summaryrefslogtreecommitdiff
path: root/drivers/net/wireless/ath/ath12k
AgeCommit message (Collapse)AuthorFilesLines
2023-12-15Revert "wifi: ath12k: use ATH12K_PCI_IRQ_DP_OFFSET for DP IRQ"Karthikeyan Periyasamy1-4/+4
This reverts commit 1f1f7d548a00ebe50808cb1f580df9693e194a7c. The commit caused bootup failure on QCN9274 hw2.0 platform. Incorrect hardcode DP irq offset overwrite the CE irq, which caused the driver to miss the mandatory bootup message from the firmware through the CE interrupt. This occurs because the CE count differs between platforms. The revert has no impact since the original change was based on an incorrect assumption. Log: ath12k_pci 0000:06:00.0: fw_version 0x1011001d fw_build_timestamp 2022-12-02 01:16 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 ath12k_pci 0000:06:00.0: failed to receive control response completion, polling.. ath12k_pci 0000:06:00.0: Service connect timeout ath12k_pci 0000:06:00.0: failed to connect to HTT: -110 ath12k_pci 0000:06:00.0: failed to start core: -110 Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://msgid.link/20231214053215.2087308-1-quic_periyasa@quicinc.com
2023-12-12wifi: ath12k: fix the issue that the multicast/broadcast indicator is not ↵Lingbo Kong1-2/+2
read correctly for WCN7850 We observe some packets are discarded in ieee80211_rx_handlers_result function for WCN7850. This is because the way to get multicast/broadcast indicator with RX_MSDU_END_INFO5_DA_IS_MCBC & info5 is incorrect. It should use RX_MSDU_END_INFO13_MCAST_BCAST & info13 to get multicast/broadcast indicator. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Lingbo Kong <quic_lingbok@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231206141759.5430-1-quic_lingbok@quicinc.com
2023-12-05wifi: ath12k: avoid repeated wiphy access from hwKarthikeyan Periyasamy1-35/+37
Currently repeated access of wiphy data from mac80211 hw structure is happen inside the mac80211 registration helper functions. So optimize these helper functions by storing wiphy data locally and accessing it directly. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231201013735.2292313-1-quic_periyasa@quicinc.com
2023-12-05wifi: ath12k: fix and enable AP mode for WCN7850Kang Yang2-3/+4
For AP mode, the peer is created earlier in ath12k_mac_op_add_interface() but ath12k_mac_op_assign_vif_chanctx() will try to create peer again. Then an error will return which makes AP mode startup fail. Kernel log: [ 5017.665006] ath12k_pci 0000:04:00.0: failed to create peer after vdev start delay: -22 wpa_supplicant log: Failed to set beacon parameters Interface initialization failed wls1: interface state UNINITIALIZED->DISABLED wls1: AP-DISABLED wls1: Unable to setup interface. Failed to initialize AP interface wls1: interface state DISABLED->DISABLED wls1: AP-DISABLED So fix this check and enable AP mode for WCN7850, as now AP mode works normally. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121022459.17209-1-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: set IRQ affinity to CPU0 in case of one MSI vectorKang Yang1-1/+23
With VT-d disabled on Intel platform, ath12k gets only one MSI vector. In that case, ath12k does not free IRQ when doing suspend, hence the kernel has to migrate it to CPU0 (if it was affine to other CPUs) and allocates a new MSI vector. However, ath12k has no chance to reconfig it to HW srngs during this phase, thus ath12k fails to resume. This issue can be fixed by setting IRQ affinity to CPU0 before request_irq is called. With such affinity, migration will not happen and thus the vector keeps unchanged during suspend/resume. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-8-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: do not restore ASPM in case of single MSI vectorKang Yang1-1/+4
Current code enables ASPM by default, it allows MHI to enter M2 state. In case of one MSI vector, system hang is observed if ath12k does MHI register reading in this state. The workaround here is to prevent MHI from entering M2 state, this can be done by disabling ASPM if only one MSI vector is used. When using 32 vectors ASPM is enabled as before. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-7-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: add support one MSI vectorKang Yang2-15/+52
