From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <a4c3e91d8044dc47b504fab2bf11aa54@kw.quanstro.net>
References: <97d3416a6d7cadb18e8723fff3d2e2a8@hamnavoe.com>
	<da2c7d30ad147fca47014c09d339aa1b@kw.quanstro.net>
	<81727c869e97f501968f868ee8794852@kw.quanstro.net>
	<AANLkTimysByILLkzs6Fs5XfqxmXcQNB3tazBtgbIhHiw@mail.gmail.com>
	<a4c3e91d8044dc47b504fab2bf11aa54@kw.quanstro.net>
From: Venkatesh Srinivas <me@acm.jhu.edu>
Date: Mon, 21 Jun 2010 12:42:26 -0400
Message-ID: <AANLkTikuFZdW-YR8pyDgXQ05Ymvyq6kcSecGGE51q8ju@mail.gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Content-Type: multipart/alternative; boundary=0016e6d77e5395570604898cfe27
Subject: Re: [9fans] interesting timing tests
Topicbox-Message-UUID: 36132cf6-ead6-11e9-9d60-3106f5b1d025

--0016e6d77e5395570604898cfe27
Content-Type: text/plain; charset=UTF-8

On Mon, Jun 21, 2010 at 10:40 AM, erik quanstrom <quanstro@quanstro.net>wrote:

> void
> lock(ulong *l)
> {
>        ulong old;
>        ushort next, owner;
>
>        old = _xadd(l, 1);
>        for(;;){
>                next = old;
>                owner = old>>16;
>                old = *l;
>                if(next == owner)
>                        break;
>        }
> }
>
> void
> unlock(ulong *l)
> {
>        _xadd(l, 1<<16);
> }


Sure, that's reasonable in C; (i wasn't sure how to do it in asm for 8_a_,
that was what I was asking about). Just also remember to provide xadd; the
distribution 8a and 8l didn't support it last I checked.

Just another observation, we can bypass the load of old in the uncontended
case if we reverse old = *l and the compare/break in lock.

Anyway, thoughts on this lock?

-- vs

--0016e6d77e5395570604898cfe27
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div class=3D"gmail_quote">On Mon, Jun 21, 2010 at 10:40 AM, erik quanstrom=
 <span dir=3D"ltr">&lt;<a href=3D"mailto:quanstro@quanstro.net">quanstro@qu=
anstro.net</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=
=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">

void<br>
lock(ulong *l)<br>
{<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0ulong old;<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0ushort next, owner;<br>
<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0old =3D _xadd(l, 1);<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0for(;;){<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0next =3D old;<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0owner =3D old&gt;&g=
t;16;<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0old =3D *l;<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0if(next =3D=3D owne=
r)<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0break;<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0}<br>
}<br>
<br>
void<br>
unlock(ulong *l)<br>
{<br>
 =C2=A0 =C2=A0 =C2=A0 =C2=A0_xadd(l, 1&lt;&lt;16);<br>
}</blockquote><div><br></div><div>Sure, that&#39;s reasonable in C; (i wasn=
&#39;t sure how to do it in asm for 8_a_, that was what I was asking about)=
. Just also remember to provide xadd; the distribution 8a and 8l didn&#39;t=
 support it last I checked.</div>

<div><br></div><div>Just another observation, we can bypass the load of old=
 in the uncontended case if we reverse old =3D *l and the compare/break in =
lock.=C2=A0</div><div><br></div><div>Anyway, thoughts on this lock?</div><d=
iv>

<br></div><div>-- vs</div><div><br></div></div>

--0016e6d77e5395570604898cfe27--