From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com ([192.55.52.93]:12382 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964792AbeE2Osw (ORCPT ); Tue, 29 May 2018 10:48:52 -0400 Date: Tue, 29 May 2018 22:58:25 +0800 From: Philip Li To: Matthew Wilcox Cc: Greg Kroah-Hartman , Christoph Hellwig , kbuild test robot , "Darrick J. Wong" , x86@kernel.org, viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org, kbuild-all@01.org Subject: Re: [kbuild-all] [vfs:hch.aio 7/40] include/linux/syscalls.h:233:18: warning: 'sys_io_pgetevents' alias between functions of incompatible types 'long int(aio_context_t, long int, long int, struct io_event *, struct timespec *, const struct __aio_sigset *)' and 'long int(long int, long int, long int, long int, long int, long int)' Message-ID: <20180529145825.GA19550@intel.com> References: <201805291127.t3LJlHRg%fengguang.wu@intel.com> <20180529072001.GA393@lst.de> <20180529072024.GA21375@kroah.com> <20180529080507.GA4990@intel.com> <20180529112931.GA24842@bombadil.infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180529112931.GA24842@bombadil.infradead.org> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tue, May 29, 2018 at 04:29:31AM -0700, Matthew Wilcox wrote: > On Tue, May 29, 2018 at 04:05:07PM +0800, Philip Li wrote: > > On Tue, May 29, 2018 at 09:20:24AM +0200, Greg Kroah-Hartman wrote: > > > I was seeing this type of mess when trying to build with gcc-8.1, and > > > could not figure it out either. So I just gave up and went back to > > > gcc-7 for the time being. I don't think it's anything new you did here. > > thanks, we will follow up to exclude this pattern in report in bot side. > > I don't think that's the right solution here. People just don't want > to see errors/warnings that they didn't introduce. It'd be unfortunate > to start skipping warnings about syscalls entirely. this is valid concern. Usually when using new compiler, we will need tune the bot a little to exclude certain issues, some are input by users. Of course, we will limit pattern to be very specific to avoid missing real issues. Based on feedback, i will switch back to gcc-7.3, and does more homework to understand the diffs of gcc-8.1.