=2 nss_offload=1` if you set nss_offload=off and frame_mode=2, it will CRASH if you set nss_offload=on and frame_mode=2, it will RUN if you set nss_offload=off and frame_mode=1, it will RUN ath11k_nss: fix n2h high_water_core0/wifi_pool_buf These were commented out, but looks like they are needed to prevent lock ups with heavy usage apps (users report in torrenting) ath11k_nss: Renumber ath11k patches in the range commit 3c7cc4b725ea406f19b736427034e3bdb436aedc Author: Yuvasree Sivasankaran <[email protected]> AuthorDate: Thu Jan 4 11:25:56 2024 +0530 Commit: Yuvasree Sivasankaran <[email protected]> CommitDate: Wed Jan 3 22:53:51 2024 -0800 wifi: ath11k: Renumber ath11k patches in the range In ath11k, patches are not maintained in the range and not sequential. Renumber the patches sequential and in the range. Change-Id: I77c51c0f5bf9f94863db4ef364b156e14465a60c Signed-off-by: Yuvasree Sivasankaran <[email protected]> ath11k_nss: Add mac hw flag to avoid tx queue in mac80211 commit 4e9b5f7f0d1ed40dbf3208f7ed4448e49b4a4ac1 Author: Yuvasree Sivasankaran <[email protected]> AuthorDate: Wed Dec 6 12:20:59 2023 +0530 Commit: Yuvasree Sivasankaran <[email protected]> CommitDate: Mon Dec 18 12:52:33 2023 +0530 wifi: mac80211: Add mac hw flag to avoid tx queue in mac80211 Queue SKB in mac80211 become mandatory from latest 6.1 kernel. Because of this queuing, there will be performance degradation. Add hw flag option to enable tx queue in Driver/Hardware. Driver/hardware can register for HAS_TX_QUEUE HW flag and avoid tx queuing in mac80211. Add same HW flag checks to avoid accessing skb queues which will be NULL or invalid and also NULL checks for sta txqs for NULL or invalid access. ath11k_nss: add the HTC+ / iPhone fix commit ccdca73cd65723c3cb63c17edc95c4c43318cb38 Author: John Crispin <[email protected]> AuthorDate: Sun Jul 9 17:12:34 2023 +0200 Commit: John Crispin <[email protected]> CommitDate: Thu Aug 31 16:08:34 2023 +0200 mac80211: add the HTC+ / iPhone fix Signed-off-by: John Crispin <[email protected]> ath11k_nss: ath-next fix connection failure due to unexpected peer delete Currently ath11k_mac_op_unassign_vif_chanctx() deletes peer but ath11k_mac_op_assign_vif_chanctx() doesn't create it. This results in connection failure if MAC80211 calls drv_unassign_vif_chanctx() and drv_assign_vif_chanctx() during AUTH and ASSOC, see below log: [ 102.372431] wlan0: authenticated [ 102.372585] ath11k_pci 0000:01:00.0: wlan0: disabling HT/VHT/HE as WMM/QoS is not supported by the AP [ 102.372593] ath11k_pci 0000:01:00.0: mac chanctx unassign ptr ffff895084638598 vdev_id 0 [ 102.372808] ath11k_pci 0000:01:00.0: WMI vdev stop id 0x0 [ 102.383114] ath11k_pci 0000:01:00.0: vdev stopped for vdev id 0 [ 102.384689] ath11k_pci 0000:01:00.0: WMI peer delete vdev_id 0 peer_addr 20:e5:2a:21:c4:51 [ 102.396676] ath11k_pci 0000:01:00.0: htt peer unmap vdev 0 peer 20:e5:2a:21:c4:51 id 3 [ 102.396711] ath11k_pci 0000:01:00.0: peer delete resp for vdev id 0 addr 20:e5:2a:21:c4:51 [ 102.396722] ath11k_pci 0000:01:00.0: mac removed peer 20:e5:2a:21:c4:51 vdev 0 after vdev stop [ 102.396780] ath11k_pci 0000:01:00.0: mac chanctx assign ptr ffff895084639c18 vdev_id 0 [ 102.400628] wlan0: associate with 20:e5:2a:21:c4:51 (try 1/3) [ 102.508864] wlan0: associate with 20:e5:2a:21:c4:51 (try 2/3) [ 102.612815] wlan0: associate with 20:e5:2a:21:c4:51 (try 3/3) [ 102.720846] wlan0: association with 20:e5:2a:21:c4:51 timed out The peer delete logic in ath11k_mac_op_unassign_vif_chanctx() is introduced by commit b4a0f54156ac ("ath11k: move peer delete after vdev stop of station for QCA6390 and WCN6855") to fix firmware crash issue caused by unexpected vdev stop/peer delete