一. binder调试信息
1.1 binder_thread
调用方法:print_binder_thread
|
1 |
thread 8980: l 12 //tid=8980，looper=12 |
关于looper状态值:
|
1 2 3 4 5 6 |
BINDER_LOOPER_STATE_REGISTERED = 0x01, // 创建注册线程BC_REGISTER_LOOPER BINDER_LOOPER_STATE_ENTERED = 0x02, // 创建主线程BC_ENTER_LOOPER BINDER_LOOPER_STATE_EXITED = 0x04, // 已退出 BINDER_LOOPER_STATE_INVALID = 0x08, // 非法 BINDER_LOOPER_STATE_WAITING = 0x10, // 等待中 BINDER_LOOPER_STATE_NEED_RETURN = 0x20, // 需要返回 |
所以0x12 = BINDER_LOOPER_STATE_ENTERED | BINDER_LOOPER_STATE_WAITING，代表的是等待就绪状态且由为binder主线程. 简单说,looper值, 十位为1代表处于binder_thread_read()状态, 个位为1代表已注册的binder线程,个位为2代表binder主线程.
1.2 binder_node
关于binder_node的输出信息:print_binder_node()
例如:
|
1 |
node 3079465: u0000005593cc3540 c0000005593f37030 hs 1 hw 1 ls 0 lw 0 is 1 iw 1 proc 8963 |
含义:
- debug_id = 3079465
- ptr = u0000005593cc3540
- cookies = c0000005593f37030
- has_strong_ref(hs) = 1
- has_weak_ref(hw) = 1
- local_strong_refs(ls) = 0
- local_weak_refs(lw) = 0
- internal_strong_refs(is) = 1
- count: (node->refs总引用次数) = 1
- proc: (node->refs->proc->pid) = 8963
1.3 binder_ref
调用方法: print_binder_ref()
|
1 2 |
//含义：ref `debug_id`:`desc` `node` `node->debug_id` `strong` `weak` `death` ref 340122: desc 1 node 340121 s 1 w 1 d ffffffc04f90a340 |
输出binder_ref结构体成员变量:
node是指当 ref->node->proc为空则代表node已死亡,采用deadnode,否则node.death: 应用注册死亡通知时，此域不为空.
1.4 binder_buffer
调用方法: print_binder_buffer()
|
1 2 |
//含义：buffer `debug_id`:`data` `data_size`:`offsets_size` `transaction` buffer 3473176: ffffff8007700218 size 0:0 delivered |
输出的便是binder_buffer结构体的成员变量,其中transaction不为空,则为active,否则为delivered.
1.5 binder_transaction
print_binder_transaction
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
static void print_binder_transaction(struct seq_file *m, const char *prefix, struct binder_transaction *t) { seq_printf(m, "%s %d: %p from %d:%d to %d:%d code %x flags %x pri %ld r%d", prefix, t->debug_id, t, t->from ? t->from->proc->pid : 0, t->from ? t->from->pid : 0, t->to_proc ? t->to_proc->pid : 0, t->to_thread ? t->to_thread->pid : 0, t->code, t->flags, t->priority, t->need_reply); if (t->buffer == NULL) { seq_puts(m, " buffer free\n"); return; } if (t->buffer->target_node) seq_printf(m, " node %d", t->buffer->target_node->debug_id); seq_printf(m, " size %zd:%zd data %p\n", t->buffer->data_size, t->buffer->offsets_size, t->buffer->data); } |
接下来,分别说说系统中几个常见进程:surfaceflinger, mediaserver, servicemanager,system_server
二. mediaserver
|
1 2 3 4 5 |
USER PID PPID VSIZE RSS WCHAN PC NAME media 8814 1 128376 16068 binder_thr 00f6da2db0 S /system/bin/mediaserver media 9304 8814 128376 16068 binder_thr 00f6da2db0 S Binder_1 media 9305 8814 128376 16068 binder_thr 00f6da2db0 S Binder_2 media 22609 8814 128376 16068 binder_thr 00f6da2db0 S Binder_3 |
2.1 stats
|
1 2 3 4 5 6 7 8 9 |
proc 8814 threads: 8 requested threads: 0+2/15 ready threads 4 free async space 520192 nodes: 13 refs: 14 s 14 w 14 buffers: 0 pending transactions: 0 |
进程mediaserver:
- 处于ready状态的binder线程个数为4:
- BC_ENTER_LOOPER创建2个binder线程
- BC_REGISTER_LOOPER创建2个binder线程
2.2 proc/8814
|
1 2 3 4 5 6 7 8 9 |
proc 8814 thread 8814: l 12 thread 9294: l 00 thread 9296: l 00 thread 9297: l 00 thread 9299: l 00 thread 9304: l 12 thread 9305: l 11 thread 22609: l 11 |
三. servicemanager
单线程的进程:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
system 435 1 3872 1040 binder_thr 7fb462bcd4 S /system/bin/servicemanager proc 435 threads: 1 requested threads: 0+0/0 ready threads 1 free async space 65536 nodes: 1 refs: 130 s 130 w 0 buffers: 0 pending transactions: 0 proc 435 thread 435: l 12 node 1: u0000000000000000 c0000000000000000 hs 1 hw 1 ls 1 lw 1 is 72 iw 72 proc ... |
该binder_thread是由BC_ENTER_LOOPER所创建的binder主线程, 只有一个binder_node对象, 被72个进程所使用.servicemanager作为服务管家,被系统大量进程所使用.