On some platforms it's not possible to allocate 32 MSI vectors for various reasons, maybe kernel configuration, VT-d disabled, buggy BIOS etc. So ath12k was not able to use WCN7850 PCI devices on those platforms. Add support for one MSI vector to solve that. In case of one MSI vector, interrupt migration needs to be disabled. This is because when interrupt migration happens, the msi_data may change. However, msi_data is already programmed to rings during initial phase and ath12k has no way to know that msi_data is changed during run time and reprogram again. In case of one MSI vector, MHI subsystem should not use IRQF_NO_SUSPEND as WCN7850 doesn't set this flag too. Ath12k doesn't need to leave IRQ enabled in suspend state. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-6-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: refactor multiple MSI vector implementationKang Yang2-8/+42
This is to prepare for one MSI vector support. IRQ enable and disable of CE and DP are done only in case of multiple MSI vectors. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-5-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: use ATH12K_PCI_IRQ_DP_OFFSET for DP IRQKang Yang1-4/+4
Like ATH12K_PCI_IRQ_CE0_OFFSET, define ATH12K_PCI_IRQ_DP_OFFSET for DP to save the IRQ instead of base_vector from MSI config. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-4-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: add CE and ext IRQ flag to indicate irq_handlerKang Yang2-0/+18
Add two flags to indicate whether IRQ handler for CE and DP can be called. This is because in one MSI vector case, interrupt is not disabled in hif_stop and hif_irq_disable. So if interrupt is disabled, MHI interrupt is disabled too. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-3-quic_kangyang@quicinc.com
2023-12-05wifi: ath12k: get msi_data again after request_irq is calledKang Yang1-0/+30
The reservation mode of interrupts in kernel assigns a dummy vector when the interrupt is allocated and assigns a real vector when the request_irq is called. The reservation mode helps to ease vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. So on reservation mode, the msi_data may change after request_irq is called, then it will lead to spurious interrupt. But when VT-d in BIOS is enabled and ath12k can get 32 MSI vectors, ath12k always get the same msi_data before and after request_irq. So in case of one MSI vector, ath12k need read msi_data again after request_irq is called, and then the correct msi_data is programmed into WCN7850 hardware components. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231121021304.12966-2-quic_kangyang@quicinc.com
2023-12-01wifi: ath12k: avoid repeated hw access from arKarthikeyan Periyasamy3-83/+90
Currently, the helper functions are accessing mac80211 hw data from the radio (ar) structure repeatedly. So optimize these helper functions by storing mac80211 hw data locally and accessing it directly. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231120235812.2602198-3-quic_periyasa@quicinc.com
2023-12-01wifi: ath12k: Optimize the mac80211 hw data accessKarthikeyan Periyasamy2-10/+10
Currently mac80211 hw data is accessed by convert the hw to radio (ar) structure and then radio to hw structure which is not necessary in some places where mac80211 hw data is already present. So in that kind of places avoid the conversion and directly access the mac80211 hw data. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231120235812.2602198-2-quic_periyasa@quicinc.com
2023-12-01wifi: ath12k: add 320 MHz bandwidth enumsMuna Sinada6-3/+16
Add 320 MHz bandwidth as a new bandwidth enum for ATH12k driver. This is extending existing bandwidth related enums to include 320 MHz. This is a precursor to supporting 320 MHz in the future. Sanity test performed to confirm that there was no impact in existing bandwidths. Additionally update QuIC copyright to include 2023 in hal_rx.h. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Muna Sinada <quic_msinada@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231116221839.1303170-1-quic_msinada@quicinc.com
2023-11-30wifi: ath12k: Update Qualcomm Innovation Center, Inc. copyrightsJeff Johnson21-21/+21
Update the copyright for all ath12k files modified on behalf of Qualcomm Innovation Center, Inc. in 2023. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231128-ath12kcopyrights-v1-1-be0b7408cbac@quicinc.com
2023-11-29wifi: ath12k: refactor DP Rxdma ring structureKarthikeyan Periyasamy4-39/+46
Currently data path Rxdma ring structure store the IDR buffer and lock. These IDR handling is needed only for SW cookie conversion and not needed for HW cookie conversion. REO Rxdma ring use the HW cookie conversion and monitor Rxdma ring use the SW cookie conversion. Since idr not needed for REO Rxdma ring, remove the IDR data entity from the data path Rxdma ring structure. Introduce the new data path ring structure for monitor rxmda rings since it need IDR data entity. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231111043934.20485-5-quic_periyasa@quicinc.com
