Incorrect calculation of path fraction value in MergeAppend

On 2025-May-07, Andrei Lepikhov wrote:

Hi,

Because there was no answer to the issue [1], I'll add it to the next
commitfest. Feel free to reject it if my logic on choosing the merge append
path is wrong.

If this is a bug new in pg18, you should add an item to the "open items"
list here: https://wiki.postgresql.org/wiki/Open_Items The RMT will
then take over and pester the committer/author to ensure the bug is
fixed.

If it's not new in pg18, then adding a commitfest entry in the "bug
fixes" category is the right thing. I'd suggest to CC the committer and
author of the patch that caused the bug; with the amount of traffic in
pgsql-hackers these days, it's all but guaranteed that people will miss
things.

--
Álvaro Herrera Breisgau, Deutschland — https://www.EnterpriseDB.com/
"Pido que me den el Nobel por razones humanitarias" (Nicanor Parra)

On 7/5/2025 09:57, Álvaro Herrera wrote:

On 2025-May-07, Andrei Lepikhov wrote:

Hi,

Because there was no answer to the issue [1], I'll add it to the next
commitfest. Feel free to reject it if my logic on choosing the merge append
path is wrong.

If this is a bug new in pg18, you should add an item to the "open items"
list here: https://wiki.postgresql.org/wiki/Open_Items The RMT will
then take over and pester the committer/author to ensure the bug is
fixed.

It is not a new bug.

If it's not new in pg18, then adding a commitfest entry in the "bug
fixes" category is the right thing. I'd suggest to CC the committer and
author of the patch that caused the bug; with the amount of traffic in
pgsql-hackers these days, it's all but guaranteed that people will miss
things.

Yep, that's what I wanna do. Procedure of adding new entries to
commitfest has always been close to magic, but at the moment it is not
working on my laptop at all - indexing issues?

--
regards, Andrei Lepikhov

On 2025-May-07, Andrei Lepikhov wrote:

If it's not new in pg18, then adding a commitfest entry in the "bug
fixes" category is the right thing. I'd suggest to CC the committer and
author of the patch that caused the bug; with the amount of traffic in
pgsql-hackers these days, it's all but guaranteed that people will miss
things.

Yep, that's what I wanna do. Procedure of adding new entries to commitfest
has always been close to magic, but at the moment it is not working on my
laptop at all - indexing issues?

Weird. Are you doing it here?
https://commitfest.postgresql.org/53/new/

--
Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/
“Cuando no hay humildad las personas se degradan” (A. Christie)

On 7/5/2025 11:12, Álvaro Herrera wrote:

On 2025-May-07, Andrei Lepikhov wrote:

If it's not new in pg18, then adding a commitfest entry in the "bug
fixes" category is the right thing. I'd suggest to CC the committer and
author of the patch that caused the bug; with the amount of traffic in
pgsql-hackers these days, it's all but guaranteed that people will miss
things.

Yep, that's what I wanna do. Procedure of adding new entries to commitfest
has always been close to magic, but at the moment it is not working on my
laptop at all - indexing issues?

Weird. Are you doing it here?
https://commitfest.postgresql.org/53/new/

Yes,
The magic is how it finds a link to the thread. This time it doesn't
find it by link:
/messages/by-id/c0c19730-7fec-4c78-bff8-b87ee42b709f@gmail.com
or just a thread-id:
c0c19730-7fec-4c78-bff8-b87ee42b709f@gmail.com

--
regards, Andrei Lepikhov

On 2025-May-07, Andrei Lepikhov wrote:

Yes,
The magic is how it finds a link to the thread. This time it doesn't find it
by link:
/messages/by-id/c0c19730-7fec-4c78-bff8-b87ee42b709f@gmail.com
or just a thread-id:
c0c19730-7fec-4c78-bff8-b87ee42b709f@gmail.com

Weird, that doesn't seem to work very well if you try to search for
stuff. But if you just specify the Message-Id without clicking "search"
then it works. So here you go:
https://commitfest.postgresql.org/patch/5742/

--
Álvaro Herrera Breisgau, Deutschland — https://www.EnterpriseDB.com/
"Uno puede defenderse de los ataques; contra los elogios se esta indefenso"

=?utf-8?Q?=C3=81lvaro?= Herrera <alvherre@kurilemu.de> writes:

Weird, that doesn't seem to work very well if you try to search for
stuff. But if you just specify the Message-Id without clicking "search"
then it works. So here you go:
https://commitfest.postgresql.org/patch/5742/

Yeah, that "search" button is misleading as heck. Ignore it and
just paste in the message-ID of the patch email, and you should
be good --- at least, I've never had trouble with that method.

regards, tom lane