On Thu, Aug 09, 2012 at 03:50:12PM +0200, Gerd Hoffmann wrote: > On 08/09/12 15:41, Alon Levy wrote: > > On Mon, Aug 06, 2012 at 09:31:19AM +0200, Gerd Hoffmann wrote: > >>> diff --git a/configure b/configure > >>> index cef0a71..5fcd315 100755 > >>> --- a/configure > >>> +++ b/configure > >>> @@ -2630,7 +2630,7 @@ EOF > >>> spice_cflags=$($pkg_config --cflags spice-protocol spice-server > >>> 2>/dev/null) > >>> spice_libs=$($pkg_config --libs spice-protocol spice-server > >>> 2>/dev/null) > >>> if $pkg_config --atleast-version=0.8.2 spice-server >/dev/null 2>&1 && > >>> \ > >>> - $pkg_config --atleast-version=0.8.1 spice-protocol > /dev/null 2>&1 > >>> && \ > >>> + $pkg_config --atleast-version=0.12.0 spice-protocol > /dev/null > >>> 2>&1 && \ > >> > >> I'd prefer to not require cutting-edge spice bits to build ... > > > > I only increased the protocol requirement, not the server. > > Distros tend to update those in sync, so this doesn't help much for the > average user, this change would likely cause spice detection fail on > almost every released distro ...
You're right. This would be solved with a submodule. > > > Meanwhile I'll add a define based on the pkg-config spice-protocol > > version, which I can test from hw/qxl.c, since the commit adding > > QXL_IO_MONITORS_CONFIG_ASYNC didn't introduce any new define I can > > check. > > Just checking "SPICE_SERVER_VERSION >= 0x000b01" doesn't work? I'd > expect spice-server 0.11.1+ having a dependency on a recent enougth > spice-protocol so you can expect QXL_IO_MONITORS_CONFIG_ASYNC being > present then, no? Since spice-server now uses submodules, it doesn't check for an external spice-protocol, so no. > > thanks, > Gerd >