summaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)Author
2025-11-24staging: fbtft: core: fix potential memory leak in fbtft_probe_common()Jianglei Nie
fbtft_probe_common() allocates a memory chunk for "info" with fbtft_framebuffer_alloc(). When "display->buswidth == 0" is true, the function returns without releasing the "info", which will lead to a memory leak. Fix it by calling fbtft_framebuffer_release() when "display->buswidth == 0" is true. Fixes: c296d5f9957c ("staging: fbtft: core support") Signed-off-by: Jianglei Nie <niejianglei2021@163.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Abdun Nihaal <abdun.nihaal@gmail.com> Link: https://patch.msgid.link/20251112192235.2088654-1-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24staging: gpib: Destage gpibDave Penkler
Move the gpib drivers out of staging and into the "real" part of the kernel. This entails: - Remove the gpib Kconfig menu and Makefile build rule from staging. - Remove gpib/uapi from the header file search path in subdir-ccflags of the gpib Makefile - move the gpib/uapi files to include/uapi/linux - Move the gpib tree out of staging to drivers. - Remove the word "Linux" from the gpib Kconfig file. - Add the gpib Kconfig menu and Makefile build rule to drivers Signed-off-by: Dave Penkler <dpenkler@gmail.com> Link: https://patch.msgid.link/20251117144021.23569-5-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24staging: gpib: Fix SPDX license for gpib headersDave Penkler
As part destaging the gpib drivers we need to add the WITH Linux-syscall-note to the SPDX license in the gpib common includes as these will be part of the kernel headers. Signed-off-by: Dave Penkler <dpenkler@gmail.com> Link: https://patch.msgid.link/20251117144021.23569-4-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24staging: gpib: Update TODO fileDave Penkler
The comments have now been fixed so remove the item about comments from the TODO file. Signed-off-by: Dave Penkler <dpenkler@gmail.com> Link: https://patch.msgid.link/20251117144021.23569-3-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24staging: gpib: Change // comments in uapi header fileDave Penkler
A ppc64 randconfig build complained about the // comments in the uapi header file: error: // comments are not allowed in this language [-Werror,-Wcomment] Change the // comments to /* */ format. Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202511141203.sbp27oAR-lkp@intel.com/ Signed-off-by: Dave Penkler <dpenkler@gmail.com> Link: https://patch.msgid.link/20251117144021.23569-2-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24firmware: stratix10-svc: fix make htmldocs warning for stratix10_svcDinh Nguyen
Fix this warning that was generated from "make htmldocs": WARNING: drivers/firmware/stratix10-svc.c:58 struct member 'intel_svc_fcs' not described in 'stratix10_svc' Fixes: e6281c26674e ("firmware: stratix10-svc: Add support for FCS") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Closes: https://lore.kernel.org/linux-next/20251106145941.37920e97@canb.auug.org.au/ Signed-off-by: Dinh Nguyen <dinguyen@kernel.org> Link: https://patch.msgid.link/20251114185815.358423-1-dinguyen@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24firmware: stratix-svc: fix make htmldocs warningDinh Nguyen
Stephen Rothwell reports htmldocs warnings when merging char-misc tree: WARNING: drivers/firmware/stratix10-svc.c:58 This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst * Total number of transaction IDs, which is a combination of WARNING: drivers/firmware/stratix10-svc.c:302 This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst * svc_mem_lock protects access to the svc_data_mem list for Fixes: bcb9f4f07061 ("firmware: stratix10-svc: Add support for async communication") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Closes: https://lore.kernel.org/linux-next/20251114153347.16001109@canb.auug.org.au/ Signed-off-by: Dinh Nguyen <dinguyen@kernel.org> Link: https://patch.msgid.link/20251114185815.358423-2-dinguyen@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-11-24Merge tag 'icc-6.19-rc1' of ↵Greg Kroah-Hartman
