MIME-Version: 1.0
References: <CACG=ezaWg7_nt-8ey4aKv2w9LcuLthHknwCawmBgEeTnJrJTcw@mail.gmail.com>
 <ff143b24-a093-40da-9833-d36b83726bdf@iki.fi> <CAGjGUAKO1GCzG5wBMt5RosWo0PatgFpYY=Gjgt77tN2brNe=Bg@mail.gmail.com>
 <CACG=ezYokoiumOFnqUfg_ffHD5s8T+6iHYfzKLfa=QQ-1pNrBg@mail.gmail.com>
 <CACG=ezY9xq73jcX_EjVqx5-f90nbQ9PyhFCTW2fwFCS2wmNiFw@mail.gmail.com>
 <CACG=eza+27CfLBobJJccRhXrA3He6c1irAnoyTtSC1-z9UXLrg@mail.gmail.com>
 <CAPpHfduczcop9s6gKUpLGgFUe2y4ERGMJx6SS6Kp+s-kQPwMjg@mail.gmail.com>
 <CACG=ezbye4g_ERNqE=gBcvQ0YypRaVENhNUu8xrs4PL12UdnUA@mail.gmail.com>
 <CACG=ezaMncd0-BcGHBgsSR2eqHfrz9WznHGLKX8biz6zu-azGw@mail.gmail.com>
 <5ecc35f5-1111-47fc-8a02-36d89490a50d@iki.fi> <CACG=ezb9XTvd3ZmS0y8gUunx_wBBdJO7ou+BfCOnnA5jE-11vg@mail.gmail.com>
 <CACG=ezYFNqGjsxF6Vb2CHF6JzKcjhAFauaFm9js0nu_3Ngcdkw@mail.gmail.com>
In-Reply-To: <CACG=ezYFNqGjsxF6Vb2CHF6JzKcjhAFauaFm9js0nu_3Ngcdkw@mail.gmail.com>
From: wenhui qiu <qiuwenhuifx@gmail.com>
Date: Fri, 25 Oct 2024 11:38:57 +0800
Message-ID: <CAGjGUA+dcV7veaCV1H65vCNsbS++nT8=ho772gDvsXUW9H7eXQ@mail.gmail.com>
Subject: Re: POC: make mxidoff 64 bits
To: Maxim Orlov <orlovmg@gmail.com>
Cc: Heikki Linnakangas <hlinnaka@iki.fi>, Alexander Korotkov <aekorotkov@gmail.com>, 
	Postgres hackers <pgsql-hackers@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="0000000000008a326e062544dccc"
Archived-At: <https://www.postgresql.org/message-id/CAGjGUA%2BdcV7veaCV1H65vCNsbS%2B%2BnT8%3Dho772gDvsXUW9H7eXQ%40mail.gmail.com>
Precedence: bulk

--0000000000008a326e062544dccc
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

HI Maxim Orlov
> After a bit of thought, I've realized that to be conservative here is the
way to go.
>We can reuse a maximum of existing logic. I mean, we can remove offset
wraparound "error logic" and reuse "warning logic". But set the threshold
for "warning >logic" to a much higher value. For now, I choose 2^32-1. In
other world, legit logic, in my view, here would be to trigger autovacuum
if the number of offsets (i.e. >difference nextOffset - oldestOffset)
exceeds 2^32-1. PFA patch set.
good point ,Couldn't agree with you more. xid64 is the solution to the
wraparound problem,The previous error log is no longer meaningful ,But we
might want to refine the output waring log a little(For example, checking
the underlying reasons why age has been increasing),Though we don't have to
worry about xid wraparound

+/*
+ * Multixact members warning threshold.
+ *
+ * If difference bettween nextOffset and oldestOffset exceed this value, w=
e
+ * trigger autovacuum in order to release the disk space if possible.
+ */
+#define MULTIXACT_MEMBER_AUTOVAC_THRESHOLD UINT64CONST(0xFFFFFFFF)
Can we refine this annotation a bit? for example
+/*
+ * Multixact members warning threshold.
+ *
+ * If difference bettween nextOffset and oldestOffset exceed this value, w=
e
+ * trigger autovacuum in order to release the disk space ,reduce table
bloat if possible.
+ */
+#define MULTIXACT_MEMBER_AUTOVAC_THRESHOLD UINT64CONST(0xFFFFFFFF)

