ROX-9409: sets DRIVER_NAME in collector CMakeLists

[Stringy]

Description

A detailed explanation of the changes in your PR.

Feel free to remove this section if it is overkill for your PR, and the title of your PR is sufficiently descriptive.

Checklist

• Investigated and inspected CI test results
• Updated documentation accordingly

Automated testing

• Added unit tests
• Added integration tests
• Added regression tests

If any of these don't apply, please comment below.

Testing Performed

TODO(replace-me)
Use this space to explain how you tested your PR, or, if you didn't test it, why you did not do so. (Valid reasons include "CI is sufficient" or "No testable changes")
In addition to reviewing your code, reviewers must also review your testing instructions, and make sure they are sufficient.

For more details, ref the Confluence page about this section.

[ghost]

COLLECTOR_TAG=3.8.x-79-g71d9891f97
COLLECTOR_BUILDER_TAG=cache

Results for Performance Benchmarks on build #251959

Kernel	Method	Without Collector Time (secs)	With Collector Time (secs)	Baseline median (secs)	Collector median (secs)	PValue
cos.cos-85-lts	ebpf	139.667	400.862	158.55	202.5	🔴
cos.cos-89-lts	ebpf	144.515	434.698	182.27	233.56	🔴
cos.cos-beta	ebpf	230.023	452.573	181.55	203.1	🔴
cos.cos-dev	ebpf	162.929	336.802	187.73	203.53	🔴
cos.cos-stable	ebpf	221.057	447.054	197.85	221.51	🔴
flatcar.flatcar-stable	ebpf	201.675	436.472	208.76	214.79	🔴
flatcar.flatcar-stable	module	204.849	224.534	204.44	206.5	🟢
garden-linux.garden-linux	ebpf	244.398	468.9	230.59	251.63	🔴
garden-linux.garden-linux	module	252.455	473.839	226.88	249.69	🔴
rhel.rhel-7	ebpf	98.632	296.855	173.52	212.23	🔴
rhel.rhel-7	module	246.71	279.454	183.38	208.09	🟢
rhel.rhel-8	ebpf	215.555	464.046	161.54	201.51	🔴
rhel.rhel-8	module	219.341	239.939	160.78	180.39	🟢
suse.sles-12	module	170.56	203.378	121.75	140.51	🔴
suse.sles-15	ebpf	97.531	318.267	122.24	162.71	🔴
suse.sles-15	module	173.93	195.676	120.48	138.34	🟢
ubuntu-os-pro.ubuntu-pro-1804-lts	ebpf	188.005	457.849	165.7	206.6	🔴
ubuntu-os-pro.ubuntu-pro-1804-lts	module	127.537	156.169	161.8	186.74	🟢
ubuntu-os.ubuntu-1804-lts	ebpf	118.697	372.288	149.2	201.05	🔴
ubuntu-os.ubuntu-1804-lts	module	197.111	227.75	162.92	186.49	🟢
ubuntu-os.ubuntu-2004-lts	ebpf	209.898	447.169	150.75	175.6	🔴
ubuntu-os.ubuntu-2004-lts	module	207.053	228.222	182.46	204.48	🟢
ubuntu-os.ubuntu-2204-lts	ebpf	217.478	452.727	204.84	230.01	🔴
ubuntu-os.ubuntu-2204-lts	module	217.261	232.969	201.46	211.76	🟢

[Molter73]

Can you change the commit in the falcosecurity-libs sub-module so it points to the head of the branch on stackrox/falcosecurity-libs#3? We'll roll it back before merging, but it would be nice to see how it interacts with CI.

We should also run our CI with no-cache and dockerized-no-cache to ensure compilation of drivers is not broken by this change.

[Molter73]

Looking good! Just have a couple of questions on some of the versioning stuff.

[Molter73]

Just one more comment but it shouldn't be a blocker. LGTM!

[Molter73]

I was just going through the CI run on a PR that builds onto these changes and noticed that the legacy drivers build procedure would be broken, since the prepare-src script now sets DRIVER_NAME but legacy drivers still expect the PROBE_NAME variable when running cmake. I think the easiest fix would be to simply set both variables in the cmake command inside prepare-src, the one that doesn't belong in the build will simply be ignored. Another possibility would be to use cmake -LA as detailed in this SO answer, grep'ing the output and adjust the arguments as needed, but I think this might be a little bit overkill.

We should also add the build-legacy-probes label to the PR and re-run CI to double check.

[Molter73]

LGTM!

[robbycochran]

@Stringy It looks like this change degraded the performance of the ebpf probe significantly based on the reported collector times and this regression remains on the custom probe for rhel7. I suspect that the culprit may be this commit which removed the usage of g_bpf_drop_syscalls.