Subject: Re: Driver_data is probably zero in serverworks IDE driver
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
To: Marton Balint <cus@fazekas.hu>
Cc: linux-kernel@vger.kernel.org, akpm@osdl.org, torvalds@osdl.org
In-Reply-To: <Pine.LNX.4.61.0609091840170.30278@cinke.fazekas.hu>
References: <Pine.LNX.4.61.0609091840170.30278@cinke.fazekas.hu>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Date: Sat, 09 Sep 2006 21:36:37 +0100
Message-Id: <1157834197.6877.72.camel@localhost.localdomain>
Mime-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1249
Lines: 33

Ar Sad, 2006-09-09 am 19:19 +0200, ysgrifennodd Marton Balint:
> Please take a look at commit f201f5046ddaeeccb036bdf6848549bf5cb51bb1.
> This commit introduced the usage of the PCI_DEVICE macro, but this macro 
> does not set class and class_mask so I think now we set .class instead of 
> .driver_data. The drivers that are also affected by this commit may 
> have similar problems.

Yes it looks to be the case, and PCI_DEVICE_CLASS() macros don't help as
they assign in differing orders.

The following drivers appear to be broken from this

Aec62xx.c			All	
Serverworks.c			OSB4/CSB5/CSB6/CSB6-2/HT1000


Actual breakage will affect CSB6/CSB6-2/HT1000, and all the AEC6xxx
chips in various ways.

Unfortunately it appears I missed the bug, nobody checked the diff and
nobody ever ran -mm with these chipsets so it somehow got through. I'll
look at rolling a patch but it might be Monday. Until then anyone
rolling a 2.6.18 tree should revert the PCI table changes in serverworks
and aec62xx.

Alan


-
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/