MacOS10.4 thread stopping code (was: Deadlock in Class::forName on OSX)

Tue May 30 23:52:00 GMT 2006

On 2006年5月31日, Juerg Lehni wrote:
> Regarding your concerns about breaking older versions: As far as I can 
> remember, GCJ on Mac only works well since 10.4, as some Apple weirdness in 
> the compiler / linker slowed startup time down a lot in older versions, and 
> broke some of the fundamental features in GCJ (Andreas Tobler can tell more 
> about this). Basically it made GCJ on these versions of OS X only usefull for 
> playing around a bit and testing some things, but not more.
>> So if 10.4 would allow to use more standard pthread mechanisms and using 
> these would mean to break compatibility with 10.3 and lower, I wonder how bad 
> this really would be.

This isn't exactly true. It all depends what you mean with 'works well'. I've 
had PyLucene pretty happily working on OS X since 10.3 with gcj 3.4.x.
gcj 3.4.x works well on 10.4 and now, with a bunch of patches, gcj 4.0.2 even 
works on Intel Mac 10.4.
There is a darwinports of gcj 4.1.0 which I managed to build on ppc 10.4 and 
run PyLucene on just fine.
I'm not sure of the wisdom of breaking 10.3 gcj support at the moment, but 
there's always gcj 3.4.x, I guess.
Actually, at the moment gcj 3.4.x works the best (except on intel mac) because 
I can statically link my python extension .so with libgcj.a. I cannot do this 
with gcj 4.x (I can link a .so but it doesn't run, it crashes during JVM 
init).
Andi..