Re: Is enq: TX - row lock contention a wait or syscall

That is nice to know but ultimately not very important. When analyzing a trace file, still the best and most reliable method of tuning an application, what you want to know is how much time is spent waiting and what is the application waiting for. The fact that the wait events are implemented as as semtimedop call is ultimately not very relevant. Furthermore, it's probably not that call on threaded architecture, like Windows. That makes me ask: is anyone using threaded architecture on Linux? Many fewer system calls should provide some benefits, as far as CPU is concerned.

On 1/20/20 9:43 AM, Jonathan Lewis wrote:
> The answer to your question is at the end of the 3rd paragraph of the blog note:
>
> "The event called enqueue: that's a semtimedop call."
>
> Many years ago there was just one event relating to enqueues, and it was called "enqueue". Now there are several hundred very specific "enqueue waits", of which "enq: TX - row lock contention". So yes, it's a syscall, but the full text of the name also gives some background to what prompted the syscall.
>
> Regards
> Jonathan Lewis

-- 
Mladen Gogala
Database Consultant
Tel: (347) 321-1217

--
http://www.freelists.org/webpage/oracle-l