Re:GCC 4.7 adds transactional memory extensions for C/C++
by CGamesPlay on Monday November 21, 2011 @12:58PM
(#3262025)
This is pretty cool. Is it done naively and using one global lock, or is it be more intelligent? Can GCC identify what memory location require locking for a given transaction, and lock just those? What is the granularity?
[ Parent
| on HN ]
Re:GCC 4.7 adds transactional memory extensions for C/C++
by roxtar on Monday November 21, 2011 @08:29PM
(#3263626)
It's smarter than a global lock (come on!). I would suggest reading the article which describes the implementation [1].
[1]: http://www.velox-project.eu/velox-transactional-memory-stack (courtesy scott_s)