netdev
[Top] [All Lists]

Re: [RFC] Wireless extensions rethink

To: jt@xxxxxxxxxx
Subject: Re: [RFC] Wireless extensions rethink
From: Jeff Garzik <jgarzik@xxxxxxxxx>
Date: Thu, 17 Jun 2004 15:34:22 -0400
Cc: netdev@xxxxxxxxxxx, Gertjan van Wingerde <gwingerde@xxxxxxx>, sfeldma@xxxxxxxxx, jkmaline@xxxxxxxxx
In-reply-to: <20040617191338.GD32216@xxxxxxxxxxxxxxxxxx>
References: <40D08769.3070106@xxxxxxx> <20040616204248.GA23617@xxxxxxxxxxxxxxxxxx> <40D0BD5B.201@xxxxxxxxx> <20040616223316.GA29618@xxxxxxxxxxxxxxxxxx> <40D0D265.3070804@xxxxxxxxx> <20040617174717.GA30460@xxxxxxxxxxxxxxxxxx> <40D1E185.2010201@xxxxxxxxx> <40D1E24C.8090802@xxxxxxxxx> <20040617185815.GB32216@xxxxxxxxxxxxxxxxxx> <40D1EAD9.6090403@xxxxxxxxx> <20040617191338.GD32216@xxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040510
Jean Tourrilhes wrote:
On Thu, Jun 17, 2004 at 03:02:49PM -0400, Jeff Garzik wrote:

Jean Tourrilhes wrote:

On Thu, Jun 17, 2004 at 02:26:20PM -0400, Jeff Garzik wrote:


Note that the above is only a first step. Through the standard Linux development process -- evolution -- each hook can be pared down to precisely what each call needs. The above allows for a quick transition of drivers, while keeping them working.

        Jeff


        Have you looked at the patch I sent you ? In which way does it
fails to meet your need ?


The three major problems I listed in a previous email are still present...


        Are we talking of the same patch ? I'm talking of this patch :
http://marc.theaimsgroup.com/?l=linux-netdev&m=108749580004668&w=2
        I reattached the patch below. It's short enough.

Your patch is half the job -- it allows development of a type-specific interface... but it does nothing to address the problems with the underlying type-opaque interface.

The creation of the type-specific interface replaces the type-opaque interface, not layers on top of it.

So while this patch may be useful in early development, it does not allow the direct exposure of core wireless code to the type-specific interfaces, and as such, it can paper over problems that would be immediately obviously if the type-specific interface were the only one to exist.


Also there is a fourth -- WE doesn't work 100% when you have a 32-bit userland and a 64-bit kernel.


        Since when ? What made you change your mind ?
        Please check :
http://marc.theaimsgroup.com/?l=linux-netdev&m=107894322418086&w=2

The general API, yes. But most driver-private interfaces will fail miserably through 32/64-bit translation.

        Jeff



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