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/djakov/icc into char-misc-next Georgi writes: interconnect changes for 6.19 This pull request contains the interconnect changes for the 6.19-rc1 merge window. The core and driver changes are listed below. Core changes: - kbps_to_icc() macro optimization Driver changes: - Switch all Qualcomm RPMh interconnect drivers to use the dynamic node IDs and drop support for non-dynamic ID allocation - Add new driver and BWMON support for the Kaanapali SoC - Add QoS support for the SM6350 SoC - Add QoS support for the SA8775p SoC - Fix missing link from SNOC_PNOC to the USB 2 on MSM8996 SoC that includes also a dts change that has been acked by the maintainer - Drop the QPIC interconnect and BCM nodes for the SDX75 SoC, as these should be handled by the rpmh-clk driver - Other misc fixes Signed-off-by: Georgi Djakov <djakov@kernel.org> * tag 'icc-6.19-rc1' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/djakov/icc: (40 commits) interconnect: qcom: sm6350: enable QoS configuration interconnect: qcom: sm6350: Remove empty BCM arrays interconnect: qcom: icc-rpmh: Get parent's regmap for nested NoCs dt-bindings: interconnect: qcom,sm6350-rpmh: Add clocks for QoS dt-bindings: interconnect: qcom-bwmon: Document Kaanapali BWMONs interconnect: qcom: icc-rpmh: drop support for non-dynamic IDS interconnect: qcom: sm8750: convert to dynamic IDs interconnect: qcom: sm8650: convert to dynamic IDs interconnect: qcom: sm8550: convert to dynamic IDs interconnect: qcom: sm8450: convert to dynamic IDs interconnect: qcom: sm8350: convert to dynamic IDs interconnect: qcom: sm8150: convert to dynamic IDs interconnect: qcom: sm7150: convert to dynamic IDs interconnect: qcom: sm6350: convert to dynamic IDs interconnect: qcom: sdx75: convert to dynamic IDs interconnect: qcom: sdx65: convert to dynamic IDs interconnect: qcom: sdx55: convert to dynamic IDs interconnect: qcom: sdm670: convert to dynamic IDs interconnect: qcom: sc7180: convert to dynamic IDs interconnect: qcom: sar2130p: convert to dynamic IDs ...
2025-11-24Merge tag 'coresight-next-v6.19' of ↵Greg Kroah-Hartman
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/coresight/linux into char-misc-next Suzuki writes: coresight: Updates for Linux v6.19 The changes for Linux v6.19 include : - Support for static TPDM - Fixes to TMC-ETR with CATU where buffer wasn't available to CATU in perf mode - Clean ups to the component operations to accept coresight_path - Fixes to the ETM4x/ETM3x driver Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com> * tag 'coresight-next-v6.19' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/coresight/linux: coresight: etm4x: Remove the state_needs_restore flag coresight: etm4x: Remove the redundant DSB coresight: etm4x: Properly control filter in CPU idle with FEAT_TRF coresight: etm4x: Add context synchronization before enabling trace coresight: etm4x: Correct polling IDLE bit coresight: etm3x: Always set tracer's device mode on target CPU coresight: etm4x: Always set tracer's device mode on target CPU coresight: Change device mode to atomic type coresight: change the sink_ops to accept coresight_path coresight: change helper_ops to accept coresight_path coresight: tmc: add the handle of the event to the path coresight: tpdm: remove redundant check for drvdata coresight: tpdm: add static tpdm support dt-bindings: arm: document the static TPDM compatible coresight: ETR: Fix ETR buffer use-after-free issue
2025-11-24RAS/AMD/ATL: Replace bitwise_xor_bits() with hweight16()Nikolay Borisov
Doing hweight16() and checking whether the lsb is set is functionally equivalent to what bitwise_xor_bits() does. In addition, it results in better generated code as before gcc would inline the function 4 times. With hweight16(), the resulting code boils down to 2 instructions - POPCNT and AND, and all relevant CPUs support POPCNT. An alternative would have been to use the __builtin_parity() function provided by both Clang/GCC, however under some circumstances the compiler can choose not to inline it but generate a library call which is unsupported in the kernel. No functional changes. [ bp: Massage commit message. ] Signed-off-by: Nikolay Borisov <nik.borisov@suse.com> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Link: https://patch.msgid.link/20251124142517.1708451-1-nik.borisov@suse.com
2025-11-24perf: arm_spe: Add support for filtering on data sourceJames Clark
SPE_FEAT_FDS adds the ability to filter on the data source of packets. Like the other existing filters, enable filtering with PMSFCR_EL1.FDS when any of the filter bits are set. Each bit position of the 64 bit filter maps to numerical data sources 0-63 described by bits[0:5] in the data source packet (although the full range of data source is 16 bits so higher value data sources can't be filtered on). The filter is an OR of all the filter bits, so for example clearing filter bits 0 and 3 only includes packets from data sources 0 OR 3. Invert the filter given by userspace so that the default value of 0 is equivalent to including all values (no filtering). This allows us to skip adding a new format bit to enable filtering and still support excluding all data sources which would have been a filter value of 0 if not for the inversion. Tested-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Leo Yan <leo.yan@arm.com> Signed-off-by: James Clark <james.clark@linaro.org> Signed-off-by: Will Deacon <will@kernel.org>
2025-11-24USB: serial: option: add Foxconn T99W760Slark Xiao
T99W760 is designed based on Qualcomm SDX35 (5G redcap) chip. There are three serial ports to be enumerated: Modem, NMEA and Diag. test evidence as below: T: Bus=03 Lev=01 Prnt=01 Port=03 Cnt=01 Dev#= 4 Spd=5000 MxCh= 0 D: Ver= 3.20 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=0489 ProdID=e123 Rev=05.15 S: Manufacturer=QCOM S: Product=SDXBAAGHA-IDP _SN:39A8D3E4 S: SerialNumber=39a8d3e4 C: #Ifs= 6 Cfg#= 1 Atr=a0 MxPwr=896mA I: If#= 0 Alt= 0 #EPs= 1 Cls=02(commc) Sub=0e Prot=00 Driver=cdc_mbim E: Ad=82(I) Atr=03(Int.) MxPS= 64 Ivl=32ms I: If#= 1 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim E: Ad=01(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=81(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=02(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=83(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=84(I) Atr=03(Int.) MxPS= 10 Ivl=32ms I: If#= 3 Alt= 0 #EPs= 1 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none) E: Ad=85(I) Atr=03(Int.) MxPS= 64 Ivl=32ms I: If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=03(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=86(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=87(I) Atr=03(Int.) MxPS= 10 Ivl=32ms I: If#= 5 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option E: Ad=04(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms E: Ad=88(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms 0&1: MBIM, 2:Modem, 3:GNSS(non-serial port), 4: NMEA, 5:Diag Signed-off-by: Slark Xiao <slark_xiao@163.com> Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold <johan@kernel.org>
2025-11-24perf/imx_ddr: Add support for PMU in DB (system interconnects)Joakim Zhang
There is a PMU in DB, which has the same function with PMU in DDR subsystem, the difference is PMU in DB only supports cycles, axid-read, axid-write events. e.g. perf stat -a -e imx8_db0/axid-read,axi_mask=0xMMMM,axi_id=0xDDDD,axi_port=0xPP,axi_channel=0xH/ cmd perf stat -a -e imx8_db0/axid-write,axi_mask=0xMMMM,axi_id=0xDDDD,axi_port=0xPP,axi_channel=0xH/ cmd Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com> Signed-off-by: Frank Li <Frank.Li@nxp.com> Signed-off-by: Will Deacon <will@kernel.org>
2025-11-24perf/imx_ddr: Get and enable optional clksFrank Li
Get and enable optional clks because fsl,imx8dxl-db-pmu have two clocks. Signed-off-by: Frank Li <Frank.Li@nxp.com> Signed-off-by: Will Deacon <will@kernel.org>
2025-11-24perf/imx_ddr: Move ida_alloc() from ddr_perf_init() to ddr_perf_probe()Frank Li
Move ida_alloc() from helper ddr_perf_init() into ddr_perf_probe() to clarify why ida_free() must be called at the error path. Add return value check for ida_alloc(). Rename label 'cpuhp_state_err' to 'idr_free' to make the code clearer, since two error paths now jump to this label. Signed-off-by: Frank Li <Frank.Li@nxp.com> Signed-off-by: Will Deacon <will@kernel.org>
2025-11-24spi: Fix potential uninitialized variable in probe()Dan Carpenter
