Malloc improvements

[Anton Blanchard]

Hi DJ,

> Hmmm... not sure why that test case is worse in my branch, the whole
> point of my work is to add a lockless fast path.  I'll have to
> investigate that some more.  Conveniently, I have a trace feature in
> there that I'm also working on ;-)

I took a look at this, and malloc(8) is trying to grab from tc_idx 2,
but free() is returning it to tc_idx 3.

Inside malloc(), size2tidx() rounds the request up to 32 bytes, but in
free() size is 48 bytes, size-SIZE_SZ is 40, and this causes us to
return it to tc_idx 3. Did we add 16B of padding instead of 8?

Anton