[Top] [All Lists]

Re: [PATCH] Improve behaviour of Netlink Sockets

To: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Subject: Re: [PATCH] Improve behaviour of Netlink Sockets
From: "David S. Miller" <davem@xxxxxxxxxxxxx>
Date: Tue, 28 Sep 2004 21:02:25 -0700
Cc: hadi@xxxxxxxxxx, pablo@xxxxxxxxxxx, davem@xxxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <20040929032724.GA26959@xxxxxxxxxxxxxxxxxxx>
References: <20040928024614.GA9911@xxxxxxxxxxxxxxxxxxx> <1096340772.8659.51.camel@xxxxxxxxxxxxxxxx> <20040928032321.GB10116@xxxxxxxxxxxxxxxxxxx> <1096343125.8661.96.camel@xxxxxxxxxxxxxxxx> <20040928035921.GA10675@xxxxxxxxxxxxxxxxxxx> <1096367787.8662.146.camel@xxxxxxxxxxxxxxxx> <20040928111159.GA18421@xxxxxxxxxxxxxxxxxxx> <1096374736.8663.217.camel@xxxxxxxxxxxxxxxx> <20040929001327.GB25293@xxxxxxxxxxxxxxxxxxx> <1096426324.1044.129.camel@xxxxxxxxxxxxxxxx> <20040929032724.GA26959@xxxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
On Wed, 29 Sep 2004 13:27:24 +1000
Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> wrote:

> On Tue, Sep 28, 2004 at 10:52:05PM -0400, jamal wrote:
> >
> > > You're right.  rtmsg_info() is using GOODISZE unnecessarily.  I'll
> > > write up a patch.
> > 
> > But why? ;->
> So that the alloc_skb() is slightly less likely to fail.  The dumpers
> gain a lot by using GOODSIZE since they can fill it up.  As rtmsg_info
> has no chance of getting anywhere near GOODSIZE we should provide a
> more accruate estimate.
> It also means that with netlink_trim() you'll save a realloc/copy.
> Signed-off-by: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
> Dave, you can stop reading now :)

Hehe, ok.  Patch applied :-)


<Prev in Thread] Current Thread [Next in Thread>