From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH v2 1/2] virtio: stop using legacy struct vring Date: Mon, 6 Apr 2020 11:57:43 -0400 Message-ID: <20200406115653-mutt-send-email-mst@kernel.org> References: <20200406153245.127680-1-mst@redhat.com> <20200406153245.127680-2-mst@redhat.com> <20200406155602.GA160445@kroah.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20200406155602.GA160445@kroah.com> Sender: linux-kernel-owner@vger.kernel.org To: Greg Kroah-Hartman Cc: linux-kernel@vger.kernel.org, Sudeep Dutt , Ashutosh Dixit , Arnd Bergmann , Andy Shevchenko , Darren Hart , Vadim Pasternak , Jason Wang , platform-driver-x86@vger.kernel.org, virtualization@lists.linux-foundation.org List-Id: virtualization@lists.linuxfoundation.org On Mon, Apr 06, 2020 at 05:56:02PM +0200, Greg Kroah-Hartman wrote: > On Mon, Apr 06, 2020 at 11:35:23AM -0400, Michael S. Tsirkin wrote: > > struct vring (in the uapi directory) and supporting APIs are kept > > around to avoid breaking old userspace builds. > > It's not actually part of the UAPI - it was kept in the UAPI > > header by mistake, and using it in kernel isn't necessary > > and prevents us from making changes safely. > > In particular, the APIs actually assume the legacy layout. > > > > Add struct vring_s (identical ATM) and supporting > > legacy APIs and switch everyone to use that. > > How are we going to know that "struct vring_s" is what we need/want to > use? What does "_s" mean? > > "struct vring_kernel"? > > naming is hard... > > greg k-h Hmm. I guess I can just add an ifdef so kernel doesn't see the UAPI version anymore ... -- MST