2023-11-29wifi: ath12k: avoid explicit HW conversion argument in Rxdma replenishKarthikeyan Periyasamy3-53/+31
Currently Rxdma replenish require HW conversion argument which is unnecessary argument since ath12k driver configures the Rxdma only in HW conversion. To optimize the rx data path per packet, avoid the explicit unnecessary argument and condition check in the rx replenish. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231111043934.20485-4-quic_periyasa@quicinc.com
2023-11-29wifi: ath12k: avoid explicit RBM id argument in Rxdma replenishKarthikeyan Periyasamy3-11/+4
Currently all Rxdma replenish callers pass the same return buffer manager id argument, so make it implicitly. To optimize the rx data path per packet, avoid the explicit unnecessary argument in Rxdma replenish function. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231111043934.20485-3-quic_periyasa@quicinc.com
2023-11-29wifi: ath12k: avoid explicit mac id argument in Rxdma replenishKarthikeyan Periyasamy3-13/+11
Currently all Rxdma replenish callers pass zero for the mac id argument, so make it as zero implicitly. To optimize the rx data path per packet, avoid the explicit unnecessary argument in Rxdma replenish function. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231111043934.20485-2-quic_periyasa@quicinc.com
2023-11-29wifi: ath12k: fix the error handler of rfkill configKarthikeyan Periyasamy1-2/+4
When the core rfkill config throws error, it should free the allocated resources. Currently it is not freeing the core pdev create resources. Avoid this issue by calling the core pdev destroy in the error handler of core rfkill config. Found this issue in the code review and it is compile tested only. Fixes: 004ccbc0dd49 ("wifi: ath12k: add support for hardware rfkill for WCN7850") Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231111040107.18708-1-quic_periyasa@quicinc.com
2023-11-29wifi: ath12k: use select for CRYPTO_MICHAEL_MICJames Prestwood1-1/+1
Let ath12k select this option automatically which makes building more intuitive if the user enables this driver (rather than the driver not building unless CRYPTO_MICHAEL_MIC is explicitly enabled). Further investigation shows that ath11k and ath12k are the only who use 'depends on' with CRYPTO_MICHAEL_MIC: ./drivers/net/wireless/intel/ipw2x00/Kconfig: select CRYPTO_MICHAEL_MIC ./drivers/net/wireless/intersil/hostap/Kconfig: select CRYPTO_MICHAEL_MIC ./drivers/net/wireless/intersil/orinoco/Kconfig: select CRYPTO_MICHAEL_MIC ./drivers/net/wireless/ath/ath11k/Kconfig: depends on CRYPTO_MICHAEL_MIC ./drivers/net/wireless/ath/ath12k/Kconfig: depends on CRYPTO_MICHAEL_MIC ./drivers/staging/rtl8192e/Kconfig: select CRYPTO_MICHAEL_MIC ./drivers/staging/ks7010/Kconfig: select CRYPTO_MICHAEL_MIC Signed-off-by: James Prestwood <prestwoj@gmail.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231113153544.282461-2-prestwoj@gmail.com
2023-11-22wifi: ath12k: Consolidate WMI peer flagsJeff Johnson1-42/+19
Currently wmi.h has two separate set of definitions for peer flags. One set of flags is defined in enum wmi_tlv_peer_flags, and, except for the last three, are named WMI_TLV_PEER_*. The other set of flags are defined as macros, and are named WMI_PEER_*. The last three macros have the same name as the last three wmi_tlv_peer_flags enumerators. The code only uses the WMI_PEER_* names; the WMI_TLV_PEER_* names are unused. So as a first step in consolidation, remove all the WMI_TLV_PEER_* names. But since having an enum to define all the flags is actually a good thing since that provides a handle by which to refer to the entire set of flags, recast the WMI_PEER_* macros into enumerators. Compile tested only. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231106-ath-peer-flags-v1-4-781e83b7e8e8@quicinc.com
2023-11-22wifi: ath12k: Remove obsolete struct wmi_peer_flags_map *peer_flagsJeff Johnson1-2/+0
Currently both struct ath12k_wmi_pdev and struct ath12k_wmi_base define: const struct wmi_peer_flags_map *peer_flags; But that member is not used, and in fact, struct wmi_peer_flags_map is not defined within ath12k; these are obsolete remnants inherited from ath11k. So remove them. Compile tested only. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231106-ath-peer-flags-v1-2-781e83b7e8e8@quicinc.com
2023-11-22wifi: ath12k: Remove struct ath12k::opsJeff Johnson1-2/+1
