函数逻辑报告

Linux Kernel

v5.5.9

Brick Technologies Co., Ltd

Source Code:kernel\bpf\cpumap.c Create Date:2022-07-27 14:41:44
Last Modify:2020-03-12 14:18:49 Copyright©Brick
首页 函数Tree
注解内核,赢得工具下载SCCTEnglish

函数名称：cpu_map_kthread_run

函数原型：static int cpu_map_kthread_run(void *data)

返回类型：int

参数：

类型参数名称
void *data
251  rcpu等于data
253  set_current_state(睡眠态)
260  当非kthread_should_stop - should this kthread return now?* When someone calls kthread_stop() on your kthread, it will be woken* and this will return true. You should then return, and your return* value will be passed through to kthread_stop().或非Test ring empty status without taking any locks循环
261  drops等于0, sched等于0
264  gfp等于__GFP_ZERO按位或DOC: Useful GFP flag combinations* Useful GFP flag combinations* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~* Useful GFP flag combinations that are commonly used. It is recommended* that subsystems start with one of these combinations and then set/clear
268  如果Test ring empty status without taking any locks则
269  set_current_state(睡眠态)
271  如果Test ring empty status without taking any locks则
272  进程调度
273  sched等于1
274  否则
277  否则
278  sched等于cond_resched()
286  n等于ptr_ring_consume_batched( Queue with potential multi-producers, and single-consumer kthread , frames, CPUMAP_BATCH)
288  以i小于n循环
289  f等于frames[i]
290  page等于virt_to_page(kaddr) returns a valid pointer if and only if* virt_addr_valid(kaddr) returns true.(f)
296  3dnow prefetch to get an exclusive cache line.* Useful for spinlocks to avoid one state transition in the* cache coherency protocol:
299  m等于kmem_cache_alloc_bulk(skbuff_head_cache, gfp, n, skbs)
300  如果此条件成立可能性小(为编译器优化)(m == 0)则
301  以i小于n循环skbs[i] = NULL
303  drops等于n
306  local_bh_disable()
307  以i小于n循环
308  xdpf等于frames[i]
309  skb等于skbs[i]
312  skb等于cpu_map_build_skb(rcpu, xdpf, skb)
313  如果非skb则
314  xdp_return_frame(xdpf)
315  继续下一循环
319  ret等于netif_receive_skb_core(skb)
320  如果ret恒等于packet dropped 则drops自加
324  trace_xdp_cpumap_kthread( Back reference to map , n, drops, sched)
326  local_bh_enable()
328  set_current_state() includes a barrier so that the write of current->state* is correctly serialised wrt the caller's subsequent test of whether to* actually sleep:* for (;;) {* set_current_state(TASK_UNINTERRUPTIBLE);* if (!need_sleep)* break;* (就绪态)
330  put_cpu_map_entry(rcpu)
331  返回:0