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


Re: [Xen-devel] [PATCH] Non-polling trace record access

To: Rob Gardner <rob.gardner@xxxxxx>
Subject: Re: [Xen-devel] [PATCH] Non-polling trace record access
From: Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>
Date: Thu, 6 Apr 2006 10:27:32 +0100
Cc: xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Thu, 06 Apr 2006 02:27:18 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <442D6CE6.70704@xxxxxx>
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/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <442D6CE6.70704@xxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx

On 31 Mar 2006, at 18:54, Rob Gardner wrote:

This is the kernel side code to implement a virtual irq that gets sent when the xen trace buffers become half-full. This allows userland tools such as xentrace and xenmon to avoid polling for new trace records. A future patch will include support for this in XenMon, along with various other enhancements and bug fixes.

 This patch has been tested on x86, x86-64, and x86 SMP machines.

Please re-sync the patch with current tip, and also I think your interrupt holdoff logic is unnecessary. I think you can get rid of last_virq_sent and test '(buf->prod - buf->cons) == t_buf_highwater'. No need for '>=' because the producer is incremented by one each time trace() is called so you are bound to equal t_bug_highwater before you become greater than it.

 -- Keir

Xen-devel mailing list

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