Currently struct ath12k defines the following member: struct ieee80211_ops *ops; This is being flagged by checkpatch.pl: WARNING: struct ieee80211_ops should normally be const The original plan was to add the const qualifier. However, it turns out this is actually unused, so remove it. No functional changes, compile tested only. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231106-ath12k-remove-ieee80211_ops-v1-2-d72cef1a855b@quicinc.com
2023-10-31wifi: ath12k: drop NULL pointer check in ath12k_update_per_peer_tx_stats()Ma Ke1-3/+0
Since 'user_stats' is a fixed-size array of 'struct htt_ppdu_user_stats' in 'struct htt_ppdu_stats', any of its member can't be NULL and so relevant check may be dropped. Signed-off-by: Ma Ke <make_ruc2021@163.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231013074711.2202850-1-make_ruc2021@163.com
2023-10-25Merge ath-next from git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.gitKalle Valo16-68/+169
ath.git patches for v6.7. Major changes: ath12k * QCN9274: mesh support ath11k * firmware-2.bin support
2023-10-25wifi: ath12k: Introduce and use ath12k_sta_to_arsta()Jeff Johnson4-11/+16
Currently, the logic to return an ath12k_sta pointer, given a ieee80211_sta pointer, uses typecasting throughout the driver. In general, conversion functions are preferable to typecasting since using a conversion function allows the compiler to validate the types of both the input and output parameters. ath12k already defines a conversion function ath12k_vif_to_arvif() for a similar conversion. So introduce ath12k_sta_to_arsta() for this use case, and convert all of the existing typecasting to use this function. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231019-upstream-ath12k_sta_to_arsta-v1-1-06f06f693338@quicinc.com
2023-10-25wifi: ath12k: fix htt mlo-offset event lockingJohan Hovold1-2/+5
The ath12k active pdevs are protected by RCU but the htt mlo-offset event handling code calling ath12k_mac_get_ar_by_pdev_id() was not marked as a read-side critical section. Mark the code in question as an RCU read-side critical section to avoid any potential use-after-free issues. Compile tested only. Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Cc: stable@vger.kernel.org # v6.2 Signed-off-by: Johan Hovold <johan+linaro@kernel.org> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231019113650.9060-3-johan+linaro@kernel.org
2023-10-25wifi: ath12k: fix dfs-radar and temperature event lockingJohan Hovold1-1/+10
The ath12k active pdevs are protected by RCU but the DFS-radar and temperature event handling code calling ath12k_mac_get_ar_by_pdev_id() was not marked as a read-side critical section. Mark the code in question as RCU read-side critical sections to avoid any potential use-after-free issues. Note that the temperature event handler looks like a place holder currently but would still trigger an RCU lockdep splat. Compile tested only. Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Cc: stable@vger.kernel.org # v6.2 Signed-off-by: Johan Hovold <johan+linaro@kernel.org> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231019113650.9060-2-johan+linaro@kernel.org
2023-10-25wifi: ath12k: rename the sc naming convention to abKarthikeyan Periyasamy2-11/+11
In PCI and HAL interface layer module, the identifier sc is used to represent an instance of ath12k_base structure. However, within ath12k, the convention is to use "ab" to represent an SoC "base" struct. So change the all instances of sc to ab. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231018153008.29820-3-quic_periyasa@quicinc.com
2023-10-25wifi: ath12k: rename the wmi_sc naming convention to wmi_abKarthikeyan Periyasamy1-11/+11
In WMI layer module, the identifier wmi_sc is used to represent an instance of ath12k_wmi_base structure. However, within ath12k, the convention is to use "ab" to represent an SoC "base" struct. So change the all instances of wmi_sc to wmi_ab. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00125-QCAHKSWPL_SILICONZ-1 Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231018153008.29820-2-quic_periyasa@quicinc.com
2023-10-23wifi: mac80211: rename ieee80211_tx_status() to ieee80211_tx_status_skb()Kalle Valo1-2/+2
