[PHP-BUG] Req #65858 [NEW]: json_encode could consider __get magic method
From: leonardo at ebussola dot com
Operating system: all
PHP version: Irrelevant
Package: JSON related
Bug Type: Feature/Change Request
Bug description:json_encode could consider __get magic method
Description:
When I use json_encode on a class, this get the properties values
directly and bypass the __get magic method.
It would be great if I could control the properties with __get
Test script:
---
class Example {
public foo;
public function __construct() {
$this->foo = 'bar;
}
public function __get($param) {
return $this->$param . ' modified';
}
}
$foo = new Example();
echo json_encode($foo);
Expected result:
{"foo":"bar modified"}
Actual result:
--
{"foo":"bar"}
--
Edit bug report at https://bugs.php.net/bug.php?id=65858&edit=1
--
Try a snapshot (PHP 5.4):
https://bugs.php.net/fix.php?id=65858&r=trysnapshot54
Try a snapshot (PHP 5.5):
https://bugs.php.net/fix.php?id=65858&r=trysnapshot55
Try a snapshot (trunk):
https://bugs.php.net/fix.php?id=65858&r=trysnapshottrunk
Fixed in SVN: https://bugs.php.net/fix.php?id=65858&r=fixed
Fixed in release: https://bugs.php.net/fix.php?id=65858&r=alreadyfixed
Need backtrace: https://bugs.php.net/fix.php?id=65858&r=needtrace
Need Reproduce Script: https://bugs.php.net/fix.php?id=65858&r=needscript
Try newer version: https://bugs.php.net/fix.php?id=65858&r=oldversion
Not developer issue:https://bugs.php.net/fix.php?id=65858&r=support
Expected behavior: https://bugs.php.net/fix.php?id=65858&r=notwrong
Not enough info:
https://bugs.php.net/fix.php?id=65858&r=notenoughinfo
Submitted twice:
https://bugs.php.net/fix.php?id=65858&r=submittedtwice
register_globals: https://bugs.php.net/fix.php?id=65858&r=globals
PHP 4 support discontinued: https://bugs.php.net/fix.php?id=65858&r=php4
Daylight Savings: https://bugs.php.net/fix.php?id=65858&r=dst
IIS Stability: https://bugs.php.net/fix.php?id=65858&r=isapi
Install GNU Sed:https://bugs.php.net/fix.php?id=65858&r=gnused
Floating point limitations: https://bugs.php.net/fix.php?id=65858&r=float
No Zend Extensions: https://bugs.php.net/fix.php?id=65858&r=nozend
MySQL Configuration Error: https://bugs.php.net/fix.php?id=65858&r=mysqlcfg
Bug #65822 [Csd]: crash on shutdown because of zend extension cleanup order
Edit report at https://bugs.php.net/bug.php?id=65822&edit=1
ID: 65822
User updated by:askalski at gmail dot com
Reported by:askalski at gmail dot com
Summary:crash on shutdown because of zend extension cleanup
order
Status: Closed
Type: Bug
Package:Reproducible crash
Operating System: Linux
PHP Version:5.4.20
Assigned To:laruence
Block user comment: N
Private report: N
New Comment:
Thanks; wasn't aware this had been fixed on APC trunk. Just noticed now that
APC has been moved from subversion to git - this made my day.
Unaware of the already-committed fixes, I had started on a parallel development
path of configuration to disable opcode caching and interned strings. One
thing I did in my version of the patch which you might consider:
In PHP_MINIT_FUNCTION(apc), I have it call zend_get_extension("Zend OPcache")
to see if opcache is loaded, and if so, force-disable opcode caching and
interned strings.
It's a minor convenience, but could help to avoid questions in the future.
Previous Comments:
[2013-10-08 11:20:31] larue...@php.net
actually, this is a knew issue.
and already be fixed both in opcache:
https://github.com/php/php-src/commit/0aa342e903bf012efe03db1e9f1fe4ed54225e76
and apc: http://svn.php.net/viewvc?view=revision&revision=330859
please try with the latest snapshot
[2013-10-07 18:30:24] askalski at gmail dot com
Turns out it's not so simple. It never is...
Modules need to be shutdown before the global function/class tables are
destroyed. Extensions need to be shutdown after. Consequently there is no
valid way to interleave extension and module shutdown.
