From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gregory Etelson Subject: Re: [PATCH v2] igb_uio: issue FLR during open and release of device file Date: Tue, 11 Jul 2017 08:42:49 +0300 Message-ID: <1645416.4FjkIAPa5l@polaris> References: <1497260285-27536-1-git-send-email-shijith.thotton@caviumnetworks.com> <5cf77ba2-8c29-ca31-1141-63ca01eff9f5@intel.com> <1912538.AEa4I0QC60@polaris> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: Shijith Thotton , dev@dpdk.org, Thomas Monjalon , Stephen Hemminger , Jianfeng Tan , Wenzhuo Lu To: Ferruh Yigit Return-path: Received: from mail-wr0-f180.google.com (mail-wr0-f180.google.com [209.85.128.180]) by dpdk.org (Postfix) with ESMTP id E06AE3257 for ; Tue, 11 Jul 2017 07:42:52 +0200 (CEST) Received: by mail-wr0-f180.google.com with SMTP id 77so167155976wrb.1 for ; Mon, 10 Jul 2017 22:42:52 -0700 (PDT) In-Reply-To: <1912538.AEa4I0QC60@polaris> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Hello Ferruh, Both patches [1] http://dpdk.org/dev/patchwork/patch/26633/ and [2] http://dpdk.org/dev/patchwork/patch/25061/ have failed the same test. This is kind of strange because [2] has already passed that test numerous times. I'll recalibrate my cluster and run the test again. Besides that, [1] does the job Regards, Gregory On Monday, 10 July 2017 06:07:45 IDT Gregory Etelson wrote: Hello Ferruh, I could not reproduce server crash with the patch. However, some tests report ixgbe_vf_pmd and i40e_vf_pmd do not receive and transmit frames after process restart, although PMD initialization completed successfully Is there a way to collect PF firmware dump for investigation ? Regards, Gregory On Friday, 7 July 2017 18:10:40 IDT Ferruh Yigit wrote: > On 7/7/2017 12:13 PM, Shijith Thotton wrote: > > Set UIO info device file operations open and release. Call pci reset > > function inside open and release to clear device state at start and end. > > Copied this behaviour from vfio_pci kernel module code. With this patch, > > it is not mandatory to issue FLR by PMD's during init and close. > > > > Bus master enable and disable are added in open and release respectively > > to take care of device DMA. > > > > Signed-off-by: Shijith Thotton > > Gregory, > > Would you mind testing this one? > > Thanks, > ferruh > >