Message-ID: <3F4A0916.3010906@basmevissen.nl>
Date: Mon, 25 Aug 2003 15:03:18 +0200
From: Bas Mevissen <ml@basmevissen.nl>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624
MIME-Version: 1.0
To: Kendall Bennett <KendallB@scitechsoft.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Kernel module symbol versioning?
References: <3F461DF4.26519.490C74E@localhost>
In-Reply-To: <3F461DF4.26519.490C74E@localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1405
Lines: 37

Kendall Bennett wrote:

> Hi,
> 
> I have been reading the book "Linux Device Drivers, 2nd Edition" and have 
> some questions about symbol versioning. In Chapter 11 it mentions that 
> you can use the <linux/modversions.h> header file to compile your module 
> with symbol versions enabled, so that your module will load on multiple 
> kernels and fail if the symbol CRC's do not match. I tested this out on a 
> simple test module, but this module fails to load unless I pass the '-f' 
> flag to insmod (Red Hat 7.3 and 8.0). 
> 
> Is there a way to compile the module so that insmod will only complain if 
> there is a version conflict? Or do you always have to use -f in this case 
> to force the module to load? If you have to do that, will -f still fail 
> to load if the versioned symbols don't match?
> 

Isn't the confict cause here (just) that one is compiled with gcc 2.x 
and the other with gcc 3.x?

In other words, retry compiling this module on one system with the two 
different kernels, as you want to compare kernel source level interface 
  differences and not kernel ABI (Application Binairy Interface IIRC) 
differences.

Regards,

Bas.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/