[libvirt] [PATCH v2 02/22] qemu: Introduce qemuBlockJobUpdate

[John Ferlan]

On 06/10/2015 07:31 AM, Jiri Denemark wrote:
> On Wed, Jun 10, 2015 at 07:21:27 -0400, John Ferlan wrote:
>>
>>
>> On 06/02/2015 08:34 AM, Jiri Denemark wrote:
>>> The wrapper is useful for calling qemuBlockJobEventProcess with the
>>> event details stored in disk's privateData, which is the most likely
>>> usage of qemuBlockJobEventProcess.
>>>
>>> Signed-off-by: Jiri Denemark <jdenemar at redhat.com>
>>> ---
> ...
>>> diff --git a/src/qemu/qemu_blockjob.c b/src/qemu/qemu_blockjob.c
>>> index 098a43a..605c2a5 100644
>>> --- a/src/qemu/qemu_blockjob.c
>>> +++ b/src/qemu/qemu_blockjob.c
> ...
>>> @@ -218,9 +243,7 @@ qemuBlockJobSyncEnd(virQEMUDriverPtr driver,
>>>      if (diskPriv->blockJobSync && diskPriv->blockJobStatus != -1) {
>>>          if (ret_status)
>>>              *ret_status = diskPriv->blockJobStatus;
>>> -        qemuBlockJobEventProcess(driver, vm, disk,
>>> -                                 diskPriv->blockJobType,
>>> -                                 diskPriv->blockJobStatus);
>>> +        qemuBlockJobUpdate(driver, vm, disk);
>>
>> ^^
>> This doesn't get the returned status...
> 
> qemuBlockJobUpdate returns the original value of
> diskPriv->blockJobStatus, which is already stored to *ret_status above.
> Not to mention that one of the following patches will completely remove
> ret_status from qemuBlockJobSyncEnd.
> 

I was just perusing to 'learn' a bit about the code... I hadn't looked
forward yet.

In essence it's a void function though... I see that qemuBlockJobUpdate
checks/sets blockJobStatus = -1 and the same setting is done right after
the call.

John