Re: Add client connection check during the execution of the query

On Sat, Aug 3, 2019 at 4:40 AM Konstantin Knizhnik
<k(dot)knizhnik(at)postgrespro(dot)ru> wrote:
> On 18.07.2019 6:19, Tatsuo Ishii wrote:
> > So the performance is about 5% down with the feature enabled in this
> > case. For me, 5% down is not subtle. Probably we should warn this in
> > the doc.
>
> I also see some performance degradation, although it is not so large in
> my case (I used pgbench with scale factor 10 and run the same command as
> you).
> In my case difference is 103k vs. 105k TPS is smaller than 2%.

I didn't test, but hopefully the degradation is fixed by commit 09cf1d52?

> If OS detected closed connection, it should return POLLHUP, should not it?
> I am not sure if it is more portable or more efficient way - just seems
> to be a little bit more natural way (from my point of view) to check if
> connection is still alive.

That's if you're sleeping inepoll etc. This patch is for CPU-bound
backends, running a long query. We need to do something special to
find out if the kernel knows that the connection has been closed.

I've done a quick rebase of this the patch and added it to the
commitfest. No other changes. Several things were mentioned earlier
that still need to be tidied up.