Because this issue likely stems from APC meddling with PHP internals in a
manner in which modules were never intended to meddle, I see no real reason to
act upon this ticket... with one exception:
One thing that still ought to be fixed is zend_shutdown_extensions() should
apply zend_extension_shutdown over the zend_extensions list in *reverse* order,
not forward order. (This would involve adding a zend_llist_reverse_apply_*
family of functions to zend_llist.c)
[2013-10-03 16:00:22] askalski at gmail dot com
Spent some time digging into this. It's not as simple as calling
zend_shutdown_extensions before zend_destroy_modules; there are a few issues at
play which complicate matters.
The two Zend Extensions I investigated (OpCache and XDebug) implement both the
zend_extension and zend_module API. Both of these extensions load the
zend_module portion by calling zend_startup_module in the zend_extension
startup function. Also, it is possible to load modules at runtime using the
dl() userland function.
Consequently, the shutdown order is not a simple matter of "extensions first,
then modules". Because startup order of extensions and modules can be
interleaved, this precise order must be recorded during initialization. The
implementation might be something as simple as a stack of enumerated values: {
MODULE, MODULE, MODULE, MODULE, EXTENSION, MODULE, EXTENSION, MODULE }
[2013-10-02 22:09:12] askalski at gmail dot com
Description:
php_module_startup() initializes modules first, extensions second.
However, zend_shutdown() destroys them in the same order (modules first,
extensions second), rather than in stack order as one would expect.
Furthermore, it seems (based on reading the zend_startup_extensions()
zend_shutdown_extensions() functions) that if multiple zend extensions are
loaded, they are destroyed in the wrong order as well.
Multiple modules work fine; they are destroyed in stack order.
To reproduce the issue, load an extension and module which both override the
same Zend structure. For example, loading both OpCache 7.0.2 and APC 3.1.13
will cause a segfault on shutdown because of improper cleanup order of
orig_interned_strings_start, old_interned_strings_start, and
compiler_globals.interned_strings_start.
I'm aware that the example sounds like a bizarre combination of modules and
extensions here; I'm reporting the bug because it points at an issue in PHP
itself. The specific use case for loading both APC and OpCache is to use
OpCache for opcodes and APC with apc.cache_by_default=0 for the
apc_store/apc_fetch userland functions.
Test script:
---
zend_extension=/usr/lib64/php/modules/opcache.so
extension=apc.so
Load both OpCache and APC in mod_php in Apache "prefork" mode. Send SIGTERM to
one of the workers (or simply send enough requests to make Apache reap the
worker), and watch for the Segmentation fault in Apache's error_log.
Bug #53467 [Com]: Phar cannot compress large archives
Edit report at https://bugs.php.net/bug.php?id=53467&edit=1
ID: 53467
Comment by: Tr at visPaul dot me
Reported by:mep_eisen at web dot de
Summary:Phar cannot compress large archives
Status: Open
Type: Bug
Package:PHAR related
Operating System: Windows 7 - 64
PHP Version:5.3.3
Block user comment: N
Private report: N
New Comment:
I also am seeing this problem in Fedora 19 Php 5.5.1. I found 2 workarounds and
I'm curious if they works for others:
1. Don't attempt to compress the Phar. The compression seems to be the root of
the issue and throws an exception: "unable to create temporary file"
2. Bump the number of max files that you can open by editing
/etc/security/limits.conf or similar.
For whatever reason it seems that the Phar's compress method needs to open all
of the files at once. Perhaps a Php Dev can shed some light on this and
confirm/deny that it is a bug.
Previous Comments:
[2013-02-27 16:59:11] cicerc...@php.net
Still present with PHP 5.4 as well (with Ubuntu 12.04).
PHP 5.4.12-1~quantal+1 (cli) (built: Feb 25 2013 19:19:48)
Copyright (c) 1997-2013 The PHP Group
Zend Engine v2.4.0, Copyright (c) 1998-2013 Zend Technologies
with Xdebug v2.2.1, Copyright (c) 2002-2012, by Derick Rethans
[2012-07-27 12:10:16] mep_eisen at web dot de
This bug is still present. The workarounds on large phar files are resulting
high cpu-load and packaging phar in more than 30 mins.
