From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [PATCH] efi/cper: Fix endianness of PCI class code Date: Fri, 26 May 2017 10:01:28 +0200 Message-ID: References: <20170506090755.GB19740@wunner.de> <20170510084143.GA16261@wunner.de> <20170525123047.GA4172@wunner.de> <20170525124445.GA4181@wunner.de> <20170525125650.GA4196@wunner.de> <20170526060843.GA4446@wunner.de> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Return-path: In-Reply-To: <20170526060843.GA4446@wunner.de> Sender: linux-pci-owner@vger.kernel.org To: Lukas Wunner Cc: Ard Biesheuvel , "linux-efi@vger.kernel.org" , Ashok Raj , linux-pci , Huang Ying List-Id: linux-efi@vger.kernel.org On Fri, May 26, 2017 at 8:08 AM, Lukas Wunner wrote: > On Thu, May 25, 2017 at 06:07:35AM -0700, Ard Biesheuvel wrote: >> On 25 May 2017 at 05:56, Lukas Wunner wrote: >> >> Or does 'guarantee' mean 'I tested it and it works'? >> > >> > I tested it with x86_64 (le) and ppc32 (be) and it works. >> > I don't have an aarch64 machine available here. >> > >> >> Good. > > Good to merge then? Please remove the bit fields and find a way to do the patch without them. We really don't want any bitfields in structures that are interpreted by hardware: The rules for struct packing are hard enough to understand without bitfields, so please don't give any bad examples even if you have proved from the C99 spec that your specific case is ok (which so far you have refused to do). Arnd