From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sowmini Varadhan Subject: Re: [PATCH net-next] sunvnet: improve error handling when a remote crashes Date: Mon, 26 Jan 2015 15:29:46 -0500 Message-ID: <54C6A3BA.3060004@oracle.com> References: <54C6911B.5040501@oracle.com> <20150126195403.GE6437@oracle.com> <54C6A154.3090906@oracle.com> Reply-To: sowmini.varadhan@oracle.com Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: David Miller , netdev@vger.kernel.org To: David L Stevens Return-path: Received: from userp1040.oracle.com ([156.151.31.81]:48607 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753120AbbAZU3s (ORCPT ); Mon, 26 Jan 2015 15:29:48 -0500 In-Reply-To: <54C6A154.3090906@oracle.com> Sender: netdev-owner@vger.kernel.org List-ID: On 01/26/2015 03:19 PM, David L Stevens wrote: >> I think you'll have a higher reclaim rate for finding !READY if you >> started at dr->cons instead: dr->cons is the one that was last ACK'ed, >> and that ack would only have been sent after the peer had marked the >> descriptor as DONE. (consumer would have had a chance to read more >> descriptors, by the time the tx-reclaim loop goes around) : > The descriptors indices we look at are unchanged by this patch. Understood, and what I suggested was an optimization for tx-reclaim. > It should only happen if there is active traffic when a remote > crashes, but if a remote is giving us garbage in other cases, I think an > admin would want to check that out. This allows one DomU to spam up everyone else's log file by crashing periodically. In any case, the contents of the message do not in any way suggest that an admin (of the DomU, who probably has no control over other DomU's anyway) should go and baby-sit some other DomU that just crashed. ymmv. --Sowmini