From: jacktby jacktby <jacktby@gmail.com>
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_120D8671-5F2E-41E8-960C-090C0B9F6EF1"
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3731.600.7\))
Subject: Re: Buffer Cache Problem
Date: Fri, 10 Nov 2023 22:38:57 +0800
References: <059DE243-F4B0-41CC-84BE-9362FC5FB99A@gmail.com>
 <CAEze2WjGH6ccJW1Jv+JLhn=uFec8HAZCOJikbzHQifQMaR9dpw@mail.gmail.com>
 <A8B7B88B-75B9-40EB-91E1-13E2E48F6A77@gmail.com>
To: pgsql-hackers@lists.postgresql.org
In-Reply-To: <A8B7B88B-75B9-40EB-91E1-13E2E48F6A77@gmail.com>
Message-Id: <8ADBC847-2D66-4EFB-89AF-76338B8B46F4@gmail.com>
Archived-At: <https://www.postgresql.org/message-id/8ADBC847-2D66-4EFB-89AF-76338B8B46F4%40gmail.com>
Precedence: bulk


--Apple-Mail=_120D8671-5F2E-41E8-960C-090C0B9F6EF1
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8


> 2023=E5=B9=B411=E6=9C=8810=E6=97=A5 22:31=EF=BC=8Cjacktby jacktby =
<jacktby@gmail.com> =E5=86=99=E9=81=93=EF=BC=9A
>=20
> In the bus_internal.h,I see
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D
>  Note: Buffer header lock (BM_LOCKED flag) must be held to examine or =
change  tag, state or wait_backend_pgprocno fields.
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D
> As we all know, this buffer header lock is implemented by a bit in =
state filed, and this state field is a atomic_u32 type, so in fact we =
don=E2=80=99t need to=20
> hold buffer lock when we update state, this comment has error=EF=BC=8Cri=
ght?
Oh, sorry this is true, in fact we never acquire a spin lock when update =
the state.=

--Apple-Mail=_120D8671-5F2E-41E8-960C-090C0B9F6EF1
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"overflow-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: =
after-white-space;"><br><div><blockquote =
type=3D"cite"><div>2023=E5=B9=B411=E6=9C=8810=E6=97=A5 22:31=EF=BC=8Cjackt=
by jacktby &lt;jacktby@gmail.com&gt; =E5=86=99=E9=81=93=EF=BC=9A</div><br =
class=3D"Apple-interchange-newline"><div><meta http-equiv=3D"content-type"=
 content=3D"text/html; charset=3Dutf-8"><div style=3D"overflow-wrap: =
break-word; -webkit-nbsp-mode: space; line-break: after-white-space;">In =
the bus_internal.h,I see<div>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<span =
style=3D"caret-color: rgb(0, 0, 0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
</span><span style=3D"caret-color: rgb(0, 0, =
0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</span><span =
style=3D"caret-color: rgb(0, 0, =
0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</span></div><div>&nbsp;Note: =
Buffer header lock (BM_LOCKED flag) must be held to examine or change =
&nbsp;tag, state or wait_backend_pgprocno =
fields.</div><div>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<span =
style=3D"caret-color: rgb(0, 0, 0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
</span><span style=3D"caret-color: rgb(0, 0, =
0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</span><span =
style=3D"caret-color: rgb(0, 0, =
0);">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</span></div><div><font>As =
we all know, this buffer header lock is implemented by a bit in state =
filed, and this state field is a atomic_u32 type, so in fact we don=E2=80=99=
t need to&nbsp;</font></div><div><font>hold buffer lock when we update =
state, this comment has =
error=EF=BC=8Cright?</font></div></div></div></blockquote></div>Oh, =
sorry this is true, in fact we never acquire a spin lock when update the =
state.</body></html>=

--Apple-Mail=_120D8671-5F2E-41E8-960C-090C0B9F6EF1--