make htmldocs warns: Documentation/driver-api/80211/mac80211:109: ./include/net/mac80211.h:5170: WARNING: Duplicate C declaration, also defined at mac80211:1117. Declaration is '.. c:function:: void ieee80211_tx_status (struct ieee80211_hw *hw, struct sk_buff *skb)'. This is because there's a function named ieee80211_tx_status() and a struct named ieee80211_tx_status. This has been discussed previously but no solution found: https://lore.kernel.org/all/20220521114629.6ee9fc06@coco.lan/ There's also a bug open for three years with no solution in sight: https://github.com/sphinx-doc/sphinx/pull/8313 So I guess we have no other solution than to a workaround this in the code, for example to rename the function to ieee80211_tx_status_skb() to avoid the name conflict. I got the idea for the name from ieee80211_tx_status_noskb() in which the skb is not provided as an argument, instead with ieee80211_tx_status_skb() the skb is provided. Compile tested only. Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231012114229.2931808-2-kvalo@kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-18wifi: ath12k: Remove ath12k_base::bd_apiJeff Johnson2-4/+4
Currently struct ath12k_base defines the member bd_api. However, this member is only accessed within ath12k_core_fetch_bdf(). Since the scope is local just to that one function, remove it from ath12k_base and instead just use a local stack variable. No functional changes, compile tested only. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231013-ath11k_bd_api-v1-2-3fefe4629706@quicinc.com
2023-10-18wifi: ath12k: Enable Mesh support for QCN9274Ramya Gnanasekar1-2/+4
Currently QCN9274 supports only AP and station interface modes. Add interface type mesh to ath12k_hw_params for QCN9274 to provide support for mesh mode as well. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0-02903-QCAHKSWPL_SILICONZ-1 Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231013070007.25597-2-quic_rgnanase@quicinc.com
2023-10-18wifi: ath12k: register EHT mesh capabilitiesRamya Gnanasekar1-0/+45
The capabilities for the EHT mesh are generated from the capabilities reported by the firmware. But the firmware only reports the overall capabilities and not the one which are specific for mesh. Capabilities which requires infrastructure setup with a main STA(AP) controlling operations are not needed for mesh and hence remove these capabilities from the list. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0-02903-QCAHKSWPL_SILICONZ-1 Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231013070007.25597-3-quic_rgnanase@quicinc.com
2023-10-12wifi: ath12k: fix invalid m3 buffer addressWen Gong1-0/+1
This is to fix m3 buffer reuse issue as m3_mem->size isn't set to zero in the free function, which leads invalid m3 downloading to firmware and firmware crashing. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Wen Gong <quic_wgong@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230721055305.20420-4-quic_wgong@quicinc.com
2023-10-12wifi: ath12k: add ath12k_qmi_free_resource() for recoveryWen Gong3-0/+8
ath12k_qmi_free_target_mem_chunk() and ath12k_qmi_m3_free() is static in qmi.c, they are needed for recovery, export them in a new function Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Wen Gong <quic_wgong@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230721055305.20420-3-quic_wgong@quicinc.com
2023-10-12wifi: ath12k: configure RDDM size to MHI for device recoveryWen Gong3-0/+9
RDDM is Ram Dump Debug Module which is used to debug issues when the firmware encounters an error. The rddm_size is needed by the firmware while MHI goes to the RDDM state. Provide the size to MHI subsystem so that the firmware restart works when the firmware crashes. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Wen Gong <quic_wgong@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230721055305.20420-2-quic_wgong@quicinc.com
2023-10-12wifi: ath12k: add parsing of phy bitmap for reg rulesAditya Kumar Singh3-2/+23
Certain regulatory domains could put restrictions on phy mode operation. For example, in a few countries HE/EHT Operation is not allowed. For such countries, firmware indicates this via phy bitmap in each reg rule. Currently, there is no logic to parse this info and then pass it on to the cfg80211/regulatory. Add parsing of this phy bitmap from the regulatory channel change event and then accordingly map it to cfg80211/regulatory flags and pass it on to it. While at it, correct typo in debug print: s/dsf/dfs. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231004092818.25130-1-quic_adisi@quicinc.com
2023-10-10wifi: ath12k: Consistently use ath12k_vif_to_arvif()Jeff Johnson1-22/+22
Helper function ath12k_vif_to_arvif() exists to retrieve a struct ath12k_vif from a struct ieee80211_vif. However, in multiple places this logic is open-coded with inline typecasting. Since the typecasting prevents the compiler from type-checking the source and destination, update the driver to consistently use the helper function. No functional changes, compile tested only. Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231004-ath12k_vif_to_arvif-v1-1-3f38f6882d33@quicinc.com
