netdev
[Top] [All Lists]

Re: cls_u32 compile failure in current 2.6.12-rc1+BK tree

To: jamal <hadi@xxxxxxxxxx>
Subject: Re: cls_u32 compile failure in current 2.6.12-rc1+BK tree
From: Thomas Graf <tgraf@xxxxxxx>
Date: Mon, 28 Mar 2005 14:58:02 +0200
Cc: Meelis Roos <mroos@xxxxxxxx>, netdev <netdev@xxxxxxxxxxx>
In-reply-to: <1112014261.1089.1018.camel@jzny.localdomain>
References: <Pine.SOC.4.61.0503272249040.5086@math.ut.ee> <1111972035.1091.946.camel@jzny.localdomain> <Pine.SOC.4.61.0503281017060.21333@math.ut.ee> <1112012657.1089.1013.camel@jzny.localdomain> <20050328123801.GY3086@postel.suug.ch> <1112014261.1089.1018.camel@jzny.localdomain>
Sender: netdev-bounce@xxxxxxxxxxx
* jamal <1112014261.1089.1018.camel@xxxxxxxxxxxxxxxx> 2005-03-28 07:51
> On Mon, 2005-03-28 at 07:38, Thomas Graf wrote:
> > * jamal <1112012657.1089.1013.camel@xxxxxxxxxxxxxxxx> 2005-03-28 07:24
> > > If we are going to make u32 an exception then this code should be
> > > surrounded by #ifdef ..ACTION..
> > 
> > The same condition is needed for tcf_exts_dump 
> 
> This one seems fine even for u32, no?

Not sure what you mean, I don't want any ifdefs or actions bits
in the classifiers itself anymore. Everytime we dump the actions
we must be able to also dump the statistics so I think we can
safely move tcf_exts_dump_stats into tcf_exts_dump and hide all
dumping in that single API.

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