From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlad Buslov Subject: Re: [PATCH] net: sched: don't disable bh when accessing action idr Date: Sat, 19 May 2018 13:12:49 +0300 Message-ID: References: <1526658324-6570-1-git-send-email-vladbu@mellanox.com> Mime-Version: 1.0 Content-Type: text/plain Cc: Linux Kernel Network Developers , Jamal Hadi Salim , Jiri Pirko , David Miller , LKML To: Cong Wang Return-path: Received: from mail-eopbgr10041.outbound.protection.outlook.com ([40.107.1.41]:19808 "EHLO EUR02-HE1-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752103AbeESKNE (ORCPT ); Sat, 19 May 2018 06:13:04 -0400 In-reply-to: Sender: netdev-owner@vger.kernel.org List-ID: On Sat 19 May 2018 at 02:59, Cong Wang wrote: > On Fri, May 18, 2018 at 8:45 AM, Vlad Buslov wrote: >> Underlying implementation of action map has changed and doesn't require >> disabling bh anymore. Replace all action idr spinlock usage with regular >> calls that do not disable bh. > > Please explain explicitly why it is not required, don't let people > dig, this would save everyone's time. Underlying implementation of actions lookup has changed from hashtable to idr. Every current action implementation just calls act_api lookup function instead of implementing its own lookup. I asked author of idr change if there is a reason to continue to use _bh versions and he replied that he just left them as-is. > > Also, this should be targeted for net-next, right? Right. > > Thanks.