Re: [squid-users] Memory pools: why use them?

[moved to squid-dev]

On Wed, 1 Sep 2004, Robert Collins wrote:

> On Tue, 2004-08-31 at 17:13 +0200, Henrik Nordstrom wrote:
>> On Tue, 31 Aug 2004, Matus UHLAR - fantomas wrote:
>>
>>> in what way? to have that behavior permanent, or to keep things at library
>>> malloc? Does squid handle its memory in such efficient way that using
>>> malloc/free would have strong performance impact?
>>
>> Just to take away the "memory_pools on/off" configuration directive as it
>> does not make much sense to have this direcive. The configure
>> --disable-mempools directive is sufficient and serves a real purpose.
>>
>> The main reason why the configuration directive exists is to initially
>> make it easier to proof that the use of memory pools do make a benefit.
>> This is already well proven. In fact the configuration directive probably
>> should had gone away even before the first STABLE release with memory
>> pools support, remaining only as a configure --disable-mempools directive.
>>
>> The --disable-mempools configure directive is still needed, but for other
>> reasons (debugging).
>
> Actually, I was thinking the other way around: the 3.0 MemPools support
> multiple pool allocators: so we can have a trivial pool that is just
> malloc/free always compiled in and available. Then in squid.conf. the
> disable_mempools command goes away, and instead we change the default
> implementation - possibly in squid.conf, but more likely a one line
> change in MemPools.cc.

If you ask me this kind of configuration belongs at compile time. I see no
value of having a squid.conf directive for selecting which "malloc"
implementation to use.

Regards
Henrik
Received on Wed Sep 01 2004 - 10:58:19 MDT