tree: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS head: acf72db532a7791fde53af3dc7b4d0a6adcf6eff commit: 713cfd2684fa5ea08b144d92b9858b932c0f1705 [20703/21579] sched: Introduce smart grid scheduling strategy for cfs config: x86_64-randconfig-001-20240125 (https://download.01.org/0day-ci/archive/20240127/202401271837.ND9zSaBY-lkp@i...) compiler: clang version 17.0.6 (https://github.com/llvm/llvm-project 6009708b4367171ccdbf4b5905cb6a803753fe18) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240127/202401271837.ND9zSaBY-lkp@i...) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp@intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202401271837.ND9zSaBY-lkp@intel.com/ All errors (new ones prefixed by >>): In file included from kernel/sched/core.c:8: In file included from kernel/sched/sched.h:39: In file included from include/linux/blkdev.h:16: include/linux/pagemap.h:425:21: warning: cast from 'int (*)(struct file *, struct page *)' to 'filler_t *' (aka 'int (*)(void *, struct page *)') converts to incompatible function type [-Wcast-function-type-strict] 425 | filler_t *filler = (filler_t *)mapping->a_ops->readpage; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from kernel/sched/core.c:8: kernel/sched/sched.h:1249:15: warning: cast from 'void (*)(struct rq *)' to 'void (*)(struct callback_head *)' converts to incompatible function type [-Wcast-function-type-strict] 1249 | head->func = (void (*)(struct callback_head *))func; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ kernel/sched/core.c:1587:6: warning: no previous prototype for function 'sched_set_stop_task' [-Wmissing-prototypes] 1587 | void sched_set_stop_task(int cpu, struct task_struct *stop) | ^ kernel/sched/core.c:1587:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 1587 | void sched_set_stop_task(int cpu, struct task_struct *stop) | ^ | static kernel/sched/core.c:2741:10: warning: cast from 'void (*)(struct callback_head *)' to 'void (*)(struct rq *)' converts to incompatible function type [-Wcast-function-type-strict] 2741 | func = (void (*)(struct rq *))head->func; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ kernel/sched/core.c:3771:35: warning: no previous prototype for function 'preempt_schedule_irq' [-Wmissing-prototypes] 3771 | asmlinkage __visible void __sched preempt_schedule_irq(void) | ^ kernel/sched/core.c:3771:22: note: declare 'static' if the function is not intended to be used outside of this translation unit 3771 | asmlinkage __visible void __sched preempt_schedule_irq(void) | ^ | static

...

kernel/sched/core.c:5845:2: error: implicit declaration of function 'tg_update_affinity_domains' [-Werror,-Wimplicit-function-declaration] 5845 | tg_update_affinity_domains(cpu, 1); | ^ kernel/sched/core.c:5904:2: error: implicit declaration of function 'tg_update_affinity_domains' [-Werror,-Wimplicit-function-declaration] 5904 | tg_update_affinity_domains(cpu, 0); | ^ kernel/sched/core.c:5976:2: error: implicit declaration of function 'init_auto_affinity' [-Werror,-Wimplicit-function-declaration] 5976 | init_auto_affinity(&root_task_group); | ^ kernel/sched/core.c:5976:2: note: did you mean 'irq_set_affinity'? include/linux/interrupt.h:292:1: note: 'irq_set_affinity' declared here 292 | irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) | ^ kernel/sched/core.c:5976:22: error: use of undeclared identifier 'root_task_group'; did you mean 'task_group'? 5976 | init_auto_affinity(&root_task_group); | ^~~~~~~~~~~~~~~ | task_group kernel/sched/sched.h:1444:34: note: 'task_group' declared here 1444 | static inline struct task_group *task_group(struct task_struct *p) | ^ kernel/sched/core.c:6029:32: warning: variable 'ptr' set but not used [-Wunused-but-set-variable] 6029 | unsigned long alloc_size = 0, ptr; | ^ 6 warnings and 4 errors generated.

vim +/tg_update_affinity_domains +5845 kernel/sched/core.c 5831 5832 int sched_cpu_activate(unsigned int cpu) 5833 { 5834 struct rq *rq = cpu_rq(cpu); 5835 struct rq_flags rf; 5836 5837 #ifdef CONFIG_SCHED_SMT 5838 /* 5839 * When going up, increment the number of cores with SMT present. 5840 */ 5841 if (cpumask_weight(cpu_smt_mask(cpu)) == 2) 5842 static_branch_inc_cpuslocked(&sched_smt_present); 5843 #endif 5844 set_cpu_active(cpu, true);

