Re: [PATCH 4/7] Add a memory pool library

[Jonathan Nieder <jrnieder@gmail.com>]

Hi Ram,

Ramkumar Ramachandra wrote:

> I've never used talloc. What basis can we use for the comparison?

Ah, a little explanation is in order.

You see, I read the commit message, having just looked at the treap
implementation, read the word “allocator”, and decided that a
mini-malloc was exactly the sort of thing I did not want to read right
then.  Hence the silly comment about some allocator that pools memory
for heterogeneous-sized blocks (which is very nice and convenient, but
not so relevant!).  Sorry --- I should not have been so lazy.

In fact, this patch just maintains a growing array of never-freed
fixed-size blocks.  It is very similar to the existing alloc.c but
with some small differences:

 . uses a single memory area that grows with realloc, so it can be
   freed for valgrind-cleanness if one wants.

 . grows a little more aggressively, as David pointed out

And it is much simpler than I was thinking, so honestly, I don’t
even mind the code duplication so much.

For interest, the analogous CCAN module is
http://ccan.ozlabs.org/info/block_pool.html

> On a related note, note that while compiling, a lot of
> unused functions are generated which pop up as compiler warnings

Maybe __attribute__((__unused__)) could work.

Sorry for the nonsense,
Jonathan