2023-10-10wifi: ath12k: remove redundant memset() in ath12k_hal_reo_qdesc_setup()Dmitry Antipov1-2/+0
Since 'ath12k_dp_rx_peer_tid_setup()' is the only place where 'struct hal_rx_reo_queue' object is allocated with 'kzalloc()', call to 'memset()' in 'ath12k_hal_reo_qdesc_setup()' may be dropped. Compile tested only. Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231002182856.131254-1-dmantipov@yandex.ru
2023-10-09Merge ath-next from git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.gitKalle Valo8-19/+169
ath.git patches for v6.7. Major changes: ath12k * read board data variant name from SMBIOS
2023-10-06Merge tag 'wireless-next-2023-10-06' of ↵Jakub Kicinski15-77/+429
git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next Kalle Valo says: ==================== wireless-next patches for v6.7 The first pull request for v6.7, with both stack and driver changes. We have a big change how locking is handled in cfg80211 and mac80211 which removes several locks and hopefully simplifies the locking overall. In drivers rtw89 got MCC support and smaller features to other active drivers but nothing out of ordinary. Major changes: cfg80211 - remove wdev mutex, use the wiphy mutex instead - annotate iftype_data pointer with sparse - first kunit tests, for element defrag - remove unused scan_width support mac80211 - major locking rework, remove several locks like sta_mtx, key_mtx etc. and use the wiphy mutex instead - remove unused shifted rate support - support antenna control in frame injection (requires driver support) - convert RX_DROP_UNUSABLE to more detailed reason codes rtw89 - TDMA-based multi-channel concurrency (MCC) support iwlwifi - support set_antenna() operation - support frame injection antenna control ath12k - WCN7850: enable 320 MHz channels in 6 GHz band - WCN7850: hardware rfkill support - WCN7850: enable IEEE80211_HW_SINGLE_SCAN_ON_ALL_BANDS to make scan faster ath11k - add chip id board name while searching board-2.bin * tag 'wireless-next-2023-10-06' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next: (272 commits) wifi: rtlwifi: remove unreachable code in rtl92d_dm_check_edca_turbo() wifi: rtw89: debug: txpwr table supports Wi-Fi 7 chips wifi: rtw89: debug: show txpwr table according to chip gen wifi: rtw89: phy: set TX power RU limit according to chip gen wifi: rtw89: phy: set TX power limit according to chip gen wifi: rtw89: phy: set TX power offset according to chip gen wifi: rtw89: phy: set TX power by rate according to chip gen wifi: rtw89: mac: get TX power control register according to chip gen wifi: rtlwifi: use unsigned long for rtl_bssid_entry timestamp wifi: rtlwifi: fix EDCA limit set by BT coexistence wifi: rt2x00: fix MT7620 low RSSI issue wifi: rtw89: refine bandwidth 160MHz uplink OFDMA performance wifi: rtw89: refine uplink trigger based control mechanism wifi: rtw89: 8851b: update TX power tables to R34 wifi: rtw89: 8852b: update TX power tables to R35 wifi: rtw89: 8852c: update TX power tables to R67 wifi: rtw89: regd: configure Thailand in regulation type wifi: mac80211: add back SPDX identifier wifi: mac80211: fix ieee80211_drop_unencrypted_mgmt return type/value wifi: rtlwifi: cleanup few rtlxxxx_set_hw_reg() routines ... ==================== Link: https://lore.kernel.org/r/87jzrz6bvw.fsf@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-10-05wifi: ath12k: fix debug messagesKalle Valo1-1/+1
In ath12k the debug messages were broken, no matter setting what value to the debug_mask module parameter would not get the debug messages printed. The issue is that __ath12k_dbg() uses dev_dbg() to print the debug messages which requires either enabling CONFIG_DYNAMIC_DEBUG or DEBUG symbol in the driver. ath12k is supposed to use debug_mask module to control whether debug messages are printed or not. Using both CONFIG_DYNAMIC_DEBUG and debug_mask parameter does not make any sense so switch to using dev_printk(), just like ath11k does. Now it's enough just to debug_mask module parameter to get the debug messages. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231003150132.187875-1-kvalo@kernel.org
2023-10-02wifi: ath12k: add keep backward compatibility of PHY mode to avoid firmware ↵Wen Gong2-2/+66
