From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41206) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dcSLD-0005nr-VJ for qemu-devel@nongnu.org; Tue, 01 Aug 2017 04:15:49 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dcSLA-0003F1-P0 for qemu-devel@nongnu.org; Tue, 01 Aug 2017 04:15:47 -0400 Received: from mx1.redhat.com ([209.132.183.28]:50974) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dcSLA-0003EK-HR for qemu-devel@nongnu.org; Tue, 01 Aug 2017 04:15:44 -0400 Date: Tue, 1 Aug 2017 10:15:35 +0200 From: Cornelia Huck Message-ID: <20170801101535.16f4d486@gondolin> In-Reply-To: <20170728141309.14275-2-marcandre.lureau@redhat.com> References: <20170728141309.14275-1-marcandre.lureau@redhat.com> <20170728141309.14275-2-marcandre.lureau@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v2 1/2] build-sys: add --disable-vhost-user List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?UTF-8?B?TWFyYy1BbmRyw6k=?= Lureau Cc: qemu-devel@nongnu.org, philippe.mathieu.daude@gmail.com, mst@redhat.com, Christian Borntraeger , Alexander Graf On Fri, 28 Jul 2017 16:13:08 +0200 Marc-Andr=C3=A9 Lureau wrote: > Learn to compile out vhost-user. Keep it enabled by default on > non-win32, that is assumed to be POSIX. Fail if trying to enable it on > win32. >=20 > When trying to make a vhost-user netdev, it gives the following error: >=20 > -netdev vhost-user,id=3Dfoo,chardev=3Dchr-test: Parameter 'type' expects = a netdev backend type >=20 > And similar error with the HMP/QMP monitors. >=20 > While at it, rename CONFIG_VHOST_NET_TEST CONFIG_VHOST_USER_NET_TEST > since it's a vhost-user specific variable. >=20 > Signed-off-by: Marc-Andr=C3=A9 Lureau > --- > hw/virtio/virtio-pci.c | 4 ++-- > configure | 29 +++++++++++++++++++++++++++-- > default-configs/pci.mak | 2 +- > default-configs/s390x-softmmu.mak | 2 +- > tests/Makefile.include | 6 +++--- > 5 files changed, 34 insertions(+), 9 deletions(-) >=20 (...) > diff --git a/configure b/configure > index 987f59ba88..efec1a613e 100755 > --- a/configure > +++ b/configure > @@ -306,6 +306,7 @@ tcg=3D"yes" > vhost_net=3D"no" > vhost_scsi=3D"no" > vhost_vsock=3D"no" > +vhost_user=3D"" > kvm=3D"no" > hax=3D"no" > rdma=3D"" > @@ -1282,6 +1283,15 @@ for opt do > ;; > --enable-vxhs) vxhs=3D"yes" > ;; > + --disable-vhost-user) vhost_user=3D"no" > + ;; > + --enable-vhost-user) > + vhost_user=3D"yes" > + if test "$mingw32" =3D "yes" ; then > + echo "ERROR: vhost-user isn't available on win32" > + exit 1 error_exit? > + fi > + ;; > *) > echo "ERROR: unknown option $opt" > echo "Try '$0 --help' for more information" (...) > diff --git a/default-configs/s390x-softmmu.mak b/default-configs/s390x-so= ftmmu.mak > index b227a36179..51191b77df 100644 > --- a/default-configs/s390x-softmmu.mak > +++ b/default-configs/s390x-softmmu.mak > @@ -1,6 +1,6 @@ > CONFIG_PCI=3Dy > CONFIG_VIRTIO_PCI=3Dy > -CONFIG_VHOST_USER_SCSI=3D$(CONFIG_LINUX) > +CONFIG_VHOST_USER_SCSI=3D$(and $(CONFIG_VHOST_USER),$(CONFIG_LINUX)) Huh. I wonder if anyone actually tried this on s390x? (The change is fine in the context of this patch, of course.) > CONFIG_VIRTIO=3Dy > CONFIG_SCLPCONSOLE=3Dy > CONFIG_TERMINAL3270=3Dy