Hi Marc-André, Am 09.07.2015 um 16:37 schrieb Marc-André Lureau: > On Fri, Jun 26, 2015 at 6:02 PM, Marc-André Lureau > <marcandre.lur...@gmail.com> wrote: >>> Also, see the discussion on my RFC: CONFIG_LINUX is wrong here. It would >>> need to depend on KVM/eventfd, and for that there is no easy config >>> option available here, which stopped my patch at the time. >> >> >> Right, I don't have good answer for that. I'd need more help. > > Actually, ivshmem doesn't use ioeventfd by default. And my tests do > not enable it either. > > So the test runs fine with kvm disabled on my host. Is that enough then? > > I couldn't find the discussion on "RFC: CONFIG_LINUX", do you have a pointer?
Pretty sure I pointed you to my RFC already: http://patchwork.ozlabs.org/patch/336367/ Paolo wrote: "I think ivshmem depends on ioeventfd" And mst suggested: "maybe it's easier to implement eventfd APIs" Which I didn't do, and am not aware anyone else did, but maybe I missed it? I've dug out my old patch with the tempfile name fix but didn't get to implementing that missing query while dealing with the qom-cpu pull. Regards, Andreas -- SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton; HRB 21284 (AG Nürnberg)
