Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:47786
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain mozo.jp from 209.85.221.191 cause and error)
MIME-Version: 1.0
In-Reply-To: <1941231697.20100401163215@gmail.com>
References: <1941231697.20100401163215@gmail.com>
Date: Mon, 5 Apr 2010 18:46:22 +0900
Message-ID: <s2lcd1fb7541004050246u2ec3f419xf2382be992383e0c@mail.gmail.com>
To: internals@lists.php.net
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PHP-DEV] php and multithreading (additional arguments)
From: mozo@mozo.jp (Moriyoshi Koizumi)

I used to play with TSRM days ago and successfully implemented
userland threading support using GNU Pth.  It's just a proof of
concept and I did it for fun.

If interested, check out
http://github.com/moriyoshi/php-src/tree/PHP_5_3-threading/ and
read http://github.com/moriyoshi/php-src/blob/PHP_5_3-threading/ext/threadi=
ng/README
for detail (not much information though).

Also note that the language syntax was extended there so it would
support golang-like message passing.

<?php
function sub($i, $ch) {
    for (;;) {
        // receive the message from $ch
        $a =3D <- [$ch];
        printf("%d: %s\n", $i, $a);
    }
}

$ch =3D thread_message_queue_create();
for ($i =3D 0; $i < 10; $i++) {
    thread_create('sub', $i, $ch);
}

$i =3D 0;
for (;;) {
    // send $i to $ch
    [$ch] <- $i++;
    usleep(50000);
}
?>

Moriyoshi

On Thu, Apr 1, 2010 at 11:32 PM, speedy <speedy.spam@gmail.com> wrote:
> Hello PHP folks,
>
> =A0 =A0 =A0I've seen this discussed previously, and would like to add a f=
ew
> =A0 =A0 =A0arguments for the multi-threading side vs. async processing,
> =A0 =A0 =A0which I've seen mentioned as a viable alternative:
>
> =A0 =A0 =A01. Imagine that from time to time, some background processing =
takes 1
> =A0 =A0 =A0second of CPU time - w/o multithreading, all your async operat=
ions,
> =A0 =A0 =A0like accepting a connection to a socket, aio or others are bas=
ically
> =A0 =A0 =A0stalled. So, async is a good approach, but does not work as a =
magic
> =A0 =A0 =A0wand for all problem spaces. Alternatively, you could fork and=
 then do the
> =A0 =A0 =A0processing, but then the state syncing of the forked backgroun=
d processing
> =A0 =A0 =A0results with the main thread requires a whole new protocol / s=
witching to
> =A0 =A0 =A0interprocess communication, which makes such developments unne=
cessarily
> =A0 =A0 =A0hard. Threads exist for a _reason_ not only because they sound=
 cool.
>
> =A0 =A0 =A02. Without thread support, it is not possible to use multi-cor=
e processing
> =A0 =A0 =A0paradigms directly from PHP - which means PHP relies on extern=
al frameworks for
> =A0 =A0 =A0that feature, which, in a sense, makes it a non-general-purpos=
e language.
> =A0 =A0 =A0It _could become_ a general purpouse tool, if it had proper mu=
lti-threading
> =A0 =A0 =A0support built-in.
>
> =A0 =A0 =A0I, personally, considered developing websockets / nanoserv ser=
ver stack with PHP
> =A0 =A0 =A0and bumped into the multithreading limitation - AFAIK it is th=
e only big feature
> =A0 =A0 =A0separating PHP from the general purpouse languages. Everything=
 else is well
> =A0 =A0 =A0integrated with lots of external libraries/modules giving rise=
 to potential rapid
> =A0 =A0 =A0application development while using it.
>
> =A0 =A0 =A0Cheers and let me know about your thoughts, and potential core=
 implementation
> =A0 =A0 =A0issues regarding developing this language feature.
>
> --
> Best regards,
> =A0speedy =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0mailto:speed=
y.spam@gmail.com
>
>
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: http://www.php.net/unsub.php
>
>