If the device tree is messed up, then potentially the "protocol" string could potentially be uninitialized. The property is supposed to default to "motorola" so if the of_property_read_string() function returns -EINVAL then default to "motorola". Fixes: 059f545832be ("spi: add support for microchip "soft" spi controller") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Prajna Rajendra Kumar <prajna.rajendrakumar@microchip.com> Link: https://patch.msgid.link/aSQPkfkiJ0w-FJMW@stanley.mountain Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: add support for RZ/T2H and RZ/N2HCosmin Tanislav
Compared to the previously supported RZ/V2H, the Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have a smaller FIFO, no resets, and only two clocks: PCLKSPIn and PCLK. PCLKSPIn, being the clock from which the SPI transfer clock is generated, is the equivalent of the TCLK clock from RZ/V2H. They also support generating the SPI transfer clock from PCLK. PCLKSPIn supports multiple dividers, generating multiple possible frequencies from its parent. To handle this, do the following changes. Use the minimum frequency of SPI clock to calculate the SPI controller's min_speed_hz, and the maximum frequency to calculate max_speed_hz. Add a new function, rzv2h_rspi_find_rate_variable(), which is used for the .find_tclk_rate() callback, and which supports handling clocks with a variable rate, with the following overall logic. Iterate through all possible BRDV values. For each BRDV, calculate two different SPRs, one for the clock's minimum frequency, and one for the maxmimum, and iterate through each SPR between them. If the minimum SPR is higher than the upper SPR limit, the minimum rate is too high to achieve the requested SPI frequency, skip to the next BRDV. For each SPR, calculate a rate and let the clock framework round it to the closest supported rate of the clock. The rate and SPR that generate a transfer frequency closest to the requested SPI transfer frequency will be picked. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-12-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: add support for loopback modeCosmin Tanislav
Add support for loopback mode for debugging purposes, allowing us to test the SPI controller at the maximum SPI transfer clock without being limited by external wiring. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-10-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: add support for variable transfer clockCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have a more complicated clocking setup for the SPI transfer clock than RZ/V2H, as the clock from which it is generated supports multiple dividers. To prepare for adding support for these SoCs, do the following changes. Use the minimum frequency of SPI clock to calculate the SPI controller's min_speed_hz, and the maximum frequency to calculate max_speed_hz. Apply the clock rate found by the .find_tclk_rate() to the found clock. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-9-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: add support for using PCLK for transfer clockCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs support generating the SPI transfer clock from PCLK, with the quirk that SPR 0 is not supported, causing the highest achievable SPI transfer frequency to be 31.25MHz. Add support for generating the SPI transfer clock from PCLK. Renesas RZ/V2H (R9A09G057) also has the BPEN bit used to enable this option in the datasheet, but it is not explicitly documented and there's no details about its limitations as there are on RZ/T2H. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-8-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: make transfer clock rate finding chip-specificCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have a more complicated clocking setup for the SPI transfer clock than RZ/V2H, as the clock from which it is generated supports multiple dividers. To prepare for adding support for these SoCs, split out the logic for finding the SPR and BRDV for a fixed clock into rzv2h_rspi_find_rate_fixed(), and add and use a .find_tclk_rate() callback into the chip-specific structure. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-7-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: avoid recomputing transfer frequencyCosmin Tanislav
Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have a more complicated algorithm for calculating the optimal SPI transfer frequency compared to RZ/V2H, as the clock from which the SPI frequency is generated supports multiple dividers. Cache the requested transfer frequency and skip calling rzv2h_rspi_setup_clock() if it matches the last used one to prepare for adding support for variable clock frequency handling. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-6-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: move register writes out of rzv2h_rspi_setup_clock()Cosmin Tanislav
In preparation for caching the last requested transfer frequency, move register writes outside of rzv2h_rspi_setup_clock(). The transfer list is iterated to determine the speed of the transfer and the bits per word. The speed of the transfer is used to compute SPR and BRDV inside rzv2h_rspi_setup_clock(). BRDV and SPB are stored in the SPCMD register. Move the transfer iteration earlier, move the SPR and BRDV writing out of rzv2h_rspi_setup_clock(), consolidate writing BRDV and SPB into the initial write to the SPCMD register. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-5-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: make clocks chip-specificCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have different clocks compared to RZ/V2H. Set the number of clocks and the name of the transfer clock in the chip-specific structure to prepare for adding support for them. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-4-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: make FIFO size chip-specificCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs have a different FIFO size compared to RZ/V2H. Add a chip-specific structure, and set the FIFO size inside it, to prepare for adding support for them. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-3-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24spi: rzv2h-rspi: make resets optionalCosmin Tanislav
The Renesas RZ/T2H (R9A09G077) and RZ/N2H (R9A09G087) SoCs don't have reset lines for the SPI peripheral, make them optional to prepare for adding support for them. Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com> Link: https://patch.msgid.link/20251119161434.595677-2-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
2025-11-24mt76: mt7615: Fix memory leak in mt7615_mcu_wtbl_sta_add()Zilin Guan
In mt7615_mcu_wtbl_sta_add(), an skb sskb is allocated. If the subsequent call to mt76_connac_mcu_alloc_wtbl_req() fails, the function returns an error without freeing sskb, leading to a memory leak. Fix this by calling dev_kfree_skb() on sskb in the error handling path to ensure it is properly released. Fixes: 99c457d902cf9 ("mt76: mt7615: move mt7615_mcu_set_bmc to mt7615_mcu_ops") Signed-off-by: Zilin Guan <zilin@seu.edu.cn> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251113062415.103611-1-zilin@seu.edu.cn Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt792x: fix wifi init fail by setting MCU_RUNNING after CLC loadQuan Zhou
Set the MT76_STATE_MCU_RUNNING bit only after mt7921_load_clc() has successfully completed. Previously, the MCU_RUNNING state was set before loading CLC, which could cause conflict between chip mcu_init retry and mac_reset flow, result in chip init fail and chip abnormal status. By moving the state set after CLC load, firmware initialization becomes robust and resolves init fail issue. Signed-off-by: Quan Zhou <quan.zhou@mediatek.com> Reviewed-by: druth@chromium.org Link: https://patch.msgid.link/19ec8e4465142e774f17801025accd0ae2214092.1763465933.git.quan.zhou@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: Strip whitespace from build ddateMario Limonciello (AMD)
On systems I have with mt7925 cards I've been noticing a blank line in my kernel logs. IE: ``` [ 17.294105] mt7925e 0000:c3:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250721232852a [ 17.314233] r8169 0000:c4:00.0 enp196s0f0: Link is Down ``` This is because the build_date from the header has a newline character as does the dev_info() print. As the firmware isn't guaranteed to always have a newline but the print is, copy the firmware build date to a temporary variable and strip any whitespace from it before showing it in the logs. Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Link: https://patch.msgid.link/20251120155829.3494747-1-superm1@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: Add missing locking in mt7996_mac_sta_rc_work()Lorenzo Bianconi
Grab the mt76 mutex running mt7996_mac_sta_rc_work() since it is required by mt7996_mcu_add_rate_ctrl routine. Fixes: 28d519d0d493a ("wifi: mt76: Move RCU section in mt7996_mcu_add_rate_ctrl_fixed()") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251118-mt7996-rc-work-missing-mtx-v1-1-0739c493a6cb@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: skip ieee80211_iter_keys() on scanning link removeLorenzo Bianconi
mt7996_vif_link_remove routine is executed by mt76_scan_complete() without holding the wiphy mutex triggering the following lockdep warning. WARNING: CPU: 0 PID: 72 at net/mac80211/key.c:1029 ieee80211_iter_keys+0xe4/0x1a0 [mac80211] CPU: 0 UID: 0 PID: 72 Comm: kworker/u32:2 Tainted: G S 6.18.0-rc5+ #27 PREEMPT(full) Tainted: [S]=CPU_OUT_OF_SPEC Hardware name: Default string Default string/SKYBAY, BIOS 5.12 02/15/2023 Workqueue: phy3 mt76_scan_work [mt76] RIP: 0010:ieee80211_iter_keys+0xe4/0x1a0 [mac80211] Code: 4c 48 83 c4 10 5b 5d 41 5c 41 5d 41 5e 41 5f c3 48 8b 47 48 be ff ff ff ff 48 8d 78 68 e8 b4 eb 1e e1 85 c0 0f 85 49 ff ff ff 4c 8b ab 90 1a 00 00 48 8d 83 90 RSP: 0018:ffffc900002f7cb0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff888127e00ee0 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff888127e00788 RDI: ffff88811132b5c8 RBP: ffffffffa0ddf400 R08: 0000000000000001 R09: 000000009dcc1dac R10: 0000000000000001 R11: ffff88811132b5a0 R12: ffffc900002f7d00 R13: ffff8882581e6a80 R14: ffff888127e0afc8 R15: ffff888158832038 FS: 0000000000000000(0000) GS:ffff8884da486000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000030a0fd90 CR3: 0000000002c52004 CR4: 00000000003706f0 Call Trace: <TASK> ? lock_acquire+0xc2/0x2c0 mt7996_vif_link_remove+0x64/0x2b0 [mt7996e] mt76_put_vif_phy_link+0x41/0x50 [mt76] mt76_scan_complete+0x77/0x100 [mt76] mt76_scan_work+0x2eb/0x3f0 [mt76] ? process_one_work+0x1e5/0x6d0 process_one_work+0x221/0x6d0 worker_thread+0x19a/0x340 ? rescuer_thread+0x450/0x450 kthread+0x108/0x220 ? kthreads_online_cpu+0x110/0x110 ret_from_fork+0x1c6/0x220 ? kthreads_online_cpu+0x110/0x110 ret_from_fork_asm+0x11/0x20 </TASK> irq event stamp: 45471 hardirqs last enabled at (45477): [<ffffffff813d446e>] __up_console_sem+0x5e/0x70 hardirqs last disabled at (45482): [<ffffffff813d4453>] __up_console_sem+0x43/0x70 softirqs last enabled at (44500): [<ffffffff81f2ae0c>] napi_pp_put_page+0xac/0xd0 softirqs last disabled at (44498): [<ffffffff81fa32a0>] page_pool_put_unrefed_netmem+0x290/0x3d0 ---[ end trace 0000000000000000 ]--- Fix the issue skipping ieee80211_iter_keys() for scanning links in mt7996_vif_link_remove routine since we have not uploaded any hw keys for these links. Fixes: 04414d7bba78 ("wifi: mt76: mt7996: delete vif keys when requested") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Tested-by: Ben Greear <greearb@candelatech.com> Link: https://patch.msgid.link/20251115-mt7996-key-iter-link-remove-fix-v1-1-4f3f4e1eaa78@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: skip deflink accounting for offchannel linksLorenzo Bianconi
Do not take into account offchannel links for deflink accounting. Fixes: a3316d2fc669f ("wifi: mt76: mt7996: set vif default link_id adding/removing vif links") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Tested-by: Ben Greear <greearb@candelatech.com> Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-4-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: Move mt76_abort_scan out of mt76_reset_device()Lorenzo Bianconi
