[xorg-memory] DRI GEM objects leak detected, before 1071 and after 1371

[Brendan Donegan]

On some machines, the xorg-memory test is failing with the following sort of error:

 DRI GEM objects leak detected, before 1071 and after 1371

With different values usually. It's difficult to debug this one remotely since the test doesn't seem to appreciate not being in an X session.

[Brendan Donegan (brendan-donegan)]

This is still happening:

https://certification.canonical.com/hardware/200908-3578/submission/ofrW1sgRqB94x3M/results?form.status=FAIL&form.status=UNRESOLVED

[Brendan Donegan (brendan-donegan)]

This seems to be still happening on a number of systems, even in the final release:

[Desktops]
200910-4493
201004-5592
201005-5775
201004-5597
201011-6865
201005-5771
201010-6632
200910-4293
200908-3578
201011-6700
201008-6444

[Laptops]
201003-5427
201004-5575
201010-6645
201003-5428
201004-5572
201004-5578

(This is not an exhaustive list, there are more)

[Brendan Donegan (brendan-donegan)]

I still see instances of this failure in the final testing - setting importance to medium.

[Ara Pulido (ara)]

Moving to checkbox

[Jeff Lane  (bladernr)]

I discussed this with Bryce and Alberto and the consensus is that this is something we can get rid of. They said it would only be useful if we were interested in

A: running X through valgrind to debug it (can't do through checkbox, and we probably won't do anyway)
B: escalating it every time as a bug, again, something we probably don't care to do as it doesn't seem to cause any issues.

I propose just removing the test.

[Jeff Lane  (bladernr)]

adding the -cert task to remember to change the whitelists.

[Jeff Lane  (bladernr)]

fixed in checkbox and checkbox-certification