On 04/09/2018 03:10 PM, Jim Fehlig wrote:
On 04/09/2018 08:32 AM, Daniel P. Berrangé wrote:
On Fri, Apr 06, 2018 at 02:44:54PM -0600, Jim Fehlig wrote:
In preparation of removing the legacy Xen driver, move the
sexpr2xml tests from WITH_XEN to WITH_LIBXL. Even though the
legacy driver will be removed, we'll want to maintain the ability
to convert sexpr to XML. Requires fixing up the tests to account
for different behavior of Xen vs libxl post parse functions.

There is some test file fallout due to differences in handling
of default values between xend and libxl.

diff --git a/tests/sexpr2xmldata/sexpr2xml-boot-grub.xml b/tests/sexpr2xmldata/sexpr2xml-boot-grub.xml
index b9a8716b2..4b9f535fc 100644
--- a/tests/sexpr2xmldata/sexpr2xml-boot-grub.xml
+++ b/tests/sexpr2xmldata/sexpr2xml-boot-grub.xml
@@ -15,7 +15,7 @@
      <disk type='block' device='disk'>
-      <driver name='phy'/>
+      <driver name='phy' type='raw'/>
        <source dev='/dev/MainVG/GuestVG'/>
        <target dev='xvda' bus='xen'/>
@@ -24,6 +24,5 @@
      <input type='mouse' bus='xen'/>
      <input type='keyboard' bus='xen'/>
-    <memballoon model='xen'/>

For this I wonder if a better solution is actually to make the libxl
driver fill in the memballoon device. IIUC, you can't actually turn
it off, so all libxl guests will have this regardless.

Yes, good point.

So we should
just need to add the element a post-parse callback, and check for
unsupported "model" value when starting the guest

I'll send a follow up along these lines. In the meantime I've pushed this series. Good riddance xend...

Sorry, I should have put a bit more thought into this approach since it caused a lot of unnecessary test file churn. Regardless, the follow-up has been sent



libvir-list mailing list

Reply via email to