5845 tg_update_affinity_domains(cpu, 1); 5846 5847 if (sched_smp_initialized) { 5848 sched_domains_numa_masks_set(cpu); 5849 cpuset_cpu_active(); 5850 } 5851 5852 /* 5853 * Put the rq online, if not already. This happens: 5854 * 5855 * 1) In the early boot process, because we build the real domains 5856 * after all CPUs have been brought up. 5857 * 5858 * 2) At runtime, if cpuset_cpu_active() fails to rebuild the 5859 * domains. 5860 */ 5861 rq_lock_irqsave(rq, &rf); 5862 if (rq->rd) { 5863 BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); 5864 set_rq_online(rq); 5865 } 5866 rq_unlock_irqrestore(rq, &rf); 5867 5868 update_max_interval(); 5869 5870 return 0; 5871 } 5872 5873 int sched_cpu_deactivate(unsigned int cpu) 5874 { 5875 int ret; 5876 5877 set_cpu_active(cpu, false); 5878 /* 5879 * We've cleared cpu_active_mask, wait for all preempt-disabled and RCU 5880 * users of this state to go away such that all new such users will 5881 * observe it. 5882 * 5883 * Do sync before park smpboot threads to take care the rcu boost case. 5884 */ 5885 synchronize_rcu_mult(call_rcu, call_rcu_sched); 5886 5887 #ifdef CONFIG_SCHED_SMT 5888 /* 5889 * When going down, decrement the number of cores with SMT present. 5890 */ 5891 if (cpumask_weight(cpu_smt_mask(cpu)) == 2) 5892 static_branch_dec_cpuslocked(&sched_smt_present); 5893 #endif 5894 5895 if (!sched_smp_initialized) 5896 return 0; 5897 5898 ret = cpuset_cpu_inactive(cpu); 5899 if (ret) { 5900 set_cpu_active(cpu, true); 5901 return ret; 5902 } 5903 sched_domains_numa_masks_clear(cpu); 5904 tg_update_affinity_domains(cpu, 0); 5905 return 0; 5906 } 5907 5908 static void sched_rq_cpu_starting(unsigned int cpu) 5909 { 5910 struct rq *rq = cpu_rq(cpu); 5911 5912 rq->calc_load_update = calc_load_update; 5913 update_max_interval(); 5914 } 5915 5916 int sched_cpu_starting(unsigned int cpu) 5917 { 5918 sched_rq_cpu_starting(cpu); 5919 sched_tick_start(cpu); 5920 return 0; 5921 } 5922 5923 #ifdef CONFIG_HOTPLUG_CPU 5924 int sched_cpu_dying(unsigned int cpu) 5925 { 5926 struct rq *rq = cpu_rq(cpu); 5927 struct rq_flags rf; 5928 5929 /* Handle pending wakeups and then migrate everything off */ 5930 sched_ttwu_pending(); 5931 sched_tick_stop(cpu); 5932 5933 rq_lock_irqsave(rq, &rf); 5934 if (rq->rd) { 5935 BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); 5936 set_rq_offline(rq); 5937 } 5938 migrate_tasks(rq, &rf); 5939 BUG_ON(rq->nr_running != 1); 5940 rq_unlock_irqrestore(rq, &rf); 5941 5942 calc_load_migrate(rq); 5943 update_max_interval(); 5944 nohz_balance_exit_idle(rq); 5945 hrtick_clear(rq); 5946 return 0; 5947 } 5948 #endif 5949 5950 void __init sched_init_smp(void) 5951 { 5952 sched_init_numa(); 5953 5954 /* 5955 * There's no userspace yet to cause hotplug operations; hence all the 5956 * CPU masks are stable and all blatant races in the below code cannot 5957 * happen. The hotplug lock is nevertheless taken to satisfy lockdep, 5958 * but there won't be any contention on it. 5959 */ 5960 cpus_read_lock(); 5961 mutex_lock(&sched_domains_mutex); 5962 sched_init_domains(cpu_active_mask); 5963 mutex_unlock(&sched_domains_mutex); 5964 cpus_read_unlock(); 5965 5966 /* Move init over to a non-isolated CPU */ 5967 if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_FLAG_DOMAIN)) < 0) 5968 BUG(); 5969 sched_init_granularity(); 5970 5971 init_sched_rt_class(); 5972 init_sched_dl_class(); 5973 5974 sched_smp_initialized = true; 5975 5976 init_auto_affinity(&root_task_group); 5977 } 5978

-- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki