[PATCH v6 3/4] Reduce CAS in malloc spinlocks

[DJ Delorie]

"H.J. Lu" <hjl.tools@gmail.com> writes:
> I didn't pursue it further since I couldn't show how much it would
> improve performance.

Yeah, that's one of the big problems with malloc is benchmarking - it's
easy to write a bad benchmark, and hard to capture what real-world
applications are doing.  And any benchmark complex enough to be relevent
in this case would be chaotic enough to lose the tiny changes in the
noise.