[2011-09-05 06:15:37] d dot kreuer at kremedia dot de
I have the same problem under Ubuntu 10.04.3 with PHP 5.3.5
PHP 5.3.5-1ubuntu7.2ppa1~lucid with Suhosin-Patch (cli) (built: May 7 2011
03:15:14)
Copyright (c) 1997-2009 The PHP Group
Zend Engine v2.3.0, Copyright (c) 1998-2010 Zend Technologies
with the ionCube PHP Loader v4.0.7, Copyright (c) 2002-2011, by ionCube Ltd.
with Suhosin v0.9.29, Copyright (c) 2007, by SektionEins GmbH
[2010-12-03 20:31:20] mep_eisen at web dot de
Description:
I tried to create a large phar (exactly 2347 files). However compressing the
phar simply fails. That may be a similar problem as closed pecl bug:
http://pecl.php.net/bugs/bug.php?id=13727
Smaller phars are working. If using the commented section (compress them by
hand) it does work but this is really slow.
Test script:
---
startBuffering();
//$localDir =
realpath('D:\\Dev\\ws\\mavenphp-flow3\\FLOW3\\target\\classes\\Packages\\This\\FLOW3');
//$iter = new RecursiveIteratorIterator(new
RecursiveDirectoryIterator(realpath('D:\\Dev\\ws\\mavenphp-flow3\\FLOW3\\target\\classes\\Packages\\This\\FLOW3')));
//foreach ($iter as $file)
//{
// $pathName = $file->getPathname();
// $localName = substr($pathName, strlen($localDir));
// $phar->addFile($pathName, $localName);
// echo "adding $localName".PHP_EOL;
// $phar[$localName]->compress(Phar::GZ);
//}
$phar->buildFromIterator(new RecursiveIteratorIterator(new
RecursiveDirectoryIterator(realpath('D:\\Dev\\ws\\mavenphp-flow3\\FLOW3\\target\\classes\\Packages\\This\\FLOW3'))),
realpath('D:\\Dev\\ws\\mavenphp-flow3\\FLOW3\\target\\classes\\Packages\\This\\FLOW3'));
echo "packed ".$phar->count()." files".PHP_EOL;
echo "compressing".PHP_EOL;
$phar->compressFiles(Phar::GZ);
$phar->setStub('');
$phar->stopBuffering();
Expected result:
expected to get the phar.
Actual result:
--
BadMethodCallException: unable to create temporary file in D:\Dev\ws\mavenphp-fl
ow3\FLOW3\target\packagePhar.php on line 20
Call Stack:
0.0011 332160 1. {main}() D:\Dev\ws\mavenphp-flow3\FLOW3\target\packag
ePhar.php:0
5.15701068784 2. Phar->compressFiles() D:\Dev\ws\mavenphp-flow3\FLOW3\
target\packagePhar.php:20
--
Edit this bug report at https://bugs.php.net/bug.php?id=53467&edit=1
Req #53785 [Opn->Csd]: Way to query a X.509 certificate fingerprint
Edit report at https://bugs.php.net/bug.php?id=53785&edit=1
ID: 53785
Updated by: m...@php.net
Reported by:BenBE at geshi dot org
Summary:Way to query a X.509 certificate fingerprint
-Status: Open
+Status: Closed
Type: Feature/Change Request
Package:OpenSSL related
Operating System: Debian
PHP Version:5.3.5
-Assigned To:
+Assigned To:mike
Block user comment: N
Private report: N
New Comment:
Feature by Tjerk Meesters in master.
Previous Comments:
[2012-01-27 09:29:41] msn at searchy dot nl
The fingerprint is nothing more than the md5 or sha1 sum of the certificate.
When you take ther certificate, strip it of the begin and end tag, base64
decode the content and md5 that, you'll get the fingerprint.
Sample code:
$newcert = preg_replace("/-BEGIN CERTIFICATE-|-END
CERTIFICATE-/","",$cert);
$b64 = base64_decode($newcert);
echo "MD5 fingerprint: " . md5($b64) . "\n";
[2011-01-19 06:38:49] BenBE at geshi dot org
Description:
When reading a X.509 certificate file (or fetching the certificate from an SSL
connection) there is no straight forward way to query the certificate's
fingerprint as shown by browsers when viewing the site's certificate.
