From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Monjalon Subject: Re: [PATCH] eal: register rte_panic user callback Date: Wed, 07 Mar 2018 09:32:14 +0100 Message-ID: <304114136.g7uiPYdxRp@xps> References: <1520360928-9375-1-git-send-email-arnon@qwilt.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: bruce.richardson@intel.com, dev@dpdk.org To: Arnon Warshavsky Return-path: Received: from out1-smtp.messagingengine.com (out1-smtp.messagingengine.com [66.111.4.25]) by dpdk.org (Postfix) with ESMTP id 3F61B2C8 for ; Wed, 7 Mar 2018 09:32:40 +0100 (CET) In-Reply-To: <1520360928-9375-1-git-send-email-arnon@qwilt.com> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Hi, 06/03/2018 19:28, Arnon Warshavsky: > The use case addressed here is dpdk environment init > aborting the process due to panic, > preventing the calling process from running its own tear-down actions. Thank you for working on this long standing issue. > A preferred, though ABI breaking solution would be > to have the environment init always return a value > rather than abort upon distress. Yes, it is the preferred solution. We should not use exit (panic & co) inside a library. It is important enough to break the API. I would be in favor of accepting such breakage in 18.05. > This patch defines a couple of callback registration functions, > one for panic and one for exit > in case one wishes to distinguish between these events. > Once a callback is set and panic takes place, > it will be called prior to calling abort. > > Maiden voyage patch for Qwilt and myself. Are you OK to visit the other side of the solution?