From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ferruh Yigit Subject: Re: [PATCH v3 00/11] ethdev: fix race conditions in iterator and notifications Date: Fri, 11 May 2018 02:14:34 +0100 Message-ID: References: <20180509094337.26112-1-thomas@monjalon.net> <20180510235836.1099-1-thomas@monjalon.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: matan@mellanox.com, arybchenko@solarflare.com, stephen@networkplumber.org To: Thomas Monjalon , dev@dpdk.org Return-path: Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by dpdk.org (Postfix) with ESMTP id E2C741BC49 for ; Fri, 11 May 2018 03:14:37 +0200 (CEST) In-Reply-To: <20180510235836.1099-1-thomas@monjalon.net> Content-Language: en-US List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On 5/11/2018 12:58 AM, Thomas Monjalon wrote: > We have discovered some race conditions when using the port iterator > and/or the notifications recently added. > The work was done mostly with failsafe but some bugs could > be reproduced with other drivers. > These 11 patches are fixing all these issues. > The PMDs are modified to call a new function after probing each port. > > v3: > * address Andrew comments on failsafe patch 10 > * address Stephen comment about moving check in a function (is_allocated) > * rename rte_eth_dev_allocated_nolock to _rte_eth_dev_allocated > > v2 (address Gaetan comments): > * rename rte_eth_dev_allocated_lock_free to rte_eth_dev_allocated_nolock > * do not use uint32_t for avoiding port id iteration overflow > (should be fixed separately with BUILD_BUG_ON) > > > Quick survey: > Do you think it is too late to merge these fixes in 18.05? > Or do you agree that these bugs really deserve to be fixed > by this series in 18.05-rc3? > Gaetan, Andrew and Stephen have replied positively. > > > Matan Azrad (4): > ethdev: allow ownership operations on unused port > ethdev: add lock to port allocation check > net/failsafe: fix sub-device ownership race > ethdev: fix port removal notification timing > > Thomas Monjalon (7): > ethdev: fix debug log of owner id > net/failsafe: fix sub-device visibility > ethdev: add doxygen comments for each state > drivers/net: use higher level of probing helper for PCI > ethdev: add probing finish function > ethdev: fix port visibility before initialization > ethdev: fix port probing notification Series applied to dpdk-next-net/master, thanks.