Move mt76_abort_scan routine out of mt76_reset_device() in order to avoid a possible deadlock since mt76_reset_device routine is running with mt76 mutex help and mt76_abort_scan_complete() can grab mt76 mutex in some cases. Fixes: b36d55610215a ("wifi: mt76: abort scan/roc on hw restart") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Tested-by: Ben Greear <greearb@candelatech.com> Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-3-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: move mt7996_update_beacons under mt76 mutexLorenzo Bianconi
Move mt7996_update_beacons routine inside mt76 mutex critical section in mt7996_mac_reset_work() in order to run mt7996_vif_conf_link() in mt7996_mcu_add_beacon routine. Fixes: f30906c55a400 ("wifi: mt76: mt7996: disable beacons when going offchannel") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Tested-by: Ben Greear <greearb@candelatech.com> Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-2-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: grab mt76 mutex in mt7996_mac_sta_event()Lorenzo Bianconi
Grab mt76 mutex in mt7996_mac_sta_event routine in order to rely on mt76_dereference() utility macro. Fixes: ecd72f9695e7e ("wifi: mt76: mt7996: Support MLO in mt7996_mac_sta_event()") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Tested-by: Ben Greear <greearb@candelatech.com> Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-1-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7925: ensure the 6GHz A-MPDU density cap from the hardware.Michael Lo
Set the 6GHz HE A-MPDU density from the hardware capability instead of a hardcoded value, ensuring accurate capability reporting. Signed-off-by: Michael Lo <michael.lo@mediatek.com> Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com> Link: https://patch.msgid.link/20251106092151.1061648-1-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix EMI rings for RROShayne Chen
The RRO EMI rings only need to be allocated when WED is not active. This patch fixes command timeout issue for the setting of WED off and RRO on. Fixes: 3a29164425e9 ("wifi: mt76: mt7996: Add SW path for HW-RRO v3.1") Co-developed-by: Rex Lu <rex.lu@mediatek.com> Signed-off-by: Rex Lu <rex.lu@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-12-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix using wrong phy to start in mt7996_mac_restart()Shayne Chen
Pass the correct mt7996_phy to mt7996_run(). Fixes: 0a5df0ec47f7 ("wifi: mt76: mt7996: remove redundant per-phy mac80211 calls during restart") Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-11-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix MLO set key and group key issuesShayne Chen
This patch fixes the following key issues: - Pass correct link BSS to mt7996_mcu_add_key(), and use HW beacon protection mode for mt7990 chipset - Do not do group key deletion for GTK and IGTK due to FW design, the delete key command will delete all group keys of a link BSS - For deleting BIGTK, FW adds a new flow, but the "sec->add" field should be filled with "SET_KEY". Note that if BIGTK is not deleted, it will cause beacon decryption issue when switching from an AP interface to a station interface Fixes: 0c45d52276fd ("wifi: mt76: mt7996: fix setting beacon protection keys") Co-developed-by: Allen Ye <allen.ye@mediatek.com> Signed-off-by: Allen Ye <allen.ye@mediatek.com> Co-developed-by: Peter Chiu <chui-hao.chiu@mediatek.com> Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Link: https://patch.msgid.link/20251106064203.1000505-10-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix MLD group index assignmentShayne Chen
Fix extender mode and MBSS issues caused by incorrect assignment of the MLD group and remap indices. Fixes: ed01c310eca9 ("wifi: mt76: mt7996: Fix mt7996_mcu_bss_mld_tlv routine") Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-9-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: use correct link_id when filling TXD and TXPShayne Chen
Obtain the correct link ID and, if needed, switch to the corresponding wcid before populating the TX descriptor and TX payload. Rules for link id: - For QoS data of MLD peers (excluding EAPOL), select the primary or secondary wcid based on whether the TID is odd or even to meet FW/HW requirements - For other packets, use IEEE80211_TX_CTRL_MLO_LINK if specified (such as multicast and broadcast packets) Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-8-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: set link_valid field when initializing wcidShayne Chen
