Received: from eggs.gnu.org ([2001:4830:134:3::10]:34546) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XXbKr-0002Ip-AZ for qemu-devel@nongnu.org; Fri, 26 Sep 2014 15:37:51 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XXbKk-00058m-5f for qemu-devel@nongnu.org; Fri, 26 Sep 2014 15:37:44 -0400 Received: from mx1.redhat.com ([209.132.183.28]:60635) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XXbKj-00056R-UT for qemu-devel@nongnu.org; Fri, 26 Sep 2014 15:37:38 -0400 Date: Fri, 26 Sep 2014 16:37:23 -0300 From: Eduardo Habkost Message-ID: <20140926193723.GC3113@thinpad.lan.raisama.net> References: <1409344310-5441-1-git-send-email-ehabkost@redhat.com> <1409344310-5441-11-git-send-email-ehabkost@redhat.com> <54258103.9090406@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <54258103.9090406@redhat.com> Subject: Re: [Qemu-devel] [PATCH v2 10/17] accel: Move qtest accel registration to qtest.c List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: Michael Mueller , Marcel Apfelbaum , "Michael S. Tsirkin" , Alexander Graf , qemu-devel@nongnu.org, Christian Borntraeger , "Jason J. Herne" , Andreas =?iso-8859-1?Q?F=E4rber?= On Fri, Sep 26, 2014 at 05:06:43PM +0200, Paolo Bonzini wrote: > Il 29/08/2014 22:31, Eduardo Habkost ha scritto: > > As qtest_availble() returns 1 only when CONFIG_POSIX is set, keep > > setting AccelClass.available to keep current behavior (this is different > > from what we did for KVM and Xen). > > Hmm... as extra patches, could we only register the type on CONFIG_POSIX > and thus get rid of available completely? I am a bit confused about qtest support on Windows: is the "-qtest" option supposed to be working (without accel=qtest) on !CONFIG_POSIX? I see that you made qtest_init() not be a NOP anymore on !CONFIG_POSIX on commit d4fce24f3a59eda081cdf2e38e7001591b95d173, but I don't know if it makes sense to try to use it on Windows. I was going to simply not compile accel.o at all on !CONFIG_POSIX, but I don't know if it would break use cases that are expected to work. -- Eduardo