Re: [PATCH] rcu: Reword help of RCU_TRACE option
From: Nikolay Borisov
Date: Wed Oct 05 2016 - 10:18:18 EST
On 10/05/2016 05:03 PM, Paul E. McKenney wrote:
> On Wed, Oct 05, 2016 at 10:06:21AM +0300, Nikolay Borisov wrote:
>> Explicitly state that enabling RCU_TRACE enables more
>> tracepoints and not just "additional tracing".
>> Signed-off-by: Nikolay Borisov <kernel@xxxxxxxx>
>> Hello Paul,
>> Following our latest conversation re. enabling RCU tracing
>> I had to actually go and look into the code to see which
>> option enables the additional tracepoints. I wasn't sure
>> while reading the help of TRACE_RCU. So here is a patch
>> which explicitly states that this enables additional
>> Apply it if you deem necessary it might be just me that
>> thinks the description is not verbose enough.
>> lib/Kconfig.debug | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
>> index cab7405f48d2..f8c3468ec7ef 100644
>> --- a/lib/Kconfig.debug
>> +++ b/lib/Kconfig.debug
>> @@ -1429,8 +1429,8 @@ config RCU_TRACE
>> depends on DEBUG_KERNEL
>> select TRACE_CLOCK
>> - This option provides tracing in RCU which presents stats
>> - in debugfs for debugging RCU implementation.
>> + This option enables additional tracepoints in RCU which present
>> + stats in debugfs for debugging RCU implementation.
> How about something like this?
> This option provides tracing in RCU which presents stats
> in debugfs for debugging RCU implementation. It also enables
> additional tracepoints for ftrace-style event tracing.
So long as it's explicitly stated that it also enables tracepoints it's
all fine by me. Apparently I'm not the only one who got confused. In a
question re. CONFIG_RCU_TRACE here is what someone said on IRC:
"what's tracepoints do you mean? this option enables the additional
Care to apply it with the revised wording or shall I resend?
> Thanx, Paul
>> Say Y here if you want to enable RCU tracing
>> Say N if you are unsure.