4c9acdf153
ARM stuff.
14 years ago
4796a41e4e
Finished implementing the new runtime APIs.
14 years ago
4e3e71c750
Tweak 64-bit support for small objects.
15 years ago
8af52846b7
Small object (hidden in a pointer) support.
15 years ago
4ffb19230f
Make sure that the statics used to store internal tables are marked as roots.
15 years ago
0cb2ea01bc
Simplify the finalization code slightly.
15 years ago
d51500184c
More cleanups. Stop using __ identifiers for internal stuff now that it's properly marked private and we don't have to worry about conflicts.
15 years ago
6c7cf4a5a6
Add visibility attributes on all internal functions so that we aren't ever exporting them outside of libobjc.
15 years ago
95ef5a93f9
Always use the first version of a class that's loaded twice until the merging code is a bit better tested.
15 years ago
ee7817cf57
Implement a hook that allows the compiler to register the aliases used with the
...
@compatibility_alias directive so that they can be resolved at runtime.
15 years ago
faa741b133
Added some header documentation for runtime functions.
15 years ago
e0719a9c62
Added Objective-C++ exception handling support. Allows throwing Objective-C objects with throw or @throw and catching them with catch() or @catch.
15 years ago
5df6a42107
Log a warning when loading the same class twice.
15 years ago
db53012509
Disabled loaded class test.
15 years ago
ca2c7d1d9d
Stop treating NSObject as magic and do what OS X does with regard to root objects and metaclasses.
15 years ago
5574585b0f
Don't load the same class twice.
15 years ago
8db63b976f
Initial work on developer mode (allows classes to be replaced at run time with new versions, by cooperating IDEs). This support is part of ongoing work to provide fix-and-continue functionality with Clang / Cling and LanguageKit.
15 years ago
f66a69561a
Fix metametaclass setup.
15 years ago
76bb7762ff
Properly resolve the metaclass's isa pointer.
15 years ago
b6abe490ab
Remove some obsolete comments.
16 years ago
ed1c9d41eb
libobjc2: Add a pile of NULL checks. In general, any public function should be able to handle NULL for any object/class/selector/pointer to opaque structure without blowing up.
16 years ago
42a545fc7c
In some irritating circumstances, the load order can cause a +load message to trigger other code, which sends messages with selectors that have not been mapped yet.
...
The runtime was handling the message lookup correctly in this case, but only as a side-effect of the fallback code from when type-dependent dispatch is not possible. This resulted in a confusing warning message, telling you that you were calling a method with an incorrect signature, when the two signatures that it printed were the same.
This case is now handled correctly, so the warning disappears (unless you really are calling a method with the wrong signature and have TDD enabled).
16 years ago
9986294a51
Fix setting the superclass for metaclasses correctly.
16 years ago
d3122bcfa4
Make sure BOOLs in class tests are always YES or NO.
...
Make self in +load the correct value (the class, not the metaclass).
16 years ago
666c84c129
Test the correct class when resolving superclasses.
16 years ago
0e374db1db
Added support for type-dependent dispatch to libobjc2.
16 years ago
79899f8a46
Fixed unresolved class list loss when the root class gets resolved.
...
Hence not all classes were previously resolved at load time.
Was breaking +forwardInvocation: because unresolved classes such as
GSFFInvocation were not receiving +load right when gnustep-base was loaded.
16 years ago
779b28abeb
Lots of tidying, removing legacy stuff.
16 years ago
bb8cafaa15
Finished rewriting the loader. It's now cleanly separated into logically-separate components, so there's a chance it might actually be maintainable...
16 years ago
4ea82e14da
Fixed hash functions.
...
Tweaked class resolve function to make sure that it keeps scanning the list as long as one or more classes is resolved while scanning it.
16 years ago
13964517ef
Small 64bit fix.
16 years ago
de71ebb6b1
Rewritten class loading.
16 years ago
a1329e22de
A bit more class table tidying.
16 years ago
7f6f798074
Rewrote some of the class table code to be slightly cleaner.
16 years ago
09ed66f819
Fixed use of uninitialized variable.
16 years ago
e85975a36e
Don't make NSObject a subclass of NSObject. Thanks to Truls Becken!
16 years ago
a35ece500f
Rewrote objc_getClassList() to use class table directly.
...
Fixed missing UNLOCK() from last commit.
16 years ago
0abdc2a574
Small cleanups from last commit.
16 years ago
8ce32fd1c9
Rewrote class resolving code. Now it only resolves classes that actually need resolving, rather than iterating over the entire class table.
...
(This is part 1. Part 2 will lazily resolve classes.)
16 years ago
c1a637a990
Allowed hash tables to dynamically resize. This uses toydispatch to free the old version of the table after copying all of the data into the new one, after waiting enough time to allow all potential readers to finish. There are theoretical cases when this can fail, but hopefully no practical ones. The theoretical cases can be removed by making the garbage thread run at a lower priority than all of the other threads in the system.
...
The protocols table and the class table will now grow as required. Protocols are now added to the protocols table as they are created, so you should get something sensible and introspectable back when you look up a protocol by name.
16 years ago
bdf97cf64e
Added hopscotch table implementation and rewrote the class table to use it. This is done so that the class table and protocol tables can share the same code (currently there is no protocol table).
...
Note that concurrent resizing has not yet been implemented. That means that there is a hard limit on the number of classes that can be loaded. This is currently set to quite a small number for testing, to stress the hash table. If you're experiencing problems as a result, then please let me know and I will increase it.
16 years ago
theraven
thebeing
ericwa
qmathe