On 1/23/19 7:06 AM, Max Reitz wrote:
> On 21.01.19 21:55, Eric Blake wrote:
>> On 12/21/18 5:47 PM, Max Reitz wrote:
>>> In some cases, we may want to deal with qemu-nbd errors (e.g. by
>>> launching it in a different configuration until it no longer throws
>>> any).  In that case, we do not want its output ending up in the test
>>> output.
>>>
>>> It may still be useful for handling the error, though, so add a new
>>> function that works basically like qemu_nbd(), only that it returns the
>>> qemu-nbd output instead of making it end up in the log.  In contrast to
>>> qemu_img_pipe(), it does still return the exit code as well, though,
>>
>> In contrast to qemu_nbd(),
> 
> But qemu_nbd() does return the exit code.  qemu_img_pipe() doesn't.
> 

Oh, I see where I got confused. I thought you were comparing the new
code [qemu_nbd_pipe] to itself; not to the pre-existing qemu_img_pipe
that is not touched by this patch, but which served as the model you
copied after.

>>> because that is even more important for error handling.
>>>
>>> Signed-off-by: Max Reitz <mreitz@redhat.com>
>>> ---
>>>  tests/qemu-iotests/iotests.py | 14 ++++++++++++++
>>>  1 file changed, 14 insertions(+)
>>>

>> Reviewed-by: Eric Blake <eblake@redhat.com>
> 
> Thanks for reviewing!

Although my questions on patch 3 call into question whether you even
need this patch, if your only use of it was to find a free port.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org