Following kernel DEADLOCK warning noticed on arm64 Qcom db845c device while booting the device with kselftest merge config build image [1] & [2]. metadata: git_ref: master git_repo: https://gitlab.com/Linaro/lkft/mirrors/torvalds/linux-mainline git_sha: 672c0c5173427e6b3e2a9bbb7be51ceeec78093a git_describe: v5.18-rc5 kernel_version: 5.18.0-rc5 kernel-config: https://builds.tuxbuild.com/28a2o8LHusNjoIzC0DyjxMuD9pA/config vmlinux.xz: https://builds.tuxbuild.com/28a2o8LHusNjoIzC0DyjxMuD9pA/vmlinux.xz System.map: https://builds.tuxbuild.com/28a2o8LHusNjoIzC0DyjxMuD9pA/System.map artifact-location: https://builds.tuxbuild.com/28a2o8LHusNjoIzC0DyjxMuD9pA toolchain: gcc-11 Test log: --------- [ 12.978152] cpu cpu0: EM: created perf domain [ 12.984226] remoteproc remoteproc0: request_firmware failed: -2 [ 12.991986] [ 12.996140] ============================================ [ 13.001506] WARNING: possible recursive locking detected [ 13.006874] 5.18.0-rc5 #1 Not tainted [ 13.010584] -------------------------------------------- [ 13.015946] kworker/u16:9/254 is trying to acquire lock: [ 13.019488] qcom_q6v5_pas remoteproc-cdsp: supply cx not found, using dummy regulator [ 13.021314] ffff09e7811f8cf8 (&irq_desc_lock_class){-.-.}-{2:2}, at: __irq_get_desc_lock+0x64/0xa4 [ 13.021342] [ 13.021342] but task is already holding lock: [ 13.021344] ffff09e785f4c0f8 [ 13.037483] qcom_q6v5_pas remoteproc-cdsp: supply px not found, using dummy regulator [ 13.038298] (&irq_desc_lock_class [ 13.044607] cfg80211: Loading compiled-in X.509 certificates for regulatory database [ 13.047108] ){-.-.}-{2:2}, at: __irq_get_desc_lock+0x64/0xa4 [ 13.047116] [ 13.047116] other info that might help us debug this: [ 13.074565] cfg80211: Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7' [ 13.078588] Possible unsafe locking scenario: [ 13.078588] [ 13.078591] CPU0 [ 13.078593] ---- [ 13.078595] lock(&irq_desc_lock_class); [ 13.078601] lock(&irq_desc_lock_class); [ 13.078607] [ 13.078607] *** DEADLOCK *** [ 13.078607] [ 13.078609] May be due to missing lock nesting notation [ 13.078609] [ 13.078612] 6 locks held by kworker/u16:9/254: [ 13.078616] #0: [ 13.085399] platform regulatory.0: Direct firmware load for regulatory.db failed with error -2 [ 13.091177] ffff09e780018d38 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x1e8/0x6e0 [ 13.091196] #1: [ 13.093692] cfg80211: failed to load regulatory.db [ 13.096138] ffff800008de3dd0 (deferred_probe_work){+.+.}-{0:0}, at: process_one_work+0x1e8/0x6e0 [ 13.157611] #2: ffff09e7812e2988 (&dev->mutex){....}-{3:3}, at: __device_attach+0x40/0x1d0 [ 13.166066] #3: ffffa9861bcd0b70 (cpu_hotplug_lock){++++}-{0:0}, at: cpus_read_lock+0x1c/0x2c [ 13.174783] #4: ffff09e780f90918 (subsys mutex#8){+.+.}-{3:3}, at: subsys_interface_register+0x64/0x150 [ 13.184374] #5: ffff09e785f4c0f8 (&irq_desc_lock_class){-.-.}-{2:2}, at: __irq_get_desc_lock+0x64/0xa4 [ 13.193869] [ 13.193869] stack backtrace: [ 13.198274] CPU: 3 PID: 254 Comm: kworker/u16:9 Not tainted 5.18.0-rc5 #1 [ 13.205137] Hardware name: Thundercomm Dragonboard 845c (DT) [ 13.210856] Workqueue: events_unbound deferred_probe_work_func [ 13.216759] Call trace: [ 13.219238] dump_backtrace+0xc0/0x134 [ 13.223032] show_stack+0x24/0x7c [ 13.226381] dump_stack_lvl+0x8c/0xb8 [ 13.230090] dump_stack+0x18/0x34 [ 13.233442] __lock_acquire+0xbc0/0x2180 [ 13.237406] lock_acquire.part.0+0xe0/0x230 [ 13.241639] lock_acquire+0x68/0x84 [ 13.245164] _raw_spin_lock_irqsave+0x88/0x170 [ 13.249659] __irq_get_desc_lock+0x64/0xa4 [ 13.253806] enable_irq+0x40/0xb0 [ 13.257158] lmh_enable_interrupt+0x38/0x44 [lmh] [ 13.261924] irq_enable+0x4c/0xa0 [ 13.265277] __irq_startup+0x80/0xb0 [ 13.268896] irq_startup+0x84/0x174 [ 13.272432] __enable_irq+0x7c/0x90 [ 13.275967] enable_irq+0x54/0xb0 [ 13.279320] qcom_cpufreq_ready+0x2c/0x40 [ 13.283382] cpufreq_online+0x7dc/0xa50 [ 13.287263] cpufreq_add_dev+0xd4/0xf0 [ 13.291053] subsys_interface_register+0x138/0x150 [ 13.295899] cpufreq_register_driver+0x180/0x300 [ 13.300572] qcom_cpufreq_hw_driver_probe+0xe0/0x150 [ 13.305588] platform_probe+0x74/0x100 [ 13.309378] really_probe+0x18c/0x3e0 [ 13.313086] __driver_probe_device+0x124/0x1a0 [ 13.317577] driver_probe_device+0x44/0x110 [ 13.321809] __device_attach_driver+0xb0/0x154 [ 13.326302] bus_for_each_drv+0x84/0xe0 [ 13.330178] __device_attach+0xe0/0x1d0 [ 13.334056] device_initial_probe+0x20/0x30 [ 13.338288] bus_probe_device+0xac/0xb4 [ 13.342166] deferred_probe_work_func+0xc8/0x120 [ 13.346840] process_one_work+0x280/0x6e0 [ 13.350892] worker_thread+0x80/0x434 [ 13.354597] kthread+0x10c/0x120 [ 13.357863] ret_from_fork+0x10/0x20 [ 13.415473] xhci_hcd 0000:01:00.0: Adding to iommu group 6 [ 13.423508] cpu cpu4: EM: created perf domain Reported-by: Linux Kernel Functional Testing <lkft@xxxxxxxxxx> -- Linaro LKFT https://lkft.linaro.org [1] https://lkft.validation.linaro.org/scheduler/job/4975803#L3721 [2] https://qa-reports.linaro.org/lkft/linux-mainline-master/build/v5.18-rc5/testrun/9321027/suite/linux-log-parser/test/check-kernel-warning-4975803/log