The output of openssl_x509_parse doesn't contain the fingerprint while
openssl_x509_export might contain it in a hard to parse string representation.
--
Edit this bug report at https://bugs.php.net/bug.php?id=53785&edit=1
Bug #64101 [Com]: SoapClient weirdness when passed undefined connection_timeout
Edit report at https://bugs.php.net/bug.php?id=64101&edit=1
ID: 64101
Comment by: nampuom at gmail dot com
Reported by:jeffdafoe at gmail dot com
Summary:SoapClient weirdness when passed undefined
connection_timeout
Status: Open
Type: Bug
Package:SOAP related
Operating System: CentOS linux, Debian stable
PHP Version:5.3.21
Block user comment: N
Private report: N
New Comment:
I confirm this bug.
Linux CentOS PHP 5.4.20 affected
Linux Fentoo PHP 5.3.18 also affected
It was very hard to locate this bug.
Also we see this behavior:
If try access to sub-element of non-existent array, then parent array
initialize to 0.
Test script:
--
#$_SESSION['application'] = array();
$_SESSION['application']['id'] = 123;
Expected result:
--
array(
'application'=>array('id'=>1)
)
Actual result:
--
array(
'application'=>0
)
Previous Comments:
[2013-07-05 15:03:17] 2013 at birth-online dot de
Here's an old bug report from the Ubuntu bugtracker from May 2012 and today we
stumbled upon this bug again using PHP 5.3.10-1ubuntu3.6
(latest version for Ubuntu 12.04LTS):
EXAMPLE CODE
'www.silversolutions.de',
'uri' => 'www.silversolutions.de', 'connection_timeout' => $timeout ) );
$header = array();
$header["header_general"]["order_type"] = "SHOP";
print_r( $header );
?>
COMMAND
---
php -n test.php
(NO CONFIG IS USED!!)
ACTUAL OUTPUT
-
Array
(
[header_general] => Array
(
[order_type] => SHOP
)
)
Warning: Cannot use a scalar value as an array in /home/mab/test.php on line 11
Array
(
[header_general] => 0
)
EXPECTED OUTPUT
---
Array
(
[header_general] => Array
(
[order_type] => SHOP
)
)
Array
(
[header_general] => Array
(
[order_type] => SHOP
)
)
This is on
Linux johndoe 3.2.0-24-virtual #37-Ubuntu SMP Wed Apr 25 10:17:19 UTC 2012
x86_64 x86_64 x86_64 GNU/Linux.
and
Linux johndoe 3.2.0-38-virtual #61-Ubuntu SMP Tue Feb 19 12:37:47 UTC 2013
x86_64 x86_64 x86_64 GNU/Linux
On a desktop machine with
Linux johndoe 3.2.0-24-generic #38-Ubuntu SMP Tue May 1 16:18:50 UTC 2012
x86_64 x86_64 x86_64 GNU/Linux
it works as expected.
[2013-01-29 21:00:24] jeffdafoe at gmail dot com
Description:
When SoapClient is passed an undefined variable in the connection_timeout hash
key, it seems that all global variables are initialized to 0.
I can reproduce this on multiple linux distributions and PHP versions. I can
also repro it using any WSDL. It only happens when connection_timeout is set
to an uninitialized variable, I am not able to reproduce it in any other case.
I discovered it by accident but figured I'd report it due to the unusual and
seemingly wide impact of the resultant behavior.
