AthCUDA Warning Fixes, master branch (2020.07.07.)

While building AthExCUDA for an unrelated development recently, I had to realise that when CUDA is not available for the build, the code produces a lot of warnings. These changes are meant to fix all of those.

It's worth noting that this just highlights that we will need to turn on more rigorous warnings for the CUDA compilation in AtlasCompilerSettings.cmake...

Pinging @leggett and @fwinkl.

This merge request affects 3 packages:

• Control/AthCUDA/AthCUDACore
• Control/AthCUDA/AthCUDAKernel
• Control/AthCUDA/AthCUDAServices

Adding @ssnyder as watcher

added Core master review-pending-level-1 labels

CI Result FAILURE (hash 1c3b8142)

	Athena	AthSimulation	AnalysisBase	AthGeneration
externals
cmake
make
required tests
optional tests

Full details available on this CI monitor view
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
AthGeneration: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 16637]

Has the following been happening in other MRs as well yesterday?

...
[CI_TEST_WRAPPER] minutes elapsed: 10 
[CI_TEST_WRAPPER] minutes elapsed: 11 
[CI_TEST_WRAPPER] minutes elapsed: 12 
2020-07-07T122832 CEST TrigValSteering ERROR    RootComp killed on timeout of 300 seconds
art-result: 9 RootComp
...

http://atlas-computing.web.cern.ch/atlas-computing/links/distDirectory/ci/CIWebArea/nicos_web_areaMRCIbuilds64BC7G8AthenaOpt/NICOS_TestLog_MR-34502-2020-07-07-14-16/ReleaseTests___Trigger_athena_MC-test__Trigger_athena_MC-test__m.html

Naturally, I don't believe that this MR has anything to do with that failure...

Hmm... we saw this a week ago but it was supposed to be fixed by ATR-21652. Have to look at it again, but yes, it's definitely not related to your MR.

Is it worth trying the CI again?

Since it's definitely not related, I don't see the point of retrying.

So then, is anyone going to accept or merge this? Unfortunately the shifters seem to be ignoring it because of the failed test.

Let me mark it as review-approved then. Complain if you disagree.

added review-approved label and removed review-pending-level-1 label

Clear, approving.

Xiaozhong (L1)

merged

mentioned in commit 2d91e9db

added sweep:ignore label