crash In a special WCN7855 firmware release the EHT (IEEE 802.11be) support has been disabled for size reduction. Currently ath12k always enables EHT PHY mode during vdev start but with the special firmware that will cause a firmware crash during vdev start in firmware initialisation. This is because the firmware will use the EHT mode to allocate resources but as the EHT mode is not available in the firmware, there's an internal conflict and the firmware will crash. To fix the crash check the WMI_TLV_SERVICE_11BE flag to see if the firmware supports EHT. If EHT is not supported downgrade the PHY mode to HE (IEEE 802.11ax). This does not impact QCN9274, because WMI_SERVICE_11BE is always enabled for QCN9274, then eht_cap->has_eht will always set for it, and the logic of this patch will not take effect and the PHY mode will not down grade for it. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Wen Gong <quic_wgong@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230913105757.17744-1-quic_wgong@quicinc.com
2023-10-02wifi: ath12k: add read variant from SMBIOS for download board dataWen Gong3-1/+94
This is to read variant from SMBIOS such as read from DT, the variant string will be used to one part of string which used to search board data from board-2.bin. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Wen Gong <quic_wgong@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230905105637.10230-1-quic_wgong@quicinc.com
2023-10-02wifi: ath12k: do not drop data frames from unassociated stationsHari Chandrakanthan1-11/+1
>From 'IEEE Std 802.11-2020 section 11.3.4.1': If STA A in an infrastructure BSS receives a Class 2 or Class 3 frame from STA B that is not authenticated with STA A (i.e., the state for STA B is State 1), STA A shall discard the frame. If the frame has an individual address in the Address 1 field, the MLME of STA A shall send a Deauthentication frame to STA B. When data frames from unassociated stations are received by an AP, the AP is supposed to send a deauthentication/disassociation frame with reason code "Class 2 frame received from nonauthenticated STA" or "Class 3 frame received from nonassociated STA". But ath12k AP doesn't send deauthentication/disassociation frames, when it receives data frames from unassociated stations. The ath12k driver drops the data frames from unassociated station and the upper layer(mac80211/hostapd) is not aware of such event. Hence deauthentication/disassociation frame is not sent to that particular station by the AP. To address this issue, allow the data frames from the unassociated stations to reach mac80211 so that mac80211 can send NL80211_CMD_UNEXPECTED_FRAME event to userspace(hostapd) and hostapd upon receiving the event will send the deauthentication/disassociation frame with proper reason code. The data frame from unassociated stations gets dropped in mac80211. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0-02903-QCAHKSWPL_SILICONZ-1 Signed-off-by: Hari Chandrakanthan <quic_haric@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/1689749074-14676-1-git-send-email-quic_haric@quicinc.com
2023-10-02wifi: ath12k: mhi: fix potential memory leak in ath12k_mhi_register()Ma Ke1-4/+7
mhi_alloc_controller() allocates a memory space for mhi_ctrl. When some errors occur, mhi_ctrl should be freed by mhi_free_controller() and set ab_pci->mhi_ctrl = NULL. We can fix it by calling mhi_free_controller() when the failure happens and set ab_pci->mhi_ctrl = NULL in all of the places where we call mhi_free_controller(). Signed-off-by: Ma Ke <make_ruc2021@163.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230922021036.3604157-1-make_ruc2021@163.com
2023-09-28Merge ath-next from git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.gitKalle Valo15-71/+423
ath.git patches for v6.7. Major changes: ath12k * WCN7850: enable 320 MHz channels in 6 GHz band * WCN7850: hardware rfkill support * WCN7850: enable IEEE80211_HW_SINGLE_SCAN_ON_ALL_BANDS to make scan faster ath11k * add chip id board name while searching board-2.bin
2023-09-28wifi: ath12k: add msdu_end structure for WCN7850Kang Yang2-20/+83
WCN7850 and QCN9274 currently use the same structure rx_msdu_end_qcn9274 for msdu_end. But content of msdu_end on WCN7850 is different from that of QCN9274. Need to update it for WCN7850, otherwise will get the wrong values when using it. For example, TID is no longer in WCN7850's msdu_end. But ath12k_dp_rx_process_err() and ath12k_dp_rx_process_wbm_err() still get TID from msdu_end. So an uncertain value will be used in these two functions on WCN7850. Therefore, add new structure rx_msdu_end_wcn7850 for WCN7850. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 Signed-off-by: Kang Yang <quic_kangyang@quicinc.com> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230911093054.74943-1-quic_kangyang@quicinc.com