Messages in this thread | | | Date | Fri, 28 Oct 2011 10:34:09 +0800 | From | Li Zefan <> | Subject | Re: [GIT PULL rcu/next] RCU commits for 3.1 |
| |
>>>> >>>> And I cannot reproduce after merging into 3.1. :-( >>> >>> Here's another one i just got with latest -tip: >>> >>> PM: Adding info for No Bus:vcsa2 >>> >>> =============================== >>> [ INFO: suspicious RCU usage. ] >>> ------------------------------- >>> include/linux/cgroup.h:548 suspicious rcu_dereference_check() usage! >>> >>> other info that might help us debug this: >>> >>> >>> rcu_scheduler_active = 1, debug_locks = 0 >>> 1 lock held by true/655: >>> #0: (&sig->cred_guard_mutex){+.+.+.}, at: [<810d1bd7>] prepare_bprm_creds+0x27/0x70 >>> >>> stack backtrace: >>> Pid: 655, comm: true Not tainted 3.1.0-tip-01868-g1271bd2-dirty #161079 >>> Call Trace: >>> [<81abe239>] ? printk+0x18/0x1a >>> [<81064920>] lockdep_rcu_suspicious+0xc0/0xd0 >>> [<8108aa02>] perf_event_enable_on_exec+0x1d2/0x1e0 >>> [<81063764>] ? __lock_release+0x54/0xb0 >>> [<8108cca8>] perf_event_comm+0x18/0x60 >>> [<810d1abd>] ? set_task_comm+0x5d/0x80
void set_task_comm(struct task_struct *tsk, char *buf) { task_lock(tsk); ... task_unlock(tsk); perf_event_comm(tsk); }
see, perf_event_comm() is called after releasing task_lock.
perf_event_comm() perf_event_enable_on_exec() perf_cgroup_sched_out() perf_cgroup_from_task() task_subsys_state()
No proper lock is held, hence the warning.
>>> [<81af622d>] ? _raw_spin_unlock+0x1d/0x40 >>> [<810d1ac4>] set_task_comm+0x64/0x80 >>> [<810d25fd>] setup_new_exec+0xbd/0x1d0 >>> [<810d1b61>] ? flush_old_exec+0x81/0xa0 >>> [<8110753e>] load_elf_binary+0x28e/0xa00 >>> [<810d2101>] ? search_binary_handler+0xd1/0x1d0 >>> [<81063764>] ? __lock_release+0x54/0xb0 >>> [<811072b0>] ? load_elf_library+0x260/0x260 >>> [<810d2108>] search_binary_handler+0xd8/0x1d0 >>> [<810d2060>] ? search_binary_handler+0x30/0x1d0 >>> [<810d242f>] do_execve_common+0x22f/0x2a0 >>> [<810d24b2>] do_execve+0x12/0x20 >>> [<81009592>] sys_execve+0x32/0x70 >>> [<81af7752>] ptregs_execve+0x12/0x20 >>> [<81af76d4>] ? sysenter_do_call+0x12/0x36 >>> >>> Note that the backtrace suggests that perf was used - and indeed on >>> that testbox i have this in rc.local: >>> >>> /home/mingo/bin/perf stat true & >>> >>> ... which i forgot about, completely. >>> >>> If you try 'perf stat true' can you trigger the warning perhaps? >> >> Ah! I will install this into my KVM image and see what happens. >> Your /home/mingo/bin/perf is a script that does "perf stat true" >> in a loop? > > no, it's just plain 'perf' installed locally. >
| |