malloc: performance improvements and bugfixes

[Paul Eggert]

Thanks for doing all the work and bringing it to our attention. A couple 
of comments on the process:

On 01/25/2016 04:24 PM, Joern Engel wrote:
> I happen to prefer the kernel coding style over GNU coding style.

Nevertheless, let's keep the GNU coding style for glibc. In some places 
the existing code doesn't conform to that style but that can be fixed as 
we go.

> I believe there are some applications that
> have deeper knowledge about malloc-internal data structures than they
> should (*cough*emacs).  As a result it has become impossible to change
> the internals of malloc without breaking said applications

This underestimates Emacs. :-)

Emacs "knows" so much about glibc malloc's internal data structures that 
Emacs should do the right thing if glibc removes the hooks in question. 
Of course we should test the resulting combination. However, the point 
is that Emacs's usage of glibc malloc internals shouldn't ossify glibc 
malloc.