Re: [Qemu-devel] [PATCH V2 for-2.3] util/qemu-config: fix regression of qmp_query_command_line_options

[Marcel Apfelbaum <marcel@redhat.com>]

On 04/01/2015 08:29 PM, Eric Blake wrote:
> On 04/01/2015 10:47 AM, Marcel Apfelbaum wrote:
>> Commit 49d2e64 (machine: remove qemu_machine_opts global list)
>> made machine options specific to machine sub-type, leaving
>> the qemu_machine_opts desc array empty. Sadly this is the place
>> qmp_query_command_line_options is looking for supported options.
>>
>> As a fix for for 2.3 the machine_qemu_opts (the generic ones)
>> are restored only for qemu-config scope.
>> We need to find a better fix for 2.4.
>>
>> Reported-by: Tony Krowiak <akrowiak@linux.vnet.ibm.com>
>> Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
>> ---
>
>> +        },{
>> +            .name = "firmware",
>> +            .type = QEMU_OPT_STRING,
>> +            .help = "firmware image",
>> +        },{
>> +            .name = "iommu",
>> +            .type = QEMU_OPT_BOOL,
>> +            .help = "Set on/off to enable/disable Intel IOMMU (VT-d)",
>> +        },{
>> +            .name = "suppress-vmdesc",
>> +            .type = QEMU_OPT_BOOL,
>> +            .help = "Set on to disable self-describing migration",
>> +        },
>
> No longer a strict superset of the Fedora 21 qemu-kvm, which had:
>
>              "parameters": [
>                  {
>                      "name": "max-ram-below-4g",
>                      "help": "maximum ram below the 4G boundary (32bit
> boundary)"
> ,
>                      "type": "size"
>                  },
>                  {
>                      "name": "kvm-type",
>                      "help": "Specifies the KVM virtualization mode (HV,
> PR)",
>                      "type": "string"
>                  },
>                  {
>                      "name": "firmware",
>                      "help": "firmware image",
>                      "type": "string"
>                  },
>
> (remembering that query-command-line-options does things in reverse
> order).  I don't think iommu and suppress-vmdesc hurt to add, but we
> shouldn't lose kvm-type or max-ram-below-4g, if those were advertised at
> the point prior to the QemuOpts conversion.  (I didn't actually
> research, though, whether I'm comparing against downstream Fedora
> qemu-kvm patches instead of upstream...)
Hmm, in 2.3 kvm-type and max-ram-below-4g are machine specific.
I only took the ones from hw/core/machine.c that are common to all machines.
As you said, this approach does the best to not lie to libvirt...

Thanks,
Marcel
>
>