[PATCH] KVM test: Disable HPET on windows timedrift tests

[PATCH] KVM test: Disable HPET on windows timedrift tests

[Lucas Meneghel Rodrigues]

By default, HPET is enabled on qemu and no time drift
mitigation is being made for it. So, add -no-hpet
if qemu supports it, during windows timedrift tests.

Signed-off-by: Lucas Meneghel Rodrigues <lmr@redhat.com>
---
 client/tests/kvm/kvm_vm.py             |    9 +++++++++
 client/tests/kvm/tests_base.cfg.sample |    2 ++
 2 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/client/tests/kvm/kvm_vm.py b/client/tests/kvm/kvm_vm.py
index ff69aed..4856ebe 100755
--- a/client/tests/kvm/kvm_vm.py
+++ b/client/tests/kvm/kvm_vm.py
@@ -292,6 +292,12 @@ class VM:
             return (" -chardev file,id=testlog,path=%s"
                     " -device testdev,chardev=testlog" % filename)
 
+        def add_no_hpet(help):
+            if has_option(help, "no-hpet"):
+                return " -no-hpet"
+            else:
+                return ""
+
         # End of command line option wrappers
 
         if name is None: name = self.name
@@ -429,6 +435,9 @@ class VM:
         if params.get("testdev") == "yes":
             qemu_cmd += add_testdev(help, self.get_testlog_filename())
 
+        if params.get("disable_hpet") == "yes":
+            qemu_cmd += add_no_hpet(help)
+
         # If the PCI assignment step went OK, add each one of the PCI assigned
         # devices to the qemu command line.
         if self.pci_assignable:
diff --git a/client/tests/kvm/tests_base.cfg.sample b/client/tests/kvm/tests_base.cfg.sample
index c678898..7f59bfc 100644
--- a/client/tests/kvm/tests_base.cfg.sample
+++ b/client/tests/kvm/tests_base.cfg.sample
@@ -1047,6 +1047,8 @@ variants:
         stress_boot:
             alive_test_cmd = systeminfo
         timedrift:
+            # Timedrift compensation on Windows with hpet does not happen
+            disable_hpet = yes
             extra_params += " -rtc-td-hack"
             time_command = "echo TIME: %date% %time%"
             time_filter_re = "(?<=TIME: \w\w\w ).{19}(?=\.\d\d)"
-- 
1.7.0.1

Re: [PATCH] KVM test: Disable HPET on windows timedrift tests

[Avi Kivity]

On 06/30/2010 06:39 PM, Lucas Meneghel Rodrigues wrote:
> By default, HPET is enabled on qemu and no time drift
> mitigation is being made for it. So, add -no-hpet
> if qemu supports it, during windows timedrift tests.
>
>    

Hm, you're compensating for a qemu bug by not testing it.

Can we have an XFAIL for this test instead?

-- 
error compiling committee.c: too many arguments to function

Re: [PATCH] KVM test: Disable HPET on windows timedrift tests

[Lucas Meneghel Rodrigues]

On Thu, 2010-07-01 at 17:42 +0300, Avi Kivity wrote:
> On 06/30/2010 06:39 PM, Lucas Meneghel Rodrigues wrote:
> > By default, HPET is enabled on qemu and no time drift
> > mitigation is being made for it. So, add -no-hpet
> > if qemu supports it, during windows timedrift tests.
> >
> >    
> 
> Hm, you're compensating for a qemu bug by not testing it.
> 
> Can we have an XFAIL for this test instead?

Certainly we can. In actuality, that's what's being done on our internal
autotest server - this particular test is linked to the upstream bug
https://bugs.launchpad.net/qemu/+bug/599958

We've discussed about this issue this morning, it boils down to the way
people are more comfortable with handling this issue. My first thought
was to disable HPET until someone come up with a time drift mitigation
strategy for it.

But your approach makes more sense, unless someone has something else to
say about it, I'll drop the patch from autotest shortly.

Lucas

Re: [PATCH] KVM test: Disable HPET on windows timedrift tests

[Dor Laor]

On 07/01/2010 07:05 PM, Lucas Meneghel Rodrigues wrote:
> On Thu, 2010-07-01 at 17:42 +0300, Avi Kivity wrote:
>> On 06/30/2010 06:39 PM, Lucas Meneghel Rodrigues wrote:
>>> By default, HPET is enabled on qemu and no time drift
>>> mitigation is being made for it. So, add -no-hpet
>>> if qemu supports it, during windows timedrift tests.
>>>
>>>
>>
>> Hm, you're compensating for a qemu bug by not testing it.
>>
>> Can we have an XFAIL for this test instead?
>
> Certainly we can. In actuality, that's what's being done on our internal
> autotest server - this particular test is linked to the upstream bug
> https://bugs.launchpad.net/qemu/+bug/599958
>
> We've discussed about this issue this morning, it boils down to the way
> people are more comfortable with handling this issue. My first thought
> was to disable HPET until someone come up with a time drift mitigation
> strategy for it.
>
> But your approach makes more sense, unless someone has something else to
> say about it, I'll drop the patch from autotest shortly.

Actually we should do both - XFAIL when hpet is used and in addition 
(and even more importantly) test other clock sources by disabling hpet.

>
> Lucas
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html