Permalink
Commits on Jan 8, 2016
  1. Update "Compatibility"

    laruence committed Jan 8, 2016
Commits on Apr 14, 2015
  1. back to dev

    laruence committed Apr 14, 2015
  2. Prepare for 7.0.5

    laruence committed Apr 14, 2015
Commits on Apr 3, 2015
  1. Safer fix

    laruence committed Apr 3, 2015
  2. Fixed test fail for 5.2

    laruence committed Apr 3, 2015
Commits on Apr 1, 2015
Commits on Mar 31, 2015
  1. Fixed bug #68677 (Use After Free in OPcache)

    (cherry picked from commit 777c39f4042327eac4b63c7ee87dc1c7a09a3115)
    laruence committed with kaplanlior Jan 8, 2015
Commits on Mar 10, 2015
  1. Merge branch 'master' of github.com:zendtech/ZendOptimizerPlus

    Conflicts:
    	tests/bug69125.phpt
    dstogov committed Mar 10, 2015
  2. Merge branch PHP-5.5

    dstogov committed Mar 10, 2015
Commits on Mar 4, 2015
  1. Fixed test

    laruence committed Mar 4, 2015
Commits on Mar 3, 2015
  1. Merge branch 'opcache' of ../php-5.5

    Conflicts:
    	Optimizer/zend_optimizer.c
    	tests/bug69038.phpt
    laruence committed Mar 3, 2015
  2. Make killing a locker an ERROR event

    At least in apache, killing a locker with SIGKILL from inside opcache prevents
    the child process from emitting a corresponding log event to the access or
    error logs, therefore requests are termed with no record.  This makes is
    incredibly difficult to debug problems espically on systems with high counts
    of cache restarts.  Since INFO level is far too noisy for production use,
    switch the killing to a higher level so we can see it easier.
    nathanhruby committed with laruence Mar 3, 2015
  3. Make killing a locker an ERROR event

    At least in apache, killing a locker with SIGKILL from inside opcache prevents
    the child process from emitting a corresponding log event to the access or
    error logs, therefore requests are termed with no record.  This makes is
    incredibly difficult to debug problems espically on systems with high counts
    of cache restarts.  Since INFO level is far too noisy for production use,
    switch the killing to a higher level so we can see it easier.
    nathanhruby committed with laruence Mar 3, 2015
  4. Merge pull request #206 from nathanhruby/nrh/locker-kill-logging

    Make killing a locker an ERROR event.
    laruence committed Mar 3, 2015
Commits on Feb 27, 2015
Commits on Feb 26, 2015
  1. Make killing a locker an ERROR event.

    At least in apache, killing a locker with SIGKILL from inside opcache prevents
    the child process from emitting a corresponding log event to the access or
    error logs, therefore requests are termed with no record.  This makes is
    incredibly difficult to debug problems espically on systems with high counts
    of cache restarts.  Since INFO level is far too noisy for production use,
    switch the killing to a higher level so we can see it easier.
    nathanhruby committed Feb 26, 2015
Commits on Feb 25, 2015
  1. better name

    laruence committed Feb 25, 2015
  2. better name

    laruence committed Feb 25, 2015
  3. Improve fix for #69038

    laruence committed Feb 25, 2015
  4. Improve fix for #69038

    laruence committed Feb 25, 2015
Commits on Feb 23, 2015
  1. Fixed build for 5.4-

    laruence committed Feb 23, 2015
Commits on Feb 20, 2015
  1. Merge branch PHP-5.5 of master

    dstogov committed Feb 20, 2015
  2. Fixed type mismatch

    dstogov committed Feb 20, 2015
  3. Fixed type mismatch

    dstogov committed Feb 20, 2015
Commits on Feb 17, 2015
Commits on Feb 9, 2015
  1. Shut up, my lovely compiler; I do not like your warnings

    Conflicts:
    	Zend/zend_object_handlers.c
    bwoebi committed with laruence Apr 11, 2014
  2. Shut up, my lovely compiler; I do not like your warnings

    Conflicts:
    	Zend/zend_object_handlers.c
    bwoebi committed with laruence Apr 11, 2014
Commits on Jan 22, 2015