Test script:
---
$myi = null;
print 'Before: myi->foo=' . $myi->foo . ' fakevar=' . $fakevar . '
fakeobj->prop=' . $fakeobj->prop . "\n";
$url = 'http://soap.amazon.com/schemas2/AmazonWebServices.wsdl';
$sc = new SoapClient($url,
array(
'connection_timeout'=>$foo
)
);
print 'After: myi->foo=' . $myi->foo . ' fakevar=' . $fakevar . '
fakeobj->prop=' . $fakeobj->prop . "\n";
Expected result:
myi->foo= fakevar= fakeobj->prop=
Actual result:
--
myi->foo=0 fakevar=0 fakeobj->prop=0
--
Edit this bug report at https://bugs.php.net/bug.php?id=64101&edit=1
Bug #65729 [Opn->Csd]: CN_match gives false positive
Edit report at https://bugs.php.net/bug.php?id=65729&edit=1 ID: 65729 Updated by: m...@php.net Reported by:datib...@php.net Summary:CN_match gives false positive -Status: Open +Status: Closed Type: Bug Package:OpenSSL related Operating System: Linux PHP Version:master-Git-2013-09-21 (Git) -Assigned To: +Assigned To:mike Block user comment: N Private report: N New Comment: Merged. Previous Comments: [2013-09-21 08:27:27] datib...@php.net Description: When the CN_match option is used to verify "foo.test.com.sg" and the server certificate CN is "*.test.com" it will succeed erroneously. Test script: --- $context = stream_context_create(['ssl' => [ 'verify_peer' => true, 'allow_self_signed' => true, 'CN_match' => 'foo.test.com.sg', ]]); $s = file_get_contents($url, 'rt', $context); var_dump($s); Expected result: Warning: file_get_contents(): Peer certificate CN=`*.test.com' did not match expected CN=`foo.test.com.sg' in /path/to/script.php on line 12 Warning: file_get_contents(): Failed to enable crypto in /Users/tjerk/work/ssl/ssl.php on line 11 Warning: file_get_contents(https://localhost:4433): failed to open stream: operation failed in /path/to/script.php on line 11 bool(false) Actual result: -- No errors. -- Edit this bug report at https://bugs.php.net/bug.php?id=65729&edit=1
Bug #65845 [Com]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Comment by: guy dot cesaro at gmail dot com
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
Hello Laruence,
I've tried with php-cli (windows and ubuntu), the output is ok.
Antoher try with : Apache/2.4.6 (Ubuntu) + PHP Version
5.5.4-1+debphp.org~raring+1 and the output is not ok.
Can you please try the script with http server ?
Previous Comments:
[2013-10-08 11:05:44] larue...@php.net
Please try using this snapshot:
http://snaps.php.net/php5.5-latest.tar.gz
For Windows:
http://windows.php.net/snapshots/
please try with the latest snapshot of php5.5 in git
[2013-10-08 11:02:28] larue...@php.net
hmm, I run this with opcache enable, and fully optimizer.
seems the output is expected?
"
$ php55 -d opcache.enable=1 -d opcache.optimization_level=0x /tmp/1.php
This should echo empty string:
This should echo "tutu" string: tutu
"
[2013-10-08 10:44:11] brunobergot at gmail dot com
Me again, here is a last test script that work without changing
opcache.optimization_level :
APC BUG
<
// This should echo "tutu" string: >tutu<
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string:
>".vide($Pile['vars'][$zzz=(string)'toto'] = 'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:30:33] brunobergot at gmail dot com
Hi again Laruence,
We found that the problem comes form the cast in $Pile['vars'][(string)'toto'].
With this news test script it works as expected :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
var_dump($table);
var_dump($cle);
var_dump($table[$cle]);
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars']['toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:17:30] brunobergot at gmail dot com
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
https://bugs.php.net/bug.php?id=65845
--
Edit this bug report at https://bugs.php.net/bug.php?id=65845&edit=1
Bug #65776 [Opn]: Incorrect line endings causes segfault in convert.quoted-printable-encode
Edit report at https://bugs.php.net/bug.php?id=65776&edit=1
ID: 65776
Updated by: larue...@php.net
Reported by:mrubinsk at horde dot org
Summary:Incorrect line endings causes segfault in
convert.quoted-printable-encode
Status: Open
Type: Bug
Package:Streams related
Operating System: OSX, Ubuntu
PHP Version:5.4.20
-Assigned To:
+Assigned To:moriyoshi
Block user comment: N
Private report: N
New Comment:
hey Moriyoshi:
seems these codes are obviously wrong, if in_pp is null but the lb_ptr is
smaller than lb_cnt, then it must segfault in the later "*in_pp";
php-5.4/ext/standard/filters.c:805
if ((in_pp == NULL || in_left_p == NULL) && (lb_ptr >=lb_cnt)) {
return PHP_CONV_ERR_SUCCESS;
}
ps = (unsigned char *)(*in_pp);
Previous Comments:
[2013-09-27 18:50:58] mrubinsk at horde dot org
Description:
When using the convert.quoted-printable-encode stream filter, if the stream
contains a single bare CR line ending at the end of the stream when
line-break-chars is set to CRLF this can cause a segfault with certain values
of line-length.
