On 02/23/2017 04:57 PM, Marc Hartmayer wrote:
On Thu, Feb 23, 2017 at 03:33 PM +0100, Michal Privoznik
<mprivozn(a)redhat.com> wrote:
> On 02/23/2017 10:44 AM, Marc Hartmayer wrote:
>> Fix incorrect jump labels in error paths as the stop jump is only
>> needed if the driver has already changed the state. For example
>> 'virAtomicIntInc(&driver->nactive)' will be 'reverted' in
the
>> qemuProcessStop call.
>>
>> Signed-off-by: Marc Hartmayer <mhartmay(a)linux.vnet.ibm.com>
>> Reviewed-by: Boris Fiuczynski <fiuczy(a)linux.vnet.ibm.com>
>> ---
>> src/qemu/qemu_process.c | 13 ++++++++-----
>> 1 file changed, 8 insertions(+), 5 deletions(-)
>>
>> diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c
>> index ea10fff..a57d136 100644
>> --- a/src/qemu/qemu_process.c
>> +++ b/src/qemu/qemu_process.c
>> @@ -4755,9 +4755,12 @@ qemuProcessInit(virQEMUDriverPtr driver,
>> */
>> VIR_DEBUG("Setting current domain def as transient");
>> if (virDomainObjSetDefTransient(caps, driver->xmlopt, vm) < 0)
>> - goto stop;
>> + goto cleanup;
>>
>> - if (!(flags & VIR_QEMU_PROCESS_START_PRETEND)) {
>> + if (flags & VIR_QEMU_PROCESS_START_PRETEND) {
>> + if (qemuDomainSetPrivatePaths(driver, vm) < 0)
>> + goto cleanup;
>
> This should have been goto stop; After SetDefTransien() succeeds,
> everything must goto stop in order to call qemuProcessStop which undoes
> the SetDefTransient(). I will fix it before pushing.
Hmm why? In qemuProcessStop() the driver->nactive will be decreased and
I think that is definitely not what we want if we haven't increased
that value... (for the VIR_QEMU_PROCESS_START_PRETEND path)
Ah, that's a good point. On the other hand, we need to call
virDomainObjRemoveTransientDef(). So I guess the proper solution is to:
if (qemuDomainSetPrivatePaths(driver, vm) < 0) {
virDomainObjRemoveTransientDef(vm);
goto cleanup;
}
I'll post the patch shortly.
Michal