Re: pending output optimization: was: [PATCH v3 3/3] printk: use the lockless ringbuffer

[John Ogness]

On 2020-06-25, Petr Mladek <pmladek@xxxxxxxx> wrote:
>> --- a/kernel/printk/printk.c
>> +++ b/kernel/printk/printk.c
>> @@ -2009,9 +2056,9 @@ asmlinkage int vprintk_emit(int facility, int level,
>>  
>>  	/* This stops the holder of console_sem just where we want him */
>>  	logbuf_lock_irqsave(flags);
>> -	curr_log_seq = log_next_seq;
>> +	pending_output = !prb_read_valid(prb, console_seq, NULL);
>>  	printed_len = vprintk_store(facility, level, dict, dictlen, fmt, args);
>> -	pending_output = (curr_log_seq != log_next_seq);
>> +	pending_output &= prb_read_valid(prb, console_seq, NULL);
>
> This will stop working after we remove the locks. Consoles will be
> able to handle messages while the new one is being added. There will
> be no gurantee that someone is still hadling the previously pending
> output.
>
> Please, always handle consoles when printed_len is not zero!!!
>
> The pending output was just an optimization added recently. Nobody
> requested it. It was just an idea that made sense.

OK. I will insert a patch before this one that reverts commit
3ac37a93fa92 ("printk: lock/unlock console only for new logbuf
entries"). Then there is no @pending_output for me to implement and it
will be clear that this series is changing/reverting some printk
behavior.

John Ogness