sequence. Actually for a STA interface peer should be deleted in ath11k_mac_op_sta_state() when STA's state changes from IEEE80211_STA_NONE to IEEE80211_STA_NOTEXIST, which also coincides with current peer creation design that peer is created during IEEE80211_STA_NOTEXIST -> IEEE80211_STA_NONE transition. So move peer delete back to ath11k_mac_op_sta_state(), also stop vdev before deleting peer to fix the firmware crash issue mentioned there. In this way the connection failure mentioned here is also fixed. Also do some cleanups in patch "wifi: ath11k: remove invalid peer create logic", and refactor in patches "wifi: ath11k: rename ath11k_start_vdev_delay()" and "wifi: ath11k: avoid forward declaration of ath11k_mac_start_vdev_delay()". Tested this patch set using QCA6390 and WCN6855 on both STA and SAP interfaces. Basic connection and ping work well. Baochen Qiang (4): wifi: ath11k: remove invalid peer create logic wifi: ath11k: rename ath11k_start_vdev_delay() wifi: ath11k: avoid forward declaration of ath11k_mac_start_vdev_delay() wifi: ath11k: fix connection failure due to unexpected peer delete drivers/net/wireless/ath/ath11k/mac.c | 564 +++++++++++++------------- 1 file changed, 288 insertions(+), 276 deletions(-) ath11k_nss: Revert support for beacon_tx_mode ath11k_nss: Update release fix dependancies ath11k_nss: mgmt and data ack rssi update Data ACK RSSI : Advertise NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT feature support for accounting and notifying "last ack signal" and "avg ack signal" to user space through NL interface. Enabled data ack rssi support for ethernet mode. Mgmt ACK RSSI: Enabled support for Tx-ACK RSSI in HTT over Management packets. ath11k_nss: add missing support to enable/disable bss color collision detection ath11k_nss: FW Initiated Dynamic MU-EDCA Implementing the updating of firmware initiated dynamic MU-EDCA parameters in Beacon IE. Firmware routinely checks its clients and updates its MU-EDCA values every 3 seconds. Firmware is tuning MU-EDCA parameters to improve performance. As part of this process, the firmware informs host about new MU-EDCA values utilizing WMI_MUEDCA_PARAMS_CONFIG_EVENTID. FW expectation is that host will update MU-EDCA parameters in the Beacon IE. ath11k_nss: refresh patches + cleanup Makefile ath11k_nss: Set correct pbufs for 1GB profile ath11k_nss: Revert setting fw_mem_mode for IPQ807x Leave it set to '0', as it will kernel panic with 2K skb patch. This flag was incorrectly assumed to save memory on 1G platforms. ath11k_nss: remove leftover max_tx_ring it was not removed when applying patch to replace it with 'max_tx_ring = DP_TCL_NUM_RING_MAX +1' ath11k_nss: Import bugfix patches ath11k_nss: Experimental build for IPQ6018 This will require setting the correct `ATH11K_MEM_PROFILE_XXX` for your platform. Setting it to value lower/higher than physically available will cause NULL virtual address kernel panics. I believe this setting was not originally meant to reduce memory footprint of 1G+ platforms, but to account for platforms that were 512M or less. Will require tweaking to allow for the old behvaior on 1G+ IPQ807x, while still saving memory for platforms <= 512M. commit d2d5a0d Author: Qosmio <[email protected]> AuthorDate: Sat Mar 9 11:59:20 2024 -0500 Commit: Sean Khan <[email protected]> CommitDate: Fri Oct 11 11:13:03 2024 -0400 ath11k_nss: fix typo in 512M memory profile ath11k_nss: remove SFE patch 718-e-mac80211-Deliver-the-frame-to-driver-tx-ops-directly It is not relevant to NSS builds and only meant for SFE. ath11k_nss: remove unecessary patches Color collision should be left on by default, as it's a primary feature of 802.11AX. ath11k_nss: fix spacing