On Thu, Jun 14, 2018 at 05:56:15PM +0200, Ján Tomko wrote:
On Thu, Jun 14, 2018 at 02:28:02PM +0200, Erik Skultety wrote:
The whole <sev> element was optional which it shouldn't be as if the platform doesn't support SEV we should format <sev supported='no'/>.
Signed-off-by: Erik Skultety <eskultet@redhat.com> --- docs/schemas/domaincaps.rng | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/docs/schemas/domaincaps.rng b/docs/schemas/domaincaps.rng index e25201fc68..ed29823548 100644 --- a/docs/schemas/domaincaps.rng +++ b/docs/schemas/domaincaps.rng @@ -185,9 +185,7 @@ <ref name='gic'/> <ref name='vmcoreinfo'/> <ref name='vmgenid'/> - <optional> - <ref name='sev'/> - </optional> + <ref name='sev'/> </interleave> </element> </define> @@ -213,12 +211,15 @@
<define name='sev'> <element name='sev'> - <element name='cbitpos'> - <data type='unsignedInt'/> - </element> - <element name='reducedPhysBits'> - <data type='unsignedInt'/> - </element> + <ref name='supported'/> + <optional> + <element name='cbitpos'> + <data type='unsignedInt'/> + </element> + <element name='reduced-phys-bits'>
Actually, s/reduced-phys-bits/reducedPhysBits/
Wow, how did that one got in there?! I didn't even notice this bit to be honest (since I basically only did a simple code movement), I guess it could have only happened as part of resolving some merge conflicts with an older branch I got, so thanks for pointing that out, I'll adjust. Erik