Changing the line-length in the test script, or setting line-break-chars to
"\r" prevents the segfault.
Test script:
---
$data = fopen('php://temp', 'r+');
fwrite($data, "test\r\ntest\r\n\r");
$stream = fopen("php://temp", 'r+');
stream_filter_append($stream, 'convert.quoted-printable-encode',
STREAM_FILTER_WRITE, array('line-length' => 5, 'line-break-chars' => "\r\n"));
rewind($data);
stream_copy_to_stream($data, $stream);
Expected result:
The data should be copied from the $data stream to the $stream stream, applying
the convert filter.
Actual result:
--
Segfault.
--
Edit this bug report at https://bugs.php.net/bug.php?id=65776&edit=1
Bug #65822 [Opn->Csd]: crash on shutdown because of zend extension cleanup order
Edit report at https://bugs.php.net/bug.php?id=65822&edit=1
ID: 65822
Updated by: larue...@php.net
Reported by:askalski at gmail dot com
Summary:crash on shutdown because of zend extension cleanup
order
-Status: Open
+Status: Closed
Type: Bug
Package:Reproducible crash
Operating System: Linux
PHP Version:5.4.20
-Assigned To:
+Assigned To:laruence
Block user comment: N
Private report: N
New Comment:
actually, this is a knew issue.
and already be fixed both in opcache:
https://github.com/php/php-src/commit/0aa342e903bf012efe03db1e9f1fe4ed54225e76
and apc: http://svn.php.net/viewvc?view=revision&revision=330859
please try with the latest snapshot
Previous Comments:
[2013-10-07 18:30:24] askalski at gmail dot com
Turns out it's not so simple. It never is...
Modules need to be shutdown before the global function/class tables are
destroyed. Extensions need to be shutdown after. Consequently there is no
valid way to interleave extension and module shutdown.
Because this issue likely stems from APC meddling with PHP internals in a
manner in which modules were never intended to meddle, I see no real reason to
act upon this ticket... with one exception:
One thing that still ought to be fixed is zend_shutdown_extensions() should
apply zend_extension_shutdown over the zend_extensions list in *reverse* order,
not forward order. (This would involve adding a zend_llist_reverse_apply_*
family of functions to zend_llist.c)
[2013-10-03 16:00:22] askalski at gmail dot com
Spent some time digging into this. It's not as simple as calling
zend_shutdown_extensions before zend_destroy_modules; there are a few issues at
play which complicate matters.
The two Zend Extensions I investigated (OpCache and XDebug) implement both the
zend_extension and zend_module API. Both of these extensions load the
zend_module portion by calling zend_startup_module in the zend_extension
startup function. Also, it is possible to load modules at runtime using the
dl() userland function.
Consequently, the shutdown order is not a simple matter of "extensions first,
then modules". Because startup order of extensions and modules can be
interleaved, this precise order must be recorded during initialization. The
implementation might be something as simple as a stack of enumerated values: {
MODULE, MODULE, MODULE, MODULE, EXTENSION, MODULE, EXTENSION, MODULE }
[2013-10-02 22:09:12] askalski at gmail dot com
Description:
php_module_startup() initializes modules first, extensions second.
However, zend_shutdown() destroys them in the same order (modules first,
extensions second), rather than in stack order as one would expect.
Furthermore, it seems (based on reading the zend_startup_extensions()
zend_shutdown_extensions() functions) that if multiple zend extensions are
loaded, they are destroyed in the wrong order as well.
Multiple modules work fine; they are destroyed in stack order.
To reproduce the issue, load an extension and module which both override the
same Zend structure. For example, loading both OpCache 7.0.2 and APC 3.1.13
will cause a segfault on shutdown because of improper cleanup order of
orig_interned_strings_start, old_interned_strings_start, and
compiler_globals.interned_strings_start.
I'm aware that the example sounds like a bizarre combination of modules and
extensions here; I'm reporting the bug because it points at an issue in PHP
itself. The specific use case for loading both APC and OpCache is to use
OpCache for opcodes and APC with apc.cache_by_default=0 for the
apc_store/apc_fetch userland functions.