ath11k_nss: Remove unnecessary TKIP bloat Remove TKIP patches that are not being used as 99% of folks are running modern encryption (AES-CCMP,SAE,etc). ath11k_nss: parameterize DP_RXDMA_REFILL_RING_SIZE memory profile ath11k_nss: Remove SFE related code Cleanup SFE (shortcut fe) related code as we're not using it on NSS ath11k_nss: idr, ampdu, and skb headroom check optimizations ath11k_nss: get valid last_rate for rx_bitrate from cpu stats ath11k_nss: Fix BCCA counter for EMA Currently BCCA counter is updated to FW via csa counter offs and beacon with new countdown is updated for every beacon tx completion event. For EMA, all EMA beacons are updated in one shot, and counter update for every tx event will mess up the actual sequence of countdown sent over the air. Allow FW to update the countdown till 1 and finalize the color change. ath11k_nss: Fix compile for TRACE feature
QSDK NSS builds utilize skbuff recycling for better handling of memory. On a Dynalink DL-WRX36 (pbuf script should be set to 'auto') a significant drop in memory usage was observed as well consistent sustained RX/TX speeds. BEFORE: echo 3 >! /proc/sys/vm/drop_caches free -m total used free shared buff/cache available Mem: 867 338 547 90 101 528 Swap: 0 0 0 AFTER: total used free shared buff/cache available Mem: 867 242 594 1 81 624 Swap: 0 0 0 NOTE: For 512MB platforms, users need to test with the following scenarios, as the patch `999-233-ath11k-Disable-rx_header-tlv-for-2K-SKB.patch` is really only testable on platforms with 512M or less RAM. 1.) Explicitly setting 'ATH11K_MEM_PROFILE_512M' on and see if system crashes on boot. 2.) Explicitly setting 'ATH11K_MEM_PROFILE_1G' 3.) Remove patches 999-233-ath11k-Disable-rx_header-tlv-for-2K-SKB.patch 999-311-ath11k-configure-nss-thread-priority-during-pdev_ini.patch And re-test with #1 and #2 It was incorrectly assumed that setting a 512M for 1G platforms would save memory, instead it needs to be explicitly set to know proper memory regions, otherwise it would cause fw crash. Signed-off-by: Sean Khan <[email protected]>
Since SKB recycler was merged into main nss-wifi branch, it is not necessary to manually tinker with pbuf script. Memory is now properly managed between NSS driver allocating/deallocating SKBs. For optimal wifi performance, especially upload, it is advised to leave the script to 'auto'. Users who use sysupgrade should manually set the uci config '/etc/config/pbuf' as it will not overwrite existing configuration.
Signed-off-by: Sean Khan <[email protected]>
not needed for NSS offload and helps save some memory
The performance of Qualcomm Crypto Engine is 10x slower than kernel based encryption via ARMv8 crypto extensions. Rather than building it into the kernel allow users to select it as module which can be tested with other HW based crypto APIs via cryptodev-linux.
This will now be the default NSS branch, and will cover both kernel 6.1 and 6.6 related changes going forward. qualcommax: NSS: fix up 'nss_region'
Disable software input output translation lookaside buffer (swiotlb) as it wastes memory on low memory platforms (512m or less) qualcommax: only disable swiotlb for platforms <= 512M Disabling swiotlb is only required for platforms with 512M or less to save 64mb.
Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: Sean Khan <[email protected]> (cherry picked from commit 4ea718f)
Signed-off-by: Sean Khan <[email protected]>
It's just log spam. It doesn't affect usage. Signed-off-by: Sean Khan <[email protected]>
1.) Clarify what radio_5g_high, and radio_5g_low mean
2.) Make txpower configurable
3.) Don't hardcode radio names in the generated config, but allow
specifiying them as variables.
