[PHP-DEV] Best practice of initializing zval array hash?

2016-10-14 Thread Yasuo Ohgaki
Hi all, I'm wandering what is the best practice of zval array hash initialization. Following code leaks 56 bytes on ZVAL_NEW_ARR() chash = php_mb_convert_encoding_recursive(HASH_OF(entry), _to_encoding, _from_encodings); if (chash) { ZVAL_NEW_ARR(&entry_tmp); // _array_init() is called. Should

[PHP-DEV] Limited access to php_version.h [Fwd: [PHP-CVS] svn: /SVNROOT/ global_avail]

2016-10-14 Thread Johannes Schlüter
Hi, I've restricted access to main/php_version.h. This should restrict wrong merges by a bit. Hope this doesn't cause issues. Current RMs should have access. This is the current list: avail|rasmus,johannes,tyrael,ab,jpauli,davey,krakjoe,dsp|php-src.git/main/php_version.h johannes --- Begin Messa

[PHP-DEV] lxr.php.net is coming soon.... finally....

2016-10-14 Thread Yasuo Ohgaki
Hi all, LXR is very useful, but the site is down for a long time. http://lxr.php.net/ Is this going to be fixed or it will be gone forever? Regards, -- Yasuo Ohgaki yohg...@ohgaki.net -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] PHP 5.6.27 is available

2016-10-14 Thread Ferenc Kovacs
Hello! The PHP development team announces the immediate availability of PHP 5.6.27. Several security related issues were fixed in this release. All PHP 5.6 users are encouraged to upgrade to this version. For source downloads of PHP 5.6.27 please visit our downloads page: http://www.php.net/downl

Re: [PHP-DEV] Performance drops after some time, PHP7-FPM + Docker

2016-10-14 Thread Rasmus Lerdorf
On Fri, Oct 14, 2016 at 9:15 PM, Jérémie BORDIER wrote: > > I don't really know how to investigate further. If you have any > pointers on how to help figuring out what's wrong, I'd love to try. > I would breakout the Linux perf command for something like this. Run it like this: perf record -p

Re: [PHP-DEV] [!] Master merged into PHP-7.1

2016-10-14 Thread Joe Watkins
Afternoon all, urm ... sorry, I'm so terrible at git ... Cheers Joe On Thu, Oct 13, 2016 at 11:07 PM, Yasuo Ohgaki wrote: > On Fri, Oct 14, 2016 at 4:51 AM, Lauri Kenttä > wrote: > > Actually I just followed the lines in the git graph: > > > > $ git log origin/PHP-7.1.0RC3 --oneline | hea

[PHP-DEV] NEUTRAL Benchmark Results for PHP Master 2016-10-14

2016-10-14 Thread lp_benchmark_robot
Results for project PHP master, build date 2016-10-14 16:17:56+03:00 commit: 77584e9 previous commit:522dbc0 revision date: 2016-10-13 15:40:52+02:00 environment:Haswell-EP cpu:Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2x18 cores, stepping 2, LLC 45 MB

[PHP-DEV] Performance drops after some time, PHP7-FPM + Docker

2016-10-14 Thread Jérémie BORDIER
Hello, I am seeing a strange performance drop on our production servers in the PHP components. It happens usually after 1-3 uptime days, the response time suddenly doubles. I already investigated to try to isolate any strange request, but found nothing. We are having 2k to 4k RPM, the stack is PH

Re: [PHP-DEV] IMPORTANT: Git reset after wrong merge

2016-10-14 Thread Christoph M. Becker
On 14.10.2016 at 04:25, Anatol Belski wrote: > I would explicitly ask Dmitry, Sara, Hui, Stas, Bob, Nikita and > Christoph to check that all the code pushed after Monday 2016-10-10 > 13:11:45 is where it's expected to be. But also everyone else > involved/interested - it would be a big help to dou

Re: [PHP-DEV] ext/gd: Deprecate png2wbmp() and jpeg2wbmp()

2016-10-14 Thread Pierre Joye
On Oct 14, 2016 1:02 AM, "Christoph M. Becker" wrote: > > Hi! > > I suggest to deprecate png2wbmp() and jpeg2wbmp() as of PHP 7.2 and to > remove the functions as of PHP 8.0. I fully agree. And I oppose to other propositions to remove them (or anything else) in 7.x. Cheers Pierre

Re: [PHP-DEV] ext/gd: Deprecate png2wbmp() and jpeg2wbmp()

2016-10-14 Thread Christoph M. Becker
On 14.10.2016 at 00:50, Kalle Sommer Nielsen wrote: > 2016-10-13 23:17 GMT+02:00 Dan Ackroyd : > >> On 13 October 2016 at 18:03, Christoph M. Becker wrote: >> >>> Implementing a fallback in userland would be trivial >> >> You should do that first, and get at least some people to use them, if >> y