I don't buy into the argument that we shouldn't start even trying to solve the thread safety issues in PHP because of some arbitrary "we can't tell" or "it's faster not to do it" sort of argument. Threads aren't exactly an archaic or edge technology, and it's just stubborn of us not to support them. Having a means by which to identify and keep track of what extensions are considered "thread safe" is the only way to take a reasonable step forward on this issue. Saying TS of an ext could change with every commit is a bogus argument (anything can), and although it's reasonable to say there is a slight speed loss to PHP itself when operating in threaded mode we are simply supporting the technology not recommending it.
On Thu, 2005-08-25 at 09:17 +0200, Marcus Boerger wrote: > Hello Andi, > > wow, now that makes me wonder if you perhaps also know a reason for? > I mean in theory it should be faster shouldn't it? Or is the problem > that we far to often use TRSMLS_FETCH() with all its disadvantages? > > best regards > marcus > > Thursday, August 25, 2005, 1:28:49 AM, you wrote: > > > Marcus, > > > You will most likely find that the "faster" Apache way with > > thread-safe PHP is slower than the slower Apache way with > > non-thread-safe PHP. And even FastCGI will be faster :) > > > Andi > > > At 12:25 PM 8/24/2005, Marcus Boerger wrote: > >>Hello John, > >> > >>Wednesday, August 24, 2005, 5:22:07 PM, you wrote: > >> > >> > >> > On Wed, 2005-08-24 at 17:41 +0300, Zeev Suraski wrote: > >> >> Maybe we can give extensions a way to indicate that they're Unicode > >> >> compatible, and assume they're not if they don't. Non-compatible > >> >> extensions will not be loaded and produce an error. > >> > >> > Not to hijack the topic, but if we are going to do something like this > >> > why not also provide these sorts of flags for things like extensions and > >> > thread safety? > >> > >>Even though a change in an external lib or a commit in our source might > >>change this - i see those cases very rarly and typically detected by the > >>maintainers easily. Our code is threadsafe and most libs are, too. > >>The other thing is the advantage and that is very big to my guesses since > >>it would allow us to go with the faster apache way finally. So i like > >>this idea pretty much. > -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
