From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.tds.net (smtp-fo.tds.net [64.8.70.105]) by dpdk.org (Postfix) with ESMTP id A3B715A8A for ; Fri, 20 Mar 2015 20:04:19 +0100 (CET) X_CMAE_Category: , , X-CNFS-Analysis: v=2.1 cv=Iv0KcdPg c=1 sm=1 tr=0 a=zDUj20tEDmPq8qaMnVvhOg==:117 a=zDUj20tEDmPq8qaMnVvhOg==:17 a=K-v-2zaBAAAA:8 a=kj9zAlcOel0A:10 a=txOcFob-AAAA:8 a=McbVWVKZKGtuChtqEy4A:9 a=CjuIK1q_8ugA:10 X-CM-Score: 0 X-Scanned-by: Cloudmark Authority Engine X-Authed-Username: amt2b3NzZW5AdGRzLm5ldA== Authentication-Results: smtp02.tds.cmh.synacor.com smtp.user=jkvossen@tds.net; auth=pass (LOGIN) Received: from [184.60.119.24] ([184.60.119.24:57537] helo=[10.0.1.17]) by mail.tds.net (envelope-from ) (ecelerity 3.6.5.45644 r(Core:3.6.5.0)) with ESMTPSA (cipher=AES128-SHA) id FC/1A-17222-D2F6C055; Fri, 20 Mar 2015 15:04:14 -0400 Mime-Version: 1.0 (Mac OS X Mail 7.3 \(1878.6\)) Content-Type: text/plain; charset=us-ascii From: Joseph Vossen In-Reply-To: Date: Fri, 20 Mar 2015 15:04:10 -0400 Content-Transfer-Encoding: quoted-printable Message-Id: References: To: Jay Rolette X-Mailer: Apple Mail (2.1878.6) Cc: DPDK Subject: Re: [dpdk-dev] soft lockup calling receive burst X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: bford@vossen.org List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 20 Mar 2015 19:04:20 -0000 [snip] > I haven't seen that, but there's a soft-lockup in KNI in DPDK 1.6. = Here's what I posted about that one a few weeks ago: >=20 > Found the problem. No patch to submit since it's already fixed in = later versions of DPDK, but thought I'd follow up with the details since = I'm sure we aren't the only ones trying to use bleeding-edge versions of = DPDK... >=20 > In kni_net_rx_normal(), it was calling netif_receive_skb() instead of = netif_rx(). The source for netif_receive_skb() point out that it should = only be called from soft-irq context, which isn't the case for KNI. >=20 > As typical, simple fix once you track it down. >=20 > Yao-Po Wang's fix: commit 41a6ebded53982107c1adfc0652d6cc1375a7db9. I noticed that posting, but KNI is not involved here...thanks!=