On Wed, Jun 04, 2014 at 09:33:12PM +0200, Tom Gundersen wrote:
> On Wed, Jun 4, 2014 at 7:48 PM, Dave Reisner <dreis...@archlinux.org> wrote:
> > 505f8da7325 left link->mac uninitialized, causing MACAddress based
> > [Match] sections to fail to match anything.
> >
> > https://bugs.freedesktop.org/show_bug.cgi?id=79638
>
> Thanks for the report and the patch.
>
> I'd much prefer if we got these potentially dynamic properties only
> from one source though (i.e., from rtnl), as we are not guaranteed any
> ordering between udev and rtnl events. In this particular case it
> seems like there can not be a problem, but I'd rather be consistent.
Fair enough.
> I pushed an alternative fix. Could you have a look to see if it looks ok to
> you?
>
Works for me. Thanks!
> > ---
> > src/network/networkd-link.c | 12 +++++++++++-
> > 1 file changed, 11 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/network/networkd-link.c b/src/network/networkd-link.c
> > index 6e6fb83..f3bab9f 100644
> > --- a/src/network/networkd-link.c
> > +++ b/src/network/networkd-link.c
> > @@ -1683,8 +1683,18 @@ int link_initialized(Link *link, struct udev_device
> > *device) {
> > if (link->state != LINK_STATE_INITIALIZING)
> > return 0;
> >
> > - if (device)
> > + if (device) {
> > + const char *mac_address = udev_device_get_sysattr_value(
> > + device, "address");
> > +
> > + if (mac_address) {
> > + struct ether_addr *hwaddr =
> > ether_aton(mac_address);
> > + if (hwaddr)
> > + memcpy(&link->mac, hwaddr, sizeof(struct
> > ether_addr));
> > + }
> > +
> > link->udev_device = udev_device_ref(device);
> > + }
> >
> > log_debug_link(link, "udev initialized link");
> >
> > --
> > 2.0.0
> >
> > _______________________________________________
> > systemd-devel mailing list
> > systemd-devel@lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/systemd-devel
_______________________________________________
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel
