[Qemu-devel] Re: Supsend/resume regression in c995b4 WAS: Re: [PATCH] Fix migration uint8 arrys handled

[Anthony Liguori <aliguori@us.ibm.com>]

On 03/22/2011 08:55 AM, Juan Quintela wrote:
> Anthony Liguori<aliguori@us.ibm.com>  wrote:
>> On 03/22/2011 07:00 AM, Jan Kiszka wrote:
>>> We had a few migration related regressions recently. Do we have
>>> sufficient test cases in autotest for them? Also for migrating from
>>> older to the latest version?
>> Autotest is too late and also not nearly rigorous enough for what
>> you're trying to catch.
>>
>> Here's how I propose we tackle this.  This patch adds a -dump-savevm
>> option that takes a version.  It spits out all of the fields we save
>> for a particular version (well, not really, but it should).  We also
>> can add type information.  The idea is that we'd write a simple test
>> case (using gtester) that ran through and dumped the schema for each
>> version.  We'd store the schema's in the tree and the test can compare
>> old schema's to the current schema to check for failure.
>>
>> This was thrown together in just a few minutes.  I'll try to put
>> together something more complete later today but I wanted to share
>> this before the call at least.
> This would be an start, althought I still think that a way to dump a
> single device, and a way to dump the state of a device in a specific
> version is needed.  Information as:
> - is this always saved
> - size of arrays
> - ....
>
> that is there is not saved.

Yeah, we can add that down the road though.  With just something as 
simple as this, we can catch quite a few regressions.

Regards,

Anthony Liguori

> Later, Juan.