Re: Additional LWLOCK_STATS statistics

On Mon, Dec 21, 2015 at 4:50 PM, Jesper Pedersen
<jesper.pedersen@redhat.com> wrote:
> On 12/18/2015 01:16 PM, Robert Haas wrote:
>>
>> Is this just for informational purposes, or is this something you are
>> looking to have committed?  I originally thought the former, but now
>> I'm wondering if I misinterpreted your intent.  I have a hard time
>> getting excited about committing something that would, unless I'm
>> missing something, pretty drastically increase the overhead of running
>> with LWLOCK_STATS...
>>
>
> Yeah, so unless other people using LWLOCK_STATS find the additional
> information of use (w/ the extra overhead), I think we can mark it as
> "Returned with feedback" or "Rejected".

Marked as rejected for this CF then, log overhead is not something to
ignore. There has been a fair amount of infrastructure work done btw
thanks to your impulse.

> Alternative, I can redo the patch requiring an additional #define - f.ex.
> LWLOCK_STATS_QUEUE_SIZES

Feel free to do so if you wish, that may be interesting to see what this gives.
-- 
Michael