Branch: Tag:

2010-10-05

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Discussed two ways to solve gc of weakly ref'ed things.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Did away with preemptive thread suspend and resume.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Sorted out the issue with concurrent gc access to pointers.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Various small changes.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Some clarifications for shared mapping/multiset data blocks.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Googled a bit on tls support. OS X is a problem. :P

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Talk about the string append case for single-refcount optimizations.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

More on weak refs.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

State compatibility as a goal.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

More on single-refcount optimizations.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Detailing asynchronous lock space pointer access.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

Various small changes.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

More details on which pike stack frames the micro-gc has to scan.

2010-10-05 22:46:22 by Martin Stjernholm <mast@lysator.liu.se>

TLS on OS X is a no-go in the foreseeable future. :(

2010-10-05 22:46:21 by Martin Stjernholm <mast@lysator.liu.se>

Various changes based on feedback and new ideas.

2010-10-05 22:46:21 by Martin Stjernholm <mast@lysator.liu.se>

Described new gc approach.

2010-10-05 22:46:21 by Martin Stjernholm <mast@lysator.liu.se>

Initial version.