netdev
[Top] [All Lists]

[PATCH 6/5] sis900 printk audit

To: Francois Romieu <romieu@xxxxxxxxxxxxx>
Subject: [PATCH 6/5] sis900 printk audit
From: Daniele Venzano <webvenza@xxxxxxxxx>
Date: Sun, 23 Jan 2005 10:10:49 +0100
Cc: Jeff Garzik <jgarzik@xxxxxxxxx>, NetDev <netdev@xxxxxxxxxxx>
In-reply-to: <20050122222618.GA24461@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Mail-followup-to: Francois Romieu <romieu@xxxxxxxxxxxxx>, Jeff Garzik <jgarzik@xxxxxxxxx>, NetDev <netdev@xxxxxxxxxxx>
References: <20050122214657.18649.66345@xxxxxxxxxxxxxxxxxxxxx> <20050122214659.18649.83119@xxxxxxxxxxxxxxxxxxxxx> <20050122222618.GA24461@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.6+20040907i
On Sat, Jan 22, 2005 at 11:26:18PM +0100, Francois Romieu wrote:
> Daniele Venzano <webvenza@xxxxxxxxx> :
> > -                   printk(KERN_INFO "Cannot allocate mem for struct 
> > mii_phy\n");
> > +                   printk(KERN_WARNING "Cannot allocate mem for struct 
> > mii_phy\n");
> 
> -> printk(KERN_WARNING "%s: Cannot allocate mem for struct mii_phy\n", ...) ?
> 
> Btw it may make sense to use pci_name() instead of the name associated with
> the struct net_device until it is known for sure that the net_device will not
> be removed due to a failure of the probe() method. 

Attached a patch on top of the previous ones.

It only fixes the missing prefix. I understand that pci_name could be a
better choice here, but I want to see what other drivers are doing in
their probe() before changing policy.

Signed-off-by: Daniele Venzano <webvenza@xxxxxxxxx>

-- 
-----------------------------
Daniele Venzano
Web: http://teg.homeunix.org

Attachment: sis900_c_89.diff
Description: Text document

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