Re: [PATCH v2][GIT PULL] tracing: Prevent unloadable modules from using trace_bprintk()

[Li Zefan <lizf@cn.fujitsu.com>]

Steven Rostedt wrote:
> On Fri, 2010-10-22 at 13:30 +0800, Li Zefan wrote:
> 
>> In fact tracepoint is free from this bug, because we'll empty the ring
>> buffer if the unloading module has tracepoints in it.
> 
> Hehe, and I should know, I wrote that code :-)
> 

I reported that bug. ;)

> 
>> So for trace_bprintk, why can't we do the same thing? If a module has
>> trace_bprintk calls in it, just empty the ring buffer when unloading
>> module.
>>
>> And that's as simple as something like this:
>>
>> diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c
>> index 2547d88..103987f 100644
>> --- a/kernel/trace/trace_printk.c
>> +++ b/kernel/trace/trace_printk.c
>> @@ -80,6 +80,13 @@ static int module_trace_bprintk_format_notify(struct notifier_block *self,
>>  
>>  		if (val == MODULE_STATE_COMING)
>>  			hold_module_trace_bprintk_format(start, end);
>> +		else if (val == MODULE_STATE_GOING) {
>> +			/*
>> +			 * It is safest to reset the ring buffer if the
>> +			 * module being unloaded uses trace_bprintk.
>> +			 */
>> +			tracing_reset_current_online_cpus();
>> +		}
>>  	}
>>  	return 0;
>>  }
> 
> This could definitely work.
> 
> But this prevents developers using trace_printk() in their exit() code.
> We should probably add a trace_mod_printk() or something that just
> forces the slow version that does not require flushing the ring buffer
> on removal. As long as there's not a single trace_printk() in the
> module, and it only uses the alternative, then this should work.
> 

Agreed. Add this trace_mod_printk() and add some comments to
explain why and when use it.