Re: Merge fix for #54918 (r311748) to 5.4?
| From: | David Soria Parra | Date: | Fri, 30 Sep 2011 13:06:39 +0000 |
| Subject: | Re: Merge fix for #54918 (r311748) to 5.4? | ||
| References: | 1 2 | Groups: | php.internals |
| Request: | Send a blank email to internals+get-55674@lists.php.net to get a copy of this message | ||
On 2011-09-26, Stas Malyshev <smalyshev@sugarcrm.com> wrote:
> Hi!
>
> On 9/26/11 1:42 AM, Pierre Joye wrote:
>> It may be good to merge
>> http://svn.php.net/viewvc?view=revision&revision=311748
>> to 5.4 too.
>> This commit is about issues similar to the one describe in #54918.
>>
>> Any reason not to do it before next beta/rc?
>
> I'm not sure I understand this patch completely. I understand that if
> one of the MINITs fails, this patch just continues. But we have tons of
> code that assumes all basic MINITs worked - we always use variables
> initialized in these MINITs without checking, etc. So does it really
> make sense to continue if one of basic MINITs fail? On regular module
> loading, MINIT failure is E_CORE_ERROR, why not here?
I'm not sure about this too. Is there a reason why we continue to MINIT
the other basic modules if one fails ?