Thanks

Maxim Orlov <orlovmg@gmail.com> =E4=BA=8E2024=E5=B9=B410=E6=9C=8823=E6=97=
=A5=E5=91=A8=E4=B8=89 23:55=E5=86=99=E9=81=93=EF=BC=9A

> After a bit of thought, I've realized that to be conservative here is the
> way to go.
>
> We can reuse a maximum of existing logic. I mean, we can remove offset
> wraparound "error logic" and reuse "warning logic". But set the threshold
> for "warning logic" to a much higher value. For now, I choose 2^32-1. In
> other world, legit logic, in my view, here would be to trigger autovacuum
> if the number of offsets (i.e. difference nextOffset - oldestOffset)
> exceeds 2^32-1. PFA patch set.
>
> --
> Best regards,
> Maxim Orlov.
>

--0000000000008a326e062544dccc
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">HI Maxim Orlov<div>&gt; After a bit of thought, I&#39;ve r=
ealized that to be conservative here is the way to go.</div>&gt;We can reus=
e a maximum of existing logic. I mean, we can remove offset wraparound &quo=
t;error logic&quot; and reuse &quot;warning logic&quot;. But set the thresh=
old for &quot;warning &gt;logic&quot; to a much higher value. For now, I ch=
oose 2^32-1. In other world, legit logic, in my view, here would be to trig=
ger autovacuum if the number of offsets (i.e. &gt;difference nextOffset - o=
ldestOffset) exceeds 2^32-1. PFA patch set.<div>good point ,Couldn&#39;t ag=
ree with you more.=C2=A0xid64 is the solution to the wraparound=C2=A0proble=
m,The previous error log is no longer meaningful ,But we might want to refi=
ne the output waring log a little(For example, checking the underlying reas=
ons why age has been increasing),Though we don&#39;t have to worry about xi=
d wraparound<br></div><div><br></div><div>+/*<br>+ * Multixact members warn=
ing threshold.<br>+ *<br>+ * If difference bettween nextOffset and oldestOf=
fset exceed this value, we<br>+ * trigger autovacuum in order to release th=
e disk space if possible.<br>+ */<br>+#define MULTIXACT_MEMBER_AUTOVAC_THRE=
SHOLD		UINT64CONST(0xFFFFFFFF)<br></div><div>Can we refine this annotation =
a bit? for example<br></div><div>+/*<br>+ * Multixact members warning thres=
hold.<br>+ *<br>+ * If difference bettween nextOffset and oldestOffset exce=
ed this value, we<br>+ * trigger autovacuum in order to release the disk sp=
ace ,reduce table bloat=C2=A0if possible.<br>+ */<br>+#define MULTIXACT_MEM=
BER_AUTOVAC_THRESHOLD UINT64CONST(0xFFFFFFFF)<br></div><div><br></div><div>=
Thanks</div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"=
gmail_attr">Maxim Orlov &lt;<a href=3D"mailto:orlovmg@gmail.com">orlovmg@gm=
ail.com</a>&gt; =E4=BA=8E2024=E5=B9=B410=E6=9C=8823=E6=97=A5=E5=91=A8=E4=B8=
=89 23:55=E5=86=99=E9=81=93=EF=BC=9A<br></div><blockquote class=3D"gmail_qu=
ote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-st=
yle:solid;border-left-color:rgb(204,204,204);padding-left:1ex"><div dir=3D"=
ltr"><div>After a bit of thought, I&#39;ve realized that to be conservative=
 here is the way to go.<br><br>We can reuse a maximum of existing logic. I =
mean, we can remove offset wraparound &quot;error logic&quot; and reuse &qu=
ot;warning logic&quot;. But set the threshold for &quot;warning logic&quot;=
 to a much higher value. For now, I choose 2^32-1. In other world, legit lo=
gic, in my view, here would be to trigger autovacuum if the number of offse=
ts (i.e. difference nextOffset - oldestOffset) exceeds 2^32-1. PFA patch se=
t.</div><br><span class=3D"gmail_signature_prefix">-- </span><br><div dir=
=3D"ltr" class=3D"gmail_signature"><div dir=3D"ltr"><div>Best regards,<br><=
/div><div>Maxim Orlov.</div></div></div></div>
</blockquote></div>

--0000000000008a326e062544dccc--