malloc: performance improvements and bugfixes

[Florian Weimer]

On 01/26/2016 04:21 AM, Jörn Engel wrote:

> And just in case it wasn't clear, all the above is my personal opinion
> and not that of my employer.  Maybe my employer sees enough merit in
> getting code upstream to handle the legal work.  I do not and will not
> pursue this any further.

Even if you cannot contribute code directly, performance testing of
malloc changes by others is extremely helpful.  In the past, lack of
credible, real-world benchmarking prevented us from making any
fundamental changes to the malloc code.

Florian