[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: uninitialized value on error path (rev3)

[Martin Peres]

On 18/04/17 12:14, Dan Carpenter wrote:
> On Tue, Apr 18, 2017 at 10:58:44AM +0300, Jani Nikula wrote:
>> On Mon, 17 Apr 2017, Dan Carpenter <dan.carpenter at oracle.com> wrote:
>>> On Fri, Apr 14, 2017 at 08:13:43PM -0000, Patchwork wrote:
>>>> == Series Details ==
>>>>
>>>> Series: drm/i915: uninitialized value on error path (rev3)
>>>> URL   : https://patchwork.freedesktop.org/series/23038/
>>>> State : success
>>>
>>> These patchwork emails are sort of useless.  I wouldn't have sent the
>>> patch if it couldn't compile which is basically all this is testing.
>>> It's not exercising the failure path.
>>
>> Yeah, the CI is not smart enough to know that just by looking at the
>> patch, so it'll test them all. ;)
>>
> 
> But could it just send responses if something fails?  Or it could be a
> much shorter email:
> 
>      CI test successful.
> 
>      Use X-mailer-blahblah to filter successful tests.

I guess that could be an interesting thing to add. But what prevents you
from dropping these emails if you don't want them? Their subject always
starts with "Fi.CI.BAT: success".

> 
> We don't need to expose all the other internal information unless
> something fails.

The purpose of this email is to say the patch series actually got
tested, making it easier for the person pushing the series that it
passed the basic requirements. The alternative would be to ask
people to check on patchwork every time, which is cumbersome.

One reason also to "expose the internals" is that machines come
and go as we often, we only have one machine of old platforms. So, a
patch checking a fix for gen4 with no actual gen4 HW testing it would be
pointless.

</ my two cents>