Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:113047
To: Aaron Piotrowski <aaron@trowski.com>, php internals
	<internals@lists.php.net>
Thread-Topic: [PHP-DEV] [RFC] Fibers
Thread-Index: AQHW1JH3Qg2ZcNIGwEee0GXThHxj4KpEclKt
Date: Tue, 2 Feb 2021 03:04:19 +0000
Message-ID:
 <DM6PR07MB6618A2605514704DAA0833F6F9B59@DM6PR07MB6618.namprd07.prod.outlook.com>
References: <E4BE4688-3CD1-4CE8-A818-5A862F0293F5@trowski.com>
In-Reply-To: <E4BE4688-3CD1-4CE8-A818-5A862F0293F5@trowski.com>
Accept-Language: en-CA, en-US
Content-Language: en-CA
x-ms-exchange-messagesentrepresentingtype: 1
x-tmn: [aMTuRCUXOlx47j6V5vCf1c9vjESDxmSoYFTECnIDGB0i2LV2t5xntwwRHvNaQghH]
x-ms-publictraffictype: Email
x-incomingheadercount: 44
x-eopattributedmessage: 0
x-ms-office365-filtering-correlation-id: c8da35a6-3631-4209-0acf-08d8c7273c15
x-ms-traffictypediagnostic: DM6NAM10HT059:
x-microsoft-antispam: BCL:0;
x-microsoft-antispam-message-info:
 t5rOkM19Tbv81QsyfNRAWjCH5caeo7DdxyGnJMYRkzcCrLKGm20yUsj9frgf2b0yFj7v87oAm2w0jP1a7lXiR0ZzPZ74TAohLE3HgAP4T0d1+6NDyICv1tACtTO6s0cd2dmz2zODYaVbDivf6lWEtE7MF5cMuX4rAgGWUJN86TsW0xUcRdT3YLv2dTO/oUFuSliDNrQCmTjvVysI/OvCHAPoKsNWbYFIzdsb67SqSi0KPZiqTgZ/zD5ZsisNa1gTS3jOKHbcviRKBFfvX1H8iJOIB839mVmWw7FqjzbYCgfJUzOGbTfgX25W3EDjBoLouI2KgSZtOk5LI6RFkDC0SkNgS3yPTYDNxpvHo0gGiSOpIOYxuopgqU7gfWbnPiplR4WloXl1NRN8Jgho3+lGbVRO4MIa8HQnIgSDQQ6h9rVT8zKQXREwAtG9lHUaoA9L
x-ms-exchange-antispam-messagedata:
 RKLkPmSyhrZD6X3Z4W7pVZ1E0cjptC9DvFpsyRe7tGUudrZ9o+2SjFBRJmRsTo+wNjuspj/SQYug/tiGhjEPFlGzqas9ygPZVVpJM9D1wy69pf710nYxd8F5Wx5gDlBC5D8usfbS4XPbE/ScbOvvZLbCczR0819qVCdiXzr0xYmJAMz3/d0oLLX0Uer4h64JJUI3ud/PUVtLNTYAFa9NNQ==
x-ms-exchange-transport-forked: True
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: hotmail.com
X-MS-Exchange-CrossTenant-AuthAs: Anonymous
X-MS-Exchange-CrossTenant-AuthSource: DM6NAM10FT058.eop-nam10.prod.protection.outlook.com
X-MS-Exchange-CrossTenant-RMS-PersistedConsumerOrg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-CrossTenant-Network-Message-Id: c8da35a6-3631-4209-0acf-08d8c7273c15
X-MS-Exchange-CrossTenant-originalarrivaltime: 02 Feb 2021 03:04:19.8179
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Internet
X-MS-Exchange-CrossTenant-id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa
X-MS-Exchange-CrossTenant-rms-persistedconsumerorg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM6NAM10HT059
Subject: Re: [PHP-DEV] [RFC] Fibers
From: tysonandre775@hotmail.com (tyson andre)

Hi Aaron Piotrowski,=0A=
=0A=
> I would like to introduce an RFC for adding full-stack fibers to PHP: htt=
ps://wiki.php.net/rfc/fibers=0A=
>=0A=
> Fibers are primarily used to implement green-threads or coroutines for as=
ynchronous I/O. Fibers are similar to threads, except fibers exist within a=
 single thread and require cooperative scheduling of the fibers by the proc=
ess. Since fibers do not require a full CPU context switch, they are lightw=
eight and more performant than multi-processing or threading for awaiting I=
/O.=0A=
>=0A=
> An implementation as an extension is at https://github.com/amphp/ext-fibe=
r=0A=
>=0A=
> Fibers are a complex feature. The RFC contains many examples and links to=
 code using fibers to help explain and demonstrate what is possible, howeve=
r I=92m certain many more questions and concerns will arise. Looking forwar=
d to feedback and discussion.=0A=
=0A=
I took a look at the amphp/ext-fiber repo a few weeks ago - nothing stood o=
ut as a major concern but I'm only moderately familiar with threading,=0A=
A minor comment is that it'd be easier to read the phpt test cases if the `=
--EXPECTF--`=0A=
section included the file's basename in the test output of error traces ins=
tead of just `%s`, etc.=0A=
(e.g. tests/002-throw.phpt)=0A=
=0A=
```=0A=
Stack trace:=0A=
#0 %s(%d): {closure}()=0A=
#1 %s(%d): Loop->tick()=0A=
#2 %s(%d): Loop->run()=0A=
...=0A=
```=0A=
=0A=
I didn't see in the test cases/rfc:=0A=
How do the `zend_try`/`zend_catch`/`zend_end_try` macros and `zend_error_no=
return` macros get handled after a fatal error?=0A=
I'm not 100% sure - I think those use setjmp/longjmp internally - will fata=
l errors continue to work if there's a fatal error from within a fiber.=0A=
(e.g. `require_once` on a file with a compile-time fatal error such as dupl=
icate parameters).=0A=
I forget exactly how they work, e.g. in the context of a web server before =
this RFC - at a glance I'd guess an unrecoverable fatal error would cause t=
he worker to shut down.=0A=
Would being in a different fiber and different C stack interfere with the s=
hutdown process for fatal errors? (I guess the macros could be changed to m=
ake that switch to the main fiber if needed to fix that)=0A=
=0A=
Thanks,=0A=
-Tyson Andre=0A=