This ensures the upper layer uses the correct link ID during packet processing. Fixes: dd82a9e02c05 ("wifi: mt76: mt7996: Rely on mt7996_sta_link in sta_add/sta_remove callbacks") Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-7-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix teardown command for an MLD peerShayne Chen
For an MLD peer, we only need to call the teardown command when removing the last link, and there's no need to call mt7996_mcu_add_sta() for the earlier links. Fixes: c1d6dd5d03eb ("wifi: mt76: mt7996: Add mt7996_mcu_teardown_mld_sta rouine") Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-6-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix several fields in mt7996_mcu_bss_basic_tlv()Shayne Chen
Fix several fields in mt7996_mcu_bss_basic_tlv() that were not obtained from the correct link. Without this patch, the MLD station interface does not function properly. Fixes: 34a41bfbcb71 ("wifi: mt76: mt7996: prepare mt7996_mcu_add_dev/bss_info for MLO support") Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20251106064203.1000505-5-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: support fixed rate for link stationShayne Chen
Introduce mt7996_link_sta_add_debugfs() to extend fixed rate support for MLO link station. Co-developed-by: Howard Hsu <howard-yh.hsu@mediatek.com> Signed-off-by: Howard Hsu <howard-yh.hsu@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Link: https://patch.msgid.link/20251106064203.1000505-4-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix implicit beamforming support for mt7992Howard Hsu
Fix the ibf_timeout field for mt7996, mt7992 and mt7990 chipsets. For the mt7992, this value shall be set as 0xff, while the others shall be set as 0x18. Fixes: ad4c9a8a9803 ("wifi: mt76: mt7996: add implicit beamforming support for mt7992") Signed-off-by: Howard Hsu <howard-yh.hsu@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Link: https://patch.msgid.link/20251106064203.1000505-3-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: no need to wait ACK event for SDO commandPeter Chiu
For the SDO unified command, driver does not need to wait for ACK event so do not mark MCU_CMD_ACK in command header. Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Link: https://patch.msgid.link/20251106064203.1000505-2-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7996: fix max nss value when getting rx chainmaskStanleyYP Wang
Since wiphy->available_antennas_tx now accumulates the chainmask of all the radios of a wiphy, use phy->orig_antenna_mask to get the original max nss for comparison. Fixes: 69d54ce7491d ("wifi: mt76: mt7996: switch to single multi-radio wiphy") Signed-off-by: StanleyYP Wang <StanleyYP.Wang@mediatek.com> Signed-off-by: Shayne Chen <shayne.chen@mediatek.com> Link: https://patch.msgid.link/20251106064203.1000505-1-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: mt7925: cqm rssi low/high event notifyJack Kao
The implementation amounts to setting the driver flag IEEE80211_VIF_SUPPORTS_CQM_RSSI, and then providing mechanisms for continuously updating enough information to be able to provide notifications to userspace when RSSI drops below a certain threshold Signed-off-by: Jack Kao <jack.kao@mediatek.com> Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com> Link: https://patch.msgid.link/20251001012506.2168037-1-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
2025-11-24wifi: mt76: replace use of system_wq with system_percpu_wqMarco Crivellari
Currently if a user enqueue a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistentcy cannot be addressed without refactoring the API. For more details see the Link tag below. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") Switch to using system_percpu_wq because system_wq is going away as part of a workqueue restructuring. Suggested-by: Tejun Heo <tj@kernel.org> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com> Link: https://patch.msgid.link/20250922093243.140946-2-marco.crivellari@suse.com Signed-off-by: Felix Fietkau <nbd@nbd.name>
generated by cgit v1.3 (git 2.53.0) at 2026-10-04 16:39:47 +0000