MIME-Version: 1.0
References: 
 <CAGCUe0Kp7kTUWFjHhiE-xjZ4=UDe2fUdDVajKcjV+BT+T5d21A@mail.gmail.com>
In-Reply-To: 
 <CAGCUe0Kp7kTUWFjHhiE-xjZ4=UDe2fUdDVajKcjV+BT+T5d21A@mail.gmail.com>
From: Siddharth Kothari <sidkot@google.com>
Date: Wed, 8 Apr 2026 10:57:16 +0530
Message-ID: 
 <CAGCUe0LUtEDyy1U-KLzax3KGz6zA8T75d4VxN+3vu_9EmLMeLQ@mail.gmail.com>
Subject: Re: [PATCH] Add RetrieveInstrumentation hook for CustomScan providers
To: pgsql-hackers@lists.postgresql.org
Cc: Vaibhav Jain <jainva@google.com>, Madhukar <madhukarprasad@google.com>
Content-Type: multipart/alternative; boundary="000000000000d963bc064eec2709"
Archived-At: 
 <https://www.postgresql.org/message-id/CAGCUe0LUtEDyy1U-KLzax3KGz6zA8T75d4VxN%2B3vu_9EmLMeLQ%40mail.gmail.com>
Precedence: bulk

--000000000000d963bc064eec2709
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi everyone,

I=E2=80=99m just checking in to see if anyone has had a chance to look at t=
his or
if there=E2=80=99s any further information I should provide to help with th=
e
review. I have also added the patch to PG20-1 CF queue, the link is
https://commitfest.postgresql.org/patch/6524/.

Thanks,
Siddharth

On Wed, Feb 18, 2026 at 3:09=E2=80=AFPM Siddharth Kothari <sidkot@google.co=
m> wrote:

> Dear PostgreSQL Hackers,
>
> This email proposes a patch to enhance the CustomScan provider interface.
> The patch file,
> 0001-Add-RetrieveInstrumentationCustomScan-hook-for-Custo.patch, is
> attached.
>
> *Problem:*
>
> CustomScan providers currently lack a standard method to aggregate
> instrumentation data from parallel workers back to the leader process
> before the Dynamic Shared Memory (DSM) segment is unlinked. This makes it
> difficult to gather comprehensive performance metrics from parallelized
> custom scans.
>
> *Solution:*
>
> This patch introduces a new optional hook,
> RetrieveInstrumentationCustomScan, to the CustomExecMethods struct. This
> hook allows custom scan providers to implement logic to collect and
> consolidate instrumentation from shared memory or worker states during th=
e
> parallel query cleanup phase. This hook is invoked via the new
> ExecCustomScanRetrieveInstrumentation function, called from
> ExecParallelRetrieveInstrumentation for T_CustomScanState nodes. Since
> the hook is optional (checked for NULL before calling), it maintains full
> backward compatibility.
>
> *Testing & Compatibility:*
>
>    - The patch compiles and passes all core regression tests (make
>    check-world) on my x86_64 instance.
>    - The changes are not platform-specific.
>    - Regression Tests: This patch provides a new *capability* for custom
>    scan providers. Since the hook's functionality is only realized when
>    implemented by an extension, specific tests would naturally reside wit=
hin
>    that extension rather than in the core regression suite.
>
> This patch does not directly address a specific item on the official TODO
> list but enhances the extensibility framework.
>
> I believe this patch is complete and ready for review. I look forward to
> any feedback and am happy to make revisions. I will also add this patch t=
o
> the next CommitFest.
>
> Thank you,
>
> Siddharth Kothari
>

--000000000000d963bc064eec2709
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi everyone,=C2=A0</div><div><br></div><div>I=E2=80=
=99m just checking in to see if anyone has had a chance to look at this or =
if there=E2=80=99s any further information I should provide to help with th=
e review. I have also added the patch to PG20-1 CF queue, the link is <a hr=
ef=3D"https://commitfest.postgresql.org/patch/6524/">https://commitfest.pos=
tgresql.org/patch/6524/</a>.</div><div><br></div><div>Thanks,=C2=A0<br>Sidd=
harth</div></div><br><div class=3D"gmail_quote gmail_quote_container"><div =
dir=3D"ltr" class=3D"gmail_attr">On Wed, Feb 18, 2026 at 3:09=E2=80=AFPM Si=
ddharth Kothari &lt;<a href=3D"mailto:sidkot@google.com">sidkot@google.com<=
/a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0=
px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><=
div dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr"><p>Dear PostgreSQL Hacker=
s,</p><p>This email proposes a patch to enhance the CustomScan provider int=
erface. The patch file,=C2=A0<code>0001-Add-RetrieveInstrumentationCustomSc=
an-hook-for-Custo.patch</code>, is attached.</p><p><strong>Problem:</strong=
></p><p>CustomScan providers currently lack a standard method to aggregate =
instrumentation data from parallel workers back to the leader process befor=
e the Dynamic Shared Memory (DSM) segment is unlinked. This makes it diffic=
ult to gather comprehensive performance metrics from parallelized custom sc=
ans.</p><p><strong>Solution:</strong></p><p>This patch introduces a new opt=
ional hook,=C2=A0<code>RetrieveInstrumentationCustomScan</code>, to the=C2=
=A0<code>CustomExecMethods</code>=C2=A0struct. This hook allows custom scan=
 providers to implement logic to collect and consolidate instrumentation fr=
om shared memory or worker states during the parallel query cleanup phase. =
This hook is invoked via the new=C2=A0<code>ExecCustomScanRetrieveInstrumen=
tation</code>=C2=A0function, called from=C2=A0<code>ExecParallelRetrieveIns=
trumentation</code>=C2=A0for=C2=A0<code>T_CustomScanState</code>=C2=A0nodes=
. Since the hook is optional (checked for NULL before calling), it maintain=
s full backward compatibility.</p><p><strong>Testing &amp; Compatibility:</=
strong></p><ul><span style=3D"color:inherit"><li style=3D"margin-left:15px"=
>The patch compiles and passes all core regression tests (<code>make check-=
world</code>) on my x86_64 instance.</li></span><li style=3D"margin-left:15=
px">The changes are not platform-specific.</li><li style=3D"margin-left:15p=
x">Regression Tests:=C2=A0This patch provides a new=C2=A0<em>capability</em=
>=C2=A0for custom scan providers. Since the hook&#39;s functionality is onl=
y realized when implemented by an extension, specific tests would naturally=
 reside within that extension rather than in the core regression suite.</li=
></ul><p>This patch does not directly address a specific item on the offici=
al TODO list but enhances the extensibility framework.</p><span><p>I believ=
e this patch is complete and ready for review. I look forward to any feedba=
ck and am happy to make revisions. I will also add this patch to the next C=
ommitFest.</p><p>Thank you,</p><p>Siddharth Kothari</p></span></div>
</div>
</div>
</blockquote></div>

--000000000000d963bc064eec2709--