This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] Re: Problems with serial interrupts with IPMI 2.0 SoL

To: Jeremy Fitzhardinge <jeremy@xxxxxxxx>
Subject: [Xen-devel] Re: Problems with serial interrupts with IPMI 2.0 SoL
From: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx>
Date: Thu, 10 Dec 2009 21:55:50 -0200
Cc: Xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>, Keir Fraser <keir.fraser@xxxxxxxxxxxxx>, Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>, Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>
Delivery-date: Fri, 11 Dec 2009 06:51:15 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <4B216C12.1020001@xxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <4B216C12.1020001@xxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Em Thu, 10 Dec 2009 13:45:54 -0800
Jeremy Fitzhardinge <jeremy@xxxxxxxx> escreveu:

> Hi,
> I'm using a system with an embedded IPMI 2.0 BMC which provides SoL 
> (Supermicro X8SIL-F).  This works fine under Linux (fully interrupt 
> driven), but fails for Xen as it appears to not be sending interrupts; 
> it works OK if I set the driver to polling mode.
> I'm assuming the Xen is tickling some bug in the emulated 16550.  The 
> Linux driver has a couple of pieces of code to explicitly work around 
> bugs in the Intel AMT SoL implementation, but I don't think they're 
> coming into play here (because there are no Intel SoL PCI devices on 
> this system to trigger the NO_TXEN_TEST flag).
> I'm wondering if you have any pointers or suggestions about more subtle 
> problems?

If you don't hit NO_TXEN_TEST flag, you'll be loosing interrupts, as the
emulation of 16550 is not perfect.

As far as I remember, interrupts take too long to happen with the emulated
16550, when comparing with a normal serial.



Xen-devel mailing list

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