[openbeosnetteam] Re: Pools implementation

> Well, just running it here's what I get...
> 
> For 50 loops malloc()/free() took 3183232
>         malloc() took 3009073
>         free() took 174159
> For 50 loops pool_get()/pool_put() took 1400692 (44%)
>         pool_get() took 902635 (29%)
>         pool_put() took 498057 (285%)
> 
> These look good for me!

Is this the old pool code, or the new one?

Anyway, I have put some #ifdef in the code to use benaphores instead of 
a read/write lock (if POOL_USES_BENAPHORES is 1 at compile time):

With a block size of 256 bytes:

For 50 loops malloc()/free() took 125041
        malloc() took 108324
        free() took 16717
For 50 loops pool_get()/pool_put() took 12664 (10%)
        pool_get() took 6626 (6%)
        pool_put() took 6038 (36%)

Should I check this code in? Of course it's much faster for this test, 
but it may be slower when there is much concurrent access.
And we could go back to a read/write lock when we set 
POOL_USES_BENAPHORES to 0 in no time (could be nice to check that once 
the whole stack is done).

> Guess which one is our stack and which is BONE? Not what I would have
> expected at all really...

Nice :-)

Adios...
   Axel.

Other related posts: