Re: Major pgbench synthetic SELECT workload regression, Ubuntu 23.04+PG15

From: Melanie Plageman <melanieplageman(at)gmail(dot)com>
To: Gregory Smith <gregsmithpgsql(at)gmail(dot)com>
Cc: Andres Freund <andres(at)anarazel(dot)de>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Major pgbench synthetic SELECT workload regression, Ubuntu 23.04+PG15
Date: 2023-06-12 14:28:05
Message-ID: CAAKRu_aAD2POtuYUrhnRQ36+rU7vwfHQ_xWYDfxbgccK5SvH+w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Jun 9, 2023 at 3:28 AM Gregory Smith <gregsmithpgsql(at)gmail(dot)com> wrote:
> On Thu, Jun 8, 2023 at 9:21 PM Andres Freund <andres(at)anarazel(dot)de> wrote:
>>
>> You might need to add --no-children to the perf report invocation, otherwise
>> it'll show you the call graph inverted.
>
>
> My problem was not writing kernel symbols out, I was only getting addresses for some reason. This worked:
>
> sudo perf record -g --call-graph dwarf -d --phys-data -a sleep 1
> perf report --stdio

Do you know why using phys-data would have solved the problem in your
particular setup? I find figuring out what perf options I need
mystifying.
I end up trying random things from
https://wiki.postgresql.org/wiki/Profiling_with_perf, the perf man
page, and https://www.brendangregg.com/perf.html

The pg wiki page actually has a lot of detail. If you think your
particular problem is something others would encounter, it could be
good to add it there.

FWIW, I think it is helpful to have hackers threads like this where
people work through unexplained performance results with others in the
community.

- Melanie

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Joe Conway 2023-06-12 14:44:52 Re: BUG #17946: LC_MONETARY & DO LANGUAGE plperl - BUG
Previous Message Melanie Plageman 2023-06-12 14:09:20 Re: Wrong results from Parallel Hash Full Join