From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wg0-f52.google.com (mail-wg0-f52.google.com [74.125.82.52]) by dpdk.org (Postfix) with ESMTP id F3034376E for ; Tue, 12 May 2015 12:33:06 +0200 (CEST) Received: by wgic8 with SMTP id c8so4026855wgi.1 for ; Tue, 12 May 2015 03:33:06 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:from:date:message-id:subject:to :content-type; bh=hfqsdDrYYyp+WhJ5wYzY1/o4V+0NLQJ/qwf/Ukfsppk=; b=mib322xhfjzNlpt36mqys2xH85kRrna5qcXioWGWpLmpqJ4rV/nWr4vDNfAIzb7k3H A5KTYNCmEwh6Ecsj5CuJ4uiksNGAERV1UCOpUL7QI4JUYNmhdenGFFBomTFmrxj7irEJ zK91507kcj3cJLJ6J/yGKDffu5sQ5GMcaiBlT4exioLIwaVNGeLgiFmvWs9T8wHclfy/ b2K393L6V9pSsZvsXRvfoLIgXnb18vr27sziZmjD6vxHQc+NGTD1Ukiz6PcuGciOj7Yd EAs9yvoUR8cOmSJRN+1I9jWT2IJR/YoYgkBKzLgCaiDflEnrrNAxOCEz0oQtrQdqh0LV cQrg== X-Gm-Message-State: ALoCoQk6rbJQHpwovLKjerxa81fM+7QYZwNhw4PbxgFyFQKDuL3FAcyhvXvEYGK9Xtr6Eq7r4H9n X-Received: by 10.194.185.8 with SMTP id ey8mr7793902wjc.118.1431426786835; Tue, 12 May 2015 03:33:06 -0700 (PDT) MIME-Version: 1.0 Received: by 10.28.215.131 with HTTP; Tue, 12 May 2015 03:32:46 -0700 (PDT) From: Yotam Rubin Date: Tue, 12 May 2015 13:32:46 +0300 Message-ID: To: "dev@dpdk.org" Content-Type: text/plain; charset=UTF-8 X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: [dpdk-dev] ixgbevf TX hang X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 May 2015 10:33:07 -0000 Hello, We're using dpdk with with the ixgbevf driver on an AWS (Xen hypervisor) instance. We use the NIC to drive jumbo packets at high throughput. Occasionally, the vf experiences a non-recoverable TX hang. Specifically, all descriptors in the TX queue are used up - and remain in this state. All descriptors' writeback status don't have the IXGBE_TXD_STAT_DD set. After resetting the port, TX no longer hangs. We're using the suggested TX conf for the 82599: static struct rte_eth_txconf tx_conf = { .tx_thresh = { .pthresh = 36, /* Ring prefetch threshold */ .hthresh = 0, /* Ring host threshold */ .wthresh = 0, /* Ring writeback threshold */ }, .tx_free_thresh = 0, /* Use PMD default values */ .tx_rs_thresh = 0, /* Use PMD default values */ }; Any ideas? Thanks, Yotam