Re: [PATCH 2/2] powerpc: Improve "kvm unavailable" message

["Lukáš Doktor" <ldoktor@redhat.com>]

Dne 11.4.2016 v 14:02 Andrew Jones napsal(a):
> On Mon, Apr 11, 2016 at 01:37:53PM +0200, Lukáš Doktor wrote:
>> The message when ACCEL=kvm and kvm not available is rather confusing.
>> Let's produce (hopefully) easier to understand message.
>>
>> Signed-off-by: Lukáš Doktor <ldoktor@redhat.com>
>> ---
>>   powerpc/run | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/powerpc/run b/powerpc/run
>> index 56698e6..adffb93 100755
>> --- a/powerpc/run
>> +++ b/powerpc/run
>> @@ -16,7 +16,7 @@ if [ -c /dev/kvm ]; then
>>   fi
>>
>>   if [ "$ACCEL" = "kvm" ] && [ "$kvm_available" != "yes" ]; then
>> -	echo "skip $TESTNAME (kvm only)"
>> +	echo "skip $TESTNAME (kvm not available on your system)"
>
> This will now be inconsistent with scripts/runtime.bash. Also,
> the 'kvm only' message conveys that this test can only run with
> kvm, as opposed to also with tcg. While I agree it's terse, we've
> now lost that information with the new message. Additionally, I
> would say 'this system' instead of 'your system' in the message.
>
> I think Radim has some plans for giving run script error messages
> an overhaul. Maybe he can address this one with that series too?
> Anyway, for now, I would suggest we drop this patch.
>
I thought the message probably has some inner-context, but for me it was 
super confusing. Even more because I'm using

     ACCEL=kvm,kvm-type=PR ...

in some of my power testing, which works well, but when I tried

     ACCEL=kvm ...

it reported test skips. I'm looking forward to some proper errors 
handling (and ways to pass SKIPs as well).

Thank you and feel free to drop this patch,
Lukáš

> Thanks,
> drew
>
>>   	exit 2
>>   fi
>>
>> --
>> 2.5.5
>>