[cairo] Overhead reduction

Clemens Eisserer linuxhippy at gmail.com
Sun May 17 11:07:51 PDT 2009

Hi Jonathan,

> Okay, that's useful to know.  An atomic-test-and-inc routine should be
> roughly the same speed as a plain i++ on most modern hardware, it just
> needs to be implemented properly, which is an utter pain.
In a real-world benchmark, I lost about 90 cycles for cmpxchg based lock+unlock.
On an old P4 its arround a few hundred cycles ;)

- Clemens

More information about the cairo mailing list