>>> On 08.12.15 at 12:02, <ian.campb...@citrix.com> wrote: > On Wed, 2015-12-02 at 02:33 -0700, Jan Beulich wrote: >> Then we should see about adding support for "efi=no-time". > > And based on what I'm reading in this thread about the reliability of the > time RS in the field it seems to me we should make it the default (on x86 > at least) and provide efi=time to opt in.
Which would get us to actively violate at least early versions of the spec (the current one doesn't appear to be as strict anymore). I'm fine making it easy to work around the issue, but I'm against us doing the wrong thing by default. Apart from that, Ross, the patch you provided breaks on systems without CMOS clock (e.g. so called legacy free ones). This definitely can't be a compile time thing. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