Test script:
---
zend_extension=/usr/lib64/php/modules/opcache.so
extension=apc.so
Load both OpCache and APC in mod_php in Apache "prefork" mode. Send SIGTERM to
one of the workers (or simply send enough requests to make Apache reap the
worker), and watch for the Segmentation fault in Apache's error_log.
Expected result:
No crash.
Actual result:
--
Segmentation fault.
--
Edit this bug report at https://bugs.php.net/bug.php?id=65822&edit=1
Bug #65845 [Fbk]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Updated by: larue...@php.net
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
Please try using this snapshot:
http://snaps.php.net/php5.5-latest.tar.gz
For Windows:
http://windows.php.net/snapshots/
please try with the latest snapshot of php5.5 in git
Previous Comments:
[2013-10-08 11:02:28] larue...@php.net
hmm, I run this with opcache enable, and fully optimizer.
seems the output is expected?
"
$ php55 -d opcache.enable=1 -d opcache.optimization_level=0x /tmp/1.php
This should echo empty string:
This should echo "tutu" string: tutu
"
[2013-10-08 10:44:11] brunobergot at gmail dot com
Me again, here is a last test script that work without changing
opcache.optimization_level :
APC BUG
<
// This should echo "tutu" string: >tutu<
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string:
>".vide($Pile['vars'][$zzz=(string)'toto'] = 'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:30:33] brunobergot at gmail dot com
Hi again Laruence,
We found that the problem comes form the cast in $Pile['vars'][(string)'toto'].
With this news test script it works as expected :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
var_dump($table);
var_dump($cle);
var_dump($table[$cle]);
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars']['toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:17:30] brunobergot at gmail dot com
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 05:56:39] larue...@php.net
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
hmm, are you able to get the compiled SPIP view file? it must be a PHP script
which can be used to reproduce this problem?
The remainder of the comments for this report are too long. To view
the r
Bug #65845 [Fbk]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Updated by: larue...@php.net
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
hmm, I run this with opcache enable, and fully optimizer.
seems the output is expected?
"
$ php55 -d opcache.enable=1 -d opcache.optimization_level=0x /tmp/1.php
This should echo empty string:
This should echo "tutu" string: tutu
"
Previous Comments:
[2013-10-08 10:44:11] brunobergot at gmail dot com
Me again, here is a last test script that work without changing
opcache.optimization_level :
APC BUG
<
// This should echo "tutu" string: >tutu<
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string:
>".vide($Pile['vars'][$zzz=(string)'toto'] = 'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:30:33] brunobergot at gmail dot com
Hi again Laruence,
We found that the problem comes form the cast in $Pile['vars'][(string)'toto'].
With this news test script it works as expected :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
var_dump($table);
var_dump($cle);
var_dump($table[$cle]);
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars']['toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:17:30] brunobergot at gmail dot com
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 05:56:39] larue...@php.net
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
hmm, are you able to get the compiled SPIP view file? it must be a PHP script
which can be used to reproduce this problem?
[2013-10-06 18:13:34] bruno+php at ahennezel dot info
Wrong package : Opcache
The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
https://bugs.php.net/bug.php?id=65845
--
Edit this bug report at https://bugs.
Bug #65845 [Com]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Comment by: brunobergot at gmail dot com
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
Me again, here is a last test script that work without changing
opcache.optimization_level :
APC BUG
<
// This should echo "tutu" string: >tutu<
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string:
>".vide($Pile['vars'][$zzz=(string)'toto'] = 'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
Previous Comments:
[2013-10-08 10:30:33] brunobergot at gmail dot com
Hi again Laruence,
We found that the problem comes form the cast in $Pile['vars'][(string)'toto'].
With this news test script it works as expected :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
var_dump($table);
var_dump($cle);
var_dump($table[$cle]);
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars']['toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 10:17:30] brunobergot at gmail dot com
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 05:56:39] larue...@php.net
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
hmm, are you able to get the compiled SPIP view file? it must be a PHP script
which can be used to reproduce this problem?
[2013-10-06 18:13:34] bruno+php at ahennezel dot info
Wrong package : Opcache
[2013-10-06 18:10:34] bruno+php at ahennezel dot info
Typo : With opcache.optimization_level=0xffef the result is "value".
The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
https://bugs.php.net/bug.php?id=65845
--
Edit this bug report at https://bugs.php.net/bug.php?id=65845&edit=1
Bug #65845 [Com]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Comment by: brunobergot at gmail dot com
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
Hi again Laruence,
We found that the problem comes form the cast in $Pile['vars'][(string)'toto'].
With this news test script it works as expected :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
var_dump($table);
var_dump($cle);
var_dump($table[$cle]);
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars']['toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
Previous Comments:
[2013-10-08 10:17:30] brunobergot at gmail dot com
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
[2013-10-08 05:56:39] larue...@php.net
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
hmm, are you able to get the compiled SPIP view file? it must be a PHP script
which can be used to reproduce this problem?
[2013-10-06 18:13:34] bruno+php at ahennezel dot info
Wrong package : Opcache
[2013-10-06 18:10:34] bruno+php at ahennezel dot info
Typo : With opcache.optimization_level=0xffef the result is "value".
[2013-10-06 11:38:42] bruno+php at ahennezel dot info
Description:
With the SPIP CMS V3, the default optimization level
opcache.optimization_level=0x
and
opcache.optimization_level=0xffef
give different results.
Test script:
---
I dont know how to test directly in PHP, but here is a very short SPIP snippet
which brings out the problem. Put in the SPIP's sommaire.html skeleton :
#SET{var,value}
[(#GET{var})]
With opcache.optimization_level=0x the result is empty.
With opcache.optimization_level=0x the result is "value".
Expected result:
I expect the content "value"
Actual result:
--
The content is empty
--
Edit this bug report at https://bugs.php.net/bug.php?id=65845&edit=1
Bug #65845 [Com]: Error when Zend Opcache Optimizer is fully enabled
Edit report at https://bugs.php.net/bug.php?id=65845&edit=1
ID: 65845
Comment by: brunobergot at gmail dot com
Reported by:bruno+php at ahennezel dot info
Summary:Error when Zend Opcache Optimizer is fully enabled
Status: Feedback
Type: Bug
Package:opcache
Operating System: Linux 3.11.2-1-ARCH
PHP Version:5.5.4
Block user comment: N
Private report: N
New Comment:
Hi Laurence,
Here is a script that show the bug in action :
APC BUG
<
// This should echo "tutu" string: >tutu<
// Works as expected with opcache.optimization_level=0xffef in php.ini
function table_valeur($table, $cle, $defaut='') {
foreach (explode('/', $cle) as $k) {
$table = is_string($table) ? @unserialize($table) : $table;
if (is_object($table)) {
$table = (($k !== "") and isset($table->$k)) ? $table->$k :
$defaut;
} elseif (is_array($table)) {
$table = isset($table[$k]) ? $table[$k] : $defaut;
} else {
$table = $defaut;
}
}
return $table;
}
function vide($texte){
return "";
}
echo "This should echo empty string: >".vide($Pile['vars'][(string)'toto'] =
'tutu')."<";
echo "This should echo \"tutu\" string:
>".table_valeur($Pile['vars'],'toto')."<";
?>
Previous Comments:
[2013-10-08 05:56:39] larue...@php.net
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
hmm, are you able to get the compiled SPIP view file? it must be a PHP script
which can be used to reproduce this problem?
[2013-10-06 18:13:34] bruno+php at ahennezel dot info
Wrong package : Opcache
[2013-10-06 18:10:34] bruno+php at ahennezel dot info
Typo : With opcache.optimization_level=0xffef the result is "value".
[2013-10-06 11:38:42] bruno+php at ahennezel dot info
Description:
With the SPIP CMS V3, the default optimization level
opcache.optimization_level=0x
and
opcache.optimization_level=0xffef
give different results.
Test script:
---
I dont know how to test directly in PHP, but here is a very short SPIP snippet
which brings out the problem. Put in the SPIP's sommaire.html skeleton :
#SET{var,value}
[(#GET{var})]
With opcache.optimization_level=0x the result is empty.
With opcache.optimization_level=0x the result is "value".
Expected result:
I expect the content "value"
Actual result:
--
The content is empty
--
Edit this bug report at https://bugs.php.net/bug.php?id=65845&edit=1
