Re: [TG3]: Add AMD K8 to list of write-reorder chipsets.

Re: [TG3]: Add AMD K8 to list of write-reorder chipsets.

[Arjan van de Ven]

[-- Attachment #1: Type: text/plain, Size: 622 bytes --]

On Sun, 2005-09-18 at 14:59 -0700, Linux Kernel Mailing List wrote:
> tree 1da3df4563187bae7eec7834d1fada04f8a9271b
> parent 67e6b629212fa9ffb7420e8a88a41806af637e28
> author David S. Miller <davem@davemloft.net> Sat, 17 Sep 2005 06:59:20 -0700
> committer David S. Miller <davem@davemloft.net> Sat, 17 Sep 2005 06:59:20 -0700
> 
> [TG3]: Add AMD K8 to list of write-reorder chipsets.
> 
> Thanks to Andy Stewart for the report and testing
> debug patches from Michael Chan.


shouldn't something like this move to generic code?
(and in this case, probably as quirk that disables the chipset
"feature" ?)


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [TG3]: Add AMD K8 to list of write-reorder chipsets.

[Alan Cox]

On Llu, 2005-09-19 at 08:21 -0400, Arjan van de Ven wrote:
> shouldn't something like this move to generic code?
> (and in this case, probably as quirk that disables the chipset
> "feature" ?)

You don't normally want to disable this feature of chipsets, it just
gives the tg3 indigestion. 

Alan

Re: [TG3]: Add AMD K8 to list of write-reorder chipsets.

[David S. Miller]

From: Arjan van de Ven <arjanv@redhat.com>
Date: Mon, 19 Sep 2005 08:21:58 -0400

> shouldn't something like this move to generic code?
> (and in this case, probably as quirk that disables the chipset
> "feature" ?)

In the long term yes, but the latter part of your suggestion is
assuming this is something that can be controlled and isn't just some
chipset bug.