Subject: Re: mbuf external storage sharing
To: None <>
From: YAMAMOTO Takashi <>
List: tech-net
Date: 10/05/2004 10:43:00
> Yamamoto-san, there seem to be two more-or-less unrelated changes in this patch:
> 1. Changes to use simple_locks per external mbuf. 
> 2.  Changes to allocate, and use, a new subflavour of external-storage
> mbuf, where the `little' or cluster-header mbuf is embedded in the
> external storage.

they're very related because 1. needs 2.

> You seem to be making a bald claim that embedding the `little' mbuf
> header in external storage is the `best' approach. Can you provide
> evidence to suppor that claim?

no evidence, sorry.

> For it seems to contradict the
> conclusions in the FreeBSD-5 paper I cited earlier.

how contradict?
allocating the little header is what exactly freebsd5 does.
only difference is, freebsd5's header contains only reference count.

> >> I have no idea whether we want to follow the directions in that paper
> >> (either mbuf-level API or underlying implemntation); but we should at
> >> least think about the quantitative data.
> >
> >not relevant to my patch.
> Yes and no: it depends very much whether the desired scope of
> SMP-safeness is a naive unscalaeable approach, or a more sophisitated
> approach that scales to large numbers of CPUs (where "4" has
> historically been "large").  And I think that merits a completely
> separate discussion.

be more specific, please.
you can easily modify pool_cache to use per-cpu cache.
how is it related to my patch?