On 3/17/20 10:11 AM, Daniel P. Berrangé wrote:
On Tue, Mar 17, 2020 at 03:07:34PM +0000, Peter Maydell wrote:
On Tue, 17 Mar 2020 at 15:05, Daniel P. Berrangé <berrange@xxxxxxxxxx> wrote:
On Tue, Mar 17, 2020 at 03:00:48PM +0000, Peter Maydell wrote:
On Tue, 17 Mar 2020 at 14:57, Daniel P. Berrangé <berrange@xxxxxxxxxx> wrote:
I don't feel like -Wno-unused-function looses anything significant, as
the GCC builds will still be reporting unused functions which will
catch majority of cases.
The most interesting difference is that clang will catch unused
static inline functions which gcc does not.
That's mostly just about dead code cruft detection IIUC. That code won't
make it into the binary if it isn't used.
Indeed, but it's nice to have the dead code cruft detection. You
can always mark the function as __attribute__((unused)) if you really
mean that it might be present but not used.
The *BSDs seem to track latest glib pretty quickly. So if we got the
unused attribute into upstream glib, we would probably have about
6-9 months before we get a build platform with the fixed glib included
where we can conditionally re-enable the unused-function warning.
Looks like glib commit https://github.com/GNOME/glib/commit/b41bff1f
(2.57.2) added G_GNUC_UNUSED to all of the functions declared during
G_DEFINE_AUTOPTR_CLEANUP_FUNC. Which version of glib is on the NetBSD
machine?
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3226
Virtualization: qemu.org | libvirt.org
