Re: Pipeline mode and PQpipelineSync() - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: Pipeline mode and PQpipelineSync()
Date
Msg-id 202106291503.pvs26qmxojls@alvherre.pgsql
Whole thread Raw
In response to Re: Pipeline mode and PQpipelineSync()  (Boris Kolpackov <boris@codesynthesis.com>)
Responses Re: Pipeline mode and PQpipelineSync()  (Alvaro Herrera <alvaro.herrera@2ndquadrant.com>)
List pgsql-hackers
On 2021-Jun-29, Boris Kolpackov wrote:

> Alvaro Herrera <alvaro.herrera@2ndquadrant.com> writes:
> 
> > No luck reproducing any problems with this sequence as yet.
> 
> Can you try to recreate the call flow as implemented here (it's
> pretty much plain old C if you ignore error handling):

> https://git.codesynthesis.com/cgit/odb/libodb-pgsql/tree/odb/pgsql/statement.cxx?h=bulk#n789

Hmm, I can't see what's different there than what I get on my test
program.  Can you please do PQtrace() on the connection and send the
resulting trace file?  Maybe I can compare the traffic to understand
what's the difference.

(I do see that you're doing PQisBusy that I'm not.  Going to try adding
it next.)

Thanks

-- 
Álvaro Herrera                            39°49'30"S 73°17'W



pgsql-hackers by date:

Previous
From: gkokolatos@pm.me
Date:
Subject: Teach pg_receivewal to use lz4 compression
Next
From: Julien Rouhaud
Date:
Subject: Re: track_planning causing performance regression