4.) Set `cell_density` to 0 (don't enforce rates, let driver choose)
5.) Remove `pmk_r1_push` as `ft_psk_generate_local` already deals with
generating key locally
6.) Reduce `mesh_max_peer_links` and lower ttl
Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: Sean Khan <[email protected]>
For NSS setups, you CANNOT enable option for vlan filtering. This is done internally by NSS (via `kmod-qca-nss-drv-vlan-mgr`). Check if you have it enabled by running the following command: `uci show network | grep vlan_filtering` If it's enabled, you might see: `network.@device[0].vlan_filtering='1'` Disable it by running the following commands: ```sh uci del network.@device[0].vlan_filtering uci commit network service network restart ``` Or reboot the device You also cannot use tagging in the typical way, as the NSS does not support it. That means your config must not anything like: ``` list ports 'lan1:u*' list ports 'lan2:t' list ports 'lan3:t' ``` To set up VLANs you must: 1.) Set them up on specific ports 2.) Bridge them into interfaces (they can be left unmanaged) 3.) Create firewall rules Check `nss-setup/example/04-vlans` for more examples Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: Sean Khan <[email protected]>
Although qca-nss-ecm isn't explicitly depended on by other nss packages, it is still required for proper offloading. There are also a few init and hotplug scripts needed to mitigate issues (like disabling rx-gro-list) Signed-off-by: Sean Khan <[email protected]>
…tions - Added detailed build instructions for new users, including the use of `scripts/env` for reproducible builds. - Updated comments to provide context for each config setting and package. - Clarified steps for selecting target devices and running make commands. - Included notes on SSL preferences, performance optimizations, and optional packages. - Removed redundant or unnecessary package entries for improved readability. Signed-off-by: Sean Khan <[email protected]>
Rather than pollute `qca-nss-ecm` with logic specific to ath11k, move into its own hotplug script managed by mac80211. Also enforce `qca-nss-ecm` dependency. Signed-off-by: Sean Khan <[email protected]>
Improve NSS event callback ihandling for better clarity and type safety This patch modifies ath11k_nss_wifili_event_receive to take a context parameter , enhancing code clarity by explicitly passing the ath11k_base. A wrapper function, ath11k_nss_wifili_ext_callback_wrapper, retrieves the ath11k_base from the net_device's private data, ensuring correct context handling. Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: Sean Khan <[email protected]>
The firmware for 'ct' does not support ethernet frames, which is required for partial offload with mac80211 Signed-off-by: Sean Khan <[email protected]>
Mistakenly edited the patch by hand thinking I had incorrectly imported it from @bs. Original patch was throwing ``` peer.c:572:1: error: label at end of compound statement is a C23 extension ``` Moved spin_unlocks into exit as that's what it would have been doing anyways. Signed-off-by: Sean Khan <[email protected]>
Initial README for the example scripts/configs. Right now it's only for VLAN setup. Signed-off-by: Sean Khan <[email protected]>
Rebased against upstream commit 87033 mac80211: improve single-wiphy multi-radio support Signed-off-by: Sean Khan <[email protected]>
* 802.11 k/v/r now fully works in sae-mixed mode. So we can now switch from psk2 (WPA2) to WPA2/WPA3 mixed mode. * Clean up uncessary configs * Remove WDS example as it's not applicable in this example Signed-off-by: Sean Khan <[email protected]>
Change "GRO FRAG" to "Interface" to account for additional information of other interface features (i.e. tx-checksumming) Signed-off-by: Sean Khan <[email protected]>
- fix issue #15: minor config mistake in the new nss setup vlan readme guide - fix formatting of primary subsection displaying too large. Signed-off-by: Sean Khan <[email protected]>
Signed-off-by: Sean Khan <[email protected]> (cherry picked from commit 8efd9fedbff95990ffe192f504491fe63e9a7d55)
There is a null pointer bug caused by inconsistent handling of the `ath11k_base` structure. Previously, functions accessing `ar->ab` internally, assumed it was always valid. If `ar->ab` was not properly initialized, this led to null pointer dereferences. This patch resolves the issue by: - Explicitly passing `struct ath11k_base *ab` to functions that require it, instead of relying on `ar->ab`. - Replacing all instances of `ar->ab` with `ab` for direct access. Patch by: Sebastian Gottschall <[email protected]> Signed-off-by: Sean Khan <[email protected]> (cherry picked from commit 1f6f55f9c737ece5e3d8156bbe280f84ea4bd480)
Signed-off-by: Sean Khan <[email protected]>
Author: Felix Fietkau <[email protected]> AuthorDate: Mon Nov 18 11:21:02 2024 +0100 Commit: Felix Fietkau <[email protected]> CommitDate: Mon Nov 18 11:44:44 2024 +0100 mac80211: backport some upstream fixes Fix various issues, including potential crashes Signed-off-by: Felix Fietkau <[email protected]> Signed-off-by: Sean Khan <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
24.10 nss patch from qosmio